From lkcl at lkcl.net Thu Sep 3 00:59:28 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 3 Sep 2015 00:59:28 +0100 Subject: [Arm-netbook] libre laptop 1st and 2nd pcbs off to eurocircuits Message-ID: ok so i chose the "prototype" option - no solder mask, no silkscreen, just bare copper. the rules for the eurocircuits proto boards are... both strict and also odd. there's a minimum via size which is quite huge (24mil surround, 10mil drill), fortunately however i've been able to design all 3 pcbs around that. i believe it means that there is a big advantage when it comes to production: lower cost! :) also the 1kW IR soldering station is arriving tomorrow, it will be.... interesting to work out how to use it. i will try some experiments first... l. From lkcl at lkcl.net Thu Sep 3 02:12:20 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 3 Sep 2015 02:12:20 +0100 Subject: [Arm-netbook] libre laptop power board, needs 2oz copper, any recommendations? Message-ID: ok, i checked with the eurocircuits online system, they can't handle the power board. the LTC4155 is handling 4 amps, so the app note says that 2oz copper is needed. however, eurocircuits require a minimum track-to-track clearance of 10mil which is totally incompatible with the distance even between pads of the LTC4155 IC! therefore we conclude, simply, that eurocircuits doesn't have good enough equipment, therefore it is necessary to find an alternative PCB prototyping company. i've managed to get the track-to-track distance to 7mil, smallest via is 10mil with a 24mil ring, track width is a minimum of 8mil, and the size of the PCB is tiny: 3.4in x 1.2in. it's only two layer, components only on one side, but the copper needs to be 2oz. can anyone help with PCB manufacturing at all? tia, l. From lkcl at lkcl.net Thu Sep 3 14:08:12 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 3 Sep 2015 14:08:12 +0100 Subject: [Arm-netbook] EOMA68 interface update proposal: -GbE, +USB3.1, +GPIO Message-ID: ok, soo... i've been following the updates to USB3 (10 gigabit speeds now with USB 3.1). in combination with: * the development of both the tablet and the laptop not having room for an ethernet port * the development of lower-cost SoCs not really having ethernet * the fact that the ethernet components cost around $2 when the SoCs themselves are around $2 to $4 ... i'm coming to the conclusion that Ethernet should be removed, and the 9 lines used to * increase the USB3.0 allocation (4 wires) to USB 3.1 (8 wires) * this would leave 5 spare pins for GPIO. * at least two of those should be EINT-capable GPIOs this would make it possible to redesign the tablet PCB to a more "standard" one - removing the need for the I2C GPIO chip for example. if there are any products that need ethernet, it can be done as a USB2 or USB3 compliant external IC. the IC3128 PCB can be reduced down to around a $12 BOM, the EOMA68-A20 PCB can be reduced by a further $2.50. any thoughts or objections, much appreciated the input. l. From lasich at gmail.com Thu Sep 3 14:28:50 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Thu, 3 Sep 2015 15:28:50 +0200 Subject: [Arm-netbook] EOMA68 interface update proposal: -GbE, +USB3.1, +GPIO In-Reply-To: References: Message-ID: Depend really on your target market. For any application that your product need to be any kind of server ethernet is kind of better, more relaible compared to wifi. If your target audience is not that kind of application, or at least not in majority I guess you got it right. Just question any kind of paradigm. Another angle is that at one moment you have to stop developing and ship product, even if not perfect. Again, you decide. On Thursday, September 3, 2015, Luke Kenneth Casson Leighton wrote: > ok, soo... > > i've been following the updates to USB3 (10 gigabit speeds now with > USB 3.1). in combination with: > > * the development of both the tablet and the laptop not having room > for an ethernet port > * the development of lower-cost SoCs not really having ethernet > * the fact that the ethernet components cost around $2 when the SoCs > themselves are around $2 to $4 > > ... i'm coming to the conclusion that Ethernet should be removed, and > the 9 lines used to > > * increase the USB3.0 allocation (4 wires) to USB 3.1 (8 wires) > * this would leave 5 spare pins for GPIO. > * at least two of those should be EINT-capable GPIOs > > this would make it possible to redesign the tablet PCB to a more > "standard" one - removing the need for the I2C GPIO chip for example. > > if there are any products that need ethernet, it can be done as a USB2 > or USB3 compliant external IC. > > the IC3128 PCB can be reduced down to around a $12 BOM, the EOMA68-A20 > PCB can be reduced by a further $2.50. > > any thoughts or objections, much appreciated the input. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk -- Hrvoje Lasić Vulpes d.o.o. Gračanska 120a 10000 Zagreb Croatia tel +385 1 6152 706 tel +38598 450 603 *lasich at gmail.com hrvoje at vebbu.co * *www.vebbu.co * -------------- next part -------------- An HTML attachment was scrubbed... URL: From lkcl at lkcl.net Thu Sep 3 15:09:14 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 3 Sep 2015 15:09:14 +0100 Subject: [Arm-netbook] EOMA68 interface update proposal: -GbE, +USB3.1, +GPIO In-Reply-To: References: Message-ID: On Thu, Sep 3, 2015 at 2:28 PM, Hrvoje Lasic wrote: > Depend really on your target market. For any application that your product > need to be any kind of server ethernet is kind of better, more relaible > compared to wifi. well, USB3 ethernet is going to be reliable (and fast). even USB 3.0 is 5gbit/sec and GbE is 1gb/sec. and i had to take off SATA because it was adding extra cost. > If your target audience is not that kind of application, or at least not in > majority I guess you got it right. Just question any kind of paradigm. it's not the majority, for sure. > Another angle is that at one moment you have to stop developing and ship > product, even if not perfect. Again, you decide. yehh very true :) > > On Thursday, September 3, 2015, Luke Kenneth Casson Leighton > wrote: >> >> ok, soo... >> >> i've been following the updates to USB3 (10 gigabit speeds now with >> USB 3.1). in combination with: >> >> * the development of both the tablet and the laptop not having room >> for an ethernet port >> * the development of lower-cost SoCs not really having ethernet >> * the fact that the ethernet components cost around $2 when the SoCs >> themselves are around $2 to $4 >> >> ... i'm coming to the conclusion that Ethernet should be removed, and >> the 9 lines used to >> >> * increase the USB3.0 allocation (4 wires) to USB 3.1 (8 wires) >> * this would leave 5 spare pins for GPIO. >> * at least two of those should be EINT-capable GPIOs >> >> this would make it possible to redesign the tablet PCB to a more >> "standard" one - removing the need for the I2C GPIO chip for example. >> >> if there are any products that need ethernet, it can be done as a USB2 >> or USB3 compliant external IC. >> >> the IC3128 PCB can be reduced down to around a $12 BOM, the EOMA68-A20 >> PCB can be reduced by a further $2.50. >> >> any thoughts or objections, much appreciated the input. >> >> l. >> >> _______________________________________________ >> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk >> http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook >> Send large attachments to arm-netbook at files.phcomp.co.uk > > > > -- > Hrvoje Lasić > Vulpes d.o.o. > Gračanska 120a > 10000 Zagreb > Croatia > tel +385 1 6152 706 > tel +38598 450 603 > lasich at gmail.com > hrvoje at vebbu.co > www.vebbu.co > > > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From cand at gmx.com Thu Sep 3 16:20:46 2015 From: cand at gmx.com (Lauri Kasanen) Date: Thu, 3 Sep 2015 18:20:46 +0300 Subject: [Arm-netbook] EOMA68 interface update proposal: -GbE, +USB3.1, +GPIO In-Reply-To: References: Message-ID: <20150903182046.e249b888.cand@gmx.com> On Thu, 3 Sep 2015 15:09:14 +0100 Luke Kenneth Casson Leighton wrote: > On Thu, Sep 3, 2015 at 2:28 PM, Hrvoje Lasic wrote: > > Depend really on your target market. For any application that your product > > need to be any kind of server ethernet is kind of better, more relaible > > compared to wifi. > > well, USB3 ethernet is going to be reliable (and fast). even USB 3.0 > is 5gbit/sec and GbE is 1gb/sec. and i had to take off SATA because > it was adding extra cost. I'll believe when I see it. So far seen zero reliable USB ethernet adapters. The lack of native ethernet and sata puts me out, but perhaps I represent the small minority - get a product out, and then work on revisions for the curmudgeons ;) - Lauri From lkcl at lkcl.net Thu Sep 3 17:22:40 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 3 Sep 2015 17:22:40 +0100 Subject: [Arm-netbook] EOMA68 interface update proposal: -GbE, +USB3.1, +GPIO In-Reply-To: <20150903182046.e249b888.cand@gmx.com> References: <20150903182046.e249b888.cand@gmx.com> Message-ID: On Thu, Sep 3, 2015 at 4:20 PM, Lauri Kasanen wrote: > On Thu, 3 Sep 2015 15:09:14 +0100 > Luke Kenneth Casson Leighton wrote: > >> On Thu, Sep 3, 2015 at 2:28 PM, Hrvoje Lasic wrote: >> > Depend really on your target market. For any application that your product >> > need to be any kind of server ethernet is kind of better, more relaible >> > compared to wifi. >> >> well, USB3 ethernet is going to be reliable (and fast). even USB 3.0 >> is 5gbit/sec and GbE is 1gb/sec. and i had to take off SATA because >> it was adding extra cost. > > I'll believe when I see it. So far seen zero reliable USB ethernet > adapters. USB2 or USB3? USB2 i am keenly aware has been price-pushed so low it's really quite ridiculous. the end-result is that, yes, USB2 SATA and USB2 Eth dongles are... fun. there was word a few years ago that the Genesys Logic USB2-SATA IC needed a "hard reset" on a regular basis, it would crash so badly. but, honestly, USB2 is only 480mbit/sec - trying to saturate that with either GbE or even 30% of SATA-II bandwidth is... not going to end well, given that USB2 is a "cooperative" bus. USB3 i have much more hope for. > The lack of native ethernet and sata puts me out, but perhaps I > represent the small minority - get a product out, and then work on > revisions for the curmudgeons ;) well EOMA68 is designed to be a decade-long standard. so it's all-or-nothing, here. i'd need to work on a totally different standard, and a totally different package. well, that's the plan, anyway :) l. From lkcl at lkcl.net Sat Sep 5 20:45:36 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sat, 5 Sep 2015 20:45:36 +0100 Subject: [Arm-netbook] libre 15in laptop progress: pcbs and parts Message-ID: okaaaay sooo, whew :) spent the entire day online looking up digikey part numbers... eyes are going funny (square _and_ crossed....) but i got there. 65 items from digikey, then another 5 or so that need to be sourced from china (only) - a special narrow 3.5mm jack which can re-route both outputs to speakers if the headphone jack's not plugged in, a CM108AH USB Audio IC, GL850G hub, SY6280 USB power control IC and a SY7201 step-up converter for the 3.5 watt LCD backlight. they're all easy to get... just not outside of china. i also ordered qty2 of the PCB1 (main laptop interface board) and PCB2 (embedded controller, keyboard/mouse) as "bare prototype" boards, 0.5oz copper, from eurocircuits. the 4 boards came to a total of $EUR 100, which is pretty amazing for europe, and they were very tolerant about me making a mistake and having to cancel the order, make some changes, and re-submit even after i'd actually paid (they don't normally allow that!). i did have to redesign all the PCBs to use a 24mil surround around the 10mil drill holes - and change all the track-to-track min clearance to 7mil and the min track width to 8mil. this was the *absolute* minimum that eurocircuits "prototype pool" would allow. *sigh* i'm also going to have to go back and adjust the casework to cope with a 1.6mm PCB thickness.... i'd designed it for 1.2mm and yes, really, 0.4mm makes a huge difference, it will be about a day's work to adjust all the 3D parts. the power board (PCB3).... yyyeah i am having fun, there. the example board layout i got from linear tech's online resources says that 2oz copper is needed. *nobody* does 2oz copper with anything below 12mil track and 12mil track-to-track.... yet the distance between the pads on an LTC4155 are 7.8mil (0.2mm)... what's up with that?? soo... hmmm, i'm looking up current handling calculators online to see if i can get away with 1oz copper.... and i believe i can. which would be good, if it was the case, as i can go to eurocircuits.... BUT, i would have to make a "standard" order not prototype because the proto boards are only available in 0.5oz copper... oh! and the 1kW IR solder station arrived. it's.... much more basic than i was expecting :) there are two thermistors that you're supposed to like stick somewhere around the PCB when the 800W pre-heater and the 200W IR lamp are on.... .... i got them in the wrong place on the first board i tried (a 1in prototype board) and was fascinated to see it sizzle within about 15 seconds, this little black spot about 8mm in diameter appeared and i spent the rest of the day with the smell of cooked fibreglass up my nose... mmmm looovely :) anyway it appears that it actually works, must get some solder paste, start doing some desoldering of components on random boards as an experiment, it'll be fuuuun.... l. From lkcl at lkcl.net Tue Sep 8 20:07:44 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 8 Sep 2015 20:07:44 +0100 Subject: [Arm-netbook] ecocomputing white paper Message-ID: http://rhombus-tech.net/whitepapers/ecocomputing_07sep2015/ anyone on the IRC channel may have noticed the flood of git commits as i made several successive minor changes and used "git commit" to upload them for proper reading in-place. basically this was written based on the conversations, questions and challenges that people here on arm-netbooks have raised over the years since the list was started, so i particularly wanted to say thank you to all of you for indirectly helping in that way and making it possible to write this white paper in one day. i've been planning it for some time, but there wasn't quite a right time. l. From joem at martindale-electric.co.uk Wed Sep 9 15:48:28 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 9 Sep 2015 14:48:28 +0000 Subject: [Arm-netbook] How much to design A20 board? Message-ID: <1441810196.718.2.camel@jm-desktop> Hi, Just wondering how much is it these days ball park to pay someone like witstech or a similar design outfit to design an A20 board (like a cubie) and make 10 samples? From amery at geeks.cl Wed Sep 9 15:53:50 2015 From: amery at geeks.cl (Alejandro Mery) Date: Wed, 09 Sep 2015 14:53:50 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1441810196.718.2.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> Message-ID: Hi Joem, look at olimex boards at https://www.olimex.com/Products/OLinuXino/A20/open-source-hardware they are OSHW and so the complete design is open source. On Wed, 9 Sep 2015 at 15:51 joem wrote: > Hi, > > Just wondering how much is it these days ball park to pay someone like > witstech or a similar design outfit to design an A20 board (like a > cubie) and make 10 samples? > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk -------------- next part -------------- An HTML attachment was scrubbed... URL: From lkcl at lkcl.net Wed Sep 9 16:02:35 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 16:02:35 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1441810196.718.2.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> Message-ID: On Wed, Sep 9, 2015 at 3:48 PM, joem wrote: > Hi, > > Just wondering how much is it these days ball park to pay someone like > witstech or a similar design outfit to design an A20 board about $USD 3-4k > (like a > cubie) and make 10 samples? about $USD 2.5k l. From lkcl at lkcl.net Wed Sep 9 16:04:26 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 16:04:26 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> Message-ID: On Wed, Sep 9, 2015 at 3:53 PM, Alejandro Mery wrote: > Hi Joem, look at olimex boards at > https://www.olimex.com/Products/OLinuXino/A20/open-source-hardware they > are OSHW and so the complete design is open source. the designs are under the "attribution" style licenses, forcing you to advertise as part of the product. as such they are *not* libre licensed. if the designs were released under a GPL license it would be a different matter. l. From joem at martindale-electric.co.uk Wed Sep 9 16:03:09 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 9 Sep 2015 15:03:09 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> Message-ID: <1441811147.718.15.camel@jm-desktop> On Wed, 2015-09-09 at 14:53 +0000, Alejandro Mery wrote: > Hi Joem, look at olimex boards > at https://www.olimex.com/Products/OLinuXino/A20/open-source-hardware they are OSHW and so the complete design is open source. -- > Hi, > > Just wondering how much is it these days ball park to pay > someone like > witstech or a similar design outfit to design an A20 board > (like a > cubie) and make 10 samples? Hi Alejandro, Thank you for that link. Do you think they can design a custom board to specifications and just hand over the design files for a fee? And make 10 samples to prove it works. From joem at martindale-electric.co.uk Wed Sep 9 16:08:00 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 9 Sep 2015 15:08:00 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> Message-ID: <1441811441.718.18.camel@jm-desktop> On Wed, 2015-09-09 at 16:02 +0100, Luke Kenneth Casson Leighton wrote: > On Wed, Sep 9, 2015 at 3:48 PM, joem wrote: > > Hi, > > > > Just wondering how much is it these days ball park to pay someone like > > witstech or a similar design outfit to design an A20 board > > about $USD 3-4k Thats a good number! Do you think they will hand over the design files? I don't want to spend another 2.5k on top for the 10 samples. (Cheaper companies out there that source and populate.) > > (like a > > cubie) and make 10 samples? > > about $USD 2.5k From lkcl at lkcl.net Wed Sep 9 16:21:12 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 16:21:12 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1441811441.718.18.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> <1441811441.718.18.camel@jm-desktop> Message-ID: On Wed, Sep 9, 2015 at 4:08 PM, joem wrote: > On Wed, 2015-09-09 at 16:02 +0100, Luke Kenneth Casson Leighton wrote: >> On Wed, Sep 9, 2015 at 3:48 PM, joem wrote: >> > Hi, >> > >> > Just wondering how much is it these days ball park to pay someone like >> > witstech or a similar design outfit to design an A20 board >> >> about $USD 3-4k > > Thats a good number! Do you think they will hand over the design files? they always do. > I don't want to spend another 2.5k on top for the 10 samples. it was approximate. ... y'know... if you're looking for someone to design an A20 board, i could always do it for you, i have a working proven and very small footprint board. l. From joem at martindale-electric.co.uk Wed Sep 9 16:26:10 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 9 Sep 2015 15:26:10 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <1441811441.718.18.camel@jm-desktop> Message-ID: <1441812507.1106.4.camel@jm-desktop> > >> > Hi, > >> > > >> > Just wondering how much is it these days ball park to pay someone like > >> > witstech or a similar design outfit to design an A20 board > >> > >> about $USD 3-4k > > > > Thats a good number! Do you think they will hand over the design files? > > they always do. > > > I don't want to spend another 2.5k on top for the 10 samples. > > it was approximate. > > ... y'know... if you're looking for someone to design an A20 board, i > could always do it for you, i have a working proven and very small > footprint board. If you are offering, then name a ball park price (send to research at enemygadgets.com) I'm happy to say you deserve it more than anyone else :) From lkcl at lkcl.net Wed Sep 9 16:47:34 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 16:47:34 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1441812507.1106.4.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> <1441811441.718.18.camel@jm-desktop> <1441812507.1106.4.camel@jm-desktop> Message-ID: On Wed, Sep 9, 2015 at 4:26 PM, joem wrote: >> ... y'know... if you're looking for someone to design an A20 board, i >> could always do it for you, i have a working proven and very small >> footprint board. > > If you are offering, i am :) > then name a ball park price (send to > research at enemygadgets.com) > I'm happy to say you deserve it more than anyone else :) :) done sah. From elena.valhalla at gmail.com Wed Sep 9 18:20:14 2015 From: elena.valhalla at gmail.com (Elena ``of Valhalla'') Date: Wed, 9 Sep 2015 19:20:14 +0200 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> Message-ID: <20150909172013.GB9597@virginsteele.home.trueelena.org> On 2015-09-09 at 16:04:26 +0100, Luke Kenneth Casson Leighton wrote: > On Wed, Sep 9, 2015 at 3:53 PM, Alejandro Mery wrote: > > Hi Joem, look at olimex boards at > > https://www.olimex.com/Products/OLinuXino/A20/open-source-hardware they > > are OSHW and so the complete design is open source. > the designs are under the "attribution" style licenses, forcing you > to advertise as part of the product. as such they are *not* libre > licensed. if the designs were released under a GPL license it would > be a different matter. really? "provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code; keep intact all notices of the absence of any warranty; and give all recipients a copy of this License along with the Program." in GPLv3 "provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program." in GPLv2 Not very different from the Attribution clause in https://creativecommons.org/licenses/by-sa/4.0/legalcode The only widespread FLOSS licenses that I know of that doesn't include similar wording are a Public Domain dedication (such as CC0) and the WTFPL. -- Elena ``of Valhalla'' From lkcl at lkcl.net Wed Sep 9 19:42:25 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 19:42:25 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <20150909172013.GB9597@virginsteele.home.trueelena.org> References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> Message-ID: On Wed, Sep 9, 2015 at 6:20 PM, Elena ``of Valhalla'' wrote: > On 2015-09-09 at 16:04:26 +0100, Luke Kenneth Casson Leighton wrote: >> On Wed, Sep 9, 2015 at 3:53 PM, Alejandro Mery wrote: >> > Hi Joem, look at olimex boards at >> > https://www.olimex.com/Products/OLinuXino/A20/open-source-hardware they >> > are OSHW and so the complete design is open source. >> the designs are under the "attribution" style licenses, forcing you >> to advertise as part of the product. as such they are *not* libre >> licensed. if the designs were released under a GPL license it would >> be a different matter. > > really? yes. > "provided that you conspicuously and appropriately publish on each copy > an appropriate copyright notice; keep intact all notices stating that > this License and any non-permissive terms added in accord with section 7 > apply to the code; keep intact all notices of the absence of any > warranty; and give all recipients a copy of this License along with the > Program." in GPLv3 > > "provided that you conspicuously and appropriately publish on each copy > an appropriate copyright notice and disclaimer of warranty; keep intact > all the notices that refer to this License and to the absence of any > warranty; and give any other recipients of the Program a copy of this > License along with the Program." in GPLv2 so that's in the source code. > Not very different from the Attribution clause in > https://creativecommons.org/licenses/by-sa/4.0/legalcode attribution clauses require you to advertise *on the product*. that's utterly different from requiring to maintain the copyright notice and the fact that there is a license *in the source code*. if you're not familiar with or don't clearly understand the difference, look up the history behind why the Debian Team renamed firefox to "iceweasel". l. From lkcl at lkcl.net Wed Sep 9 19:44:00 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 19:44:00 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1441812507.1106.4.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> <1441811441.718.18.camel@jm-desktop> <1441812507.1106.4.camel@jm-desktop> Message-ID: On Wed, Sep 9, 2015 at 4:26 PM, joem wrote: >> >> > Hi, >> >> > >> >> > Just wondering how much is it these days ball park to pay someone like >> >> > witstech or a similar design outfit to design an A20 board >> >> >> >> about $USD 3-4k >> > >> > Thats a good number! Do you think they will hand over the design files? >> >> they always do. >> >> > I don't want to spend another 2.5k on top for the 10 samples. >> >> it was approximate. >> >> ... y'know... if you're looking for someone to design an A20 board, i >> could always do it for you, i have a working proven and very small >> footprint board. > > If you are offering, then name a ball park price (send to > research at enemygadgets.com) > I'm happy to say you deserve it more than anyone else :) thought about this a bit more: i need 10 samples of a revised EOMA68-A20, after i just changed the standard (again. for the last time. again). i'll do you a deal: if you can pay for the 10 samples, i'll do the design of an A20 board for you. that way, apart from anything, both of us would get a better deal on sample quantity. l. From lkcl at lkcl.net Wed Sep 9 19:57:36 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 19:57:36 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> Message-ID: On Wed, Sep 9, 2015 at 7:42 PM, Luke Kenneth Casson Leighton wrote: > if you're not familiar with or don't clearly understand the > difference, look up the history behind why the Debian Team renamed > firefox to "iceweasel". here you go: https://en.wikipedia.org/wiki/Mozilla_Corporation_software_rebranded_by_the_Debian_project so, if joe were to use the CC-attribution design, he would be required to advertise the name of the creator on the OUTSIDE OF THE PRODUCT. prominently. including on the casework if the PCB is inside a box. so, when someone bought that product, despite the fact that joe would be the one that had put the effort into marketing, sales, spent the up-front cash on prototypes (which is a considerable amount)... .... guess whom the customers are most likely to contact? not joe - the one who put the effort into getting a polished professional product into their hands. by complete contrast, for GPL'd products, all you have to do is put a little bit of paper in the box saying "contact us if you want the source code". you *DO NOT* have to put "Copyright (C) Blah Blah" on the OUTSIDE OF THE PRODUCT, just because it's got some software in it (or in the case of hardware is manufactured from design files that are GPL licensed). is that clearer? btw i was amazed and deeply impressed when i bought a TP-Link router last year, because it contained *exactly that* in the box. finally - at long last - a large company that understands its obligations under the GPL. l. From elena.valhalla at gmail.com Wed Sep 9 20:58:15 2015 From: elena.valhalla at gmail.com (Elena ``of Valhalla'') Date: Wed, 9 Sep 2015 21:58:15 +0200 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> Message-ID: <20150909195815.GA26396@virginsteele.home.trueelena.org> On 2015-09-09 at 19:42:25 +0100, Luke Kenneth Casson Leighton wrote: > > "provided that you conspicuously and appropriately publish on each copy > > an appropriate copyright notice; keep intact all notices stating that > > [...] > so that's in the source code. also in non source: section 6 says "You may convey a covered work in object code form under the terms of sections 4 and 5" (section 4 is the one quoted above) similar wording is also in GPLv2 -- Elena ``of Valhalla'' From phil at hands.com Wed Sep 9 20:59:17 2015 From: phil at hands.com (Philip Hands) Date: Wed, 09 Sep 2015 20:59:17 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> Message-ID: <871te7qs8a.fsf@whist.hands.com> Luke Kenneth Casson Leighton writes: > On Wed, Sep 9, 2015 at 7:42 PM, Luke Kenneth Casson Leighton > wrote: > >> if you're not familiar with or don't clearly understand the >> difference, look up the history behind why the Debian Team renamed >> firefox to "iceweasel". > > here you go: > > https://en.wikipedia.org/wiki/Mozilla_Corporation_software_rebranded_by_the_Debian_project That was nothing to do with copyright, nor attribution. The problem is the way that Mozilla enforces its trademark. Mozilla is (fairly reasonably) concerned that people might take one of it's trademarked programs, trojan it, and redistribute the result under the name of e.g. Firefox, thus tainting their good name. They therefore reserve the right to specify which code costitutes Firefox, etc. and want sight of any patches that are applied to allow them to determine whether they should withdraw the use of the name from the result of the patch. Debian on the other hand wants to be able to apply security patches without needing to ask Mozilla for approval, and more importantly perhaps want not to impose such restrictions on their downstreams. The use of the Ice* names is done to avoid the scenario where a security fix fails to meet with approval, and then the Debian maintainers being faced with the need to do an emergency trademark purge in order to deploy a security fix. Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] HANDS.COM Ltd. |-| http://www.hands.com/ http://ftp.uk.debian.org/ |(| Hugo-Klemm-Strasse 34, 21075 Hamburg, GERMANY -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 818 bytes Desc: not available URL: From lkcl at lkcl.net Wed Sep 9 22:04:43 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 9 Sep 2015 22:04:43 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <871te7qs8a.fsf@whist.hands.com> References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> <871te7qs8a.fsf@whist.hands.com> Message-ID: On Wed, Sep 9, 2015 at 8:59 PM, Philip Hands wrote: > Luke Kenneth Casson Leighton writes: > >> On Wed, Sep 9, 2015 at 7:42 PM, Luke Kenneth Casson Leighton >> wrote: >> >>> if you're not familiar with or don't clearly understand the >>> difference, look up the history behind why the Debian Team renamed >>> firefox to "iceweasel". >> >> here you go: >> >> https://en.wikipedia.org/wiki/Mozilla_Corporation_software_rebranded_by_the_Debian_project > > That was nothing to do with copyright, nor attribution. > > The problem is the way that Mozilla enforces its trademark. > > Mozilla is (fairly reasonably) concerned that people might take one of > its trademarked programs, trojan it, and redistribute the result under > the name of e.g. Firefox, thus tainting their good name. They therefore > reserve the right to specify which code costitutes Firefox, etc. and > want sight of any patches that are applied to allow them to determine > whether they should withdraw the use of the name from the result of the > patch. so mozilla have a total lack of trust of the debian team. that's the debian team who have software libre's interests, user's interests, their own long-standing reputation (backed up by GPG-signing) to protect, and the mozilla foundation's directors could not see fit to trust such reliable and reputable people to look after something as critical as security patches. > Debian on the other hand wants to be able to apply security patches > without needing to ask Mozilla for approval, and more importantly perhaps > want not to impose such restrictions on their downstreams. > > The use of the Ice* names is done to avoid the scenario where a security > fix fails to meet with approval, and then the Debian maintainers being > faced with the need to do an emergency trademark purge in order to > deploy a security fix. good for them. sounds like the right decision. also sounds very much like i quoted _completely_ the wrong example. any other mistakes i made that you can see, phil? :) l. From phil at hands.com Wed Sep 9 22:43:33 2015 From: phil at hands.com (Philip Hands) Date: Wed, 09 Sep 2015 22:43:33 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> <871te7qs8a.fsf@whist.hands.com> Message-ID: <87si6np8u2.fsf@whist.hands.com> Luke Kenneth Casson Leighton writes: > On Wed, Sep 9, 2015 at 8:59 PM, Philip Hands wrote: >> Luke Kenneth Casson Leighton writes: >> >>> On Wed, Sep 9, 2015 at 7:42 PM, Luke Kenneth Casson Leighton >>> wrote: >>> >>>> if you're not familiar with or don't clearly understand the >>>> difference, look up the history behind why the Debian Team renamed >>>> firefox to "iceweasel". >>> >>> here you go: >>> >>> https://en.wikipedia.org/wiki/Mozilla_Corporation_software_rebranded_by_the_Debian_project >> >> That was nothing to do with copyright, nor attribution. >> >> The problem is the way that Mozilla enforces its trademark. >> >> Mozilla is (fairly reasonably) concerned that people might take one of >> its trademarked programs, trojan it, and redistribute the result under >> the name of e.g. Firefox, thus tainting their good name. They therefore >> reserve the right to specify which code costitutes Firefox, etc. and >> want sight of any patches that are applied to allow them to determine >> whether they should withdraw the use of the name from the result of the >> patch. > > so mozilla have a total lack of trust of the debian team. No. Mozilla has a trademark policy designed to deal with abusers. Debian has a policy that requires any license to _not_ be exclusive to Debian, because that would cause trouble downstream. These two things are both reasonable, but sadly incompatible. > that's the > debian team who have software libre's interests, user's interests, > their own long-standing reputation (backed up by GPG-signing) to > protect, and the mozilla foundation's directors could not see fit to > trust such reliable and reputable people to look after something as > critical as security patches. None of that is relevant. >> Debian on the other hand wants to be able to apply security patches >> without needing to ask Mozilla for approval, and more importantly perhaps >> want not to impose such restrictions on their downstreams. >> >> The use of the Ice* names is done to avoid the scenario where a security >> fix fails to meet with approval, and then the Debian maintainers being >> faced with the need to do an emergency trademark purge in order to >> deploy a security fix. > > good for them. sounds like the right decision. > > also sounds very much like i quoted _completely_ the wrong example. > any other mistakes i made that you can see, phil? :) Since you ask: https://www.olimex.com/Products/OLinuXino/open-source-hardware "The Hardware project is released under the Creative Commons Attribution-Share Alike 3.0 United States License." http://www.gnu.org/licenses/license-list.html#ccbysa "This is a copyleft free license that is good for ..." So the olimex boards are under an FSF-approved copyleft license. I think perhaps you've conflated the word "Attribution" with the BSD 4-clause license (with its obnoxious "Advertising" clause): http://www.gnu.org/licenses/license-list.html#OriginalBSD Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] HANDS.COM Ltd. |-| http://www.hands.com/ http://ftp.uk.debian.org/ |(| Hugo-Klemm-Strasse 34, 21075 Hamburg, GERMANY -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 818 bytes Desc: not available URL: From joem at martindale-electric.co.uk Thu Sep 10 09:19:30 2015 From: joem at martindale-electric.co.uk (joem) Date: Thu, 10 Sep 2015 08:19:30 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <87si6np8u2.fsf@whist.hands.com> References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> <871te7qs8a.fsf@whist.hands.com> <87si6np8u2.fsf@whist.hands.com> Message-ID: <1441873222.3559.17.camel@jm-desktop> > Mozilla has a trademark policy designed to deal with abusers. > > Debian has a policy that requires any license to _not_ be exclusive to > Debian, because that would cause trouble downstream. > > These two things are both reasonable, but sadly incompatible. For what it may be worth, my ultimate aim is to get the thing built, and redo it in KiCAD and eventually build it and release it under GPL with full documentation as I'm paying for it. The aim is to have a GPL'd base design out there and watch all those GPL'd derivatives being built to harness it for more projects. The two or three dollar differences between boards doesn't go towards much in the way of profit today if anyone cares to look into it. So a fully GPL'd kicad design that anyone can customize won't be harming anyone with a head screwed on. Its what you customise these boards into that matters. For example PCDuino is a wild success because although its a cubieboard equivalent, it has arduino compatibility and is sold as an Arduino on steroids which it is :) There are a lot more projects like that waiting to happen me thinks. From lkcl at lkcl.net Thu Sep 10 10:20:51 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Sep 2015 10:20:51 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <87si6np8u2.fsf@whist.hands.com> References: <1441810196.718.2.camel@jm-desktop> <20150909172013.GB9597@virginsteele.home.trueelena.org> <871te7qs8a.fsf@whist.hands.com> <87si6np8u2.fsf@whist.hands.com> Message-ID: On Wed, Sep 9, 2015 at 10:43 PM, Philip Hands wrote: > I think perhaps you've conflated the word "Attribution" with the > BSD 4-clause license (with its obnoxious "Advertising" clause): that sounds about right. > http://www.gnu.org/licenses/license-list.html#OriginalBSD ta for the clarification phil. l. From lkcl at lkcl.net Thu Sep 10 18:11:11 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 10 Sep 2015 18:11:11 +0100 Subject: [Arm-netbook] libre laptop 15in pcb progress Message-ID: eurocircuits report "accepted" after manufacturing the main and the EC PCB prototypes.... they're only 1/2oz copper, literally bare copper, and for $EUR 100 including shipping 2 of each i am really happy with that. components arrived from digikey for the main PCB as well, i can start burning my eyes and singeing my hair some time next week :) l. From plugwash at p10link.net Sat Sep 12 21:43:46 2015 From: plugwash at p10link.net (peter green) Date: Sat, 12 Sep 2015 21:43:46 +0100 Subject: [Arm-netbook] libre laptop power board, needs 2oz copper, any recommendations? In-Reply-To: References: Message-ID: <55F48E82.7030108@p10link.net> On 03/09/15 02:12, Luke Kenneth Casson Leighton wrote: > ok, i checked with the eurocircuits online system, they can't handle > the power board. > It's worth actually contacting PCB vendors with your requirements. The specs listed for the online pool/quick quote type services are often not the absolute limits of what the vendor can do. > the LTC4155 is handling 4 amps, so the app note says that 2oz copper > is needed. however, eurocircuits require a minimum track-to-track > clearance of 10mil which is totally incompatible with the distance > even between pads of the LTC4155 IC! > LTC recommend 0.25 mm pads and the chip has a 0.5 mm pad pitch which leaves a 0.25 mm between pads. That is under 10 mil but only barely. > therefore we conclude, simply, that eurocircuits doesn't have good > enough equipment, therefore it is necessary to find an alternative PCB > prototyping company. > > i've managed to get the track-to-track distance to 7mil, smallest via > is 10mil with a 24mil ring, track width is a minimum of 8mil, and the > size of the PCB is tiny: 3.4in x 1.2in. it's only two layer, > components only on one side, but the copper needs to be 2oz. > > can anyone help with PCB manufacturing at all? > If you get desperate then contact zot, if they can't make it then proabablly noone can. From lasich at gmail.com Sun Sep 13 06:05:12 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Sun, 13 Sep 2015 07:05:12 +0200 Subject: [Arm-netbook] libre laptop power board, needs 2oz copper, any recommendations? In-Reply-To: <55F48E82.7030108@p10link.net> References: <55F48E82.7030108@p10link.net> Message-ID: Pls try to contact this guy: Shenzhen Xinjiaye Electronics Technology Co.,ltd 508Room Yong Hui building Bao Yuan Road Shenzhen city,China TEL:+86-755-21638081 Mobile:+86-13603083165 Skype:nbdyummy www.szxjypcb.com sales-yummy at xjypcb.cn *Global Sources Electronic Components ShowDate: 11th October 2015Place: AsiaWorld-Expo, Hong KongBooth: 9G13* *At very least you will get advice. They do prototypes but i don't know about price. You can send him gerber files to check. If you decide they make you complete prototypes,be extra careful with parts. When they buy on local market they can mess / switch parts.* On Saturday, September 12, 2015, peter green wrote: > On 03/09/15 02:12, Luke Kenneth Casson Leighton wrote: > >> ok, i checked with the eurocircuits online system, they can't handle >> the power board. >> >> > It's worth actually contacting PCB vendors with your requirements. The > specs listed for the online pool/quick quote type services are often not > the absolute limits of what the vendor can do. > >> the LTC4155 is handling 4 amps, so the app note says that 2oz copper >> is needed. however, eurocircuits require a minimum track-to-track >> clearance of 10mil which is totally incompatible with the distance >> even between pads of the LTC4155 IC! >> >> > LTC recommend 0.25 mm pads and the chip has a 0.5 mm pad pitch which > leaves a 0.25 mm between pads. > > That is under 10 mil but only barely. > >> therefore we conclude, simply, that eurocircuits doesn't have good >> enough equipment, therefore it is necessary to find an alternative PCB >> prototyping company. >> >> i've managed to get the track-to-track distance to 7mil, smallest via >> is 10mil with a 24mil ring, track width is a minimum of 8mil, and the >> size of the PCB is tiny: 3.4in x 1.2in. it's only two layer, >> components only on one side, but the copper needs to be 2oz. >> >> can anyone help with PCB manufacturing at all? >> >> > If you get desperate then contact zot, if they can't make it then > proabablly noone can. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk -- Hrvoje Lasić Vulpes d.o.o. Gračanska 120a 10000 Zagreb Croatia tel +385 1 6152 706 tel +38598 450 603 *lasich at gmail.com hrvoje at vebbu.co * *www.vebbu.co * -------------- next part -------------- An HTML attachment was scrubbed... URL: From paul at boddie.org.uk Mon Sep 14 16:47:38 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Mon, 14 Sep 2015 17:47:38 +0200 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices Message-ID: <201509141747.39869.paul@boddie.org.uk> Hello, So I was browsing the FreedomBox mailing list when I saw a mention of a plug computer being crowdfunded: http://lists.alioth.debian.org/pipermail/freedombox-discuss/2015- September/006879.html Here's the actual crowdfunding page: https://www.kickstarter.com/projects/1547898916/own-mailbox-the-first-100- confidential-mailbox And the home page of the device: https://www.own-mailbox.com/ So, it seems that the initiators of the campaign have made their own Allwinner A13 board with 256MB RAM and Ethernet support. It is open hardware, apparently, and the software is Free Software, not some kind of proprietary "snake oil" that some privacy campaigns tend to foist on their supporters. It is a bit odd that they've decided to go their own way with the hardware, though. The FreedomBox recommends existing hardware rather than focusing on one particular device: https://wiki.debian.org/FreedomBox/Hardware Another initiative just uses one of the Olimex A20-based products: http://internetcu.be/ Where the product in question looks a lot more capable than the one attempting to be funded above: https://www.olimex.com/Products/OLinuXino/A20/A20-OLinuXino-LIME/open-source- hardware Anyway, back to EOMA-68, and a quick search produced a page on the plug computer idea for EOMA-68: http://rhombus-tech.net/community_ideas/plug_computer/ It occurred to me that if only we had EOMA-68 boards out there, maybe people wouldn't be so enthusiastic to go to the trouble of making new boards and running the gauntlet of crowdfunding. There also wouldn't be the artificiality of "stretch goals" where things that should just be happening anyway are punted off into an uncertain future, dependent on the delivery of something that should be separate. Paul From lkcl at lkcl.net Tue Sep 15 18:18:26 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 15 Sep 2015 18:18:26 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: <201509141747.39869.paul@boddie.org.uk> References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Mon, Sep 14, 2015 at 4:47 PM, Paul Boddie wrote: > Hello, > > So I was browsing the FreedomBox mailing list when I saw a mention of a plug > computer being crowdfunded: > > http://lists.alioth.debian.org/pipermail/freedombox-discuss/2015- > September/006879.html > > Here's the actual crowdfunding page: > > https://www.kickstarter.com/projects/1547898916/own-mailbox-the-first-100- > confidential-mailbox > > And the home page of the device: > > https://www.own-mailbox.com/ looks great! > So, it seems that the initiators of the campaign have made their own Allwinner > A13 board with 256MB RAM and Ethernet support. It is open hardware, > apparently, and the software is Free Software, not some kind of proprietary > "snake oil" that some privacy campaigns tend to foist on their supporters. goooood. > It is a bit odd that they've decided to go their own way with the hardware, > though. well, if it's libre hardware then that's good, i hope! > Anyway, back to EOMA-68, and a quick search produced a page on the plug > computer idea for EOMA-68: > > http://rhombus-tech.net/community_ideas/plug_computer/ > > It occurred to me that if only we had EOMA-68 boards out there, maybe people > wouldn't be so enthusiastic to go to the trouble of making new boards and > running the gauntlet of crowdfunding. i knooooow :) here's the thing though: (1) i have to get the standard right for a 10-year-old and greater period (2) i have to get CPU Cards (plural) designed, sponsored, and tested (3) i have to prove that it is, indeed, simpler and lower-cost to make carrier boards. so that's what i'm doing. to explain, first: this project has an absolute top priority of being *right* (defined as "viable long-term") over being "rushed to profitability". second: it's no good having just the one CPU Card out there. people won't comprehend the modularity concept if there is only the additional cost of having a single processor available. third: even i was unable to move the micro-desktop board (which is only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - costs are around $400 for qty 5 4-layer prototype boards on a 3 week turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer prototype (bare copper) boards with larger vias is around $40 for qty 2, and around $100 for qty 5. so.... it's getting there, paul. l. From laserhawk64 at gmail.com Tue Sep 15 18:22:24 2015 From: laserhawk64 at gmail.com (Christopher Havel) Date: Tue, 15 Sep 2015 13:22:24 -0400 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Tue, Sep 15, 2015 at 1:18 PM, Luke Kenneth Casson Leighton wrote: > [...] even i was unable to move the micro-desktop board (which is > only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - > costs are around $400 for qty 5 4-layer prototype boards on a 3 week > turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer > prototype (bare copper) boards with larger vias is around $40 for qty > 2, and around $100 for qty 5. Imbecile question: is it at all reasonably possible to redesign the micro-desktop board to be a 2-layer board? -------------- next part -------------- An HTML attachment was scrubbed... URL: From lasich at gmail.com Tue Sep 15 18:41:50 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Tue, 15 Sep 2015 19:41:50 +0200 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Tue, Sep 15, 2015 at 7:22 PM, Christopher Havel wrote: > On Tue, Sep 15, 2015 at 1:18 PM, Luke Kenneth Casson Leighton < > lkcl at lkcl.net> wrote: > >> [...] even i was unable to move the micro-desktop board (which is >> only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - >> costs are around $400 for qty 5 4-layer prototype boards on a 3 week >> turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer >> prototype (bare copper) boards with larger vias is around $40 for qty >> 2, and around $100 for qty 5. > > > Imbecile question: is it at all reasonably possible to redesign the > micro-desktop board to be a 2-layer board? > > If board have high speed design it is very unlikely that you can do it properly (because you have to be careful about routing to ground, you have many lines that you have to match etc etc and when you have only 2 layers it is difficult to keep it all correct). In theory maybe it could be possible but it is a hell of a work and then you risk a lot that PCB will not be good. > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lkcl at lkcl.net Wed Sep 16 02:41:23 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Sep 2015 02:41:23 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Tue, Sep 15, 2015 at 6:41 PM, Hrvoje Lasic wrote: > > > On Tue, Sep 15, 2015 at 7:22 PM, Christopher Havel > wrote: >> >> On Tue, Sep 15, 2015 at 1:18 PM, Luke Kenneth Casson Leighton >> wrote: >>> >>> [...] even i was unable to move the micro-desktop board (which is >>> only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - >>> costs are around $400 for qty 5 4-layer prototype boards on a 3 week >>> turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer >>> prototype (bare copper) boards with larger vias is around $40 for qty >>> 2, and around $100 for qty 5. >> >> >> Imbecile question: is it at all reasonably possible to redesign the >> micro-desktop board to be a 2-layer board? >> > > If board have high speed design it is very unlikely that you can do it > properly (because you have to be careful about routing to ground, you have > many lines that you have to match etc etc and when you have only 2 layers it > is difficult to keep it all correct). In theory maybe it could be possible > but it is a hell of a work and then you risk a lot that PCB will not be > good. i took a look at a gigabit ethernet board that phil kindly sent me a while back. the layout of the differential pairs was absolutely fascinating [and the board had, obviously, passed FCC tests]. the layout involved putting ground vias exactly... something like 20mil from the differential pairs, spaced out at exactly something like... 100mil, right the way *both* sides. there were no components permitted either side for some distance either. there were no vias in the actual differential pairs, either. there were twenty sets of differential pairs like this - all with exactly the same very very clearly and meticulously laid out arrangement, with the spacing between each differential pair also very meticulously laid out. so it can be done. in the micro-desktop board, however, the actual distance that the USB differential pairs has to travel is well under 1cm. i am arranging the connectors *directly* in front of where the signals come out. i have also deliberately arranged the EOMA68 interface so that the pairs come out directly and do not require a via to "cross over" each other. so it is much less of a concern than might otherwise normally be. these aren't 10cm traces, where EM radiation would definitely be a major concern, they're literally 1cm long, and i intend to surround them with ground vias. of slightly more concern is the VGA interface (RGB/TTL being converted to analogue), which will be operating at around 75mhz (or so). however that's not 480mhz, so i am not hugely concerned. i am again getting in as many GND vias as can fit, and keeping the traces very very short. the buffer ICs actually straddle the PCMCIA interface on the other side of the board, so that the RGB/TTL signals can, with vias, go left or right, routing as appropriate, maximum trace length about... 3mm. l. From lkcl at lkcl.net Wed Sep 16 02:42:20 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Sep 2015 02:42:20 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Tue, Sep 15, 2015 at 6:22 PM, Christopher Havel wrote: > On Tue, Sep 15, 2015 at 1:18 PM, Luke Kenneth Casson Leighton > wrote: >> >> [...] even i was unable to move the micro-desktop board (which is >> only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - >> costs are around $400 for qty 5 4-layer prototype boards on a 3 week >> turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer >> prototype (bare copper) boards with larger vias is around $40 for qty >> 2, and around $100 for qty 5. > > > Imbecile question: is it at all reasonably possible to redesign the > micro-desktop board to be a 2-layer board? yes, i've done it already. i do have to increase the via sizes from 10mil + a 20mil surround to 12mil + a 24mil surround to fit the rules of a lower-cost PCB factory that i know of, but other than that, it's done already. l. From lasich at gmail.com Wed Sep 16 05:28:28 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Wed, 16 Sep 2015 06:28:28 +0200 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Wed, Sep 16, 2015 at 3:41 AM, Luke Kenneth Casson Leighton wrote: > On Tue, Sep 15, 2015 at 6:41 PM, Hrvoje Lasic wrote: > > > > > > On Tue, Sep 15, 2015 at 7:22 PM, Christopher Havel < > laserhawk64 at gmail.com> > > wrote: > >> > >> On Tue, Sep 15, 2015 at 1:18 PM, Luke Kenneth Casson Leighton > >> wrote: > >>> > >>> [...] even i was unable to move the micro-desktop board (which is > >>> only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - > >>> costs are around $400 for qty 5 4-layer prototype boards on a 3 week > >>> turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer > >>> prototype (bare copper) boards with larger vias is around $40 for qty > >>> 2, and around $100 for qty 5. > >> > >> > >> Imbecile question: is it at all reasonably possible to redesign the > >> micro-desktop board to be a 2-layer board? > >> > > > > If board have high speed design it is very unlikely that you can do it > > properly (because you have to be careful about routing to ground, you > have > > many lines that you have to match etc etc and when you have only 2 > layers it > > is difficult to keep it all correct). In theory maybe it could be > possible > > but it is a hell of a work and then you risk a lot that PCB will not be > > good. > > i took a look at a gigabit ethernet board that phil kindly sent me a > while back. the layout of the differential pairs was absolutely > fascinating [and the board had, obviously, passed FCC tests]. > > the layout involved putting ground vias exactly... something like > 20mil from the differential pairs, spaced out at exactly something > like... 100mil, right the way *both* sides. there were no components > permitted either side for some distance either. there were no vias in > the actual differential pairs, either. > > there were twenty sets of differential pairs like this - all with > exactly the same very very clearly and meticulously laid out > arrangement, with the spacing between each differential pair also very > meticulously laid out. > > so it can be done. > > in the micro-desktop board, however, the actual distance that the USB > differential pairs has to travel is well under 1cm. i am arranging > the connectors *directly* in front of where the signals come out. i > have also deliberately arranged the EOMA68 interface so that the pairs > come out directly and do not require a via to "cross over" each other. > > so it is much less of a concern than might otherwise normally be. > these aren't 10cm traces, where EM radiation would definitely be a > major concern, they're literally 1cm long, and i intend to surround > them with ground vias. > > of slightly more concern is the VGA interface (RGB/TTL being converted > to analogue), which will be operating at around 75mhz (or so). > however that's not 480mhz, so i am not hugely concerned. i am again > getting in as many GND vias as can fit, and keeping the traces very > very short. the buffer ICs actually straddle the PCMCIA interface on > the other side of the board, so that the RGB/TTL signals can, with > vias, go left or right, routing as appropriate, maximum trace length > about... 3mm. > > anyway, absolutely amazing work to fit it all like you described. > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lkcl at lkcl.net Wed Sep 16 11:20:21 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 16 Sep 2015 11:20:21 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: On Wed, Sep 16, 2015 at 5:28 AM, Hrvoje Lasic wrote: > anyway, absolutely amazing work to fit it all like you described. the complexity of that kind of 2-layer layout was one (unspoken) reason why i took off ethernet. yehh, we'll see what happens, neh? :) l. From gacuest at gmail.com Wed Sep 16 13:41:42 2015 From: gacuest at gmail.com (GaCuest) Date: Wed, 16 Sep 2015 14:41:42 +0200 Subject: [Arm-netbook] Interchangeable screen Message-ID: Well, we are designing a handheld games console based on EOMA-68. We have thought about doing an interchangeable screen. Users can buy the handheld games console without a display (more cheaper, users can use the HDMI of EOMA-68), with a 480p display or 720p display (more expensive) and upgrade it in a future. We would connect the TFT and touch screen connector on a PCB, and user may connect it with a connector to the base PCB. We have thought about use neodymium magnet (some of 2x1 mm of them) to connect the case of the display+touch panel with the case of handheld games console. But we have some doubts about how to connect both PCB. We've thought use a slim B2B connector like this: http://www.literature.molex.com/SQLImages/kelmscott/Molex/PDF_Images/987651-2331.PDF The problem is that slim B2B connector have slow mating cycles (about 30-50 times). Any can help us about a better way to connect it? Thanks.  From paul at boddie.org.uk Wed Sep 16 23:40:39 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Thu, 17 Sep 2015 00:40:39 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: <201509170040.39501.paul@boddie.org.uk> On Wednesday 16. September 2015 14.41.42 GaCuest wrote: > Well, we are designing a handheld games console based on EOMA-68. Are you involved in this effort...? http://rhombus-tech.net/community_ideas/games_console/ > We have thought about doing an interchangeable screen. Users can buy the > handheld games console without a display (more cheaper, users can use the > HDMI of EOMA-68), with a 480p display or 720p display (more expensive) and > upgrade it in a future. We would connect the TFT and touch screen > connector on a PCB, and user may connect it with a connector to the base > PCB. It certainly sounds like an interesting idea, and screens are definitely things that age most obviously (or have at least aged very obviously in the past). Plus, albeit not for a games console, having screen technology choices like e-paper would be nice for certain kinds of devices. Sadly, I don't have any opinions about connectors, but I thought I'd indicate some kind of support for what you're doing. :-) Paul From paul at boddie.org.uk Wed Sep 16 23:49:59 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Thu, 17 Sep 2015 00:49:59 +0200 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> Message-ID: <201509170049.59235.paul@boddie.org.uk> On Tuesday 15. September 2015 19.18.26 Luke Kenneth Casson Leighton wrote: > On Mon, Sep 14, 2015 at 4:47 PM, Paul Boddie wrote: > > > > It occurred to me that if only we had EOMA-68 boards out there, maybe > > people wouldn't be so enthusiastic to go to the trouble of making new > > boards and running the gauntlet of crowdfunding. > > i knooooow :) > > here's the thing though: > > (1) i have to get the standard right for a 10-year-old and greater period > (2) i have to get CPU Cards (plural) designed, sponsored, and tested > (3) i have to prove that it is, indeed, simpler and lower-cost to > make carrier boards. > > so that's what i'm doing. > > to explain, first: this project has an absolute top priority of being > *right* (defined as "viable long-term") over being "rushed to > profitability". Sure, I understand that. But what worries me a little is that experience isn't being gained to possibly refine the standard or develop for it. Of course, I'm writing this with only a superficial knowledge of what has gone on in the past, and it is possible that people have done things with previously-produced hardware that has informed the effort. > second: it's no good having just the one CPU Card out there. people > won't comprehend the modularity concept if there is only the > additional cost of having a single processor available. I understand this, too. It's also useful to have different devices they can be used in as well, and I worry that these won't come about without any cards being available. One thing in the back of my mind (and part of a long list of things that I could consider doing) is helping to design such a device, and for that I suppose I need to collect links to documentation that might help me get started. > third: even i was unable to move the micro-desktop board (which is > only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - > costs are around $400 for qty 5 4-layer prototype boards on a 3 week > turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer > prototype (bare copper) boards with larger vias is around $40 for qty > 2, and around $100 for qty 5. > > so.... it's getting there, paul. Yes, prototyping gets expensive for advanced stuff, I suppose, and since we're not part of the normal industrial operations that can do this efficiently (in terms of costs and other things), we're at a disadvantage. Still, I wonder what those of us reading this list might be able to do to move the effort forward in our own way. Paul From lkcl at lkcl.net Thu Sep 17 00:07:33 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 00:07:33 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: On Wed, Sep 16, 2015 at 1:41 PM, GaCuest wrote: > Well, we are designing a handheld games console based on EOMA-68. yay! (and yes, paul, this one: http://rhombus-tech.net/community_ideas/games_console/ ) > We have thought about doing an interchangeable screen. Users can buy the handheld games console without a display (more cheaper, users can use the HDMI of EOMA-68), with a 480p display or 720p display (more expensive) and upgrade it in a future. We would connect the TFT and touch screen connector on a PCB, and user may connect it with a connector to the base PCB. > > We have thought about use neodymium magnet (some of 2x1 mm of them) to connect the case of the display+touch panel with the case of handheld games console. But we have some doubts about how to connect both PCB. > > We've thought use a slim B2B connector like this: http://www.literature.molex.com/SQLImages/kelmscott/Molex/PDF_Images/987651-2331.PDF > > The problem is that slim B2B connector have slow mating cycles (about 30-50 times). Any can help us about a better way to connect it? honestly, i don't recommend letting users consider it to be an "end-user arbitrarily replaceable" component, but more of an "end user *serviceable*" part i.e. they definitely definitely need some sort of technical skill. encouraging that view by requiring that they use a screwdriver or other tool to get at the screen would, i feel, be something that you should deliberately push users towards so that they feel uncomfortable unless they are confident with tools and ESD precautions. if the screen could be packaged in a small self-contained unit that had absolutely no chance of physical damage, where it was possible to "slot in" to a special set of guide rails where there was absolutely no chance that the end user could possibly get it wrong, then i would say "go for it". however i think you'd agree that doing so hugely complicates the casework design and leaves you with a lot more to consider right now. my advice to you therefore would be to use an FPC cable at either end and to have the LCD module screwed down, deliberately so that it's clear that this is *absolutely not* something that you should expect a 3-year-old or a 90-year-old infirm and elderly non-technical person to be able to handle. once you have a first revision out the door, and have some cashflow, by all means come back to the end-user removable LCD concept, but right now i think you're asking too much. l. From gacuest at gmail.com Thu Sep 17 00:08:38 2015 From: gacuest at gmail.com (GaCuest) Date: Thu, 17 Sep 2015 01:08:38 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: <201509170040.39501.paul@boddie.org.uk> References: <201509170040.39501.paul@boddie.org.uk> Message-ID: En 17 de septiembre de 2015 en 0:41:08, Paul Boddie (paul at boddie.org.uk) escrito: > On Wednesday 16. September 2015 14.41.42 GaCuest wrote: > > Well, we are designing a handheld games console based on EOMA-68. > > Are you involved in this effort...? > > http://rhombus-tech.net/community_ideas/games_console/ >  Yes, we are involved in that project. Images are old (we have to change the design due to the interchangeable screen). > > We have thought about doing an interchangeable screen. Users can buy the > > handheld games console without a display (more cheaper, users can use the > > HDMI of EOMA-68), with a 480p display or 720p display (more expensive) and > > upgrade it in a future. We would connect the TFT and touch screen > > connector on a PCB, and user may connect it with a connector to the base > > PCB. > > It certainly sounds like an interesting idea, and screens are definitely > things that age most obviously (or have at least aged very obviously in the > past). Plus, albeit not for a games console, having screen technology choices > like e-paper would be nice for certain kinds of devices. > > Sadly, I don't have any opinions about connectors, but I thought I'd indicate > some kind of support for what you're doing. :-) >  Thanks for your interest. Any help/support/suggestion that you can offer is good. Very thanks. > Paul > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Thu Sep 17 00:34:15 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 00:34:15 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: <201509170049.59235.paul@boddie.org.uk> References: <201509141747.39869.paul@boddie.org.uk> <201509170049.59235.paul@boddie.org.uk> Message-ID: On Wed, Sep 16, 2015 at 11:49 PM, Paul Boddie wrote: > On Tuesday 15. September 2015 19.18.26 Luke Kenneth Casson Leighton wrote: >> On Mon, Sep 14, 2015 at 4:47 PM, Paul Boddie wrote: >> > >> > It occurred to me that if only we had EOMA-68 boards out there, maybe >> > people wouldn't be so enthusiastic to go to the trouble of making new >> > boards and running the gauntlet of crowdfunding. >> >> i knooooow :) >> >> here's the thing though: >> >> (1) i have to get the standard right for a 10-year-old and greater period >> (2) i have to get CPU Cards (plural) designed, sponsored, and tested >> (3) i have to prove that it is, indeed, simpler and lower-cost to >> make carrier boards. >> >> so that's what i'm doing. >> >> to explain, first: this project has an absolute top priority of being >> *right* (defined as "viable long-term") over being "rushed to >> profitability". > > Sure, I understand that. But what worries me a little is that experience isn't > being gained to possibly refine the standard paul, you misunderstand the concept of a simple long-term standard. it goes from "being developed and refined" to "absolutely locked irrevocably without fail absolute without fail absolute 100% in stone" with no phase in between. once there are any end-users out there, there *is* no "second chance". that's *IT*. i cannot make that complex enough to be understood and accepted. the reasons why there are no second chances are clearly laid out in the white paper i wrote. so, the opportunity, if anyone wants to get involved with "refinement" of the standard, that's what they've had over the past four years, and now it's almost over. they can talk to me, they can read the white paper (in which i describe the justification of the interface selection), and they can try to argue for additions or changes, but right now the window of opportinity to do that is closing, as i am about to send off two CPU Cards with the latest (last) revisions. anyone beyond that point in about 10-14 days time, if they want modifications to the standard, it had better come with a cheque for $5,000 attached to it in order to repay both my clients for the samples that will have been made by that time. > or develop for it. that's straightforward "critical mass", which comes once the standard's finalised, first products out the door, etc. etc. > Of course, I'm > writing this with only a superficial knowledge of what has gone on in the > past, and it is possible that people have done things with previously-produced > hardware that has informed the effort. http://rhombus-tech.net/allwinner_a10/news/ that's a brief timeline for the a10 cpu card (which became the a20 cpu card). the standard has been revised about... i think it's three times since then, maybe four. first revision was to remove SATA and replace it with a 2nd USB2. second revision was to add VREFTTL, add SD/MMC and UART, third (or possibly still part of the 2nd) was to reduce 24-pin RGB/TTL to 18-pin RGB/TTL and use the 4 spare lines for an SPI interface, also USB3 was added at some point. PWM and an extra EINT were also added. the fourth - and almost certainly final revision - has been very recent: removal of Ethernet, upgrading to being able to do USB 3.1, as well as add 2 more EINT lines and 3 more GPIOs. those interfaces have all been very carefully considered, especially when developing the ICubeCorp IC3128 CPU Card, where, due to its low pincount and being a QFP, there's *literally* only 2 spare unused pins left on the *entire* processor that don't go to the EOMA68 interface or the SD/MMC boot card. >> second: it's no good having just the one CPU Card out there. people >> won't comprehend the modularity concept if there is only the >> additional cost of having a single processor available. > > I understand this, too. It's also useful to have different devices they can be > used in as well, and I worry that these won't come about without any cards > being available. that's why i'm doing more than one product - micro-desktop as well as the laptop. i've actually designed around... 5 or 6 different products, all at different stages. the tablet is "on hold" because of lack of interest [and, because it's a complex and dense 4-layer board it's going to be tough converting it to 2-layer]. the 15in laptop is sponsored so that's ok. and the micro-desktop, what with the PCB now being 2-layer, is actually low-cost enough for me to put together a new revision on a very low budget. irony is all these will probably hit all at the same time. > One thing in the back of my mind (and part of a long list of things that I > could consider doing) :) > is helping to design such a device, and for that I > suppose I need to collect links to documentation that might help me get > started. honestly, finding the parts (ones that are affordable, not end-of-life, or fell off the back of a lorry and i do mean that literally, even though they had to have a bit of a push to fall of....) and/or finding the contacts _willing_ to source the parts - that's the hardest bit of the entire job of designing any board. oh. that and, of course, being able to get hold of the damn datasheets. one cool product that would be awesome to have would be a 3d printer controller that took an EOMA68 CPU Card. in some ways this would be a bodge-job of taking a libre-hardware-licensed PCB design, expanding its size a bit and connecting 5V power and USB up to the on-board USB interface of the Micro-Controller. that way you would have the actual computer *on* the same board that had the (usually unreliable) power line... the one that gets spiked due to 50hz mains fluctuations and lack of proper earth loops associated with USB cables,500 watt PSUs, and amateur-designed PCBs. a second cool product would be a robotics platform / educational platform. another would be to track down a suitable libre-toolchain FPGA and make a CPU Card out of it. i heard that someone has actually managed to reverse-engineer one of the commonly-available FPGAs, to the point where the toolchain is stable. saw an article about it recently. that _would_ be awesome. >> third: even i was unable to move the micro-desktop board (which is >> only 4in x 4.5in) forward because i had designed it as a 4-layer PCB - >> costs are around $400 for qty 5 4-layer prototype boards on a 3 week >> turnaround by complete contrast, a 7-day turnaround for qty 5 2-layer >> prototype (bare copper) boards with larger vias is around $40 for qty >> 2, and around $100 for qty 5. >> >> so.... it's getting there, paul. > > Yes, prototyping gets expensive for advanced stuff, I suppose, and since we're > not part of the normal industrial operations that can do this efficiently (in > terms of costs and other things), we're at a disadvantage. > > Still, I wonder what those of us reading this list might be able to do to move > the effort forward in our own way. that would be great. From maillist_arm-netbook at aross.me Thu Sep 17 02:04:47 2015 From: maillist_arm-netbook at aross.me (Alexander Stephen Thomas Ross) Date: Thu, 17 Sep 2015 02:04:47 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> <201509170049.59235.paul@boddie.org.uk> Message-ID: <55FA11AF.70508@aross.me> On 17/09/15 00:34, Luke Kenneth Casson Leighton wrote: > that's why i'm doing more than one product - micro-desktop as well as > the laptop. i've actually designed around... 5 or 6 different > products, all at different stages. the tablet is "on hold" because of > lack of interest [and, because it's a complex and dense 4-layer board > it's going to be tough converting it to 2-layer]. the 15in laptop is > sponsored so that's ok. and the micro-desktop, what with the PCB now > being 2-layer, is actually low-cost enough for me to put together a > new revision on a very low budget. awww :( ive been looking forward to the tablet, in fact ive been holding out the last 4 years cus i wanted my first tablet to be a eoma one :D. i guess lack of interest means lack of funds given to help make it... :/ hmmm... but yea, i look forward so much to buying the eoma(-68) laptop, the eoma micro-desktop, eoma tablet, heck and maybe even the eoma games console! the eoma-68 collection! yum yum! im interested in the games console and think its brilliant that theres a team working on a eoma-68 one :D, but i dont play games to so much these days... ill guess ill see how much it all adds up to in the shopping basket at the time hehe. Those past cad renders i thought looked really cool btw. looking forward to seeing how it looks now. been wondering what state its at lately. i do love reading your posts luke. i find them a great read. I even loled at the bit about the parts falling off the lorry :D From lkcl at lkcl.net Thu Sep 17 02:13:23 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 02:13:23 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: <55FA11AF.70508@aross.me> References: <201509141747.39869.paul@boddie.org.uk> <201509170049.59235.paul@boddie.org.uk> <55FA11AF.70508@aross.me> Message-ID: On Thu, Sep 17, 2015 at 2:04 AM, Alexander Stephen Thomas Ross wrote: > On 17/09/15 00:34, Luke Kenneth Casson Leighton wrote: > >> that's why i'm doing more than one product - micro-desktop as well as >> the laptop. i've actually designed around... 5 or 6 different >> products, all at different stages. the tablet is "on hold" because of >> lack of interest [and, because it's a complex and dense 4-layer board >> it's going to be tough converting it to 2-layer]. the 15in laptop is >> sponsored so that's ok. and the micro-desktop, what with the PCB now >> being 2-layer, is actually low-cost enough for me to put together a >> new revision on a very low budget. > > awww :( ive been looking forward to the tablet, in fact ive been holding > out the last 4 years cus i wanted my first tablet to be a eoma one :D. > > i guess lack of interest means lack of funds given to help make it... :/ > hmmm... remember the "or" there - ooorrr i redesign it to be 2-layer, in which case it's no longer $600 for 5 PCBs but is something like... $100 for 5 PCBs. the only problem being, they'd be 1.5mm thick prototype PCBs, not the 1.2mm that the casework is designed for. > but yea, i look forward so much to buying the eoma(-68) laptop, the eoma > micro-desktop, eoma tablet, heck and maybe even the eoma games console! > the eoma-68 collection! yum yum! :) > im interested in the games console and think its brilliant that theres a > team working on a eoma-68 one :D, but i dont play games to so much these > days... well that's ok, because as you can see from the scenarios section of the white paper, and from the whole eoma68 concept, it's rather plainly obvious that "an eoma68 games console" is NOT just "a games console". > i do love reading your posts luke. i find them a great read. I even > loled at the bit about the parts falling off the lorry :D haha goood :) From paul at boddie.org.uk Thu Sep 17 12:04:43 2015 From: paul at boddie.org.uk (Paul Boddie) Date: Thu, 17 Sep 2015 13:04:43 +0200 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: References: <201509141747.39869.paul@boddie.org.uk> <201509170049.59235.paul@boddie.org.uk> Message-ID: <201509171304.44619.paul@boddie.org.uk> On Thursday 17. September 2015 01.34.15 Luke Kenneth Casson Leighton wrote: > On Wed, Sep 16, 2015 at 11:49 PM, Paul Boddie wrote: > > > > Sure, I understand that. But what worries me a little is that experience > > isn't being gained to possibly refine the standard > > paul, you misunderstand the concept of a simple long-term standard. Not really. What I meant by "refined" is actually this: [...] > first revision was to remove SATA and replace it with a 2nd USB2. > second revision was to add VREFTTL, add SD/MMC and UART, third (or > possibly still part of the 2nd) was to reduce 24-pin RGB/TTL to 18-pin > RGB/TTL and use the 4 spare lines for an SPI interface, also USB3 was > added at some point. PWM and an extra EINT were also added. the > fourth - and almost certainly final revision - has been very recent: > removal of Ethernet, upgrading to being able to do USB 3.1, as well as > add 2 more EINT lines and 3 more GPIOs. > > those interfaces have all been very carefully considered, especially > when developing the ICubeCorp IC3128 CPU Card, where, due to its low > pincount and being a QFP, there's *literally* only 2 spare unused pins > left on the *entire* processor that don't go to the EOMA68 interface > or the SD/MMC boot card. All of this has happened before we get to the point where we call it a final standard, but what worries me is that there may be an application that hasn't yet been considered because the collective experience of trying to make devices using it is not broad enough. I must admit that this is coloured by my interests in "retrocomputing" where one can look at products that were made and then consider how they might have been improved, even by a small amount, in a way that might have made them a lot more successful. At the time, you'd have some company or other designing and manufacturing their products to a tight schedule (usually to hit the market at the best time of year), but there would be limitations discovered by the customers that would limit the competitive lifetime of the product. In any case, from what you've written, I guess we'll find out for ourselves soon enough about how successful the refinement process has been. Not that I think that it hasn't been successful enough, however. [...] > > Still, I wonder what those of us reading this list might be able to do to > > move the effort forward in our own way. > > that would be great. Any suggestions? :-) Paul From lkcl at lkcl.net Thu Sep 17 13:13:44 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 13:13:44 +0100 Subject: [Arm-netbook] Plug computer and FreedomBox-related devices In-Reply-To: <201509171304.44619.paul@boddie.org.uk> References: <201509141747.39869.paul@boddie.org.uk> <201509170049.59235.paul@boddie.org.uk> <201509171304.44619.paul@boddie.org.uk> Message-ID: On Thu, Sep 17, 2015 at 12:04 PM, Paul Boddie wrote: > On Thursday 17. September 2015 01.34.15 Luke Kenneth Casson Leighton wrote: >> On Wed, Sep 16, 2015 at 11:49 PM, Paul Boddie wrote: >> > >> > Sure, I understand that. But what worries me a little is that experience >> > isn't being gained to possibly refine the standard >> >> paul, you misunderstand the concept of a simple long-term standard. > > Not really. What I meant by "refined" is actually this: > > [...] > >> first revision was to remove SATA and replace it with a 2nd USB2. >> second revision was to add VREFTTL, add SD/MMC and UART, third (or >> possibly still part of the 2nd) was to reduce 24-pin RGB/TTL to 18-pin >> RGB/TTL and use the 4 spare lines for an SPI interface, also USB3 was >> added at some point. PWM and an extra EINT were also added. the >> fourth - and almost certainly final revision - has been very recent: >> removal of Ethernet, upgrading to being able to do USB 3.1, as well as >> add 2 more EINT lines and 3 more GPIOs. >> >> those interfaces have all been very carefully considered, especially >> when developing the ICubeCorp IC3128 CPU Card, where, due to its low >> pincount and being a QFP, there's *literally* only 2 spare unused pins >> left on the *entire* processor that don't go to the EOMA68 interface >> or the SD/MMC boot card. > > All of this has happened before we get to the point where we call it a final > standard, but what worries me is that there may be an application that hasn't > yet been considered because the collective experience of trying to make > devices using it is not broad enough. well, as well as the section on the elinux.org web site analysing and tracking half a dozen different standards and their limitations, that's what the section on "interfaces" in the white paper is for, to go through the past (several decades of computing), learn from it, predict where it's going, track that for 4 years, adjust the tracking to make sure it fits, then re-predict, re-confirm, re-track and then at some point say "ok done". it's precisely the knowledge of prior failed standards, paul, that's kept me from going, like a drunken naive wannab computaa n00b "okaay yeahh i wanna do a standurd now, let's put some inturfaciz in 4 fun, make some lolli y not?" - translating that into english it reads "create something then throw it out the door in under a year and hope it works" *NO*. i recommend reading the sections on the white paper covering the standard's development and justification (so that i don't have to repeat it here) because i go into some depth to justify each of the decisions that are made, including analysing and demonstrating how long they've each been around, and how long they are likely to stay around. > I must admit that this is coloured by my interests in "retrocomputing" where > one can look at products that were made and then consider how they might have > been improved, even by a small amount, in a way that might have made them a > lot more successful. well then, you would enjoy the anecdotes that i included in the white paper, which include some historical and hilarious examples of exactly that. > At the time, you'd have some company or other designing > and manufacturing their products to a tight schedule (usually to hit the > market at the best time of year), but there would be limitations discovered by > the customers that would limit the competitive lifetime of the product. ... or in the case of standards, not enough thought went into them, so they cause utter confusion and mental melt-down in the minds of adopters. i am not joking but the only decent standards which do not cause such complete melt-down are COM-Express and PC-104. there's no "optionalitis" in those standards (except COM-Express module sizes, and that's ok). .... https://en.wikipedia.org/wiki/PC/104#Potential_Compatibility_Issues aaargh noooo, PC-104 is.... arg. of course, because it's based around the IBM PC which of course needs 12v, 5v and 3.3v, it's all gone tits-up thanks to some carrier boards not properly supplying all the required voltages, so the CPU boards themselves provide converters to compensate... .which of course fucks everything up because the difference between the I/O levels is enough to draw current one way or the other and burn out components. this is *EXACTLY* why the Certification Mark is so damn important, so as to be able to stomp from a Great Height on anyone not properly following the EOMA68 standard. > In any case, from what you've written, I guess we'll find out for ourselves > soon enough about how successful the refinement process has been. Not that I > think that it hasn't been successful enough, however. > > [...] > >> > Still, I wonder what those of us reading this list might be able to do to >> > move the effort forward in our own way. >> >> that would be great. > > Any suggestions? :-) a couple which were in my reply (two product ideas), whilst as you hint at most stuff now has to wait until i've got the prototypes for the jz4775 and a20 cpu cards and the two products microdesktop and laptop_15in done. l. From gacuest at gmail.com Thu Sep 17 13:28:39 2015 From: gacuest at gmail.com (GaCuest) Date: Thu, 17 Sep 2015 14:28:39 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: En 17 de septiembre de 2015 en 1:07:53, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Wed, Sep 16, 2015 at 1:41 PM, GaCuest wrote: > > Well, we are designing a handheld games console based on EOMA-68. > > yay! (and yes, paul, this one: > http://rhombus-tech.net/community_ideas/games_console/ ) > > > We have thought about doing an interchangeable screen. Users can buy the handheld games > console without a display (more cheaper, users can use the HDMI of EOMA-68), with a 480p > display or 720p display (more expensive) and upgrade it in a future. We would connect > the TFT and touch screen connector on a PCB, and user may connect it with a connector to > the base PCB. > > > > We have thought about use neodymium magnet (some of 2x1 mm of them) to connect the case > of the display+touch panel with the case of handheld games console. But we have some doubts > about how to connect both PCB. > > > > We've thought use a slim B2B connector like this: http://www.literature.molex.com/SQLImages/kelmscott/Molex/PDF_Images/987651-2331.PDF > > > > The problem is that slim B2B connector have slow mating cycles (about 30-50 times). > Any can help us about a better way to connect it? > > honestly, i don't recommend letting users consider it to be an > "end-user arbitrarily replaceable" component, but more of an "end user > *serviceable*" part i.e. they definitely definitely need some sort of > technical skill. > > encouraging that view by requiring that they use a screwdriver or > other tool to get at the screen would, i feel, be something that you > should deliberately push users towards so that they feel uncomfortable > unless they are confident with tools and ESD precautions. > > if the screen could be packaged in a small self-contained unit that > had absolutely no chance of physical damage, where it was possible to > "slot in" to a special set of guide rails where there was absolutely > no chance that the end user could possibly get it wrong, then i would > say "go for it". > > however i think you'd agree that doing so hugely complicates the > casework design and leaves you with a lot more to consider right now. > > my advice to you therefore would be to use an FPC cable at either end > and to have the LCD module screwed down, deliberately so that it's > clear that this is *absolutely not* something that you should expect a > 3-year-old or a 90-year-old infirm and elderly non-technical person to > be able to handle. > > once you have a first revision out the door, and have some cashflow, > by all means come back to the end-user removable LCD concept, but > right now i think you're asking too much. Yes, obviously change the display would be for more advanced users. We would sell the console in 3 packs (without display, with 480p display, and with 720p display). And in the future, the user can update the display itself (or send the console and we update it). However, although change the screen would be for users with more knowledge, we want to make it as simple as possible, and perhaps using FPC cables and screws complicates it too. Thanks. From lkcl at lkcl.net Thu Sep 17 14:04:17 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 14:04:17 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: On Thu, Sep 17, 2015 at 1:28 PM, GaCuest wrote: > En 17 de septiembre de 2015 en 1:07:53, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > However, although change the screen would be for users with more knowledge, we want to make it as simple as possible, and perhaps using FPC cables and screws complicates it too. it's up to you. an FPC cable at least a replacement can be bought. those connectors, once damaged, it's game over. consider this: if a console is dropped - and it will get dropped - those neodymium magnets aren't going to hold, are they? that will result in those B2B connectors being ripped off the PCB, quite likely damaging both the LCD board *and* the main board. with screws, that's not going to happen, is it? and even if it did (if someone dropped the console *explicitly* after taking the screws out), well.... the FPC cable might get torn out, but it will be the weakest chain in the link and would die before taking out the actual FPC connectors. even vibration or general handling is going to cause stress on those B2B connectors unless you have some screws making absolutely, absolutely sure that the two boards are not going to move. at all. mechanically, the combination of a tiny B2B connector and neodymium magnets seems like asking for trouble. l. From gacuest at gmail.com Thu Sep 17 16:12:51 2015 From: gacuest at gmail.com (GaCuest) Date: Thu, 17 Sep 2015 17:12:51 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: En 17 de septiembre de 2015 en 15:04:24, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Thu, Sep 17, 2015 at 1:28 PM, GaCuest wrote: > > En 17 de septiembre de 2015 en 1:07:53, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > escrito: > > > However, although change the screen would be for users with more knowledge, we want > to make it as simple as possible, and perhaps using FPC cables and screws complicates > it too. > > it's up to you. an FPC cable at least a replacement can be bought. > those connectors, once damaged, it's game over. > > consider this: if a console is dropped - and it will get dropped - > those neodymium magnets aren't going to hold, are they? that will > result in those B2B connectors being ripped off the PCB, quite likely > damaging both the LCD board *and* the main board. > > with screws, that's not going to happen, is it? > > and even if it did (if someone dropped the console *explicitly* after > taking the screws out), well.... the FPC cable might get torn out, but > it will be the weakest chain in the link and would die before taking > out the actual FPC connectors. > > even vibration or general handling is going to cause stress on those > B2B connectors unless you have some screws making absolutely, > absolutely sure that the two boards are not going to move. at all. > > mechanically, the combination of a tiny B2B connector and neodymium > magnets seems like asking for trouble. Maybe you're right and the best solution is to use FFC. Perhaps the best solution is to use neodymium magnets and screws. So any user can be safe with its screen. An advanced user who often changes the screen can always remove the screws and only use neodymium magnets. > > l. > From lkcl at lkcl.net Thu Sep 17 16:21:05 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 16:21:05 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: On Thu, Sep 17, 2015 at 4:12 PM, GaCuest wrote: > Perhaps the best solution is to use neodymium magnets and screws. So any user can be safe with its screen. An advanced user who often changes the screen can always remove the screws and only use neodymium magnets. i think you'll find that even an advanced user may consider that they could conceivably drop the unit. if there was some sort of plastic "lock" so that it was impossible for the screen module to drop out without extreme force, then you would be fine. i think you should do some calculations to see what kind of strength magnet would be needed to provide "anti-shock" level of attraction, based on the weight of the components. i see "150 G" shock-force on hard drives, normally, so that i assume would be a good number to start from. so now you would need to check the weight of the LCD plus its case and PCB, then see what size neodymium magnet *at each corner* would be sufficient to provide 150 gravities (10Nm) of attractive force. that's 1500 Nm. i think you'll find that the magnets needed are far bigger than you imagine might be needed. l. From lkcl at lkcl.net Thu Sep 17 18:14:40 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Thu, 17 Sep 2015 18:14:40 +0100 Subject: [Arm-netbook] libre laptop 15in PCB 1 and 2 arrive Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/ okaaay! so the main and the EC pcbs arrived from eurocircuits, they're all silver because they're prototype-ready, covered entirely in solder, no markings or masks at all. i experimented with the USB connectors, and the results can be seen above. heating up the boards took longer than i expected (even with an 800 watt ceramic bed and a 200 watt overhead lamp) but it got there in the end, the solder melted happily on the pins of the USB connector and we're good. i can do this :) l. From gacuest at gmail.com Thu Sep 17 18:59:58 2015 From: gacuest at gmail.com (GaCuest) Date: Thu, 17 Sep 2015 19:59:58 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: En 17 de septiembre de 2015 en 17:21:12, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Thu, Sep 17, 2015 at 4:12 PM, GaCuest wrote: > > > Perhaps the best solution is to use neodymium magnets and screws. So any user can be safe > with its screen. An advanced user who often changes the screen can always remove the screws > and only use neodymium magnets. > > i think you'll find that even an advanced user may consider that they > could conceivably drop the unit. if there was some sort of plastic > "lock" so that it was impossible for the screen module to drop out > without extreme force, then you would be fine. > > i think you should do some calculations to see what kind of strength > magnet would be needed to provide "anti-shock" level of attraction, > based on the weight of the components. i see "150 G" shock-force on > hard drives, normally, so that i assume would be a good number to > start from. > > so now you would need to check the weight of the LCD plus its case > and PCB, then see what size neodymium magnet *at each corner* would be > sufficient to provide 150 gravities (10Nm) of attractive force. > that's 1500 Nm. i think you'll find that the magnets needed are far > bigger than you imagine might be needed. >  The heaviest screen we would use weighs 66 grams. > l. > From maillist_arm-netbook at aross.me Fri Sep 18 04:22:50 2015 From: maillist_arm-netbook at aross.me (Alexander Stephen Thomas Ross) Date: Fri, 18 Sep 2015 04:22:50 +0100 Subject: [Arm-netbook] Documentary: Consumed - Is Our Consumer Culture Leading to Disaster? Message-ID: <55FB838A.2060809@aross.me> https://www.youtube.com/watch?v=bOKl04TWVsU talks mainly about the people/human side of being consumed. From lkcl at lkcl.net Fri Sep 18 13:11:34 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Fri, 18 Sep 2015 13:11:34 +0100 Subject: [Arm-netbook] Documentary: Consumed - Is Our Consumer Culture Leading to Disaster? In-Reply-To: <55FB838A.2060809@aross.me> References: <55FB838A.2060809@aross.me> Message-ID: On Fri, Sep 18, 2015 at 4:22 AM, Alexander Stephen Thomas Ross wrote: > https://www.youtube.com/watch?v=bOKl04TWVsU > > talks mainly about the people/human side of being consumed. yehhh it's good to have the problem defined... personally i prefer it when there's advice offered at the end. no need to shock people who aren't really in a position to adapt to what they're being presented with, which is why i like the "thrive movement" video. they give you a brief foray into some considerable research which must have been done over several years, then, obviously, they'd sat down and gone "hell, what can we offer that would get people _out_ of this mess??". that's what i'm doing with eoma - offering a solution that gets people "out of the mess". merely pointing out the problem doesn't help, because all that happens is... people carry on living the exact same mess... just with even more stress because at the back of their mind they now *know* that what they're doing isn't going to end well.... l. From gacuest at gmail.com Sun Sep 20 23:53:45 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 00:53:45 +0200 Subject: [Arm-netbook] Handheld Games Console Message-ID: Hello everyone. Now I am going to talk about handheld games console based on EOMA-68. As I mentioned in other posts, we have finally changed some things. For example, we have made the interchangeable screen. It will be available 3 models (without display, cheaper; with 480p display; and with 720p display, more expensive). Furthermore, more advanced users can change the display. Moreover, we will design the console in 3 PCB. On one hand, the central PCB (first PCB) where all components except the controls are, and the second and third PCB where the controls are. This allows us to design the 1st PCB first, and then design the case around the 1st PCB and decide where to place the controls. After, we design the second and third PCB. You can see this concept here: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/Handheld%20Games%20Console.png Now I put the block diagram of all PCB. Block diagram of the first PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/1st%20PCB/1%20PCB%20Block%20Diagram%20Console.png Datasheet of components of first PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/1st%20PCB/ Block diagram of the second PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/2nd%20PCB/2%20PCB%20Block%20Diagram%20Console.png Datasheet of components of second PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/2nd%20PCB/ Block diagram of the third PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/3rd%20PCB/3%20PCB%20Block%20Diagram%20Console.png Datasheet of components of third PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/3rd%20PCB/ Block diagram of the 480p display PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/PCB%20Display%20480p/PCB%20480p%20Block%20Diagram%20Console.png Datasheet of components of 480p display PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/PCB%20Display%20480p/ Block diagram of the 720p display PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/PCB%20Dispaly%20720p/PCB%20720p%20Block%20Diagram%20Console.png Datasheet of components of 720p display PCB: http://george.the-petries.co.uk/shared-write-access/eoma/HandheldGamesConsole/PCB%20Dispaly%20720p/ Any help/change/suggestion is welcome. From lkcl at lkcl.net Mon Sep 21 01:16:18 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 01:16:18 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Sun, Sep 20, 2015 at 11:53 PM, GaCuest wrote: > Hello everyone. > > Now I am going to talk about handheld games console based on EOMA-68. awesome, miguel - i'll chip in tomorrow, it's late here: if anyone would like to put these below on the wiki page http://rhombus-tech.net/community_ideas/games_console/news/ that would be great, i will do it tomorrow if someone else hasn't. l. From joem at martindale-electric.co.uk Mon Sep 21 08:01:08 2015 From: joem at martindale-electric.co.uk (joem) Date: Mon, 21 Sep 2015 07:01:08 +0000 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server Message-ID: <1442818932.21135.9.camel@jm-desktop> Open sourced A13 Kicad design mail server https://www.own-mailbox.com/libre.html Found them on kickstarter: https://www.kickstarter.com/projects/1547898916/own-mailbox-the-first-100-confidential-mailbox Seems these guys have done some real hard work. As well release their design on KiCAD, they also made a distro that runs on A13 which has a modified mail server that accepts mail replies and posts it at secure https links to standard email addresses which don't offer security built in. The KiCAD designs are missing some libs due to the way the paths have been hard coded into the project files. I'm sure they will get around to repairing it. Its a more powerful design than a raspberry pi, but many times smaller. From joem at martindale-electric.co.uk Mon Sep 21 08:07:56 2015 From: joem at martindale-electric.co.uk (joem) Date: Mon, 21 Sep 2015 07:07:56 +0000 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: Message-ID: <1442819346.21135.14.camel@jm-desktop> On Thu, 2015-09-17 at 17:12 +0200, GaCuest wrote: > Perhaps the best solution is to use neodymium magnets and screws. You know I noticed that 16 gbit fiber optic transcievers are being sold for $15 in Aliexpress. Time right me thinks for someone to 3D print a 2mm plastic fiber connector and use whatever chippery inside these things to make fiber interconnect for detachable screens. At short distances, these fiber connectors should not pose a communications problem. And 2mm plastic fiber can be cut with a stanley blade without losing connection quality over such short links. From lkcl at lkcl.net Mon Sep 21 10:42:26 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 10:42:26 +0100 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server In-Reply-To: <1442818932.21135.9.camel@jm-desktop> References: <1442818932.21135.9.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 8:01 AM, joem wrote: > Open sourced A13 Kicad design mail server > > https://www.own-mailbox.com/libre.html yeh i think paul mentioned it last week. the A13's easy to do a proper libre design because it's QFP. well done them managing DDR RAM layouts with KiCAD though! dang that must have been heavy going... but they did it. hey you wanna know something ironic: the latest revision to eoma68 would allow the use of the allwinner A13 at a reasonable cost... because no more ethernet phy on-board... l. From lkcl at lkcl.net Mon Sep 21 10:48:08 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 10:48:08 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: <1442819346.21135.14.camel@jm-desktop> References: <1442819346.21135.14.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 8:07 AM, joem wrote: > On Thu, 2015-09-17 at 17:12 +0200, GaCuest wrote: > >> Perhaps the best solution is to use neodymium magnets and screws. > > You know I noticed that 16 gbit fiber optic transcievers > are being sold for $15 in Aliexpress. coooool. miguel, the main reason why i thought that joypad controls would be ok to modularise (as demonstrated successfully by that x-box controller) would be because the number of signal wires is significantly less as well as being a greatly reduced data rate. soo.... rather large (multi-millimetre) spring-loaded contacts (like in SIM cards), absolutely no problem by contrast for the lcd you're dealing with an extremely high number of parallel signals at a MHz-level transmission rate in a very compact amount of space. so you definitely can't use huge spring-loaded contacts because they will radiate EMF like buggery. *but*... if you can however as joe suggests decouple things, remove some of the extreme requirements... l. From gacuest at gmail.com Mon Sep 21 11:17:29 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 12:17:29 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: En 21 de septiembre de 2015 en 11:48:23, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 21, 2015 at 8:07 AM, joem wrote: > > On Thu, 2015-09-17 at 17:12 +0200, GaCuest wrote: > > > >> Perhaps the best solution is to use neodymium magnets and screws. > > > > You know I noticed that 16 gbit fiber optic transcievers > > are being sold for $15 in Aliexpress. > > coooool. > > miguel, the main reason why i thought that joypad controls would be > ok to modularise (as demonstrated successfully by that x-box > controller) would be because the number of signal wires is > significantly less as well as being a greatly reduced data rate. > soo.... rather large (multi-millimetre) spring-loaded contacts (like > in SIM cards), absolutely no problem > > by contrast for the lcd you're dealing with an extremely high number > of parallel signals at a MHz-level transmission rate in a very compact > amount of space. so you definitely can't use huge spring-loaded > contacts because they will radiate EMF like buggery. *but*... if you > can however as joe suggests decouple things, remove some of the > extreme requirements... >  We do not want to make a product difficult to design and expensive. Perhaps the idea of selling the console in 3 models and advanced users can change the screen being the best. Something similar to what happens on a PC. > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From joem at martindale-electric.co.uk Mon Sep 21 11:47:04 2015 From: joem at martindale-electric.co.uk (joem) Date: Mon, 21 Sep 2015 10:47:04 +0000 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server In-Reply-To: References: <1442818932.21135.9.camel@jm-desktop> Message-ID: <1442832486.22711.8.camel@jm-desktop> On Mon, 2015-09-21 at 10:42 +0100, Luke Kenneth Casson Leighton wrote: > On Mon, Sep 21, 2015 at 8:01 AM, joem wrote: > > Open sourced A13 Kicad design mail server > > > > https://www.own-mailbox.com/libre.html > > yeh i think paul mentioned it last week. the A13's easy to do a > proper libre design because it's QFP. well done them managing DDR RAM > layouts with KiCAD though! dang that must have been heavy going... > but they did it. Thanks to CERN, KiCAD now supports real time length tuning. I tried it - put KiCAD into openGL mode and then its absolute piece of cake!! (Youtube videos out there.) > hey you wanna know something ironic: the latest revision to eoma68 > would allow the use of the allwinner A13 at a reasonable cost... > because no more ethernet phy on-board... Absolute shame the A13 doesn't have sata (which i need for server use). A lot more projects could have happened without bga but with sata. From lkcl at lkcl.net Mon Sep 21 11:59:16 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 11:59:16 +0100 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server In-Reply-To: <1442832486.22711.8.camel@jm-desktop> References: <1442818932.21135.9.camel@jm-desktop> <1442832486.22711.8.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 11:47 AM, joem wrote: > Thanks to CERN, KiCAD now supports real time length tuning. > I tried it - put KiCAD into openGL mode and then > its absolute piece of cake!! (Youtube videos out there.) ahh goooood. > >> hey you wanna know something ironic: the latest revision to eoma68 >> would allow the use of the allwinner A13 at a reasonable cost... >> because no more ethernet phy on-board... > > Absolute shame the A13 doesn't have sata (which i need for server use). > A lot more projects could have happened without bga but with sata. well there is an allwinner product which has sata, it's called the A10 and the A20 :) but yes, i knowwww.... QFP not BGA... btw you know that what's inside the A13 is exactly the same as the A10, right? just a different package, less pins brought out.... l. From lkcl at lkcl.net Mon Sep 21 12:00:29 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 12:00:29 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 11:17 AM, GaCuest wrote: > We do not want to make a product difficult to design and expensive. Perhaps the idea of selling the console in 3 models and advanced users can change the screen being the best. Something similar to what happens on a PC. yeah i feel that's the most sensible and safest model to follow. l. From joem at martindale-electric.co.uk Mon Sep 21 12:24:46 2015 From: joem at martindale-electric.co.uk (joem) Date: Mon, 21 Sep 2015 11:24:46 +0000 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server In-Reply-To: References: <1442818932.21135.9.camel@jm-desktop> <1442832486.22711.8.camel@jm-desktop> Message-ID: <1442834746.22862.3.camel@jm-desktop> On Mon, 2015-09-21 at 11:59 +0100, Luke Kenneth Casson Leighton wrote: > > Absolute shame the A13 doesn't have sata (which i need for server use). > > A lot more projects could have happened without bga but with sata. > > well there is an allwinner product which has sata, it's called the > A10 and the A20 :) but yes, i knowwww.... QFP not BGA... > > btw you know that what's inside the A13 is exactly the same as the > A10, right? just a different package, less pins brought out.... Yes.. hmm.. wonder if one of us can ask Allwinner to bring out a 200+ pin LQFP package (which is out now) with an A20 inside and sata pins (and enough pins for 2GB RAM)? Can make proper servers with that. From lasich at gmail.com Mon Sep 21 12:27:08 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Mon, 21 Sep 2015 13:27:08 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: Maybe try to focus on one model in beginning, to make yourself easier job going to market than proceed with three models. I don't know much about your target audience and why you want three models (only I like your idea*,* but my vision here is extremely limited) but believe me you will have more then enough challenges on one model and by the time you have feedback from customers you will gain some traction and have more than one good idea how to improve product. Even if you have resources to make three models, maybe you can direct this resources to make prepare some good game, to make product more likable, playable etc something that could help your project in different way...Just an thought... On Mon, Sep 21, 2015 at 1:00 PM, Luke Kenneth Casson Leighton wrote: > On Mon, Sep 21, 2015 at 11:17 AM, GaCuest wrote: > > > We do not want to make a product difficult to design and expensive. > Perhaps the idea of selling the console in 3 models and advanced users can > change the screen being the best. Something similar to what happens on a PC. > > yeah i feel that's the most sensible and safest model to follow. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gacuest at gmail.com Mon Sep 21 12:49:54 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 13:49:54 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: En 21 de septiembre de 2015 en 13:27:19, Hrvoje Lasic (lasich at gmail.com) escrito: > Maybe try to focus on one model in beginning, to make yourself easier job > going to market than proceed with three models. I don't know much about > your target audience and why you want three models (only I like your idea*,* > but my vision here is extremely limited) but believe me you will have more > then enough challenges on one model and by the time you have feedback from > customers you will gain some traction and have more than one good idea how > to improve product. Even if you have resources to make three models, maybe > you can direct this resources to make prepare some good game, to make > product more likable, playable etc something that could help your project > in different way...Just an thought… Well, we will not create three different models. Simply we create a base model (without display) and based on it you can add modules with different displays. The problem is that not everyone will know to put the modules, so we will sell configurations with mounted display modules. It is similar to BananaPi with different display models available. But with a case around the display-CTP to screw to the case of the game handheld games console. > > On Mon, Sep 21, 2015 at 1:00 PM, Luke Kenneth Casson Leighton > > wrote: > > > On Mon, Sep 21, 2015 at 11:17 AM, GaCuest wrote: > > > > > We do not want to make a product difficult to design and expensive. > > Perhaps the idea of selling the console in 3 models and advanced users can > > change the screen being the best. Something similar to what happens on a PC. > > > > yeah i feel that's the most sensible and safest model to follow. > > > > l. > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook at files.phcomp.co.uk > > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 21 13:16:47 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 13:16:47 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 12:49 PM, GaCuest wrote: > En 21 de septiembre de 2015 en 13:27:19, Hrvoje Lasic (lasich at gmail.com) escrito: >> Maybe try to focus on one model in beginning, to make yourself easier job >> going to market than proceed with three models. I don't know much about >> your target audience and why you want three models (only I like your idea*,* >> but my vision here is extremely limited) but believe me you will have more >> then enough challenges on one model and by the time you have feedback from >> customers you will gain some traction and have more than one good idea how >> to improve product. Even if you have resources to make three models, maybe >> you can direct this resources to make prepare some good game, to make >> product more likable, playable etc something that could help your project >> in different way...Just an thought… > > Well, we will not create three different models. > > Simply we create a base model (without display) and based on it you can add modules with different displays. The problem is that not everyone will know to put the modules, so we will sell configurations with mounted display modules. > > It is similar to BananaPi with different display models available. But with a case around the display-CTP to screw to the case of the game handheld games console. i just thought of something which you *really* really should consider *well* before doing this type of modular design for LCDs. the cost of even a 4-layer prototype board vs a 2-layer prototype board is *literally* a 5 to 10x jump in price. $40 for 2 boards 1.5mm thick can easily become $200 to $400 for 2 4-layer boards. as you have *three* such boards... no, five because you have main PCB, power PCB, and 3 LCD modules, consider trying to make the boards 2 layer ***BEFORE*** committing any other resources to the modular concept. as you have very short signals you should be ok but for goodness sake make absolutely sure that you can route all the differential pairs (MIPI, LVDS, whatever) on a single layer, directly from the connector, directly to the IC, directly to the LCD connector. make absolutely sure that you have room to surround and separate the differential pairs with GND-tied vias. take a look at this for example: http://hands.com/~lkcl/eoma/laptop_15in/laptop_2layer_pcb1.png in the bottom right, you can see the LVDS connector, and above it the SN75LVDS83b. there appears to be some unspoken communication between the designers of all the parts so that the LVDS differential pairs line up *DIRECTLY*. i have places GND vias in between (as best i can), i am not happy that the LVDS signals have to divert around them, but the cost of boards if the vias are smaller goes up from $40 for 2 to over $150 for 2. then, look in the left corner. that's a GL850G 4-port USB Hub IC. those "GND" vias in between and surrounding the tracks go *all the way* up the board, it took several days to arrange because i had to adjust it and correct this arrangement several times. it's by no means perfect, but it is what it is. you can also see top middle, 7 wires in between GND tracks, those are the SD/MMC lines, they're getting quite fast these days so i thought it best to put GND guides around them. strictly speaking from what i gather you are supposed to actually isolate the GND tracks at the outer edges, not let them become part of a larger GND plane. also you are supposed to leave a large distance to any other components. also you are supposed to leave a large distance in between each differential pair. l. From lkcl at lkcl.net Mon Sep 21 13:50:37 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 13:50:37 +0100 Subject: [Arm-netbook] Open sourced A13 Kicad design mail server In-Reply-To: <1442834746.22862.3.camel@jm-desktop> References: <1442818932.21135.9.camel@jm-desktop> <1442832486.22711.8.camel@jm-desktop> <1442834746.22862.3.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 12:24 PM, joem wrote: > Yes.. hmm.. wonder if one of us can ask Allwinner to bring out a 200+ > pin LQFP package (which is out now) with an A20 inside and sata pins > (and enough pins for 2GB RAM)? Can make proper servers with that. that'd be cool. except... you know they restricted the data lanes to only 16-bit on the A13, to be able to reduce the pin-count, right? and there are only 14 address lines on the A20, so it might not actually be feasible... don't know. From gacuest at gmail.com Mon Sep 21 13:56:54 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 14:56:54 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: En 21 de septiembre de 2015 en 14:17:10, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 21, 2015 at 12:49 PM, GaCuest wrote: > > En 21 de septiembre de 2015 en 13:27:19, Hrvoje Lasic (lasich at gmail.com) escrito: > >> Maybe try to focus on one model in beginning, to make yourself easier job > >> going to market than proceed with three models. I don't know much about > >> your target audience and why you want three models (only I like your idea*,* > >> but my vision here is extremely limited) but believe me you will have more > >> then enough challenges on one model and by the time you have feedback from > >> customers you will gain some traction and have more than one good idea how > >> to improve product. Even if you have resources to make three models, maybe > >> you can direct this resources to make prepare some good game, to make > >> product more likable, playable etc something that could help your project > >> in different way...Just an thought… > > > > Well, we will not create three different models. > > > > Simply we create a base model (without display) and based on it you can add modules with > different displays. The problem is that not everyone will know to put the modules, so > we will sell configurations with mounted display modules. > > > > It is similar to BananaPi with different display models available. But with a case around > the display-CTP to screw to the case of the game handheld games console. > > i just thought of something which you *really* really should > consider *well* before doing this type of modular design for LCDs. > > the cost of even a 4-layer prototype board vs a 2-layer prototype > board is *literally* a 5 to 10x jump in price. $40 for 2 boards 1.5mm > thick can easily become $200 to $400 for 2 4-layer boards. > > as you have *three* such boards... no, five because you have main > PCB, power PCB, and 3 LCD modules, consider trying to make the boards > 2 layer ***BEFORE*** committing any other resources to the modular > concept. >  We have 3 PCB (first for all components and second/third for controls). Then we have two PCB for 480p display and 720p display (non-display model does not have that PCB). > as you have very short signals you should be ok but for goodness sake > make absolutely sure that you can route all the differential pairs > (MIPI, LVDS, whatever) on a single layer, directly from the connector, > directly to the IC, directly to the LCD connector. make absolutely > sure that you have room to surround and separate the differential > pairs with GND-tied vias. > > take a look at this for example: > > http://hands.com/~lkcl/eoma/laptop_15in/laptop_2layer_pcb1.png > > in the bottom right, you can see the LVDS connector, and above it the > SN75LVDS83b. there appears to be some unspoken communication between > the designers of all the parts so that the LVDS differential pairs > line up *DIRECTLY*. i have places GND vias in between (as best i > can), i am not happy that the LVDS signals have to divert around them, > but the cost of boards if the vias are smaller goes up from $40 for 2 > to over $150 for 2. > > then, look in the left corner. that's a GL850G 4-port USB Hub IC. > those "GND" vias in between and surrounding the tracks go *all the > way* up the board, it took several days to arrange because i had to > adjust it and correct this arrangement several times. it's by no > means perfect, but it is what it is. > > you can also see top middle, 7 wires in between GND tracks, those are > the SD/MMC lines, they're getting quite fast these days so i thought > it best to put GND guides around them. > > strictly speaking from what i gather you are supposed to actually > isolate the GND tracks at the outer edges, not let them become part of > a larger GND plane. also you are supposed to leave a large distance > to any other components. also you are supposed to leave a large > distance in between each differential pair. >  Our idea is to do it in 2 layers. The only problem is that the PCMCIA connector is on the bottom layer and the FFC connector for the display goes on top layer. Anyway, put the display in another PCB and attach using FFC makes the design easier because we can put the FFC connector where we want. Thanks for all the tips. I can not use the tips because I do not know about PCB designs but I hope the tips will be useful to Daniel. > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 21 14:57:40 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 14:57:40 +0100 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: On Mon, Sep 21, 2015 at 1:56 PM, GaCuest wrote: > Our idea is to do it in 2 layers. > > The only problem is that the PCMCIA connector is on the bottom layer and the FFC connector for the display goes on top layer. you shouuuld be able to get away with 1 via per connection but for goodness sake make sure there are GND vias very close by. l. From gacuest at gmail.com Mon Sep 21 15:32:23 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 16:32:23 +0200 Subject: [Arm-netbook] Interchangeable screen In-Reply-To: References: <1442819346.21135.14.camel@jm-desktop> Message-ID: En 21 de septiembre de 2015 en 15:58:02, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 21, 2015 at 1:56 PM, GaCuest wrote: > > > Our idea is to do it in 2 layers. > > > > The only problem is that the PCMCIA connector is on the bottom layer and the FFC connector > for the display goes on top layer. > > you shouuuld be able to get away with 1 via per connection but for > goodness sake make sure there are GND vias very close by. >  We can put the connector where we want, so there would be no problem. I think Daniel read this mailing list, so I assume he will see your tips. Thanks! > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From gacuest at gmail.com Mon Sep 21 16:56:02 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 17:56:02 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ I hope that I have done it well. Thanks! En 21 de septiembre de 2015 en 2:16:39, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Sun, Sep 20, 2015 at 11:53 PM, GaCuest wrote: > > Hello everyone. > > > > Now I am going to talk about handheld games console based on EOMA-68. > > awesome, miguel - i'll chip in tomorrow, it's late here: if anyone > would like to put these below on the wiki page > http://rhombus-tech.net/community_ideas/games_console/news/ that would > be great, i will do it tomorrow if someone else hasn't. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 21 17:16:58 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 17:16:58 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 21, 2015 at 4:56 PM, GaCuest wrote: > I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ > > I hope that I have done it well. Thanks! yeah that looks fantastic. hmmmm, i have an idea for you. how about moving the EC onto the buttons PCB? that way you only need an absolute minimum of connections to it. USB, power, GND... maybe BOOT and RESET errr... that's all you'd really need. that would be 6 wires, and i'm sure you could find some suitable spring-loaded connector for that. maybe even a SIM card holder (one without the "holder" bit - just the bare spring-loaded bits. as you'd be talking USB 1.0 or USB 1.1 that would hardly tickle any EMF. l. From gacuest at gmail.com Mon Sep 21 18:03:07 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 19:03:07 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 21 de septiembre de 2015 en 18:17:15, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 21, 2015 at 4:56 PM, GaCuest wrote: > > I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ > > > > I hope that I have done it well. Thanks! > > yeah that looks fantastic. > > hmmmm, i have an idea for you. how about moving the EC onto the > buttons PCB? that way you only need an absolute minimum of > connections to it. USB, power, GND... maybe BOOT and RESET errr... > that's all you'd really need. > > that would be 6 wires, and i'm sure you could find some suitable > spring-loaded connector for that. maybe even a SIM card holder (one > without the "holder" bit - just the bare spring-loaded bits. as you'd > be talking USB 1.0 or USB 1.1 that would hardly tickle any EMF. >  If we put the STM32F in the second (or third) PCB, we need to connect the second with the third PCB (the same problem). Other problem is that we lost the USB ST-Link connection for FFC developers conector. What do you think? > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 21 20:49:59 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 21 Sep 2015 20:49:59 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 21, 2015 at 6:03 PM, GaCuest wrote: > En 21 de septiembre de 2015 en 18:17:15, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: >> On Mon, Sep 21, 2015 at 4:56 PM, GaCuest wrote: >> > I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ >> > >> > I hope that I have done it well. Thanks! >> >> yeah that looks fantastic. >> >> hmmmm, i have an idea for you. how about moving the EC onto the >> buttons PCB? that way you only need an absolute minimum of >> connections to it. USB, power, GND... maybe BOOT and RESET errr... >> that's all you'd really need. >> >> that would be 6 wires, and i'm sure you could find some suitable >> spring-loaded connector for that. maybe even a SIM card holder (one >> without the "holder" bit - just the bare spring-loaded bits. as you'd >> be talking USB 1.0 or USB 1.1 that would hardly tickle any EMF. >> > > If we put the STM32F in the second (or third) PCB, we need to connect the second with the third PCB (the same problem). ? why? rright right yes, got it. ok, so yes, put STM32F on 2nd (or 3rd) PCB, link 2nd to 3rd with FFC-20P, then you only have one FFC-20P, not two, the other is replaced with a simple 4 pin (or possibly 6 pin) cable / blah blah whatever. > Other problem is that we lost the USB ST-Link connection for FFC developers conector. > > What do you think? ideas occur to me... (1) if someone wants to develop an alternative embedded controller PCB, if there's already one on the main PCB they are discouraged from adding another. it occupies the USB port... you can't really take that over unless you want to replace PCB1. (2) if someone wants to develop an alternative joypad controller PCB... say... out of discrete ICs, they are again discouraged from doing so because there's an STM32F on the main PCB, esp. as it occupies an entire USB port. (3) if the STM32F is on the joypad controller PCB, and it's pretty much "USB", you could actually consider making a *separate* joypad *product* out of it, sold for other markets.... *including* offering one for sale as a 2nd controller to be wired directly into the USB port of the games console. (4) i don't see any problem with doing product development (of the joypad controller) by taking PCB2 and PCB3, testing them *entirely separately* as a stand-alone unit. (5) i don't see any problem with doing product development (of the games console) by taking out PCB2 and PCB3, and connecting "alternative products" directly into the USB port provided for (normally) joypad control. l. From gacuest at gmail.com Mon Sep 21 22:34:13 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 21 Sep 2015 23:34:13 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 21 de septiembre de 2015 en 21:50:23, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 21, 2015 at 6:03 PM, GaCuest wrote: > > En 21 de septiembre de 2015 en 18:17:15, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > escrito: > >> On Mon, Sep 21, 2015 at 4:56 PM, GaCuest wrote: > >> > I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ > >> > > >> > I hope that I have done it well. Thanks! > >> > >> yeah that looks fantastic. > >> > >> hmmmm, i have an idea for you. how about moving the EC onto the > >> buttons PCB? that way you only need an absolute minimum of > >> connections to it. USB, power, GND... maybe BOOT and RESET errr... > >> that's all you'd really need. > >> > >> that would be 6 wires, and i'm sure you could find some suitable > >> spring-loaded connector for that. maybe even a SIM card holder (one > >> without the "holder" bit - just the bare spring-loaded bits. as you'd > >> be talking USB 1.0 or USB 1.1 that would hardly tickle any EMF. > >> > > > > If we put the STM32F in the second (or third) PCB, we need to connect the second with the > third PCB (the same problem). > > ? why? rright right yes, got it. ok, so yes, put STM32F on 2nd (or > 3rd) PCB, link 2nd to 3rd with FFC-20P, then you only have one > FFC-20P, not two, the other is replaced with a simple 4 pin (or > possibly 6 pin) cable / blah blah whatever. > > > Other problem is that we lost the USB ST-Link connection for FFC developers conector. > > > > What do you think? > > ideas occur to me... > > (1) if someone wants to develop an alternative embedded controller > PCB, if there's already one on the main PCB they are discouraged from > adding another. it occupies the USB port... you can't really take > that over unless you want to replace PCB1. > > (2) if someone wants to develop an alternative joypad controller > PCB... say... out of discrete ICs, they are again discouraged from > doing so because there's an STM32F on the main PCB, esp. as it > occupies an entire USB port. > > (3) if the STM32F is on the joypad controller PCB, and it's pretty > much "USB", you could actually consider making a *separate* joypad > *product* out of it, sold for other markets.... *including* offering > one for sale as a 2nd controller to be wired directly into the USB > port of the games console. > > (4) i don't see any problem with doing product development (of the > joypad controller) by taking PCB2 and PCB3, testing them *entirely > separately* as a stand-alone unit. > > (5) i don't see any problem with doing product development (of the > games console) by taking out PCB2 and PCB3, and connecting > "alternative products" directly into the USB port provided for > (normally) joypad control. Yes, surely you are right and it is better that way. But we have a problem with EINTs. We have at least 5 IC they need EINT: * Touch panel   * AXP209   * Headphone detect   * Accelerometer   * MicroSD   However EOMA-68 only has 4 EINTs. The only solution I can think is to put the BMA250 in the second PCB and connect directly to STM32F and configure it as a USB device. I do not know if this is easy to do for us (software development) and solve the problem of EINTs. Thanks. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From gacuest at gmail.com Tue Sep 22 13:12:11 2015 From: gacuest at gmail.com (GaCuest) Date: Tue, 22 Sep 2015 14:12:11 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 21 de septiembre de 2015 en 23:34:15, GaCuest (gacuest at gmail.com) escrito: > En 21 de septiembre de 2015 en 21:50:23, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > escrito: > > On Mon, Sep 21, 2015 at 6:03 PM, GaCuest wrote: > > > En 21 de septiembre de 2015 en 18:17:15, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > > escrito: > > >> On Mon, Sep 21, 2015 at 4:56 PM, GaCuest wrote: > > >> > I have updated it: http://rhombus-tech.net/community_ideas/games_console/news/ > > >> > > > >> > I hope that I have done it well. Thanks! > > >> > > >> yeah that looks fantastic. > > >> > > >> hmmmm, i have an idea for you. how about moving the EC onto the > > >> buttons PCB? that way you only need an absolute minimum of > > >> connections to it. USB, power, GND... maybe BOOT and RESET errr... > > >> that's all you'd really need. > > >> > > >> that would be 6 wires, and i'm sure you could find some suitable > > >> spring-loaded connector for that. maybe even a SIM card holder (one > > >> without the "holder" bit - just the bare spring-loaded bits. as you'd > > >> be talking USB 1.0 or USB 1.1 that would hardly tickle any EMF. > > >> > > > > > > If we put the STM32F in the second (or third) PCB, we need to connect the second with the > > third PCB (the same problem). > > > > ? why? rright right yes, got it. ok, so yes, put STM32F on 2nd (or > > 3rd) PCB, link 2nd to 3rd with FFC-20P, then you only have one > > FFC-20P, not two, the other is replaced with a simple 4 pin (or > > possibly 6 pin) cable / blah blah whatever. > > > > > Other problem is that we lost the USB ST-Link connection for FFC developers conector. > > > > > > What do you think? > > > > ideas occur to me... > > > > (1) if someone wants to develop an alternative embedded controller > > PCB, if there's already one on the main PCB they are discouraged from > > adding another. it occupies the USB port... you can't really take > > that over unless you want to replace PCB1. > > > > (2) if someone wants to develop an alternative joypad controller > > PCB... say... out of discrete ICs, they are again discouraged from > > doing so because there's an STM32F on the main PCB, esp. as it > > occupies an entire USB port. > > > > (3) if the STM32F is on the joypad controller PCB, and it's pretty > > much "USB", you could actually consider making a *separate* joypad > > *product* out of it, sold for other markets.... *including* offering > > one for sale as a 2nd controller to be wired directly into the USB > > port of the games console. > > > > (4) i don't see any problem with doing product development (of the > > joypad controller) by taking PCB2 and PCB3, testing them *entirely > > separately* as a stand-alone unit. > > > > (5) i don't see any problem with doing product development (of the > > games console) by taking out PCB2 and PCB3, and connecting > > "alternative products" directly into the USB port provided for > > (normally) joypad control. > > Yes, surely you are right and it is better that way. > > But we have a problem with EINTs. > > We have at least 5 IC they need EINT: > * Touch panel > * AXP209 > * Headphone detect > * Accelerometer > * MicroSD > > However EOMA-68 only has 4 EINTs. > > The only solution I can think is to put the BMA250 in the second PCB and connect directly > to STM32F and configure it as a USB device. I do not know if this is easy to do for us (software > development) and solve the problem of EINTs. > > Thanks. Also I found the following FFC cable: http://aliexpress.com/store/product/Free-Shipping-FPC-FFC-connector-cable-socket-6-pin-0-5mm-connector-for-LCD-screen-interface/428886_1251083253.html Do you recommend use that connector or other (cheaper) connector? Thanks. > > > > > l. > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook at files.phcomp.co.uk > > From lkcl at lkcl.net Tue Sep 22 14:14:58 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 22 Sep 2015 14:14:58 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Tue, Sep 22, 2015 at 1:12 PM, GaCuest wrote: >> The only solution I can think is to put the BMA250 in the second PCB and connect directly >> to STM32F and configure it as a USB device. I do not know if this is easy to do for us (software >> development) and solve the problem of EINTs. making a GPIO an EINT on the STM32F, then reading I2C, then sending a few bytes of data as a USB event, it's about... 30-40 lines of code. if it's any more than that, you're doing something *drastically* wrong. >> >> Thanks. > > Also I found the following FFC cable: http://aliexpress.com/store/product/Free-Shipping-FPC-FFC-connector-cable-socket-6-pin-0-5mm-connector-for-LCD-screen-interface/428886_1251083253.html > > Do you recommend use that connector or other (cheaper) connector? honestly it's up to you. personally i just went for standard 0.254mm connectors, because that *really* makes it easy for developers. however, they're quite big, you need to get R/A ones, and i believe those could well be more expensive than a 6pin FFC. so it is your decision, you know what the goal is, available space, target price and target market. l. > Thanks. > >> >> > >> > l. >> > >> > _______________________________________________ >> > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk >> > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook >> > Send large attachments to arm-netbook at files.phcomp.co.uk >> >> > > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From gacuest at gmail.com Tue Sep 22 16:31:13 2015 From: gacuest at gmail.com (GaCuest) Date: Tue, 22 Sep 2015 17:31:13 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 22 de septiembre de 2015 en 15:15:51, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Tue, Sep 22, 2015 at 1:12 PM, GaCuest wrote: > > >> The only solution I can think is to put the BMA250 in the second PCB and connect directly > >> to STM32F and configure it as a USB device. I do not know if this is easy to do for us (software > >> development) and solve the problem of EINTs. > > making a GPIO an EINT on the STM32F, then reading I2C, then sending a > few bytes of data as a USB event, it's about... 30-40 lines of code. > if it's any more than that, you're doing something *drastically* > wrong. >  So I suppose that the better option is to put the BMA250 on the second PCB and connect it to STM32F, right? On the other hand, we don’t need a EINT of the EOMA-68 to the STM32F, right? > >> > >> Thanks. > > > > Also I found the following FFC cable: http://aliexpress.com/store/product/Free-Shipping-FPC-FFC-connector-cable-socket-6-pin-0-5mm-connector-for-LCD-screen-interface/428886_1251083253.html > > > > Do you recommend use that connector or other (cheaper) connector? > > honestly it's up to you. personally i just went for standard 0.254mm > connectors, because that *really* makes it easy for developers. > however, they're quite big, you need to get R/A ones, and i believe > those could well be more expensive than a 6pin FFC. > > so it is your decision, you know what the goal is, available space, > target price and target market. >  Well, I think that one of the most important goal is the price. > l. > > > Thanks. > > > >> > >> > > >> > l. > >> > > >> > _______________________________________________ > >> > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > >> > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > >> > Send large attachments to arm-netbook at files.phcomp.co.uk > >> > >> > > > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook at files.phcomp.co.uk > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Tue Sep 22 17:33:28 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 22 Sep 2015 17:33:28 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Tue, Sep 22, 2015 at 4:31 PM, GaCuest wrote: > En 22 de septiembre de 2015 en 15:15:51, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: >> On Tue, Sep 22, 2015 at 1:12 PM, GaCuest wrote: >> >> >> The only solution I can think is to put the BMA250 in the second PCB and connect directly >> >> to STM32F and configure it as a USB device. I do not know if this is easy to do for us (software >> >> development) and solve the problem of EINTs. >> >> making a GPIO an EINT on the STM32F, then reading I2C, then sending a >> few bytes of data as a USB event, it's about... 30-40 lines of code. >> if it's any more than that, you're doing something *drastically* >> wrong. >> > > So I suppose that the better option is to put the BMA250 on the second PCB and connect it to STM32F, right? that's what i would do, because, if i was considering making a separate controller (as a 2nd joypad) i would want to be able to "tilt" the controller... rather than just the games console. > On the other hand, we don’t need a EINT of the EOMA-68 to the STM32F, right? correct, you don't. USB events will take care of generating an interrupt, automatically. l. From gacuest at gmail.com Tue Sep 22 19:38:15 2015 From: gacuest at gmail.com (GaCuest) Date: Tue, 22 Sep 2015 20:38:15 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: Hello everyone, I updated the block diagram. I would appreciate if someone can review it. http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_22sep2015/ Thanks! From lkcl at lkcl.net Tue Sep 22 21:56:59 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 22 Sep 2015 21:56:59 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Tue, Sep 22, 2015 at 7:38 PM, GaCuest wrote: > Hello everyone, > I updated the block diagram. I would appreciate if someone can review it. > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_22sep2015/ ok so you liked the idea of being able to make a games controller from the same PCB, then? :) it would be really simple, a piece of plastic and a USB connector - done! page 27: http://hands.com/~lkcl/eoma/kde_tablet/CM108_DataSheet_v1.6.pdf i don't see a need for an EINT to the CM108AH. ah, i see what it is: the EINT goes directly to the Audio Jack. it's the physical audio socket which has a "connection" which is broken/made by the physical insertion of a jack. that, in software, you would then notice. actually what i've done in the hardware circuits i've made, is, although yes there is a connection there, i added in a resistor bridge into the circuit to bring the volume down when speakers are engaged. the CM108AH "normally" would drive the headphones (directly), but in "speaker" mode the L / R audio goes first to the headphones (where insertion of the jack would make it "cut out"), and from there (if no jack) it goes to the resistor-divider bridge to drop the voltage to 1/10th of its level, and *that* is fed into the TDA2822(s). so actually, in the circuit that i have done, strictly speaking the EINT headphone is totally irrelevant... i just added it "just in case", also to be able to do user-alerts. l. From gacuest at gmail.com Tue Sep 22 22:43:49 2015 From: gacuest at gmail.com (GaCuest) Date: Tue, 22 Sep 2015 23:43:49 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 22 de septiembre de 2015 en 22:57:20, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Tue, Sep 22, 2015 at 7:38 PM, GaCuest wrote: > > Hello everyone, > > I updated the block diagram. I would appreciate if someone can review it. > > > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_22sep2015/ > > ok so you liked the idea of being able to make a games controller > from the same PCB, then? :) it would be really simple, a piece of > plastic and a USB connector - done! >  Yes, it's a good idea. The problem is that we are not a professional team, so we can now focus only on the handheld games console. Daniel has very little free time, so for now we are progressing very slowly. Another idea would be to create a EOMA-68 with FPGA (or similar) and use it as a pass-through card. For example, you can connect a PC Gamer to HDMI and USB of this EOMA-68, and you have a PC as a handheld games console. > page 27: > http://hands.com/~lkcl/eoma/kde_tablet/CM108_DataSheet_v1.6.pdf > > i don't see a need for an EINT to the CM108AH. > > ah, i see what it is: the EINT goes directly to the Audio Jack. > it's the physical audio socket which has a "connection" which is > broken/made by the physical insertion of a jack. that, in software, > you would then notice. > > actually what i've done in the hardware circuits i've made, is, > although yes there is a connection there, i added in a resistor bridge > into the circuit to bring the volume down when speakers are engaged. > the CM108AH "normally" would drive the headphones (directly), but in > "speaker" mode the L / R audio goes first to the headphones (where > insertion of the jack would make it "cut out"), and from there (if no > jack) it goes to the resistor-divider bridge to drop the voltage to > 1/10th of its level, and *that* is fed into the TDA2822(s). > > so actually, in the circuit that i have done, strictly speaking the > EINT headphone is totally irrelevant... i just added it "just in > case", also to be able to do user-alerts. >  So we have two options, put the EINT in the headphone jack, or put (this) second EINT in the FFC 40 pin connector in case we need it in the future on future screen. What is your recommendation? Thanks. > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Tue Sep 22 23:47:58 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 22 Sep 2015 23:47:58 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Tue, Sep 22, 2015 at 10:43 PM, GaCuest wrote: > En 22 de septiembre de 2015 en 22:57:20, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: >> On Tue, Sep 22, 2015 at 7:38 PM, GaCuest wrote: >> > Hello everyone, >> > I updated the block diagram. I would appreciate if someone can review it. >> > >> > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_22sep2015/ >> >> ok so you liked the idea of being able to make a games controller >> from the same PCB, then? :) it would be really simple, a piece of >> plastic and a USB connector - done! >> > > Yes, it's a good idea. > > The problem is that we are not a professional team, so we can now focus only on the handheld games console. Daniel has very little free time, so for now we are progressing very slowly. hey it was an easy idea... for another time, yes. > Another idea would be to create a EOMA-68 with FPGA (or similar) and use it as a pass-through card. For example, you can connect a PC Gamer to HDMI and USB of this EOMA-68, and you have a PC as a handheld games console. ah yeah of course, i forgot about that :) btw could you use plain-text editing, not a rich-text format, with paragraphs broken up at (appx) 70 chars per line, please? > So we have two options, put the EINT in the headphone jack, or put (this) second EINT in the FFC 40 pin connector in case we need it in the future on future screen. What is your recommendation? i'd say keep it on the HP jack, just in case something needs to be altered (volume etc.) when switching between headphone and speaker mode. l. From gacuest at gmail.com Wed Sep 23 14:59:38 2015 From: gacuest at gmail.com (GaCuest) Date: Wed, 23 Sep 2015 15:59:38 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 23 de septiembre de 2015 en 0:48:20, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Tue, Sep 22, 2015 at 10:43 PM, GaCuest wrote: > > En 22 de septiembre de 2015 en 22:57:20, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > escrito: > >> On Tue, Sep 22, 2015 at 7:38 PM, GaCuest wrote: > >> > Hello everyone, > >> > I updated the block diagram. I would appreciate if someone can review it. > >> > > >> > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_22sep2015/ > >> > >> ok so you liked the idea of being able to make a games controller > >> from the same PCB, then? :) it would be really simple, a piece of > >> plastic and a USB connector - done! > >> > > > > Yes, it's a good idea. > > > > The problem is that we are not a professional team, so we can now focus only on the handheld > games console. Daniel has very little free time, so for now we are progressing very slowly. > > hey it was an easy idea... for another time, yes. > > > Another idea would be to create a EOMA-68 with FPGA (or similar) and use it as a pass-through > card. For example, you can connect a PC Gamer to HDMI and USB of this EOMA-68, and you have > a PC as a handheld games console. > > ah yeah of course, i forgot about that :) > > btw could you use plain-text editing, not a rich-text format, with > paragraphs broken up at (appx) 70 chars per line, please? >  I’m using AirMail. It has plain text editing, but not line-wrap. I will do that manually. > > > So we have two options, put the EINT in the headphone jack, or put (this) second EINT in > the FFC 40 pin connector in case we need it in the future on future screen. What is your recommendation? > > i'd say keep it on the HP jack, just in case something needs to be > altered (volume etc.) when switching between headphone and speaker > mode. >  I updated the block diagram: http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ Anything more to change? Thanks! > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Wed Sep 23 15:54:29 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 23 Sep 2015 15:54:29 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: > I’m using AirMail. It has plain text editing, but not line-wrap. > I will do that manually. appreciated. > I updated the block diagram: > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > > Anything more to change? yes, one thing that's important: 1st USB is mandated to be >= speed than 2nd USB. so you need to swap 1st USB with 2nd USB. reason: the STM32F is connected to the port that may end up being 480mbit/sec and the USB Hub is connected to the port that may end up being only 11mbit/sec. certainly, for the jz4775 CPU Card (which i am sending off very shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB 1.1 for the 2nd. and i am certainly not going to increase the cost of the jz4775 CPU Card by an extra $2 worth of components to put in a USB2 Hub when the SOC itself is only $2.50!!! l. From gacuest at gmail.com Wed Sep 23 16:07:44 2015 From: gacuest at gmail.com (GaCuest) Date: Wed, 23 Sep 2015 17:07:44 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: > > > I’m using AirMail. It has plain text editing, but not line-wrap. > > I will do that manually. > > appreciated. > > > I updated the block diagram: > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > > > > Anything more to change? > > yes, one thing that's important: 1st USB is mandated to be >= speed > than 2nd USB. so you need to swap 1st USB with 2nd USB. > > reason: the STM32F is connected to the port that may end up being > 480mbit/sec and the USB Hub is connected to the port that may end up > being only 11mbit/sec. > > certainly, for the jz4775 CPU Card (which i am sending off very > shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB > 1.1 for the 2nd. > > and i am certainly not going to increase the cost of the jz4775 CPU > Card by an extra $2 worth of components to put in a USB2 Hub when the > SOC itself is only $2.50!!! >  I think that you have not properly seen the block diagram. The STM32 is connected to the second USB, the USB HUB is connected to the first USB. This right? > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Wed Sep 23 16:13:19 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 23 Sep 2015 16:13:19 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Wed, Sep 23, 2015 at 4:07 PM, GaCuest wrote: > En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: >> On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: >> >> > I’m using AirMail. It has plain text editing, but not line-wrap. >> > I will do that manually. >> >> appreciated. >> >> > I updated the block diagram: >> > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ >> > >> > Anything more to change? >> >> yes, one thing that's important: 1st USB is mandated to be >= speed >> than 2nd USB. so you need to swap 1st USB with 2nd USB. >> >> reason: the STM32F is connected to the port that may end up being >> 480mbit/sec and the USB Hub is connected to the port that may end up >> being only 11mbit/sec. >> >> certainly, for the jz4775 CPU Card (which i am sending off very >> shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB >> 1.1 for the 2nd. >> >> and i am certainly not going to increase the cost of the jz4775 CPU >> Card by an extra $2 worth of components to put in a USB2 Hub when the >> SOC itself is only $2.50!!! >> > > I think that you have not properly seen the block diagram. wiggle wiggle follow the lines with my finger pointing at the screen this time... :) > The STM32 is connected to the second USB, the USB HUB is connected > to the first USB. > > This right? yes! :) From gacuest at gmail.com Wed Sep 23 16:31:58 2015 From: gacuest at gmail.com (GaCuest) Date: Wed, 23 Sep 2015 17:31:58 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 23 de septiembre de 2015 en 17:13:37, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Wed, Sep 23, 2015 at 4:07 PM, GaCuest wrote: > > En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton (lkcl at lkcl.net) > escrito: > >> On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: > >> > >> > I’m using AirMail. It has plain text editing, but not line-wrap. > >> > I will do that manually. > >> > >> appreciated. > >> > >> > I updated the block diagram: > >> > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > >> > > >> > Anything more to change? > >> > >> yes, one thing that's important: 1st USB is mandated to be >= speed > >> than 2nd USB. so you need to swap 1st USB with 2nd USB. > >> > >> reason: the STM32F is connected to the port that may end up being > >> 480mbit/sec and the USB Hub is connected to the port that may end up > >> being only 11mbit/sec. > >> > >> certainly, for the jz4775 CPU Card (which i am sending off very > >> shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB > >> 1.1 for the 2nd. > >> > >> and i am certainly not going to increase the cost of the jz4775 CPU > >> Card by an extra $2 worth of components to put in a USB2 Hub when the > >> SOC itself is only $2.50!!! > >> > > > > I think that you have not properly seen the block diagram. > > wiggle wiggle follow the lines with my finger pointing at the screen > this time... :) > > > The STM32 is connected to the second USB, the USB HUB is connected > > to the first USB. > > > > This right? > > yes! :) >  Then, anything to change? Is this right?: http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Wed Sep 23 18:44:55 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 23 Sep 2015 18:44:55 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Wed, Sep 23, 2015 at 4:31 PM, GaCuest wrote: > En 23 de septiembre de 2015 en 17:13:37, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: >> On Wed, Sep 23, 2015 at 4:07 PM, GaCuest wrote: >> > En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton (lkcl at lkcl.net) >> escrito: >> >> On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: >> >> >> >> > I’m using AirMail. It has plain text editing, but not line-wrap. >> >> > I will do that manually. >> >> >> >> appreciated. >> >> >> >> > I updated the block diagram: >> >> > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ >> >> > >> >> > Anything more to change? >> >> >> >> yes, one thing that's important: 1st USB is mandated to be >= speed >> >> than 2nd USB. so you need to swap 1st USB with 2nd USB. >> >> >> >> reason: the STM32F is connected to the port that may end up being >> >> 480mbit/sec and the USB Hub is connected to the port that may end up >> >> being only 11mbit/sec. >> >> >> >> certainly, for the jz4775 CPU Card (which i am sending off very >> >> shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB >> >> 1.1 for the 2nd. >> >> >> >> and i am certainly not going to increase the cost of the jz4775 CPU >> >> Card by an extra $2 worth of components to put in a USB2 Hub when the >> >> SOC itself is only $2.50!!! >> >> >> > >> > I think that you have not properly seen the block diagram. >> >> wiggle wiggle follow the lines with my finger pointing at the screen >> this time... :) >> >> > The STM32 is connected to the second USB, the USB HUB is connected >> > to the first USB. >> > >> > This right? >> >> yes! :) >> > > Then, anything to change? no > Is this right?: i did say yes > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > >> _______________________________________________ >> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk >> http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook >> Send large attachments to arm-netbook at files.phcomp.co.uk > > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Sun Sep 27 16:05:37 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Sun, 27 Sep 2015 16:05:37 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress Message-ID: http://rhombus-tech.net/community_ideas/laptop_15in/news/ CM108AH, and Power ICs, all functional. l. From daniel.iglesias at gmail.com Mon Sep 28 08:52:03 2015 From: daniel.iglesias at gmail.com (Daniel Iglesias) Date: Mon, 28 Sep 2015 09:52:03 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: Hello there, sorry for not writing for so long. What would you recommend as the next step now? As I told Miguel privately, I was thinking of getting an STM32 evaluation board (they're really cheap, <15€ IIRC) and building the USB controller prototype. Once we have the STM32 reading all the values from the potentiometers, pushbuttons, etc. and reporting them over USB as a HID we would go ahead and design PCBs 2 and 3. How does this sound? Should we skip the eval board thing and go straight to designing the PCBs? 2015-09-23 19:44 GMT+02:00 Luke Kenneth Casson Leighton : > On Wed, Sep 23, 2015 at 4:31 PM, GaCuest wrote: > > En 23 de septiembre de 2015 en 17:13:37, Luke Kenneth Casson Leighton ( > lkcl at lkcl.net) escrito: > >> On Wed, Sep 23, 2015 at 4:07 PM, GaCuest wrote: > >> > En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton > (lkcl at lkcl.net) > >> escrito: > >> >> On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: > >> >> > >> >> > I’m using AirMail. It has plain text editing, but not line-wrap. > >> >> > I will do that manually. > >> >> > >> >> appreciated. > >> >> > >> >> > I updated the block diagram: > >> >> > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > >> >> > > >> >> > Anything more to change? > >> >> > >> >> yes, one thing that's important: 1st USB is mandated to be >= speed > >> >> than 2nd USB. so you need to swap 1st USB with 2nd USB. > >> >> > >> >> reason: the STM32F is connected to the port that may end up being > >> >> 480mbit/sec and the USB Hub is connected to the port that may end up > >> >> being only 11mbit/sec. > >> >> > >> >> certainly, for the jz4775 CPU Card (which i am sending off very > >> >> shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB > >> >> 1.1 for the 2nd. > >> >> > >> >> and i am certainly not going to increase the cost of the jz4775 CPU > >> >> Card by an extra $2 worth of components to put in a USB2 Hub when the > >> >> SOC itself is only $2.50!!! > >> >> > >> > > >> > I think that you have not properly seen the block diagram. > >> > >> wiggle wiggle follow the lines with my finger pointing at the screen > >> this time... :) > >> > >> > The STM32 is connected to the second USB, the USB HUB is connected > >> > to the first USB. > >> > > >> > This right? > >> > >> yes! :) > >> > > > > Then, anything to change? > > no > > > Is this right?: > > i did say yes > > > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > > > >> _______________________________________________ > >> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > >> http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > >> Send large attachments to arm-netbook at files.phcomp.co.uk > > > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook at files.phcomp.co.uk > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gacuest at gmail.com Mon Sep 28 10:09:00 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 28 Sep 2015 11:09:00 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 28 de septiembre de 2015 en 9:52:43, Daniel Iglesias (daniel.iglesias at gmail.com) escrito: > Hello there, sorry for not writing for so long. What would you recommend as > the next step now? As I told Miguel privately, I was thinking of getting an > STM32 evaluation board (they're really cheap, <15€ IIRC) and building the > USB controller prototype. Once we have the STM32 reading all the values > from the potentiometers, pushbuttons, etc. and reporting them over USB as a > HID we would go ahead and design PCBs 2 and 3. How does this sound? Should > we skip the eval board thing and go straight to designing the PCBs? >  Hello Daniel, Luke and everyone. I want to introduce you to Peter Bouda. He has contacted me because he liked the project. He is a software developer ( http://www.ubrew.it/ ) and he has offered help to develop the software of the handheld games console. He does not know a lot about hardware design, but he has told me he will try to learn more about this to help in the design of hardware. So maybe soon he can also help in the design of hardware. Thanks! > 2015-09-23 19:44 GMT+02:00 Luke Kenneth Casson Leighton : > > > On Wed, Sep 23, 2015 at 4:31 PM, GaCuest wrote: > > > En 23 de septiembre de 2015 en 17:13:37, Luke Kenneth Casson Leighton ( > > lkcl at lkcl.net) escrito: > > >> On Wed, Sep 23, 2015 at 4:07 PM, GaCuest wrote: > > >> > En 23 de septiembre de 2015 en 16:55:18, Luke Kenneth Casson Leighton > > (lkcl at lkcl.net) > > >> escrito: > > >> >> On Wed, Sep 23, 2015 at 2:59 PM, GaCuest wrote: > > >> >> > > >> >> > I’m using AirMail. It has plain text editing, but not line-wrap. > > >> >> > I will do that manually. > > >> >> > > >> >> appreciated. > > >> >> > > >> >> > I updated the block diagram: > > >> >> > > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > > >> >> > > > >> >> > Anything more to change? > > >> >> > > >> >> yes, one thing that's important: 1st USB is mandated to be >= speed > > >> >> than 2nd USB. so you need to swap 1st USB with 2nd USB. > > >> >> > > >> >> reason: the STM32F is connected to the port that may end up being > > >> >> 480mbit/sec and the USB Hub is connected to the port that may end up > > >> >> being only 11mbit/sec. > > >> >> > > >> >> certainly, for the jz4775 CPU Card (which i am sending off very > > >> >> shortly), it *DEFINITELY* has USB 2.0 for the 1st port and *ONLY* USB > > >> >> 1.1 for the 2nd. > > >> >> > > >> >> and i am certainly not going to increase the cost of the jz4775 CPU > > >> >> Card by an extra $2 worth of components to put in a USB2 Hub when the > > >> >> SOC itself is only $2.50!!! > > >> >> > > >> > > > >> > I think that you have not properly seen the block diagram. > > >> > > >> wiggle wiggle follow the lines with my finger pointing at the screen > > >> this time... :) > > >> > > >> > The STM32 is connected to the second USB, the USB HUB is connected > > >> > to the first USB. > > >> > > > >> > This right? > > >> > > >> yes! :) > > >> > > > > > > Then, anything to change? > > > > no > > > > > Is this right?: > > > > i did say yes > > > > > > > http://rhombus-tech.net/community_ideas/games_console/news/pcb_block_diagram_updated2_23sep2015/ > > > > > >> _______________________________________________ > > >> arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > >> http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > >> Send large attachments to arm-netbook at files.phcomp.co.uk > > > > > > > > > _______________________________________________ > > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > > Send large attachments to arm-netbook at files.phcomp.co.uk > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook at files.phcomp.co.uk > > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 28 11:37:34 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 28 Sep 2015 11:37:34 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 28, 2015 at 8:52 AM, Daniel Iglesias wrote: > Hello there, sorry for not writing for so long. What would you recommend as > the next step now? As I told Miguel privately, I was thinking of getting an > STM32 evaluation board (they're really cheap, <15€ IIRC) i know, it's ridiculous! i bought one from reichhelt.de because they accept dutch iDEAL payment (no other reason!) and including shipping it was like... $EUR 12 or something silly. love it. > and building the > USB controller prototype. Once we have the STM32 reading all the values from > the potentiometers, pushbuttons, etc. and reporting them over USB as a HID > we would go ahead and design PCBs 2 and 3. How does this sound? great idea, i've done the same thing, bought an EVB, i have to do the software for the laptop_15in_pcb2 as well: that's going to be a 16x8 key-scan matrix and reading an I2C touchpanel, converting both to USB-HID. > Should we skip the eval board thing and go straight to designing the PCBs? weeeellll, think about that for a moment: if there's a mistake in the design concept (STM32F072RBT6 isn't good enough for some reason) you just threw away the time and the money spent designing, assembling and then testing the PCBs. if you're _really_ confident then go for it! i am merely... foolish / optimistic: i've designed and ordered the PCBs already (and have discovered missing components already *sigh*....) btw miguel, daniel, one thing: you need to check if an FFC-6 can provide enough power to PCB2. l. From lkcl at lkcl.net Mon Sep 28 11:41:42 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 28 Sep 2015 11:41:42 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 28, 2015 at 10:09 AM, GaCuest wrote: > Hello Daniel, Luke and everyone. > > I want to introduce you to Peter Bouda. hi peter, welcome > He has contacted me because he liked the project. > > He is a software developer ( http://www.ubrew.it/ ) and he has offered help to develop the software of the handheld games console. great! > He does not know a lot about hardware design, but he has told me he will try to learn more about this to help in the design of hardware. So maybe soon he can also help in the design of hardware. well there are a couple of main tasks, peter, software-wise. the first is integration of EOMA68 into the linux kernel, but i believe the team at your end decided that they won't support EOMA68 at a software level, only taking advantage of the hardware. the second is, as daniel mentioned, the STM32F embedded controller. this i recommend using libopencm3 and i will help out by developing the firmware for the laptop_15in_pcb2 simultaneously. if you can send me an ssh public key i will arrange to add you to the git repository. also i recommend joining the mailing list here, it will make life easier. or, you guys should probably set one up for the project: alain will almost certainly be happy to help there. l. From gacuest at gmail.com Mon Sep 28 11:54:11 2015 From: gacuest at gmail.com (GaCuest) Date: Mon, 28 Sep 2015 12:54:11 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: En 28 de septiembre de 2015 en 12:37:53, Luke Kenneth Casson Leighton (lkcl at lkcl.net) escrito: > On Mon, Sep 28, 2015 at 8:52 AM, Daniel Iglesias > wrote: > > Hello there, sorry for not writing for so long. What would you recommend as > > the next step now? As I told Miguel privately, I was thinking of getting an > > STM32 evaluation board (they're really cheap, <15€ IIRC) > > i know, it's ridiculous! i bought one from reichhelt.de because they > accept dutch iDEAL payment (no other reason!) and including shipping > it was like... $EUR 12 or something silly. love it. > > > and building the > > USB controller prototype. Once we have the STM32 reading all the values from > > the potentiometers, pushbuttons, etc. and reporting them over USB as a HID > > we would go ahead and design PCBs 2 and 3. How does this sound? > > great idea, i've done the same thing, bought an EVB, i have to do > the software for the laptop_15in_pcb2 as well: that's going to be a > 16x8 key-scan matrix and reading an I2C touchpanel, converting both to > USB-HID. > > > Should we skip the eval board thing and go straight to designing the PCBs? > > weeeellll, think about that for a moment: if there's a mistake in the > design concept (STM32F072RBT6 isn't good enough for some reason) you > just threw away the time and the money spent designing, assembling and > then testing the PCBs. > > if you're _really_ confident then go for it! i am merely... foolish > / optimistic: i've designed and ordered the PCBs already (and have > discovered missing components already *sigh*....) > > > btw miguel, daniel, one thing: you need to check if an FFC-6 can > provide enough power to PCB2. What is the alternative? Thanks. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 28 12:07:49 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 28 Sep 2015 12:07:49 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 28, 2015 at 11:54 AM, GaCuest wrote: >> btw miguel, daniel, one thing: you need to check if an FFC-6 can >> provide enough power to PCB2. > > What is the alternative? anything! wires, 6-pin 0.1 or 0.2mm pitch header, FPC, anything - just remember to check the current rating of whatever you decide to use, ok? l. From lkcl at lkcl.net Mon Sep 28 12:09:45 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 28 Sep 2015 12:09:45 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: On Mon, Sep 28, 2015 at 12:07 PM, Luke Kenneth Casson Leighton wrote: > On Mon, Sep 28, 2015 at 11:54 AM, GaCuest wrote: > >>> btw miguel, daniel, one thing: you need to check if an FFC-6 can >>> provide enough power to PCB2. >> >> What is the alternative? > > anything! wires, 6-pin 0.1 or 0.2mm pitch header, FPC, anything - > just remember to check the current rating of whatever you decide to > use, ok? those SIM card holders i mentioned - absolutely anything. just check what current draw the STM32F072RBT6 is likely to use, it could be as low as 0.1A in which case you're fine... but you _need_ to check. l. From gacuest at gmail.com Mon Sep 28 17:01:04 2015 From: gacuest at gmail.com (Miguel Garcia) Date: Mon, 28 Sep 2015 18:01:04 +0200 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: 2015-09-28 13:09 GMT+02:00, Luke Kenneth Casson Leighton : > On Mon, Sep 28, 2015 at 12:07 PM, Luke Kenneth Casson Leighton > wrote: >> On Mon, Sep 28, 2015 at 11:54 AM, GaCuest wrote: >> >>>> btw miguel, daniel, one thing: you need to check if an FFC-6 can >>>> provide enough power to PCB2. >>> >>> What is the alternative? >> >> anything! wires, 6-pin 0.1 or 0.2mm pitch header, FPC, anything - >> just remember to check the current rating of whatever you decide to >> use, ok? > > those SIM card holders i mentioned - absolutely anything. just check > what current draw the STM32F072RBT6 is likely to use, it could be as > low as 0.1A in which case you're fine... but you _need_ to check. Ok, thanks. That is something that Daniel has to check. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk From lkcl at lkcl.net Mon Sep 28 21:35:03 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Mon, 28 Sep 2015 21:35:03 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: Message-ID: On Sun, Sep 27, 2015 at 4:05 PM, Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/community_ideas/laptop_15in/news/ > > CM108AH, and Power ICs, all functional. found a spare GL850G and a SY6280 from other boards, have put those on. the GL850G works (yippee!) i will test the SY6280 USB-power protection IC tomorrow. last pieces (still on order) are the LCD 40-pin connector and a single diode i missed out, then this PCB is done. on to the embedded controller board and the power / battery board after that. l. From joem at martindale-electric.co.uk Tue Sep 29 08:31:04 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 07:31:04 +0000 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: Message-ID: <1443511867.29009.10.camel@jm-desktop> On Sun, 2015-09-27 at 16:05 +0100, Luke Kenneth Casson Leighton wrote: > http://rhombus-tech.net/community_ideas/laptop_15in/news/ > > CM108AH, and Power ICs, all functional. Hmm... use the force Luke! An SMT oven is cheap to buy these days.. http://www.aliexpress.com/item/T-962-Reflow-Oven-Infrared-IC-Heater-Soldering-Machine-800W-180-x-235mm-T962-BGA-SMD/32297484903.html?spm=2114.031010208.3.2.s7Nond&ws_ab_test=201556_2,201527_2_71_72_73_74_75,201409_5 (You should buy some venting pipe from screwfix and vent the fumes out - they are nasty.) Solder paste is cheap from aliexpress. Making stencils for PCBs to apply solder paste also relatively cheap these days. From joem at martindale-electric.co.uk Tue Sep 29 09:38:05 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 08:38:05 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> Message-ID: <1443515888.10509.8.camel@jm-desktop> > > Just wondering how much is it these days ball park to pay someone like > > witstech or a similar design outfit to design an A20 board > > about $USD 3-4k > > > (like a > > cubie) and make 10 samples? > > about $USD 2.5k The first batch funding of 1.6 kilo dollars on its way Luke :) The board I want is designed to run on 18650 or 26650 batteries (very similar to a UPS configuration) and have capacitive touch LCDs. Its just right for building a million dollar 6 petabyte SSD server that fits in a space of 2 cubic meters or thereabouts consuming 10kW. You may not want to believe all that right now.. :) From lkcl at lkcl.net Tue Sep 29 10:31:34 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 29 Sep 2015 10:31:34 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: <1443511867.29009.10.camel@jm-desktop> References: <1443511867.29009.10.camel@jm-desktop> Message-ID: On Tue, Sep 29, 2015 at 8:31 AM, joem wrote: > On Sun, 2015-09-27 at 16:05 +0100, Luke Kenneth Casson Leighton wrote: >> http://rhombus-tech.net/community_ideas/laptop_15in/news/ >> >> CM108AH, and Power ICs, all functional. > > > Hmm... use the force Luke! An SMT oven is cheap to buy these days.. yes i got the T862A, the one with an 800W pre-heater and a 200W overhead lamp, i can actually see what's going on, keep an eye on the solder as it melts and poke things with a stick if necessary. > Making stencils for PCBs to apply solder paste also relatively cheap > these days. yeah i saw that, eurocircuits can do them for $EUR 40 which is a hell of a lot less than $700+ i've previously been quoted. l. From lkcl at lkcl.net Tue Sep 29 10:32:49 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 29 Sep 2015 10:32:49 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1443515888.10509.8.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> <1443515888.10509.8.camel@jm-desktop> Message-ID: On Tue, Sep 29, 2015 at 9:38 AM, joem wrote: > >> > Just wondering how much is it these days ball park to pay someone like >> > witstech or a similar design outfit to design an A20 board >> >> about $USD 3-4k >> >> > (like a >> > cubie) and make 10 samples? >> >> about $USD 2.5k > > The first batch funding of 1.6 kilo dollars on its way Luke :) > > The board I want is designed to run on 18650 or 26650 batteries (very > similar to a UPS configuration) and have capacitive touch LCDs. > > Its just right for building a million dollar 6 petabyte SSD server that > fits in a space of 2 cubic meters or thereabouts consuming 10kW. > > You may not want to believe all that right now.. :) no i get it - i added up once how many CPU Cards you could get into a single rack-mount space, it was absolutely mad. l. From lasich at gmail.com Tue Sep 29 10:47:07 2015 From: lasich at gmail.com (Hrvoje Lasic) Date: Tue, 29 Sep 2015 11:47:07 +0200 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> Message-ID: so, eurocircuits supply you PCB, you also make stencil with eurocircuits and you mount all parts? is it hard to apply solder paste over stencil? did you actually tried to use oven yet? looks like a good setup... On Tue, Sep 29, 2015 at 11:31 AM, Luke Kenneth Casson Leighton < lkcl at lkcl.net> wrote: > On Tue, Sep 29, 2015 at 8:31 AM, joem > wrote: > > On Sun, 2015-09-27 at 16:05 +0100, Luke Kenneth Casson Leighton wrote: > >> http://rhombus-tech.net/community_ideas/laptop_15in/news/ > >> > >> CM108AH, and Power ICs, all functional. > > > > > > Hmm... use the force Luke! An SMT oven is cheap to buy these days.. > > yes i got the T862A, the one with an 800W pre-heater and a 200W > overhead lamp, i can actually see what's going on, keep an eye on the > solder as it melts and poke things with a stick if necessary. > > > Making stencils for PCBs to apply solder paste also relatively cheap > > these days. > > yeah i saw that, eurocircuits can do them for $EUR 40 which is a hell > of a lot less than $700+ i've previously been quoted. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook at lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook at files.phcomp.co.uk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joem at martindale-electric.co.uk Tue Sep 29 10:47:31 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 09:47:31 +0000 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> Message-ID: <1443520054.10777.3.camel@jm-desktop> > yes i got the T862A, the one with an 800W pre-heater and a 200W > overhead lamp, i can actually see what's going on, keep an eye on the > solder as it melts and poke things with a stick if necessary. I don't dispute the benefits of a T862A, but get oven, destroy a few boards until you got it timed to working 100%, and then its a doddle to plant components, press go, and get working board with near 100% reliability. From lkcl at lkcl.net Tue Sep 29 10:57:49 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 29 Sep 2015 10:57:49 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: <1443520054.10777.3.camel@jm-desktop> References: <1443511867.29009.10.camel@jm-desktop> <1443520054.10777.3.camel@jm-desktop> Message-ID: On Tue, Sep 29, 2015 at 10:47 AM, joem wrote: > >> yes i got the T862A, the one with an 800W pre-heater and a 200W >> overhead lamp, i can actually see what's going on, keep an eye on the >> solder as it melts and poke things with a stick if necessary. > > I don't dispute the benefits of a T862A, but get oven, destroy > a few boards until you got it timed to working 100%, and then its a > doddle to plant components, press go, and get working board with > near 100% reliability. for this (first) board i need a slightly different approach, which the T862A suits better. i need to go carefully and slowly, populating only a few components at a time, testing them, inspecting the board, making sure there's no short-circuits (some of the components are 0402), verifying that the circuit is functional by applying power, and correcting it there and then if necessary. if i were to populate the entire board with components, there would be absolutely no way that i could test individual circuits or components. in some places for example resistors don't measure their actual value because of other components on the board. i made one mistake by putting in the 3.3v regulator *and* the MOSFETs for the LCD digital circuit. the 3.3v regulator circuit was fine, but the LCD digital 3.3v supply circuit wasn't. so i dismantled pretty much the entire 3.3v regulator circuit, had to cut PCB power line tracks, and generally messed up the board, taking over 2 hours to track down what was wrong. if i had put down only the 3.3v regulator and then tested that, i wouldn't have had to remove any of its components. now multiply that scenario up for an entire board and it's easy to work out why i'm doing this a bit at a time. so i'm using the T862A as a PCB development tool, not a PCB *production* tool. if i was doing small runs of PCB production, where i was absolutely 100% confident that the circuit was proven, correct, and functional, *then* yes absolutely i would get one of the T962 ovens. l. From lkcl at lkcl.net Tue Sep 29 11:02:02 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 29 Sep 2015 11:02:02 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> Message-ID: On Tue, Sep 29, 2015 at 10:47 AM, Hrvoje Lasic wrote: > so, eurocircuits supply you PCB, yes. > you also make stencil with eurocircuits not yet. > and you mount all parts? yes, by hand... with tweezers and a ridiculous-looking head-mounted magnifying glass :) > is it hard to apply solder paste over stencil? from what i understand it's just "wipe with a sponge"! have a look at the section "solder paste stencil" on this link: http://www.factoryforall.com/pcba-fabrication > did you actually tried to use oven yet? looks like a good setup... T862A yes - i'll do a video shortly of putting down one of the components. joe, how did you get on with an IR oven? l. From joem at martindale-electric.co.uk Tue Sep 29 11:26:24 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 10:26:24 +0000 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> Message-ID: <1443522387.11009.4.camel@jm-desktop> > joe, how did you get on with an IR oven? The first few attempts clog up everything. Temperature settings too low or too much solder paste everywhere. The solder paste solvents can evaporate in storage, so mix it with some liquid solder flux to make it flow (and don't breath the solder flux fumes coming out of the oven - take the time to fit proper venting before starting up the oven). Allow yourself the luxury of destroying 5 boards with cheap chips and components to practice. After that its regular as clockwork. (Plenty videos on youtube.) From joem at martindale-electric.co.uk Tue Sep 29 11:39:07 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 10:39:07 +0000 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> <1443520054.10777.3.camel@jm-desktop> Message-ID: <1443523150.11009.11.camel@jm-desktop> > so i'm using the T862A as a PCB development tool No. Nurse the oven as your tool. Its more safer and use as many times as you want. (Some items such as LEDs won't survive re-use too many times - but they are cheap to replace.) From lkcl at lkcl.net Tue Sep 29 12:16:38 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Tue, 29 Sep 2015 12:16:38 +0100 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: <1443523150.11009.11.camel@jm-desktop> References: <1443511867.29009.10.camel@jm-desktop> <1443520054.10777.3.camel@jm-desktop> <1443523150.11009.11.camel@jm-desktop> Message-ID: On Tue, Sep 29, 2015 at 11:39 AM, joem wrote: > >> so i'm using the T862A as a PCB development tool > > No. Nurse the oven as your tool. remember, joe, you've done this a _lot_, so have much more confidence in your abilities, as well as, from experience, a much higher anticipated success rate. i had no idea what i was doing, so i needed something which i could monitor closely and much quicker (like... immediate), on a much smaller area. so: i can't see inside the oven with the door closed, can i? that's what i like about the one with the lamp: it focusses the IR on a small area (about 5cm in diameter), i can get tweezers in underneath it and remove components (or adjust their positions), i can watch the solder actually melting, i can actually *see* the point when the components flow into the solder, i can get solder in with a bit of solder paste on the end to encourage the solder on the board to melt... pluuuus, i can put only a tiny amount of solder paste down, which is tolerable with the windows open. you can't do _any_ of those things with an oven: it heats up the entire board, it's an enclosed space, you have to heat up *all* the components including all the ones that you previously placed, it takes a long time to warm up and cool down... so you *have* to do all the components (or risk re-heating all of them). whereas i'm safely doing 5-10 components at a time, confident that because the tested ones are in another area away from the lamp, they won't get moved, overheated or (in the case of the TSSOP-56) warped beyond useability. when all the components i've put down are clearly visibly in full solder-contact with the board, i can shut down the lamp immediately. sometimes i've managed to get components soldered in about... 4 to 5 minutes. removing components (including SOT-23s and TSOP-8s) can be as little as 3-4 minutes. can you do that with an oven? no.... you can't see inside it, clearly, so you have to play it safe and let it heat up on a full cycle, which is what... 10 to 15 minutes at a time? you said it yourself: you destroyed several boards, first time you used the oven. i only destroyed one, and it was the first test board i'd ever done, it was only 1in x 0.5in in size (the fibreglass actually delaminated). i learned from that, and haven't damaged the laptop PCB at all. i've only destroyed around 3 components in total, one was the TSSOP-56, it actually curled up at one end due to uneven heat distribution and i wasn't watching the lamp's temperature sensor closely enough. so i'm leaving its replacement until last. now, the down-side is: it's taken 2-3 weeks of about 1 to 2 hours a day to put down all the components... and there's only about 120 maybe 150 components. that's *ridiculously* slow. ...but i don't care if it's slow: this is my first PCB since.. since.... making illegal FM transmitters back in 1983 from "the anarchist's handbook". i've done the "pay someone else to design PCBs" thing - that didn't work out. even paying someone else to assemble PCBs is relatively pricey (if you get it wrong more than zero times). so i am simplifying the task to one which, with my skill set, stands an above-average chance of success and/or is a much lower iterative cost. > Its more safer and use as many times as > you want. (Some items such as LEDs won't survive re-use too many times - > but they are cheap to replace.) found that out already :) one of the other components i destroyed was an 0805 LED - i hand-soldered the next one... l. From joem at martindale-electric.co.uk Tue Sep 29 13:22:54 2015 From: joem at martindale-electric.co.uk (joem) Date: Tue, 29 Sep 2015 12:22:54 +0000 Subject: [Arm-netbook] laptop 15in pcb1 progress In-Reply-To: References: <1443511867.29009.10.camel@jm-desktop> <1443520054.10777.3.camel@jm-desktop> <1443523150.11009.11.camel@jm-desktop> Message-ID: <1443529377.11281.9.camel@jm-desktop> On Tue, 2015-09-29 at 12:16 +0100, Luke Kenneth Casson Leighton wrote: > On Tue, Sep 29, 2015 at 11:39 AM, joem wrote: > > > >> so i'm using the T862A as a PCB development tool > > > > No. Nurse the oven as your tool. > > remember, joe, you've done this a _lot_, so have much more confidence > in your abilities, No problem Luke - we all start somewhere and learn the hard way :) So here is how i would do it. You don't have to follow any of it - its just what I would do.. Rely on the oven as no.1 tool. T862A looks like a re-balling tool which is there to apply localised heat on a heavily populated board which might otherwise disintegrate in an oven. The smallish PCB you have is great for oven - so long as buttons led and other plastic heavy or heat sensitive items are applied last. Apply solder paste sparingly. They can easily bridge if too much. Use solder flux to flow in between the pins to prevent this as much as possible. If you make a mistake, have a bucket ready, re-heat in oven, at highest temperature open oven and flick the board fast and slam to the side of the bucket. Instant de-populated board to start again :) > it's taken 2-3 weeks of about 1 to 2 hours a > day to put down all the components... and there's only about 120 maybe > 150 components. that's *ridiculously* slow. Exactly what I imagined might happen with a re-balling tool. :( From pbouda at cidles.eu Tue Sep 29 16:27:38 2015 From: pbouda at cidles.eu (Peter Bouda) Date: Tue, 29 Sep 2015 16:27:38 +0100 Subject: [Arm-netbook] Handheld Games Console In-Reply-To: References: Message-ID: <560AADEA.2060709@cidles.eu> > well there are a couple of main tasks, peter, software-wise. the > first is integration of EOMA68 into the linux kernel, but i believe > the team at your end decided that they won't support EOMA68 at a > software level, only taking advantage of the hardware. > > the second is, as daniel mentioned, the STM32F embedded controller. > this i recommend using libopencm3 and i will help out by developing > the firmware for the laptop_15in_pcb2 simultaneously. if you can send > me an ssh public key i will arrange to add you to the git repository. > > also i recommend joining the mailing list here, it will make life > easier. or, you guys should probably set one up for the project: > alain will almost certainly be happy to help there. > > l. hi, first! this is really an awesome project, and I would be happy if I can contribute even some minor stuff. I will have a look at the libopencm3 and see what I can do (will send the public key). and probably order also order one of the STM32 boards and try to connect buttons... what exactly would be the task of integrating EOMA68 into the linux kernel? for me this is mainly a way to learn about embedded system development, software and hardware side, so I am interested in anything related. Best, Peter From joem at martindale-electric.co.uk Wed Sep 30 09:12:00 2015 From: joem at martindale-electric.co.uk (joem) Date: Wed, 30 Sep 2015 08:12:00 +0000 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: References: <1441810196.718.2.camel@jm-desktop> <1443515888.10509.8.camel@jm-desktop> Message-ID: <1443600723.14016.7.camel@jm-desktop> > > The board I want is designed to run on 18650 or 26650 batteries (very > > similar to a UPS configuration) and have capacitive touch LCDs. > > > > Its just right for building a million dollar 6 petabyte SSD server that > > fits in a space of 2 cubic meters or thereabouts consuming 10kW. > > > > You may not want to believe all that right now.. :) > > no i get it - i added up once how many CPU Cards you could get into a > single rack-mount space, it was absolutely mad. Its how to wire all that that is difficult. I'm building a hypercube (with emphasis on hype :) ) with extravagantly lit LED conduits for the wiring to be passed through to reach all computing elements in a 3D mesh. The idea being if all is well, the conduits are a pleasant colour while if there are problem, the conduits light the way to the problem areas. From lkcl at lkcl.net Wed Sep 30 10:39:22 2015 From: lkcl at lkcl.net (Luke Kenneth Casson Leighton) Date: Wed, 30 Sep 2015 10:39:22 +0100 Subject: [Arm-netbook] How much to design A20 board? In-Reply-To: <1443600723.14016.7.camel@jm-desktop> References: <1441810196.718.2.camel@jm-desktop> <1443515888.10509.8.camel@jm-desktop> <1443600723.14016.7.camel@jm-desktop> Message-ID: On Wed, Sep 30, 2015 at 9:12 AM, joem wrote: >> no i get it - i added up once how many CPU Cards you could get into a >> single rack-mount space, it was absolutely mad. > > Its how to wire all that that is difficult. > > I'm building a hypercube (with emphasis on hype :) ) with extravagantly > lit LED conduits for the wiring to be passed through to reach all > computing elements in a 3D mesh. The idea being if all is well, the > conduits are a pleasant colour while if there are problem, the conduits > light the way to the problem areas. niiice. btw do consider this as well, it expands up to infinite size: https://en.wikipedia.org/wiki/Nonblocking_minimal_spanning_switch assuming a 2-bar crossbar as the "base unit", basically to create the next power of 2 up, you: * duplicate the entire network of switches made so far * put them side-by-side * put in another "layer" * connect every ODD numbered output from the last of the old layers "straight through" * connect every EVEN numbered output to a port CROSS-WISE by adding N/2 to its port number (N is obviously the current total number of inputs and outputs) if you have a 4-port router then obviously you would quadruplicate the entire network so far, then connect the first one straight, the second one increase the port number by (N/4), the third by (N/4)*2, the fourth by (N/4)*3 if you have a 32-port hub as the base unit you could go straight to 32x32 nodes (1024) with only 2 layers. i pretty much guarantee though that regardless of what you do, you'll end up with more wires and more network switches than anything else :) l.