arm-netbook June 2014

arm-netbook@lists.phcomp.co.uk

24 participants
19 discussions

[status] eoma68-a20 and micro-desktop
by Luke Kenneth Casson Leighton 13 Jun '14

13 Jun '14

ok thank you to two sponsors, we have enough to pay outright for the micro-desktop 1st prototype and part-way for the revised eoma68-a20 cpu card. one final check of the revised cpu card is needed (as well as waiting a little bit for some extra money) then it will be off to get the PCB checked and done etc. if anyone wants to take a punt on that they are more than welcome, i am planning to do just 5 but if anyone is happy to pay say.... $175 then you can have an early CPU Card with either 1gb or 2gb of RAM, it is your choice (and risk!). i will be getting 2 with 1gbyte and 3 with 2gbyte RAM done. l.

3 4

Doubt about USB in EOMA-68.
by Miguel Garcia 13 Jun '14

13 Jun '14

Hi all, I've been looking at this: http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68#Table_of_EOMA-… Are all EOMA-68 obliged to have 2 USB 2.0 (the first on pin 30 and 64, the second on pin 34 and 68)? If in the future there is a EOMA-68 compatible with USB 3.0, would this USB 3.0 be associated with the first USB 2.0 or the second USB 2.0? We want to use the USB 3.0 to create a USB 3.0 port (with the current EOMA-68 only be compatible with USB 2.0, but we want to be compatible in the future with USB 3.0) and use the other USB 2.0 to connect internally the WIFI, the BT, the USB IC audio .... And we do not want that, in the future, some new EOMA-68 be incompatible with this because we have chosen the wrong USB. Thank you.

2 3

quick update
by Luke Kenneth Casson Leighton 12 Jun '14

12 Jun '14

* jz4775 board layout looks feasible (just). have had to squash the DDR3 address lines as there are no wiggles, only straight routing on the dev-board, the DDR3 is hard against one edge and the CPU hard against the other,i have to ask ingenic to check * micro-pc needs another review * 2nd revision EOMA68-A20 still to go to PCB factory for gerber checks * IC1 PCB pretty much done component layout. v busy sorry short status. l.

1 0

[status] eoma68-ic1t
by Luke Kenneth Casson Leighton 12 Jun '14

12 Jun '14

cpu cards being like buses, you wait for ages then 3 come along at once. i have been in touch with icubecorp, they have that awesome hybrid VPU/GPU/CPU core, and they decided to do a sub-$2.50 55nm 210-pin QFP out of it, running at 400mhz but quad-threaded. it's capable (on its own) of 720p video decode and 3D graphics *entirely* in software. the down-side is that this revision was designed as a low-entry android or embedded controller IC, so can only do up to 512mb RAM. they can do a respin but only with a large enough order. this will be one of the very first FSF-Endorseable CPU Cards and it will be the precursor to a more powerful version which we will do later. the BOM is something staggering like only $12 and that's including a SMC9514 USB-Ethernet Hub. it is juuuust about EOMA68 compliant and the only extra interface will be an SD/MMC card. it will boot from SPI NOR Flash so that will be deliberately very small (enough to hold a bootloader). l.

2 2

Re: [Arm-netbook] RK* thingies are fairly cheap, indeed, but I managed to brick one
by Erix 10 Jun '14

10 Jun '14

> Message: 3 > Date: Mon, 09 Jun 2014 21:11:26 -0400 > From: Stefan Monnier <monnier(a)iro.umontreal.ca> > To: arm-netbook(a)lists.phcomp.co.uk > Subject: Re: [Arm-netbook] ARM device - compiling linux software > Message-ID: > <jwvk38pin90.fsf-monnier+gmane.comp.hardware.netbook.arm(a)gnu.org> > Content-Type: text/plain > > >> It looks cheaper than the Wandboard? > >> http://radxa.com/specification/ > > Does look neat and yes, does appear to be cheaper. Might take a punt on > one > > actually, see if it cuts the mustard. Thanks for pointing it out. > > The RK* thingies are fairly cheap, indeed, but I managed to brick one > with a simple "dd" from Android and still haven't managed to unbrick it. > >From that point of view the Allwinner ones are very friendly (since you > can boot from uSD even with a completely broken nand flash). > > > Stefan > > Hi Stephan, I don't know if it could help you, but the Radxa Team post a tip/advice to "unbrick" their platform in case it happens. http://wiki.radxa.com/Rock/unbrick Regards Erix

3 3

A33
by Luke Kenneth Casson Leighton 04 Jun '14

04 Jun '14

On Tue, Jun 3, 2014 at 10:58 PM, Luke Kenneth Casson Leighton <lkcl(a)lkcl.net> wrote: > http://webcache.googleusercontent.com/search?q=cache:prYbhMDZblEJ:www.allwi… > > ok it's pin-compatible with the A23. with the removal of SATA and > replacement with USB it can be done. i'll ask for some PCB CAD files. > > l. > > On Tue, Jun 3, 2014 at 10:55 PM, Luke Kenneth Casson Leighton > <lkcl(a)lkcl.net> wrote: >> On Tue, Jun 3, 2014 at 10:14 PM, Paul NeoStormer >> <paulneostormer(a)gmail.com> wrote: >>> So I can't find the page where this was discussed but I was told that the >>> reason we couldn't go to a Quad-Core processor A31 was not because of pin >>> compatibility but because it utilized a different graphics processor which >>> we didn't really have open source for, a PowerVR instead of the Mali-400. We >>> today they announced the A33 which uses the same Mali-400MP2 that the A20 >>> uses, so is it possible to go to quad-core now? >> >> ooooo yes! hooray! if it's pin-compatible with the A20 we can use >> it straight away. ok just checked the A23 datasheet: it's only a 16-bit DDR3 RAM interface, one USB2 and one USB-OTG. so something like the 9514 USB-Eth hub would be needed (the 9512 apparently gets veery hot). also apparently A23 only addresses 1gbyte of RAM, we'll see if the A33 goes up to 2. about powervr, you want luc's talk about the state of software libre reverse-engineering of GPUs. it turns out he's worked (in a proprietary environment) with powervr, so he knows that it is an absolute god-awful mess (hardware and software wise). his recommendation is to let imgtec stew, and to not even attempt the reverse-engineering *at all*, it's that bad. l.

3 5

EOMA - Open Specification?
by Jonathan Frederickson 04 Jun '14

04 Jun '14

Hey all. I've been following the progress here for a while, though I wasn't subscribed at the time. Something Luke said a while back concerned me, however: "so i am very sorry to have to spell it out, but you will *never* be a customer of *any* EOMA or QiMod products, *ever*, and you will *never* be granted a license to make EOMA-compatible products. and that's not my decision, but we both have to live with that." I can understand being blacklisted as a customer, and removing all mention of EOMA, as it is (maybe?) a QiMod trademark. However, the bit about being granted a license to make EOMA-compatible products is troubling. It seems counter to the statement on the FAQ: "The EOMA-68 initiative is an "Open Specification". That means that anyone can create either CPU cards or motherboards that conform to it. Thus, it is possible for anyone to create an "Open Hardware" compliant CPU card or motherboard." (http://rhombus-tech.net/faq/#index12h2) ...which seems to imply that one wouldn't *need* a license to create compatible products. So my question is this: how open is the EOMA specification, really? Regards, Jon F.

12 33

Improv And Operation:Marketing for EMOA-*
by Alexander Stephen Thomas Ross 04 Jun '14

04 Jun '14

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 If there was late help from extra individuals marketing and promoting eoma-* standard devices like Improv and there were magically the number of pre-orders needed would it be possible to resume makeplaylive being a client? I expect they are stuffed when they can't even mention eoma-68! It's just, I've been trying to work on my Operation: Marketing (Op:Mark) but with this sad news I've not sure which direction to go. I had been plaining to write marketing material for a device. I have a huge amount to evolve and currently Op:Mark is my mess also my progress is dreadfully slow. I wanted to make more progress on it before I mentioned it to you but arr heck here it goes: http://rhombus-tech.net/op:marketing -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJTeiXhAAoJEOk7fWIzE1KNoE0H/1HajC3ssYa/m6f8Wd+aCawl G9xD7OYrwc3/7IwU7WojZ4ohRJsaoZgbAZBWmH5MrO7+M/ZvrLfRdcW6bTm16HBy iY/X04TgBwhKkvwYN1k98Qz4ckyRchttIBylRH3NLDUQ0uRXJX0SQMXIA3xVQuWj LvJJFYOgTZ4DPuzzAOE3MD7ZVedf7MK6A4EhEpQ6ImoJU86Cn5Q971Teb0anv2q0 1jmzRqmRS0UDg8iNjBiP4PTU+u5gDkVltjoVJG/SZ4Slrrrs4uMKVRIlQ1F04UUY 9Wnn4yaK/uLSGSfHoVcnEq/VD1JkgDQK4Tl5wnTiBuRuzNP+Aq9OOcv5pJiwMsw= =jbyQ -----END PGP SIGNATURE-----

17 67

EOMA68-A20 interface redesign
by Luke Kenneth Casson Leighton 03 Jun '14

03 Jun '14

https://www.skhynix.com/products/computing/view.jsp?info.ramKind=19&info.se… ok i've done the interface redesign which i will document properly later: SATA is gone, 2nd USB is in, i added one EINT and the 4th pin is a VREF output for GPIO and UART voltage levels. i also noticed that hynix do a 4gbit x8 RAM chip which is pin-compatible with the H5TQ2G83BFR, so i *think* we can go to 2gbyte of RAM. thoughts? l.

6 24

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

arm-netbook June 2014