arm-netbook

arm-netbook@lists.phcomp.co.uk

8 participants
1077 discussions

Re: [Arm-netbook] Arm processors
by rohbeck＠yahoo.com 02 Mar '17

02 Mar '17

Haha, just my thought. My first impulse was "Zen @4W, yay!".0 0 ------ Original message------From: Luke Kenneth Casson LeightonDate: Wed, Mar 1, 2017 20:22To: Linux on small ARM machines;Cc: Subject:Re: [Arm-netbook] Arm processors On Thu, Mar 2, 2017 at 12:49 AM, zap wrote:>>> On 03/01/2017 06:58 PM, John Luke Gibson wrote:>> And, Luke mentioned inquiries being made about using RISC-V in a 100%>> open core board. That's probably a more longterm slightly optimistic>> ah okay fair enough. I was just curious if he planned to reverse> engineer ones with 8gb or more in the later future. reverse-engineering i have come to the conclusion is a total - andcriminal - waste of time and effort. by the time all features are100% stable it's several YEARS down the line. look at how long agothe A64 was released, and the libdram code STILL HAS NOT BEENREVERSE-ENGINEERED. it's 200 lines of code for fuck's sake. NO. read this, zap: https://www.crowdsupply.com/eoma68/micro-desktop/updates/picking-a-processo… should give you some idea of how insane pretty much every singlefabless semiconductor company really is. they're just not payingattention: every single one of them makes at least one compromise,somewhere, and because it's an *integrated* SoC there's absolutelynothing that can be done about it.so we need to be of the order of a MILLION units to be in a positionto influence these people. and if you're going for a million units,you might as well get your own SoC custom-made. anyone knows of an open silicon H.264 and MPEG design that is capableof up to 1080p60 video decode, do let me know.l._______________________________________________arm-netbook mailing list arm-netbook@lists.phcomp.co.ukhttp://lists.phcomp.co.uk/mailman/listinfo/arm-netbookSend large attachments to arm-netbook(a)files.phcomp.co.uk

2 1

a20 board and rk3288 board
by Luke Kenneth Casson Leighton 27 Feb '17

27 Feb '17

ok so the a20 board's nand ic is recognised in software... but *not* by the eGON bootloader ROM: it's too modern. i'm going to have to find an older (lower-capacity) TSSOP-48 NAND IC. arse. the rk3288 board is *almost* up: the i2c bus for the PMICs appears to be non-operational (reason yet to be determined but most likely to be a short-circuit) which, when the linux kernel tries to initialise each of the PMICs hangs on accessing the I2C bus. when i replaced the I2C pullup resistors with 1.5k instead of 2.2k then SD/MMC refused to boot! only thing i can think of is, the short is pulling the 3.3v power rail down enough to affect the SD/MMC. i'll keep investigating. l. --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

1 0

passthrough, a20 and rk3288 cards
by Luke Kenneth Casson Leighton 24 Feb '17

24 Feb '17

ok so still waiting for the updates to go out from crowdsupply, brief news in the meantime. revision 2.7.2 pre-production test pcb run has been ordered, i will be assembling those myself, here in taiwan, using my host's equipment. btw i didn't hear from anyone about the offer to send out pre-production cards. i'm having fun with the rk3288 board: putting on 600-pin 0.6mm pitch bga ICs is tricky but doable. the first board i managed to get access via USB, and learned that it is actually possible to initialise them by uploading u-boot-spl over USB. however the DDR3 RAM would only initialise if the clockrate was limited to 200mhz. the second board i tried, unfortunately the 3.3v power output was shorted to the DDR3 RAM 1.5v line and destroyed $20 worth of RAM ICs and a $20 processor. not doing that again. the third board i am currently trying, i have to get the ACT8846 properly soldered down: it's not initialising (yet) so all outputs measure 0v. the passthrough card i am still waiting for an adapter cable in order to be able to wire a 2.4.1 revision a20 card's HDMI output to the passthrough card's HDMI input. it has to be micro-hdmi to micro-hdmi hence why an adapter had to be ordered. l. --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

10 30

Re: [Arm-netbook] passthrough, a20 and rk3288 cards
by dumblob 22 Feb '17

22 Feb '17

Hi Luke, speaking of commitments, timeframes and the equipment price, it appears to me we could develop and provide a fully working QEMU image to break the current dependecy on EOMA HW (of course, in the end it will be tested on a real HW, but the global parallel development of anything regarding EOMA will be much easier and faster using QEMU). Would it make sense to you? Cheers, -- Jan

4 6

Re: [Arm-netbook] A20 deadline
by John Luke Gibson 19 Feb '17

19 Feb '17

>john i appreciate your support, do try to be... less forthright than >i am when focussing on things :) I am afraid :> Being overly forthright will forever be my crime It is my over compensating for a perceived blindness in the world. If an extraterrestrial ever saw through it, I would be thoroughly flushed and embarrassed xD

1 0

Re: [Arm-netbook] A20 deadline
by John Luke Gibson 19 Feb '17

19 Feb '17

I think you read this: >well, thanks to some questioning last month we worked out a way to >increase (negotiate) power up to 10W, and i am working on a proposal / >concept to get an 8-core 64-bit RISC-V SoC produced. And, didn't read this thread: RK3288 PCB first prototype assembled. I hate to add, because it seems Luke was sufficiently crass, but I'd like to explain why it was justified and appropriate, even if it comes off as insult to injury, it is only meant to explain so truly harmful stress can be avoided in the future. When you are communicating on this mailing-list, you are not contacting customer support and you are not engaging with a corporation, a corporate entity, or any type of public relations entity. When/If you backed this project you did not back any person or organization. We are the organization, you included even if you are the metaphorical equivalent to a lazy employee just like I am [at least, I admit it]. By messaging on this mailing-list, you are not messaging to this organization, but-rather as a part of it, because this is an internal communication channel where we [this organization] are collaborating the planning and design of our thing. Luke, being the one whom holds the funds we have gathered to build our things, is simply a very pro-active member of this organization that has existed long before this campaign has, whom has become a trusted agent of this organization with the sole purpose of performing our grande finale in this epic. Once that role is complete Luke will ultimately return to being like the rest of us until the certification mark is registered, that is to say hopefully if it is successfully registered. We owe a lot to Luke, being the still unconventional and still unfortunately somewhat amorphous segment of the Linux community dedicated to hardware that we are, [and, that is the only definition for organization I will provide here] Luke certainly seems on the path to fulfilling such prerequisites as would complete the mandate of such temporal authority, and perhaps deserving it, as only history can tell. This organization will still exist even after this grande finale which now should seem as ever impending as it ever has been, and as such it is only natural that we talk about the hereafter of this whole project. Luke, making mention of a proposal to build a RISC-V SoC was not supposed as recommendation for this project as even a "Lazy Employee" such as yourself [please excuse the metaphor] can obviously see that wouldn't work. There are still many more epics to be written after this one, and, with the temporal authority supposed to be placed in Luke's hands shortly hereafter [re-read that until you get the pun and can laugh, because you won't understand the sentence until you get the pun; it happens with the word suppose], I can trust that it will be a great one with great OpenCores. Yes, Luke must be accountable being trusted with these funds that the fellow is, but this type of questioning now is an insult because we are supposed to be doing this for ourselves. Luke's deadline is soon, but we, being that we are doing this project for ourselves, have no deadline especially being that the sooner we rush this the less work and the more likely to fail we are. The deadline here is partly on us to help Luke build these boards the best we can remotely, and, if Luke were to fail to create a board soon after we develop a promising prototype, then we question Luke and hold accountable thereof. That very clearly hasn't happened and, in fact, the opposite has very clearly happened and, if Luke, in celebration of completing part of the trust we had in him, sings song of the next epic we shall write and you question it in this way that you have, you are not only insulting Luke severely, you are identifying yourself very opaquely as the colloquial "Lazy Employee". Please pay attention.

2 1

New Eoma68 housing idea
by HeadsUpHigh 18 Feb '17

18 Feb '17

Hi, I think this indiegogo project has a very interesting form factor. https://www.indiegogo.com/projects/gpd-pocket-7-0-umpc-laptop-ubuntu-or -win-10-os-laptop--2#/ I think this kind of form factor would be ideal for a new Eoma68 housing. Especially if there is an m.2 slot inside the housing for extra storage( via usb 3.1 to sata3), this would make for a really good extremely portable computer with the added bonuses of the upgradability of the eoma68 standard as well as no space wasted for active cooling( like that machine has).

7 19

RK3288 PCB first prototype assembled
by Luke Kenneth Casson Leighton 17 Feb '17

17 Feb '17

http://rhombus-tech.net/rock_chips/rk3288/news/ 2 updates from crowdsupply in the pipeline, i'll cf them later. also will update the latest news tomorrow when i have a chance to swap the USB cable over. all rather hilarious: i wired the OTG pins to the 2nd USB port and vice-versa. whoops. --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

1 0

A suggestion why Systemd may be bad
by John Luke Gibson 16 Feb '17

16 Feb '17

Perhaps it is the idea that a linux machine should be wholly modular and attaching a library to a critical component of the system, shouldn't be a viable strategy for popularizing one's work. When a distro is forced to carry a package due to a dependency of a dependency, or any magnitude there of, it breaks a core separation of power there. The users depend on distro's to provide reliable packages, however if a package is intentionally interweaving files to make these dependencies simply a part of the file and therefore robbing the distro's the ability to choose a different dependency should another developer or team thereof prove more reliable or more suited for their distro. Systemd in this sense would be like microsoft robbing those wishing to distinguish themselves of the ability by increasing the magnitude of difficulty in doing so. Now, keep in mind, I am not fluent in any programming language and have not audited Systemd, nor do I know anyone who has. This is based on a compiled understanding of observations expressed in arguments both infavor and against Systemd.

5 11

Re: [Arm-netbook] A suggestion why Systemd may be bad
by John Luke Gibson 15 Feb '17

15 Feb '17

Sorry, I just read your reply Mike. I think that answers any curiosity I had. I appreciate it very much. This is makes sense as I imagine it to be part of the problem GNU/Mach is trying to solve (separating components of the system better, and with more micro-managed permissions?). Anyway, above and beyond explaining, both Mike and Philip.. Thx!

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

arm-netbook