- arm-netbook - Mailman test

Re: [Arm-netbook] Side-Topic: Liberating PocketCHIP
by ronwirring＠Safe-mail.net 08 May '17

08 May '17

Is it common to do something like this against a person? -------- Original Message -------- From: Bill Kontos <vkontogpls(a)gmail.com> Apparently from: arm-netbook-bounces(a)lists.phcomp.co.uk To: Linux on small ARM machines <arm-netbook(a)lists.phcomp.co.uk> Subject: Re: [Arm-netbook] Side-Topic: Liberating PocketCHIP Date: Mon, 8 May 2017 12:14:50 +0300 > Verhaegen is one of those selected individuals who had the luxury of getting all the shit of the world thrown at their face for trying to do the right thing. He was one of the leaders in pushing amd into mainlining gpu drivers( which they have been successful to and keep working on) and got shit for that. He attempted to reverse engineer the arm mali drivers( look up lima driver) and he succeeded to some extend, then he run out of money becaue nobody was willing to help him and arm has put significant effort into destroying his life. The nda a company has to sign for getting the mali drivers requires 0 interaction with his work, therefor no company can hire him now. > > On May 8, 2017 8:02 AM, <doark(a)mail.com> wrote: > > > > I apologize for DOS'ing the list, I can only get online about once a week. > > > > On Thu, 4 May 2017 17:13:23 +0200 > > "mike.valk(a)gmail.com" <mike.valk(a)gmail.com> wrote: > > > > > > 2017-05-04 9:04 GMT+02:00 John Luke Gibson <eaterjolly(a)gmail.com>: > > > > > > > Since it seems like a trivially simple task that for some reason no > > > > one has taken up, I would like to take the opportunity to exercise a > > > > learning experience and simultaneously benefit the community, by > > > > liberating PocketCHIP by deblobbing the source and re-compiling. > > > > > > > > > > The PocketCHIP is powered by their SoM: > > > http://linux-sunxi.org/NextThingCo_CHIP > > > > > > That is apparently a Allwinner R8 pared with an external rtl8723bs > > > Wifi/BT chip. > > > > > > The R8 is a rebranded A13. > > What? I own one of those and I'm almost certain that the CPU is an A7. > > Let's boot the PocketCHIP up... > > The processor is detected as an A7. > > I'll attach the output, it would probably be interesting to see all of > > it... > > Done, it's compressed bzip2 since it's ~300KiB decompressed which is large > > for an email. > > > > <aside> > > I'm not too clever with gpg yet, so please tell me if the compressed file > > is also signed (I wanted to do that so that you could be certain that it > > was not infected en-route like happens to MS .cab files far to > > frequently (even though I don't have my key signed by anyone yet > > \me grumble grumble)). > > </aside> > > > > Unless your saying that the WiFi has a built in ARM R8 (Why)? which would > > really surprise me considering how large the processor chip is compared > > with the WiFi chip. > > > > > > > If you look at > > > https://getchip.com/pages/chip > > > > > > You'll see: > > > On the left image the RAM (Hynix) and Wifi+BT (Realtek) and Power module > > > (Allwinner AXP209) > > > On the right the SoC (R8/A13) and NAND (Samsung) > > > > > > The A13 does not need blob's to run anymore, the WiFi+BT chip does. > > > AFAIKT > > > > > > Display output needs some checking in Linux and U-boot mainline. But > > > most should be available or somewhat easily hacked in. > > > > > > GPU needs a BLOB which does not work on mainline AFAIKT. Luc Verhaegen > > > did get quite far before he burned out. > > <snip> > > > > You're not giving us enough details. Who is Verhaegen? What did he burn > > out on? > > When I first considered purchasing a PocketCHiP I read about the GPU > > not having 3D capabilities because of a binary blob. So, the CHIP folks > > hired (I think it was an extended goal of the kickstarter campaign), a > > kernel dev to add support to the Linux kernel for the GPU. I was going to > > mention this on this list before, but it's been so active... > > > > Sincerely, > > David > > > > _______________________________________________ > > arm-netbook mailing list arm-netbook(a)lists.phcomp.co.uk > > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > > Send large attachments to arm-netbook(a)files.phcomp.co.uk > >

3 2

Re: [Arm-netbook] firefly 3399 all source software disclosed?
by ronwirring＠Safe-mail.net 08 May '17

08 May '17

All software for the mali-t860 is open source? -------- Original Message -------- From: Luke Kenneth Casson Leighton <lkcl(a)lkcl.net> Apparently from: arm-netbook-bounces(a)lists.phcomp.co.uk To: Linux on small ARM machines <arm-netbook(a)lists.phcomp.co.uk> Subject: Re: [Arm-netbook] firefly 3399 all source software disclosed? Date: Sun, 7 May 2017 14:06:18 +0100 > --- > crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68 > > > On Sun, May 7, 2017 at 10:53 AM, <ronwirring(a)safe-mail.net> wrote: > > https://www.kickstarter.com/projects/1771382379/firefly-rk3399-six-core-64-… > > Has all source code for the computer been disclosed? > > as far as i was able to ascertain when i last checked: yes. u-boot > and linux kernel, no proprietary pieces at all. > > > Lkcl has says, he can make use of the rk3399 if he has the processor's data sheet. > > i already have the datasheet. it's the reference design that i need. > > > What data about the rk3399 does lkcl not have? > > i need a reference design. full schematics, full pcb layout. > > > What is the performance of the rk3399 processor compared to a x86 dual core 2ghz? > > the rk3288 already outperforms a dual-core intel atom. > > l. > > _______________________________________________ > arm-netbook mailing list arm-netbook(a)lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook(a)files.phcomp.co.uk

6 11

Re: [Arm-netbook] libre 64-bit risc-v SoC
by doark＠mail.com 08 May '17

08 May '17

I apologize for DOS'ing the list, I can only get online about once a week. On Fri, 28 Apr 2017 13:58:57 +0100 Luke Kenneth Casson Leighton <lkcl(a)lkcl.net> wrote: > > On Fri, Apr 28, 2017 at 12:47 PM, mike.valk(a)gmail.com > <mike.valk(a)gmail.com> wrote: > > > If you're trying to trans-code something that you don't have a > > co-processor/module for you're forced to CPU/GPU trans-coding. > > you may be misunderstanding: the usual way to interact with a GPU is > to use a memory buffer, drop some data in it, tell the GPU (again via > a memory location) "here, get on with it" - basically a > hardware-version of an API - and it goes an executes its *OWN* > instructions, completely independently and absolutely nothing to do > with the CPU. > > there's no "transcoding" involved because they share the same memory > bus. > > > Would a FPGA > > still be more power huns gry then? > > yes. > > > I think/hope FPGA's are more efficient for specific tasks then > > CPU/GPU's > > you wouldn't give a general-purpose task to an FPGA, and you wouldn't > give a specialist task for which they're not suited to a CPU, GPU _or_ > an FPGA: you'd give it to a custom piece of silicon. I always thought that FPGA's were good for prototyping or small fast tasks... But that's just how I learned about them. > in the case where you have something that falls outside of the custom > silicon (a newer CODEC for example) then yes, an FPGA would *possibly* > help... if and only if you have enough bandwidth. > > video is RIDICULOUSLY bandwidth-hungry. 1920x1080 @ 60fps 32bpp > is... an insane data-rate. it's 470 MEGABYTES per second. that's > what the framebuffer has to handle, so you not only have to have the > HDMI (or other video) PHY capable of handling that but the CODEC > hardware has to be able to *write* - simultaneously - on the exact > same memory bus. > <snip> Your number seemed off to me so I did the math: 1920*1080*60*4 == 497,664,000 You're off by almost 30 MiB. Most video cameras (that I've been able to locate), do 24bpp, 640x480 at 30fps, so that would make the bandwidth requirements. 27,648,000 Which should be more reasonable for an FPGA (not that I have all the specs sitting in front of me, mind you). I am assuming that "encoding video" means encoding from a video camera or a small youtube video as opposed to encoding to send to a device over, say, an HDMI cable. > > > We can always have evolution create a efficient decoder ;-) > > https://www.damninteresting.com/on-the-origin-of-circuits/ > <snip> Sincerely, David

2 1

Re: [Arm-netbook] The open risc v idea is good
by doark＠mail.com 08 May '17

08 May '17

I apologize for DOS'ing the list, I can only get online about once a week. On Wed, 3 May 2017 11:38:26 +0100 Luke Kenneth Casson Leighton <lkcl(a)lkcl.net> wrote: > On Wed, May 3, 2017 at 11:28 AM, zap <zapper(a)openmailbox.org> wrote: > > > > > > On 05/03/2017 01:01 AM, Luke Kenneth Casson Leighton wrote: > >> --- > >> crowd-funded eco-conscious hardware: > >> https://www.crowdsupply.com/eoma68 > >> > >> > >> On Tue, May 2, 2017 at 10:50 PM, zap <zapper(a)openmailbox.org> > >> wrote: > >>> I assume that's going to be the third series of libre cards right? > >>> you making a lowrisc based processor and graphics, etc, > >> more likely the fourth or fifth. all these things happen in > >> parallel. the one after the RK3288 to investigate is the RK3399 > >> (6-core). also i've heard that the iMX7 is out so am having a look > >> at that. > > > > RK3399... 11.0 ghz speed if you count all the four cores which are 2 > > ghz each and 1.5 for the other ones... > > > > wow... > > > > I wonder how much watts that processor uses... > > > > that is crazy fast. I mean really... 4GB ram too I see as the norm. > > which is good. > > > > on a different note though, perchance will the fourth or fifth series > > of libre cards support 8gb of ram or more? > > > > I bet you could even get it to 16gb if you wanted... though that > > wouldn't be needed. heh. > > the cost of the RAM ICs to do that are insane. the 1GB DDR3x16 ICs > are already $10 **EACH**. > > normally a DIMM would have 8 or 16 RAM ICs, meaning that for a 4GB > DIMM you need 8x 512mb ICs, or you could do 8GB by using 16 of them. But that would mean that an 8GiB stick for a desktop would cost $80 just for the ICs. I can find 8GiB of RAM for $60 and cheaper, I bet, if I poked my nose around. Or maybe desktops don't need ICs??? > normally you get 2 (matched) DIMMS totalling 8GB. > > if you want *32* GB you get 2 matched DIMMs, but nobody in > "mass-production" is shipping windows PCs or laptops with 32GB of RAM > (and if they are it's DDR4) You write that as though being dismissive of DDR4. Why is DDR4 currently unacceptable for one of your designs? > additionally, it would need either 2 pairs of x32 DDR RAM > controllers, or it would be necessary to do a cascade layout: a > daisy-chain of RAM ICs.... and i'm not sure it would be possible to > fit 8 RAM ICs onto the 43x78mm PCB anyway. > > bottom line it's far too expensive and quite impractical. 4GB is the > practical limit at the moment and even that's stupidly expensive > relatively speaking. > > l. Well, I bought a 4GiB RK3399 SOC card fro firefly for $200. Don't get me wrong, I love your effort, but I need RAM, CPU power, lots of ports, and more screen landscape. Hmm, maybe you should invent an OpenSource IC? Sincerely, David

2 1

Re: [Arm-netbook] libre 64-bit risc-v SoC
by David Niklas 08 May '17

08 May '17

I apologize for DOS'ing the list, I can only get online about once a week. On Fri, 28 Apr 2017 12:35:19 +0200 "mike.valk(a)gmail.com" <mike.valk(a)gmail.com> wrote: > > 2017-04-27 13:21 GMT+02:00 Luke Kenneth Casson Leighton >> <lkcl(a)lkcl.net>: > > > ok so it would seem that the huge amount of work going into RISC-V > > means that it's on track to becoming a steamroller that will squash > > proprietary SoCs, so i'm quite happy to make sure that it's > > not-so-subtly nudged in the right direction. > > > > i've started a page where i am keeping notes: > > http://rhombus-tech.net/riscv/libre_riscv/ and the general goal is to > > create a desirable mass-volume low-cost SoC, meaning that it will need > > to at least do 1080p60 video decode and have 3D graphics capability. > > oh... and be entirely libre. > > > > the plan is: > > > > * to create an absolute basic SoC, starting from lowRISC (64-bit), > > ORGFX (3D graphics) and MIAOW (OpenCL engine), in at least 90nm as a > > low-cost proof-of-concept where mistakes can be iterated through > > * provide the end-result to software developers so that they can have > > actual real silicon to work with > > * begin a first crowd-funding phase to create a 28nm (or better) > > multi-core SMP SoC Links? > > for this first phase the interfaces that i've tracked down so far are > > almost entirely from opencores.org, meaning that there really should > > be absolutely no need to license any costly hard macros. that > > *includes* a DDR3 controller (but does not include a DDR3 PHY, which > > will need to be designed): > > > > * DDR3 controller (not including PHY) > > * lowRISC contains "minion cores" so can be soft-programmed to do any > > GPIO > > * boot and debug through ZipCPU's UART (use an existing EC's on-board > > FLASH) > > > > Perhaps put it sirectly to an USB bridge. UART's on debugging hardware > is non existant. We all use FTDI dongles. > > Look like OpenCores has a module. https://opencores.org/project,usb2uart > > > > > * OpenCores VGA controller (actually it's an LCD RGB/TTL controller) > > * OpenCores ULPI USB 2.0 controller > > * OpenCores USB-OTG 1.1 PHY > > > > note that there are NO ANALOG INTERFACES in that. this is *really* > > important to avoid, because mixed analog and digital is incredibly > > hard to get right. also note that things like HDMI, SATA, and even > > ethernet are quite deliberately NOT on the list. Ethernet RMII (which > > is digital) could be implemented in software using a minion core. the > > advantage of using the opencores VGA (actually LCD) controller is: i > > already have the full source for a *complete* linux driver. Considering that analog was around *long* before digital I'm surprised that it is "Hard to get right", is there a reason for this? Isn't there a chip for just this kind of thing? > > I2C, SPI, SD/MMC, UART, EINT and GPIO - all of these can be > > software-programmed as bit-banging in the minion cores. > > > > these interfaces, amazingly, are enough to do an SoC that, if put into > > 40nm, would easily compete with some of TI's offerings, as well as the > > Allwinner R8 (aka A13). > > > > i've also managed to get alliance and coriolis2 compiled on > > debian/testing (took a while) so it *might* not be necessary even to Hmm, I can't seem to google that piece of SW. Do you have a link? > > pay for the ASIC design tooling (the cost of which is insane). > > coriolis2 includes a reasonable auto-router. i still have yet to go > > through the tutorials to see how it works. for design rules: 90nm > > design rules (stacks etc.) are actually publicly available, which > > would potentially mean that a clock rate of at least 300mhz would be > > achievable: interestingly 800mhz DDR3 RAM from 2012 used 90nm > > geometry. 65 down to 40nm would be much more preferable but may be > > hard to get. How would you get it in the first place? Is there a company dedicated to larger than industry standard (nm) silicon production for small businesses, or are you planning to buy a ... what would it be called? ... printed wafer producer? > > graphics: i'm going through the list of people who have done GPUs (or > > parts of one). MIAOW, Nyuzi, ORGFX. the gplgpu isn't gpl. it's been > > modified to "the text of the GPL license plus an additional clause > > which is that if you want to use this for commercial purposes then... > > you can't". which is *NOT* a GPL license, it's a proprietary > > commercial license! > > > > MIAOW is just an OpenCL engine but a stonking good one that's > > compatible with AMD's software. nyuzi is an experimental GPU where i > > hope its developer believes in its potential. ORGFX i am currently > > evaluating but it looks pretty damn good, and i think it is slightly > > underestimated. i could really use some help evaluating it properly. > > my feeling is that a combination of MIAOW to handle shading and ORGFX > > for the rendering would be a really powerful combination. <snip> What about Vulkan? I'm asking, because it is multithreaded, as opposed to OpenGL. I've also heard, though perhaps the person was wrong, that it is supposed to replace OpenCL.

2 1

Re: [Arm-netbook] libre 64-bit risc-v SoC
by doark＠mail.com 07 May '17

07 May '17

I apologize for DOS'ing the list, I can only get online about once a week. On Fri, 28 Apr 2017 19:54:43 +0200 Hannes Schnaitter <hannes(a)schnaitter.de> wrote: > Hi, > > I've done some research into this in the last couple of weeks. > > Am Fri, 28 Apr 2017 20:21:00 +0300 > schrieb Bill Kontos <vkontogpls(a)gmail.com>: > <snip> > I'm sad to say that it isn't even close to solved. The only two > concrete ideas I found are: > 1. Egoism. The driver of the car always wins. > 2. Utilitarianism. "The Greater Good". The best outcome for the most > people. > > There is also another one which ignores most of the problem. > 3. Random. Creating different possible crash cenarios and selecting one > at random. > <snip> > > Then there is the problem of who would by utilitarian cars. [2] > > Greetings > Hannes I would. It's what I'd do in real life. I've had similar choices before. I have no regrets about placing my own life on the line either. Sincerely, David

1 0

firefly 3399 all source software disclosed?
by ronwirring＠Safe-mail.net 07 May '17

07 May '17

https://www.kickstarter.com/projects/1771382379/firefly-rk3399-six-core-64-… Has all source code for the computer been disclosed? Lkcl has says, he can make use of the rk3399 if he has the processor's data sheet. https://www.kickstarter.com/projects/1771382379/firefly-rk3399-six-core-64-… What data about the rk3399 does lkcl not have? What is the performance of the rk3399 processor compared to a x86 dual core 2ghz?

3 2

More Libre Hardware Getting Funded
by Allan Mwenda 06 May '17

06 May '17

https://www.crowdsupply.com/gnubee/personal-cloud-1 Always lovely to see this happening.

5 12

The open risc v idea is good
by zap 05 May '17

05 May '17

I assume that's going to be the third series of libre cards right? you making a lowrisc based processor and graphics, etc, I am sure you can succeed. I look forward to still getting the second revision still though. :)

4 12

libre 64-bit risc-v SoC
by Luke Kenneth Casson Leighton 04 May '17

04 May '17

ok so it would seem that the huge amount of work going into RISC-V means that it's on track to becoming a steamroller that will squash proprietary SoCs, so i'm quite happy to make sure that it's not-so-subtly nudged in the right direction. i've started a page where i am keeping notes: http://rhombus-tech.net/riscv/libre_riscv/ and the general goal is to create a desirable mass-volume low-cost SoC, meaning that it will need to at least do 1080p60 video decode and have 3D graphics capability. oh... and be entirely libre. the plan is: * to create an absolute basic SoC, starting from lowRISC (64-bit), ORGFX (3D graphics) and MIAOW (OpenCL engine), in at least 90nm as a low-cost proof-of-concept where mistakes can be iterated through * provide the end-result to software developers so that they can have actual real silicon to work with * begin a first crowd-funding phase to create a 28nm (or better) multi-core SMP SoC for this first phase the interfaces that i've tracked down so far are almost entirely from opencores.org, meaning that there really should be absolutely no need to license any costly hard macros. that *includes* a DDR3 controller (but does not include a DDR3 PHY, which will need to be designed): * DDR3 controller (not including PHY) * lowRISC contains "minion cores" so can be soft-programmed to do any GPIO * boot and debug through ZipCPU's UART (use an existing EC's on-board FLASH) * OpenCores VGA controller (actually it's an LCD RGB/TTL controller) * OpenCores ULPI USB 2.0 controller * OpenCores USB-OTG 1.1 PHY note that there are NO ANALOG INTERFACES in that. this is *really* important to avoid, because mixed analog and digital is incredibly hard to get right. also note that things like HDMI, SATA, and even ethernet are quite deliberately NOT on the list. Ethernet RMII (which is digital) could be implemented in software using a minion core. the advantage of using the opencores VGA (actually LCD) controller is: i already have the full source for a *complete* linux driver. I2C, SPI, SD/MMC, UART, EINT and GPIO - all of these can be software-programmed as bit-banging in the minion cores. these interfaces, amazingly, are enough to do an SoC that, if put into 40nm, would easily compete with some of TI's offerings, as well as the Allwinner R8 (aka A13). i've also managed to get alliance and coriolis2 compiled on debian/testing (took a while) so it *might* not be necessary even to pay for the ASIC design tooling (the cost of which is insane). coriolis2 includes a reasonable auto-router. i still have yet to go through the tutorials to see how it works. for design rules: 90nm design rules (stacks etc.) are actually publicly available, which would potentially mean that a clock rate of at least 300mhz would be achievable: interestingly 800mhz DDR3 RAM from 2012 used 90nm geometry. 65 down to 40nm would be much more preferable but may be hard to get. graphics: i'm going through the list of people who have done GPUs (or parts of one). MIAOW, Nyuzi, ORGFX. the gplgpu isn't gpl. it's been modified to "the text of the GPL license plus an additional clause which is that if you want to use this for commercial purposes then... you can't". which is *NOT* a GPL license, it's a proprietary commercial license! MIAOW is just an OpenCL engine but a stonking good one that's compatible with AMD's software. nyuzi is an experimental GPU where i hope its developer believes in its potential. ORGFX i am currently evaluating but it looks pretty damn good, and i think it is slightly underestimated. i could really use some help evaluating it properly. my feeling is that a combination of MIAOW to handle shading and ORGFX for the rendering would be a really powerful combination. so. it's basically doable. comments and ideas welcome, please do edit the page to keep track of notes http://rhombus-tech.net/riscv/libre_riscv/ --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

10 36

lian dong san wei 3d printer
by Luke Kenneth Casson Leighton 01 May '17

01 May '17

http://rhombus-tech.net/3d_printers/lian_dong_san_wei/ starting some build instructions and also noting which parts are missing, what improvements can be made. so far i have the entire frame done and it's reaasonably stable but not as rigid as i was expecting: that's going to be down to the use of "interior" L-shaped brackets that go *inside* the 20x20 runners, as opposed to the use of triangular corner-braces which i know would make the entire thing as stiff as a... stiff thing. i've only made one assembly mistake, which is on the x-axis: the hot-end holder with the linear bearings in it is upside-down, which results in the belt being subtly shifted upwards by about 4mm.. out of true. that means that the belt is pointing at weird angles, being slackest when the print head is in the centre and under high tension *and a different length* at the edges. arse. it means a full disassembly of the z-axes *and* the x-asis sub-assemblies, and recalibration. arse. also... *sigh*... there are certain design decisions made by chris palmer (creator of the mendel90) which, when you look at other reprap-style 3d printers you just... smack your head in dismay. josef prusa, who is unfortunately worshipped for being one of the first to actually sell 3d printers, is not someone who can be said to actually be capable of sound engineering judgement, and this printer is unfortunately "inspired" by his efforts... and therefore has *vertical* x-axis rods (supporting the print-head). what that means is that the weight of the print-head *twists* the rods. i'm holding the end of the nozzle and moving it _gently_ around... and it's wobbling by about TWO millimetres. soo.... yeah. this is undergoing a redesign before committing to 10 more. also i can take the opportunity to reduce the "lever" effect (the printhead being a long way away from the middle point of the 2 x-axis rods). l. --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

1 0

first prototype microdesktop casework
by Luke Kenneth Casson Leighton 30 Apr '17

30 Apr '17

http://rhombus-tech.net/community_ideas/micro_desktop/news/ --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

5 9

Another set of EOMA 68 logos
by dumblob 28 Apr '17

28 Apr '17

Hi everybody, few ideas I spotted on the so far submitted logos inpired me to create also some proposals. All logos are carefully designed to meet the following requirements. * Support B/W rendering "out of the box" (just use e.g. the indexing algorithm in Gimp) * Be decently well readable * Be moderately intuitively understandable * Be compact in size, but not overly dense * Stay consistent (all components of the same importance shall have roughly the same visual area, size, color, and "shape kind") * Support future numbers in place of 68 * Favicon shall well resemble the logo (in this case the favicon will be a green character E) * The minimum safety zone does not need to be very big to let the logo be readable All the material I'm attaching is licensed under CC BY 4.0 . Unfortunately I again ran out of my time to contribute to this interesting project. I might find some time later (in few months rather than weeks). So feel free to modify the logos to your needs. There is one more idea from my side - add a leaf of a lime tree (as it looks a bit like a heart) to make it less "technically" looking and more resembling the eco-friendliness. Note, the attached bitmap exports are not perfect as the source SVG files are prepared for easy editing, but not for high-quality bitmap exporting with antialiasing. The source SVG files are Inkscape SVG files, so they are not optimized for viewing. Cheers! -- Jan

2 1

Re: [Arm-netbook] SoC warp soldering risk?
by Luke Kenneth Casson Leighton 28 Apr '17

28 Apr '17

--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68 On Sat, Apr 22, 2017 at 11:53 AM, <sokolgeo(a)posteo.net> wrote: > Hello. It seems like NEO900 project (http://neo900.org/) is having some > problems with SoC warp soldering memory chips to their PCB: > > http://talk.maemo.org/showpost.php?p=1525427&postcount=2850 > > https://irclog.whitequark.org/neo900/2017-03-14#19041338 specific details: http://lists.goldelico.com/pipermail/community/2017-February/001583.html http://lists.goldelico.com/pipermail/community/2017-February/001594.html > Luke - is there similar soldering risk in EOMA manufacturing? fuck no :) whenever i see "PoP" as an option for a processor i run away very very fast :) it's also why i haven't done a smart (or dumb) phone, because *without* that smaller size (due to "stacking") you simply can't make a small enough PCB for it to be socially acceptable, these days. that's why i was so excited when i found the Ingenic M150... and so disappointed when they told me they'd stopped making it. why? because the M150 has 128mb of DDR2 RAM *on board* the same 11x11mm package, as a Multi-Chip-Module. instead i've focussed on EOMA68 as the first modular design, which is *just* big enough to fit processor and RAM and PMICs onto one side of the PCB., and in some cases big enough to fit NAND or eMMC on as well. so no - the primary reason why i avoid PoP RAM is because of the MOQ quantities usually being around 1,000. *sigh* another expensive learning experience for the open / libre hardware community, there. i do wish sometimes that they'd read some of the things that i write. i've *already learned* that you don't try retro-fitting PCBs into pre-existing casework. never mind - sometimes it's better for people to do things they won't listen on and then fail. l.

5 11

SPI-based LCDs, 3D printing, RISC-V
by Luke Kenneth Casson Leighton 28 Apr '17

28 Apr '17

ok so a quick status update: * Frida, in Shenzhen, have a customer currently ordering their 2.8in FRD280J3703D SPI-based LCD. this means it's possible to place an order for less than the normal MOQ of 1,000 so we go with that one for the 15in laptop. * i calculated that for the microdesktop i have an ENTIRE MONTH of 3d-printing to do, and a YEAR for the laptop. there's an update going out about this soon but basically i need help. ideas or actual practical help. * i'm investigating RISC-V *long-term*, i will start a separate thread about this (immediately after writing this) so am reaching out to various people to see who's interested. l. --- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

11 26

Re: [Arm-netbook] System 76 - Entering Phase 3
by Bill Kontos 25 Apr '17

25 Apr '17

No, don't worry. I don't call names on anyone. As RMS said, I'm not glad that he died, but I'm glad that he's gone. I believe in the value of human life above all else. It would be immoral( imo) to apply crude logic to this. I do not celebrate the death of any human being no matter how savage or harmful that person might have been to humanity. And I sure won't celebrate Linus' death because you might hate him but it won't change the fact that we 100% rely on his work and we don't know how his death could affect future development of linux and free drivers in general. I guess it is a matter of how each individual thinks: For me it makes no sense to discuss how something will work if we can't get it to work our way in the first place. In other words: computing will be done and programs will be written, the question is do you want to be a part of the procedure and help do it your way or just sit in the corner waiting for the perfect solution to come up before you commit to it ? Because that solution might not come. Other people are much more idealists than me. That's fine. We need both types of people. But before you decide to completely ignore system76 and their efforts, I need to point out that they have indeed supplied a lot of code and have benefited the libre hardware community in general a lot. I have been looking at their history and various interviews of them and from what I've gathered: 1) They have contributed code both to Ubuntu specific packages( installer, unity) and to the kernel, that is writing DACs( as an example given) and pushing companies for open sourcing drivers. In fact if there is one that takes advantage of their effort it is Dell who according to Daniel Fore( one of their employees) takes advantage of that hardware on their Xps Developer's edition. 2) They do seem to like the cooperative nature of FOSS. At least their blog posts do show so. System76 started as a 2 man business installing linux on clevo laptops. Then they started expanding the range of their activities by working on drivers and paying a lot of attention to the details. The "purists" right now are probably thinking that the main goal is to just get it working in a libre way, but high quality is actually important too. So it might just be that they consider shipping a 100% libre computer that will be useful to their demographic something out of their reach as of right now. And when it comes to laptops it's not like they have many choices, it's intel for the cpus and nvidia for the gpus. Amd is non existent and arm for laptops of the class they are selling is still not a thing. 3) They are the biggest customer of Clevo and as such their word holds a lot of value and they do push towards making a lot of linux friendly decisions. See this comment https://www.reddit.com/r/System76/comments/611zf5/question_how_much_control… And anyway something that is not widely known: when arm netbooks started becoming a thing OEMs were putting linux on them( of course). Microsoft saw a very threatening market emerging and raged hell upon them that if they kept doing this the sales of their other models would suffer due to them removing the OEM discount on the windows license. After that they proceeded to change the model so instead of charging per machine they charge per model. So basically even if you buy e.g. a Dell Lattitude with Ubuntu preinstalled and then wipe it and replace it with an FSF approved distro YOU ARE STILL PAYING THE WINDOWS LICENSE. Now some people might disagree with me, but for me libre software is a war against oppression and is directly competing with microsoft etc( disagree as in that we do not compete but exist to fulfeel our own needs). So just for this reason, if you need to buy an x86 computer and a libreboot model is not an option it is better to buy a system76 instead of e.g. a dell or a thinkpad and then install gnu/linux on it. Anyway I guess there is also another big difference for me: Some people might think it is ok to have a binary blob e.g. on the firmware for their wifi card as long as you don't interact with it via userland and the OS and the applications are libre. I'm on the exact opposite position: I consider blobs and an operating system that limit or obscure my access to hardware to be of the most unethical nature and less so the programs themselves. Because I have paid for the hardware but an application might come for no cost, therefor I do not have any demands from the author of that application. Will I run it if it's not free ? probably not. But at the end of the day you do pay with your wallet and for that reason I am not e.g. buying nvidia hardware, even though it's the only 100% libre gpu with 3d accel. I don't care. I'd much rather support amd in their free driver efforts and instead hope that someone will pick the fact that they do not need to write any gallium3D stuff but just reverse engineer the (small) blobs that still exist, resulting in a much easier way to write a 100% libre driver compared to nuveau. My point is, do I think System76 is perfect ? No. Do I like what they do ? Yes I like it a lot. Do I think they could be more vigilant with their interactions with other companies when it comes to free drivers ? Absolutely. But I think it is a better solution than anything else( or at least has been until purism dropped the ME neutralization bomb on skylake a bit earlier). But at least give some respect to a company that started putting gnu/linux on laptops on freaking 2005 and managed to survive and thrive. Maybe because their choices of including some proprietary software actually made it possible. Heck, even purism which as a company started exactly with the purpose of selling 100% libre computers had to put up with the intel ME for some generations until they finally found a way to disable it. So now you can actually buy a skylake laptop that has the ME neutralized. Maybe if they did not attempt a 'half-assed" effort first they would never get to that point. On Sat, Apr 22, 2017 at 5:32 PM, Lyberta <lyberta(a)lyberta.net> wrote: > Bill Kontos: > > I'm tired of all the purists that will bash every company that does a > > step in the right direction but does not go all the way through, even > when > > that company was started somewhere around 2005 in a guy's basement when > the > > ecosystem was much less mature and they had almost no funding at all( or > > because they call it linux and not GNU/Linux). > > And I'm tired of companies exploiting hard work of the GNU project and > Free Software movement. I'm tired of the Open Source movement. I'm tired > of Linux, I'm tired of Linus Torvalds. He puts blobs in the kernel and > calls it acceptable. In the end we get tons of products that are > completely unusable. Check any Android phone and you'll that it is full > of proprietary blobs for everything, that is completely unusable without > them. > > Every time you call the operating system Linux, you help Open Source > movement. You help Google, you help Linux foundation, you help evil > corporations. Linux was the most damaging thing that happened to the > Free Software movement because people started mixing the terms Free > Software and Open Source. People became unaware of 4 fundamental freedoms. > > I have celebrated the death of Steve Jobs and I'm going to celebrate the > death of Linus Torvalds. And before you start calling me names, let me > tell that I was officially declared mentally disabled. But if there is > one good thing that my mental disability gave me, it's the inability to > accept compromises. I value my 4 fundamental freedoms and I'm not going > to give them away without a fight. > > > _______________________________________________ > arm-netbook mailing list arm-netbook(a)lists.phcomp.co.uk > http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook > Send large attachments to arm-netbook(a)files.phcomp.co.uk >

9 33

Certification attempt
by Brandon Ludwig 23 Apr '17

23 Apr '17

Bear with me, because I'm not in any way qualified for this, but a while back I did some work based on previous discussion about certification marks.

1 0

Re: [Arm-netbook] System 76 - Entering Phase 3
by Luke Kenneth Casson Leighton 23 Apr '17

23 Apr '17

> While someone's at it, perhaps Think Penguin should be contacted. > Actually, I just checked and they seem to have invited Luke to their > table at HOPE, so they are well aware of and interested in the project. they're the sponsor of the 15.6in libre laptop, tor :) zap, john, lyberta: yeah i'm finding it tiring, and frickin hard work. but there's something else that occurred to me almost... i think it was over eight years ago, which spurred this entire project, which i believe will help. i think what you're feeling, especially you lyberta, is a sort of "futility" of the effectiveness of the free software movement precisely *because* it's focussed *on* software, exclusively. i mentioned this before, but it's worth reiterating: chris from thinkpenguin told me the story of why he started a hardware selling company: it's because one of his first jobs was for lindows^Wlinspire as a QA tester. unlike many of the other people in the company - including the other QA testers - he quickly and *comprehensively* realised that, unlike microsoft whose dominance of hardware forces peripheral manufacturers to write drivers *BEFORE* releasing the hardware, linux kernel drivers are written AFTER release... if at all, making linspire's entire business model a hopelessly optimistic one. (linspire basically aimed to be a "preinstall" option on hard drives shipped out to OEMs, just like microsoft has done for decades). from the ridiculousness of this attempt at the time (it was far too early basically) and with HDDs going out that comprehensively even failed to *boot* on OEM's hardware in many cases, chris *clearly* saw in ways that many other people would never get to see that the only way to sell working hardware was to *pre-vet the hardware*. contrast this to linspire's approach which was: "we naively expect all hardware to just... well... work". even apple don't do that. in their early days they sold pre-approved special hardware: you bought that, they wrote the drivers themselves, it worked. and if you didn't you were on your own. what they sold was pretty much everything you needed, and it wasn't long before hardware companies started making software drivers *BEFORE* releasing hardware for apple products as well. so chris *DOES NOT* have the same level of tired hopelessness and frustrating futility that other people are feeling in the software libre world. instead however he has a constant race to find hardware that can be freed from proprietary drivers: the AR9271 took *TWO YEARS* to patiently walk the entire company (atheros) through the process of releasing the source code, justified easily by the increased sales. the problem there was that when qualcomm bought atheros the entire management and engineering team left, leaving absolutely nobody for him to contact for the revision 10 (802.11ac) hardware. likewise: he could see that time is running out for the thinkpad X.200 older intel processors, and he was getting increasingly concerned at the lack of hardware companies producing laptop products that he could endorse (because they keep using AMD or intel with proprietary BIOS components). so that's when he contacted me, to sponsor the 15.6in libre laptop. in summary: we're getting tired because there *are* no good hardware options. reverse-engineering is... well it's a nice challenge but you always get *old hardware*, not designed by you, but designed by someone else. that's.... well, i don't have to spell it out. so for goodness sake: every person who is building open and libre hardware, back them to the hilt. the powerpc-notebook team, the vero-apparatus team (if you can find their contact details that is), anyone and everyone. l.

2 3

Mail service problems
by Alain Williams 23 Apr '17

23 Apr '17

On Sat, Apr 22, 2017 at 08:59:35PM +0100, Alain Williams wrote: > Apologies for poor mail service today. There have been disk problems. The > hosting company has been working since this morning to fix things - but there > are still some unexpected down-time/reboots. > > Thanks for your patience The machine is back up ... not quite 100% - there may be some bounces later. One of the main data stores became corrupt, I had to completely wipe and restore from backup. This affected all email and mail lists. You might find that some mail sent in the last 18 hours has got lost. If really interested, some more background to what went wrong here: https://status.bytemark.org/issues/170 -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php #include <std_disclaimer.h>

1 0

Re: [Arm-netbook] System 76 - Entering Phase 3
by Luke Kenneth Casson Leighton 23 Apr '17

23 Apr '17

On Sat, Apr 22, 2017 at 12:48 PM, Allan Mwenda <allanitomwesh(a)gmail.com> wrote: > I've asked a System76 staffer if they'd work with EOMA68 i emailed the founders yesterday after being told by sales(a)system76.com their email addresses. what i want to do is to offer them an opportunity to discuss and make them aware of *all* options that i know of. the powerpc-notebook team, and more besides. l.

1 0

Mail service problems
by Alain Williams 22 Apr '17

22 Apr '17

Apologies for poor mail service today. There have been disk problems. The hosting company has been working since this morning to fix things - but there are still some unexpected down-time/reboots. Thanks for your patience -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php #include <std_disclaimer.h>

1 0

System 76 - Entering Phase 3
by Mike Leimon 22 Apr '17

22 Apr '17

Luke, et.al. I saw this blog post a couple of nights ago: http://blog.system76.com/post/159767214983/entering-phase-three and I was thinking to myself that it might not be a half bad idea to try and contact these guys and see if they might be interested in making an EOMA-68 laptop and maybe even get into the business of designing EOMA-68 CPU cards. While I think that they have more traditional computer/laptops in mind for the future, I could also see the possibility of them offering one EOMA-68 laptop for sale along a number of other traditional style laptops (why shouldn't they hedge their bets?). Allow me to pull the most relevant quote from their post to support my theory: """ * Easy to work on and expand At every step along the way we ask, “How does this decision affect serviceability”. Open it, change it, expand it. Our product will be flexible. """ To me the EOMA-68 paradigm sounds like an incredibly good fit. Since it sounds like they are probably still a couple of years away from trying to launch a line of custom laptops. Right now might be the best time to try and bring this project to their attention... heck, if they are curious enough it might even be worth it to send them an A20 CPU card and one of the mini desktops (after they are completed of course). -Mike

9 15

About, the rk processor.
by zap 22 Apr '17

22 Apr '17

something interesting I saw is that in the update picking a processor, it shows rk3188 as the rockchip processor you were going to reverse engineer. on the other hand, your rhombus-tech link shows that your looking at rk3288? Not to be annoying constantly, but I am curious are you looking at one or both? I appolgize if there is something I may have missed.

4 8

Re: [Arm-netbook] Certification Mark: My hat in the Ring
by John Luke Gibson 21 Apr '17

21 Apr '17

Didn't notice that it didn't go through. Here's the attachment.

7 11

System 76 - Entering Phase 3
by Mike Leimon 21 Apr '17

21 Apr '17

Lyberta: > Couple of points that got my attention. They say Linux instead of > GNU/Linux. Open source instead of Free/Libre Software. They also > mentioned Ubuntu. This was enough to me to stop having any respect for > them. At least based on that post. While I understand what you are saying I don't see how having one of the most Linux friendly laptop manufacturers making a laptop based on the EOMA68 standard would be a bad idea. Let's say hypothetically, they decided to make the most awesome modular EOMA68 laptop possible, dual usb-c 3.1 and the works. Even if they did ship the system with a CPU card that was running Ubuntu on it, the whole EOMA68 standard works on the premise that if you can plug it in, it will work. Therefore, couldn't you just swap out out the CPU card and put any other compatible EOMA68 CPU card in there to enjoy the benefits of a well designed laptop and run your preferred distro instead? I understand that Ubuntu isn't your distro of choice (it isn't mine either) however, I think that there aren't many Linux friendly computing companies out there, let alone those that don't sell computers running anything other than Linux. I think that it is certainly important to have principles and to stick to them when possible however, when there seems to be as much common ground between what we want to accomplish and what they want to accomplish I am thinking it might be better to at least try working together than to hunt around for reasons to be divisive. Allan Mwenda: > Well i dunno, system76 would want to make the fastest card not necessarily the freest. I'm not so sure about this. I get the impression from reading some of their blogs and whatnot that they spend quite a bit of time and effort vetting the particular components they are putting into their systems to guarantee compatibility with their Linux distribution of choice. So while I do think they put some degree of importance into making sure that their systems have good performance, I do think they are aware of what having upstream driver support means for system components. I would agree that they would probably want to get the best performance out of a CPU card that they could however, I don't think that they would be too keen to do so at the cost of selecting a chip that didn't have some degree of upstream support. l: > i sort-of thought about this overnight, and i think you're right, > allan. the hardware they're selling is typically mid to high-end x86 > hardware [plus yes they talk about ubuntu] something like EOMA200 > would be a much better bet... but then so would many of the other > (industrial) modular standards for their needs. I suppose I still don't see the harm in bringing the EOMA68 concept up to them. The worst they could say is that they aren't interested. However, they could potentially be a pretty good ally. Also, I did notice that in the past few months they added a 96 core arm server to their lineup: https://system76.com/servers/starling Which means to me that they aren't completely wedded to solely x86 hardware. Even with their tendency to build higher end laptops and systems I think that an EOMA68 based laptop could easily exist within their line-up as an upgradable chromebook alternative (something priced in the $250-500 range). That would allow them to test the EOMA68 waters while at the same time allow them to build on and expand their higher-end line of x86 systems as well. I guess all I am trying to say is, that we won't know what a person or a company will do until we ask the question.

5 5

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

arm-netbook