Really cool modular and freedom-respecting computer


To loop that back, if you accept this device as RYF then you need to be OK with the fact that AllWinner is only in a position to offer it today because they violated the GPL. Most people will be OK with it, and that's fine, just don't pretend like it isn't "blood money" that made it possible.

in other words, it's about transforming something that was previously not ok through *your* positive intentions. if someone does something bad, and you can make it right by supporting someone that also wants to make it right, what would you do?
 
You don't accept things as RYF, they either are, or not, based on whether they respect your freedoms as a user.
Whether you are fine with allwinner products, is not something you need anything to decide, you can make up your own mind.
 
It would be great if the Pyra and EOMA68 communities could work together on software and such. It would also be useful to be able to pop the card out of one device and into another depending on where you are and what you are doing. I know it is a bit late for both projects to change directions to accommodate sharing the same card, but the software side might get a boost if they are both using the same SoC, so I recommend considering the OMAP 5.

it's a good thought - we do actually have someone who's been collaborating with us to develop a hand-held EOMA68 games console because he loves the idea of upgradeability http://rhombus-tech.net/community_ideas/games_console/

i have been through the evaluation process for the OMAP5... TI unfortunately is using PowerVR, which, if you're familiar with that architecture (as i am) you'll know that it's so overly complex that the engineers *inside imgtec* are barely able to cope with it. added to their continued intransigence of remaining proprietary and we basically just have to let them get on with it until the whole thing collapses. for over a *decade* now, EVERY processor that ever uses PowerVR has had severe problems: reliability, software support, upgrade incompatibility, the list goes on and on and on. people who have to deal with it HATE PowerVR with a vengeance.

if the situation changed such as by imgtec releasing the full source of the PowerVR graphics SDK, then it would be a different story. as in *literally* a different story i can't find a link but there was that guy quoted as saying that he had "the most productive business meeting of his life, ever" when because intel's GPU source code was available and libre-licensed, and a games company had problems but *their* source code was libre-licensed, they were able to contact each other *WITHOUT NEEDING NDAs OR APPROVAL FROM LEGAL OR THEIR MANAGERs" and schedule an immediate meeting where both teams solved some serious issues in their respective codebases.

*that's* what libre means. i just... i can't bring myself to financially endorse a practice of *deliberately* causing both developers and end-users pain and suffering by giving money to companies that *actively* wish to retain a proprietary product stance, in the 21st century. we *know* it's a business model that's just not working, we've got decades of examples and case studies showing us why.

... sorry, rygD, i've been around computers for almost 40 years now. seen a lot of mistakes get made, will see a lot more... will learn from them all.
 
I'm probably out of place here, because I think gaming is a big part of this community, and maybe modelling or other uses for 3D. But I have little use for 3d. I'm interested in desktop, at most video (occasionally) or photography (simple). I didn't like PowerVR but I thought if I don't load the proprietary driver and don't use the 3D acceleration it won't mess anything. Do you think that I'm going to have problems with the OMAP 5432 in the Pyra even if I disable, not install or remove any PowerVR software ? I've heard some SOCs don't boot without a GPU driver (raspberry? broadcom? already solved?). Or maybe the PowerVR uses power, can't be turned off or causes inestability even without the driver ?
I had not heard of anything like that before (I've heard people don't appreciate the powerVR architecture).

The OMAP 5432 has a GC320 2D engine which I think is also in the novena laptop (i.MX6 quad) and I've heard there's the etnaviv driver for it so we may or may not get 2d acceleration on the Pyra which I would very much wellcome.

Also I thought the A20 did not have working free software 3D drivers yet (the Lima was not finished?). In your crowdfunding page you explain how the MALI 3d engine in the A20 gets out of the way without the drivers. Is the PowerVR more invasive or the problems you are refering to are when trying to use it ?

Since I have you here: Am I correct I need a credit or debit card to participate in your crowfounding ? Are you thinking of accepting bank transfers at some future date ?

In a more sci-fic tone: The Pyra has a separate CPU board. It would be fun if it could be adapted to fit inside an adapter case with EOMA68 interface (if power allows). The pyra design does not allow to plug a CPU card externaly at any moment like your laptop, but maybe if the adapter was possible some people could buy the CPU board to keep it inside an EOMA68 enclosure and some for their Pyra, or in case someone removes the CPU card from their Pyra for upgrading, then they could put it in this adapter case and use it inside EOMA68 devices instead of the old CPU board going to wsste... I haven't even bothered to check physical dimensions because it souns so far-fetched, but it's a funny thought anyway.

Anyway, thanks for your work. I'm impressed by how long, hard and thorough you work. Thanks for being around here too and best luck.
 
in other words, it's about transforming something that was previously not ok through *your* positive intentions. if someone does something bad, and you can make it right by supporting someone that also wants to make it right, what would you do?
This doesn't necessarily make it right though. It's a good thing, but a good thing doesn't immediately negate a bad thing. Looking back at the assassin analogy, but looking at the assassin himself: if you murder someone for money and then donate that money to charity you aren't absolved of the murder. If caught, the court isn't going to go "well, you did donate that money to charity, so I guess you're off the hook"; you'll get a trial and a sentence just like any other murderer.
I have no idea what AllWinner's "trial" and "punishment" should be, but this is probably a step in the right direction.
 
In a more sci-fic tone: The Pyra has a separate CPU board. It would be fun if it could be adapted to fit inside an adapter case with EOMA68 interface (if power allows). The pyra design does not allow to plug a CPU card externaly at any moment like your laptop, but maybe if the adapter was possible some people could buy the CPU board to keep it inside an EOMA68 enclosure and some for their Pyra, or in case someone removes the CPU card from their Pyra for upgrading, then they could put it in this adapter case and use it inside EOMA68 devices instead of the old CPU board going to wsste... I haven't even bothered to check physical dimensions because it souns so far-fetched, but it's a funny thought anyway.
I hadn't considered that, and it kinda fits with what I was thinking of with my idea. I won't be able to toss the board/card into my Pyra instantly, but if we ever get upgrades the old board can be reused. Great idea. Now we need to throw money at the EOMA68 project, and in a few years maybe get together a couple people to make a Pyra to EOMA68 adapter.

(If we could go the other way we would have a libre Pyra, although possibly not so useful for games...I can't remember what we will have binary blobs for.)
 
Hello,
I am one of the guys behind the project of the console based on EOMA-68

Obviously EOMA-68 is not a perfect project, nothing is perfect, especially when you do not have a lot of money for a big project like this (like Microsoft or Google). Are GP32, GP2X, Pandora or Pyra perfect?

The good thing is that EOMA-68 is a totally new concept and with a great future. It is obvious that for now is only the beginning and much remains to be done.

One of the important factors is that it is intended for everyone.

Luke's a good guy with good ideas. I think Luke will do this well.
 
Last edited:
This doesn't necessarily make it right though. It's a good thing, but a good thing doesn't immediately negate a bad thing.
I have no idea what AllWinner's "trial" and "punishment" should be, but this is probably a step in the right direction.

yehh if we could actually usefully pursue a court case without there being huge financial costs - bear in mind this would have to be in china - it would have been done already. retribution's not always what it's cracked up to be. plus, we're aware that there is a totally different culture and a language barrier here. many of the managers were shocked when someone inside the allwinner GPL compliance team actually put them in front of a chinese version of the GPL. but there's now thousands of employees there, with multiple independent investors carving out their own "niches" and blatantly ignoring the VP's instructions. it's a complex situation in other words.

it helps to be a bit aware of the situation in china - bunnie's "gongkai" post is well-worth reading: https://www.bunniestudios.com/blog/?p=4297
 
yehh if we could actually usefully pursue a court case without there being huge financial costs - bear in mind this would have to be in china - it would have been done already.
Aye, and this's what I mean when I say most people won't and shouldn't care: at the end of the day it's a thing that no one can do anything about, a proverbial battle lost before it can even be fought, so just carry on doing what you can.
Just don't try to ignore the seeming contradiction like onpon4 did if it ever gets pointed out.
Here's rooting for you!
 
Here's rooting for you!

thanks stan :)
[doublepost=1467602639,1467601187][/doublepost]
I'm probably out of place here, because I think gaming is a big part of this community, and maybe modelling or other uses for 3D. But I have little use for 3d. I'm interested in desktop, at most video (occasionally) or photography (simple). I didn't like PowerVR but I thought if I don't load the proprietary driver and don't use the 3D acceleration it won't mess anything. Do you think that I'm going to have problems with the OMAP 5432 in the Pyra even if I disable, not install or remove any PowerVR software ?

it's a device that writes independently to the framebuffer, so by merely switching that off (not using it) you should be fine. of course if the desktop environment requires opengl you'll be going hooorribly slow, but again that'd be a matter of replacing the x-window server (if you're using one! i don't know what software the pyra comes with).

I've heard some SOCs don't boot without a GPU driver

grrrrr. don't get me started. :)

I had not heard of anything like that before (I've heard people don't appreciate the powerVR architecture).

EVERY set of engineers who has EVER used PowerVR has ended up HATING it and regretting that they had used it. that includes intel, who made the stupid, stupid mistake some years ago in creating an intel atom processor with powervr. EVEN INTEL ended up hating it because imgtec were f*****g about demanding $150k fees to recompile the software which they themselves NO LONGER UNDERSTAND.

The OMAP 5432 has a GC320 2D engine which I think is also in the novena laptop (i.MX6 quad) and I've heard there's the etnaviv driver for it so we may or may not get 2d acceleration on the Pyra which I would very much wellcome.

ooooo! ooo! really!! that's so exciting to hear! wow! perhaps TI finally got the f*****g message after 15 years that people are absolutely f*****g sick to the back teeth of PowerVR! and the OMAP's DSP is powerful enough to be able to handle MPEG and other video decode without needing a proprietary video decode engine, wow! oo i'm so excited, it could be possible to have a decent FSF-Endorseable processor, wow!

aa..... awwwwwwwww darn it, GC320 2D not GC320 *3*D.

awww darn it. i got really excited there for a moment - etnaviv is for 3D, and the iMX6 vivant is 3D, so i thought GC320 was 3D to... awwwww :)


Also I thought the A20 did not have working free software 3D drivers yet (the Lima was not finished?).

correct. nobody's funded luc verhaegen since 2014 and he's had to take up full-time contract work in order to put food on the table. he's the only person in the world who has the knowledge and the skills to tackle MALI reverse-engineering. i don't understand why people aren't phoning him up EVERY DAY to say "here's some cash!"

In your crowdfunding page you explain how the MALI 3d engine in the A20 gets out of the way without the drivers.

it's just another device on the internal memory bus that can write to the framebuffer if it's told to.

Is the PowerVR more invasive or the problems you are refering to are when trying to use it ?

the problems with PowerVR's design are that it's actually an extremely powerful general-purpose processor, where it's software-programmed for the most part. the actual core 3D engine software was written over 15 years ago, it's extremely hard to understand, the key critical engineers who understood it have left, and anyone else coming in just f***s it up.

the problem for *end-users* is that that software-defined 3D graphics processor has full and direct access to the entire memory range of the main processor (because these are low-power, low-cost systems that have shared memory). and it goes horribly wrong, and pisses all over main memory NOT the part it's SUPPOSED to be writing to.

this is why people get sick of having PowerVR because you do something complex (like run a powerful computer game), it all goes to shit and your entire system crashes. great, huh?

all they f*****g well have to do to sort this out is give us the f*****g source code and documentation.

you should look up luc verhaegen's talk on "the state of embedded graphics" i think it's called, he is absolutely SCATHING of the piss-poor design of PowerVR.

Since I have you here: Am I correct I need a credit or debit card to participate in your crowfounding ? Are you thinking of accepting bank transfers at some future date ?

talk to support@crowdsupply.com. they certainly don't accept paypal. not after paypal started witholding huge sums of money on crowdfunding campaigns. it's not down to me regarding the bank transfers, the money all goes to crowdsupply first.

In a more sci-fic tone: The Pyra has a separate CPU board. It would be fun if it could be adapted to fit inside an adapter case with EOMA68 interface (if power allows). The pyra design does not allow to plug a CPU card externaly at any moment like your laptop, but maybe if the adapter was possible some people could buy the CPU board to keep it inside an EOMA68 enclosure and some for their Pyra, or in case someone removes the CPU card from their Pyra for upgrading, then they could put it in this adapter case and use it inside EOMA68 devices instead of the old CPU board going to wsste... I haven't even bothered to check physical dimensions because it souns so far-fetched, but it's a funny thought anyway.

honestly when things calm down a bit here i can take a look - i know that the team who built the pyra were not receptive to me being on these forums back when a third party posted a "hey there's this thing called EOMA68" message a couple years back. times change, though...

Anyway, thanks for your work. I'm impressed by how long, hard and thorough you work. Thanks for being around here too and best luck.

appreciated.
 
sulu: thank you for pointing this out. in my earlier reply (i'm just catching up here) i point out that there really isn't a lot of choice. i'm in touch with Allwinner, i have been for five years, and i can tell you that they're KEENLY aware of the harm to their reputation. but we have to be realistic.
Maybe you/we have to be. But a RYF certificate shouldn't be about realism/pragmatism. It should be about freedom as defined by the FSF.
So your project might deserve an honorable mention, as in: "Tried really hard, but unfortunately failed due to reasons that were not the project's fault."

think about this: the computer you're using, is it entirely spyware-free? it's an intel or an AMD post 2009 or post 2013 respectively, right? so it's got that "remote management" feature, right? so should you boycott intel because of that? and is it okay to criticise one company writing the message from another company's privacy-violating device?
This is a different problem. Afaik there is no "Respects Your Privacy" certificate on any AMD/Intel hardware. So there is nobody claiming anything he can't deliver (regardless of who's to blame).
Let me ask this way: If you had chosen a recent AMD/Intel CPU, would you be trying to get a RYP certificate (assuming one existed, that's as strict as the RYF certificate)?

my point is: we have to be realistic and make use of what we've got as a stepping-stone to work TOWARDS what we want. my goal is to get to the point where i see a mass-volume FSF-Endorseable processor being brought into existence.
And you have my utmost respect for that...

i don't mind how that's done,
...but that's something the RYF certification can't afford imo, without taking damage itself. If there is no possibility to build a device that truly respects FLOSS, then such a certificate should just not be issued.

bottom line:
I wish you all the best in your efforts to make Allwinner respect FLOSS. I might even support you with my money, just for the sake of reaching that goal. But I stand by my word, that I think a RYF certificate is not justified, as long as that goal is not reached.
 
If you had chosen a recent AMD/Intel CPU, would you be trying to get a RYP certificate (assuming one existed, that's as strict as the RYF certificate)?
You do realize that recent x86 CPUs that have mandatory co-processors literally capable of reading the entire system RAM, right? Using that and applying for a hypothetical privacy-respecting certification would be akin to the Raspberry Pi applying for RYF. Or, for that matter, applying for RYF with any computer using a modern x86 CPU.
 
You do realize that recent x86 CPUs that have mandatory co-processors literally capable of reading the entire system RAM, right?
That's exactly the point why I was asking that question you quoted.
Please don't feel offended, but I'm not going to argue on that with YOU any further, because I've realized that the two of US won't find common ground here.
 
You can refuse to argue with me all you want, but it's not analogous. The Libre Tea card does not include any proprietary software when it is shipped to you or depend on any proprietary software for its intended use. That's exactly what it means for a product to respect your freedom. An x86 CPU not only does not respect your privacy as it is shipped to you, it is incapable of respecting your privacy, because it refuses to even run without running a proprietary program that is capable of seeing your entire system RAM and reporting it back.
 
Arguing with people is pointless, arguing against people is fruitful.
Its the difference between an argument and a discussion (when there is one), where bickering and agreeing is at either side.

Im not sure where it all went wrong, but ill try to clarify from my assumptions.

A recent AMD or intel processor was not chosen because they do not respect your freedom, with no solution.

The RYF thus can't be issued, nor would you even possibly want to, because one follows the other, both ways.

So here we are in the real world, where AMD and intel doesnt have anything, and the RYF is the cert to gun for.

We can all wish for AMD/intel performance and RYF levels of freedom, but that isn't on the table.

You can then imagine a table, by insisting that either of them overlap somewhere, but they don't.

We have come to a point where state of the art is measured in high performance and levels of total spying.

Fortunately not all tasks need all that power, and we can have a reasonably cheap thing that covers most bases, in freedom.

-----------------------------------------------------------------------------

I have electronics and overclocking as a hobby, and even I am totally disinterested in the latest fancies, because they come with all that cancer. And thats the current market for this thing, people who care enough to get maybe an extra PC for some task they have. There are actually quite a few people like that. And we all know someone who wants to have a safe computer, that person who is terribly afraid of "all the news", rightfully so. But equally afraid of changing anything on their computer. Give them this and the support starts to make sense.

If we can bring it down, or rather, expand it, from where we see it, because we are able to, where regular people start buying this, which are the people with all their bases covered, we will have done good. And it is possible.

I reworked the campaign text to a human-readable format. And I backed the campaign. Hoping it helps.
 
Last edited:
Well, I'm not a native English speaker, and prepositions belong to my major problems. ;)
 
Arguing with people is pointless, arguing against people is fruitful.
Its the difference between an argument and a discussion (when there is one), where bickering and agreeing is at either side.
I can't say I recognize that distinction. An argument is really just any discussion around a disagreement. What makes it pointless or fruitful is more how emotional those involved are. So for example, looking back, this was a very sloppy post that I shouldn't have made (I made quite a few similarly bad posts in this thread):

https://pyra-handheld.com/boards/th...respecting-computer.77612/page-2#post-1384519

Not because I was "arguing with" someone instead of "arguing against" someone, but because I let my emotions get the better of me and didn't take the proper time to really look it over.
 
Last edited:
be a matter of replacing the x-window server (if you're using one! i don't know what software the pyra comes with).

Debian armhf, I've heard. That should mean choice.

ooooo! ooo! really!! that's so exciting to hear! wow! perhaps TI finally got the f*****g message after 15 years that people are absolutely f*****g sick to the back teeth of PowerVR! and the OMAP's DSP is powerful enough to be able to handle MPEG and other video decode without needing a proprietary video decode engine, wow! oo i'm so excited, it could be possible to have a decent FSF-Endorseable processor, wow!

aa..... awwwwwwwww darn it, GC320 2D not GC320 *3*D.

awww darn it. i got really excited there for a moment - etnaviv is for 3D, and the iMX6 vivant is 3D, so i thought GC320 was 3D to... awwwww :)

Sorry for any coitus interruptus. In case anyone gets excited again: I don't know what I'm talking about, right?
The OMAP 5432 should have one GC320 (they call it BB2D) .
But yes, GC320 should be 2D, scrolling, compositing, alpha, svg, this kind of thing. Sounds like enough for me.

If there was the software...
Indeed etnaviv is for 3D but I read 2015-10-07 Linux Weekly News:
Russell King has gotten the xf86-video-armada driver working using libetnaviv on top of the Vivante kernel driver. That uses the 2D GPU and provides acceleration for some common operations.
(link added by me, last commit in that repo: january 2016)
OMAP 5432 is not in the SOCs the etnaviv people have tried, or at least they didn't document it.
So maybe it won't work without the 3D gear in the GC2000 or GC880,but who knows ?
Or maybe the GC320 in the OMAP 5432 is somehow behind the PowerVR and can't be used without it, I don't know.
I also don't know how much documentation on the OMAP 5432 it would be necessary to try it and whether that's openly published or not.


correct. nobody's funded luc verhaegen since 2014 and he's had to take up full-time contract work in order to put food on the table. he's the only person in the world who has the knowledge and the skills to tackle MALI reverse-engineering. i don't understand why people aren't phoning him up EVERY DAY to say "here's some cash!"
I doubt it is only about money, he himself hinted at ARM pressure against any efforts at freeing Mali.
But yes, people should fund free software. We need more people able to do each task (and maybe fewer tasks by picking a smaller set of hardware to support, but I can't imagine how to coordinate that agreement).

the problem for *end-users* is that that software-defined 3D graphics processor has full and direct access to the entire memory range of the main processor (because these are low-power, low-cost systems that have shared memory). and it goes horribly wrong, and pisses all over main memory NOT the part it's SUPPOSED to be writing to.

The part about etnaviv in lwn also told of some unfinished MMU stuff to enhance security (and command queue validation).

all they f*****g well have to do to sort this out is give us the f*****g source code and documentation.
I wish they did.

talk to support@crowdsupply.com. they certainly don't accept paypal. not after paypal started witholding huge sums of money on crowdfunding campaigns. it's not down to me regarding the bank transfers, the money all goes to crowdsupply first.
Thanks, I'll try.
Isn't it nice when somebody (Leah Woods?) has already written what you'd want to say about internet payments ?
 
There are reasons why I haven't gotten myself a new system for seven years and this won't change anytime soon.
When I've gotten my system, it was the last one that was relatively silent and still came with WindowsXP since Win7 was everywhere already.
Later on I switched to Linux and today my priorities are different.

If there'd be a fast Linux system with the certificates mentioned in this thread, I'd start considering to buy it at some point but I'm completely put off by what Intel and AMD are doing by now.
my point is: we have to be realistic and make use of what we've got as a stepping-stone to work TOWARDS what we want. my goal is to get to the point where i see a mass-volume FSF-Endorseable processor being brought into existence.
This is actually a quite reasonable approach. Doing whatever you can is better than doing nothing.
Let's hope Allwinner will reconsider the worth of such certificates, not all customers are eager to sell their souls to the big corporations and want real alternatives which are really different.

I just hope those certificates stay as strict as they are since watering them down would worsen the situation in the long run.
 
Last edited:
Let me ask this way: If you had chosen a recent AMD/Intel CPU, would you be trying to get a RYP certificate (assuming one existed, that's as strict as the RYF certificate)?

hmmm, intriguing concept... i'm not sure that Dr Stallman could be convinced that the ethical considerations behind "Freedom" could be separated from the ethical considerations behind "Privacy", on the basis that only through *having* the source code under freedom-respecting software licenses do we stand a chance of verifying that people's privacy is respected.

my feeling is that you shouldn't even *try* to have one without the other, because it stems from the same root unethical behaviour.

And you have my utmost respect for that...

thanks

...but that's something the RYF certification can't afford imo, without taking damage itself. If there is no possibility to build a device that truly respects FLOSS, then such a certificate should just not be issued.

absolutely correct.

bottom line:
I wish you all the best in your efforts to make Allwinner respect FLOSS. I might even support you with my money, just for the sake of reaching that goal. But I stand by my word, that I think a RYF certificate is not justified, as long as that goal is not reached.

correct - and that's why we're not applying for RYF Certifications - at all - on AMD or Intel products... but more than that i've pretty much ruled them out entirely. i've spoken to Intel, and asked them to make a processor that doesn't have the "arbitrary remote management" feature, and have said to them that i really can't justify putting one of their processors into EOMA68 form-factor with things like "Treacherous Computing" in it... *especially* when you have to sign NDAs to get the Treacherous ICs! i mean _come on_!

hilariously i would actually fit the LX900 (the faster version of the original OLPC's processor) into EOMA68 form-factor if it would fit, but it'd be an extreeeeemely tight fit (43x70mm PCB), what with having to put the CS5536 companion IC on the PCB as well. AMD keeps trying desperately to get rid of that processor range: i believe they've been trying for about five years at least to shut it down, but for some strange reason perhaps something to do with the AMAZING level of support (thanks to the original OLPC) they just can't seem to stop taking people's money. what _will_ kill it off though is if Debian GNU/Linux stops support for 486 and (original) pentium-level processors. that would be... bad.
[doublepost=1467657610,1467657366][/doublepost]
Isn't it nice when somebody (Leah Woods?) has already written what you'd want to say about internet payments ?

this one's muuuch funnier... and on-point, and got results http://garethhayes.net/paypal-warning/
[doublepost=1467657803][/doublepost]
I just hope those certificates stay as strict as they are since watering them down would worsen the situation in the long run.

believe me when i say that the RYF Certifications are NEVVVERRRRR going to be "watered down". the FSF is a Foundation: there are legal ramification of it violating its own Charter. what kind of message would it send for them to back down or change their stance? nobody would ever trust them ever again!

there's a beautiful quote i see on people's signatures sometimes:

The reasonable Man adapts himself to the world.
The unreasonable Man adapts the world to himself.

Therefore, all progress depends on the *unreasonable* Man.
[doublepost=1467662881][/doublepost]oo, oo, i just learned that the qualcomm 410 might actually qualify as useable, it's available on arrow, there's schematics available, and much more - even the GPU has been reverse-engineered.
 
Back
Top