The future of Libre Computing: Crowd Funding Campaign Starts Now
- Anmelden oder Registrieren um Kommentare zu schreiben
https://www.youtube.com/watch?v=GspUDdNlsng great video with tarrin lupo, really funny - right in a middle of a thunderstorm. weather looked great when we started... ended up with flash flood alerts! totally cool
I am a member!
Here is an interesting interview/demonstration that Luke gave (demo is not great- if you follow the updates there are better videos of different aspects, like hardware accelerated playback for video, etc, but ultimately this video came out pretty good, except for a few rough spots / errors in speaking):
https://www.crowdsupply.com/eoma68/micro-desktop/updates/fsf-ryf-background big update, background on why i've been doing what i'm doing. links to paul boddie's insightful post as well as the valve/steam - intel collaboration a couple years back.
In a video lkcl says connecting several pc cards to achive a faster computer is not an option. https://youtu.be/GspUDdNlsng?t=1316
Because of the hardware or because software is not available?
If it is about software, what would it cost to provide it?
if you set up a beowulf cluster for example you'd do fine. or you set up a data centre where each computer card was totally independent. or if you used them as front-ends with DNS round-robin and set up an alternative NAS / SQL distributed cluster... basically it's about software (and setting it up). all the software already exists... it's all related to server / clustering. if you're expecting to have multiple computer cards and speed up *desktop* applications, forget it.
>multiple computer cards and speed up *desktop* applications
Can you elaborate about why you cannot get a faster notebook combining 2 computer cards? Does the hardware make it impossible or could software enable it?
You will probably dislike this. To me you can put a computer card into a notebook casing. But you do not get a notebook. I previously asked about the computer card's performance compared to a raspberry pi. I did not get an answer. I had a raspberry pi 2 and it was not like a x86 duo core computer. I have not tested a raspberry pi 3, maybe it is not fast like a notebook either. Should I buy a computer card it would be instead of a raspberry pi 3. I consider the computer card the raspberry pi I would want to have.
I have a Intel Duo T5750 2.00GHz, GM965/GL960 Integrated Graphics, 4gb ram notebook. It does all common tasks well. Should I want a computer card notebook, it should have a comparable performance. Could a double computer card notebook be made, I do not know if it would show a comparable performance.
The A20's processor is not going to be as good as any x86 computer you have, unless you're using really ancient stuff. But the key point is you can upgrade later.
On a side note, tonlee, you're probably not using anywhere near your current CPU's capabilities most of the time, unless you're playing games. The A20 card should be "good enough" for basic tasks. Of course it won't be as fast as typical x86 computers you can get today. It won't be anywhere near as fast. But that doesn't mean it's inadequate, necessarily.
And consider the benefit of using the Libre Tea card as your primary computer and using your current x86 computer only for tasks it is too slow to handle (like games): you've got no proprietary pieces, which means it's an actually secure system, not crippled by a backdoor.
Besides, it's not all about the first products. It's also about future products that can follow (like faster computer cards which also respect your freedom, and other possible form factors), and the long-term benefits of this new paradigm.
lkcl has already answered you: "if you're expecting to have multiple computer cards and speed up *desktop* applications, forget it".
You understand that a program running on a computer would not run faster if you connect this computer to another one through a network cable. That is at least the common situation. Indeed, it *could* run faster if the program was written (and configured) to send part of the work to the other computer through the network and get the result back.
However it far easier said than done! The whole program must be rewritten to work that way. You have technical issues to take care of (e.g., what happens if one of the two systems crashes?). However, the biggest difficulty is to actually find a way to divide the work into pieces that are, at the same time, sufficiently large (otherwise most of the computational cost would be spent taking care of the communication between the computers) but not too large (if a piece represents 90% of the work, then the other computer, which gets at most 10% of the work, will do very little and, overall, you get, at most, a 10% speedup), mostly independent from each other (otherwise the computer spends most of its time waiting for the other computer to reply), requiring little input and generating little output (again to limit the communication cost), etc. With several cores (rather than several computers) sharing a same memory, some technical issues disappear but the organizational issues (splitting the work in mostly independent pieces, balancing the workloads, synchronizing, etc.) remain. They are the most difficult issues.
It is like dividing a work between members of a team (at work, at school) to achieve it as fast as possible. The work is achieved differently if it is a one-man team or if it is a two-men team; the two-men team rarely achieves the work in half the time required by the one-man team (the ratio greatly depends on what the problem actually is) because the two men must synchronize each other, because one of the two receives a more difficult piece of the work and needs more time to achieve it (during that time, the other man waits) so that both can tackle what follows, etc. To get close to doubling the productivity when you double the team, the problem that is solved must be easy to distribute/parallelize. Not all problem are like that.
Back to computer programs: desktop applications do not achieve time-consuming works that the programmer can divide and distribute to two processes running on two cores. Most desktop applications are programs that require little horse power but must be reactive to the user inputs. There is simply not much (if anything) to win through parallelization. And, again, parallelizing is hard. Just take as an example the desktop free software application with the greatest development effort: Firefox. Until now, Firefox, like all your desktop applications, runs in one single process. It should run in two processes soon ( https://developer.mozilla.org/en-US/Firefox/Multiprocess_Firefox ), performance is one of the motivation... but the price to pay is high (rewriting a lot of the code, many hard-to-debug issues to expect, some use-case will performance will actually be worse because of message-passing, etc.): https://developer.mozilla.org/en-US/Firefox/Multiprocess_Firefox/Motivation
See https://en.wikipedia.org/wiki/Multi-core_processor#Software_effects too.
What is the point in having multi-core desktop computers then? Well, if a demanding application (say, playing a HD movie) is running, it eats up one core but you can still comfortably execute another application at the same time (it will run on another core).
Thank you.
You cannot combine 2 computer cards. You could have several cpus on one computer card?
good thought! not really practical - the power budget is a hard limit of 5W, you have to fit SoC *and* memory *and* boot storage *and* an interconnect... onto a 43 x 75mm PCB. reaally challenging :) might as well just get a samsung 8-core SoC instead :)
>key point is you can upgrade later
From where do you know this? Why should more powerful cpus become free software? To me it seems manufactures want to keep the software non free and they are getting better at locking it down. Ask libreboot. Even if many computer cards are sold, one cannot know if that would provide faster free software cpus.
>good enough" for basic tasks
The raspberry pi 2 I consider not fast enough for basic tasks.
I require a computer enables full hdd encryption or as a minimum home folder encryption. Can the computer card manage that?
>power budget is a hard limit of 5W, you have to fit SoC *and* memory *and* boot storage *and* an interconnect... onto a 43 x 75mm PCB
You have introduced these limitations. Discarding them what would the costs be to provide a mainboard with several cpus?
Please inform yourself. Both when it comes to technology (e.g., having two CPUs or two cores, twice more RAM, etc. does *not* improve the performance of one mono-threaded program, such as all current desktop applications) and when it comes to the project. EOMA68 is a *standard* to have a computer in a PCMCIA card. That means:
- different boards, with different components, possibly a multicore system on a chip, could be made in the future and anybody can replace a previous board conforming to the same standard (no need for a screwdriver);
- there are limitations when it comes to the size of the components (PCMCIA cards have the size of a credit card) and the energy they can consume.
> I require a computer enables full hdd encryption or as a minimum home folder encryption. Can the computer card manage that?
I don't see why it wouldn't. Encryption is not computationally expensive.
> From where do you know this?
because as the guardian of the EOMA68 standard and the designer of the various Computer Cards over the past 5 years (of which the EOMA68-A20 is the first one to be publicly released), i say that i will do it, therefore it shall be done. as you can see in the update i wrote yesterday i'm currently tracking two other processors and there's one more i may add to the list https://www.crowdsupply.com/eoma68/micro-desktop/updates/product-roadmap
> I require a computer enables full hdd encryption or as a minimum home folder encryption. Can the computer card manage that?
i have no idea - you're the first person to ask. it has a hardware-accelerated crypto engine so chances are high that the answer's "yes". you'll need to look up the ARM Cortex A7 Crypto Accelerator Engine, the specs should be available online. i'll put it on the list to investigate but if you would like to help find out please do subscribe to the mailing list so that other people can help investigate as well.
> You have introduced these limitations.
that's correct. it's so that there is no fans needed which introduces a mechanical point of failure, dust and dirt getting inside the appliances and so on.
> Discarding them what would the costs be to provide a mainboard with several cpus?
it's a totally different project. i'm guessing that it'd be somewhere around $50,000 in development costs as the space is incredibly tight, you'd need specialist RAM ICs (compact ones) to reduce space, you'd then be into MOQs of 1,000 units for the RAM ICs (if you can get them). it's a completely different proposition basically. so on top of those $50k NREs you'd need to find the cash for an order of between 500 to 1,000 units. it could be done... but you'd need to justify it (a totally new crowd funding campaign for example) and you'd need to find a market for it or a story for people to find it attractive to back.
About discarding the computer card limitations. It was not a suggestion. It was an off topic question. I wanted to know, if the mainboard could have any size, how big and costly a task would it be to manufacture a mainboard with several cpus?
If faster free software cpus are on the way and they can fit in a computer card, then your reasoning is valid. If the computer card's cpu keeps on getting produced, I wanted to know if a faster mainboard, several cpus, could be made.
When I look at my notebook's system monitor, much time cpu performance is low. Sometimes it is high. And often performance of the cores follow each other.
I do not know if there was a connection. The notebook's system would stall. After I installed more ram, it seldom stalls.
I did not expect the comptuer card's cpu would have an encryption device.
Would it be possible to buy the pfy laptop housing kit a couple of weeks after the campaign ends? If so, from where?
you can download the CAD source *right now* - i will have to make some modifications but it's all there.
https://www.youmagine.com/designs/libre-hardware-licensed-parametric-laptop-design
crowdsupply is slightly different, you'll be able to put in pledges after the campaign ends, and it will run indefinitely as a "shop"... *after* the campaign's run - that's all handled by crowdsupply *not* by me.
now, when i get to Taiwan, *as long as* there is spare cash i will be over-ordering on some of the components. so yes you'll be able to "buy"... but not whilst i am in the legal jurisdiction of the United States. it's a crowd-funding campaign, it's a gift economy (not "orders"), i did *NOT* bring in "products for the solicitation of orders", i did NOT bring in "samples" into the United States, it's right there on the Customs form that i declared on entering the United States on a 90 day visa waiver, so please don't ask me to "sell" you anything, okay? :)
Count me in as a backer!
Good to see you fighting the good fight on other message-boards, Luke.
thx RMK
I am a member!
First of all, thanks for the work and keep it up, it is important having 100% free hardware, with 0 obsolescense and enviromental responsability. So, really good job!
I have finally decided to back this project. I'd like to try it to substitute my old-non free raspberry which I use as general purpose hardware. My main idea is use it as media center and retro gaming console so I guess it is possible but, I also want to give it more uses for example:
How about use it as "pineapple wifi" with an usb wifi adapter? It would be possible to install libreCMC and "pineapple firmware" on it? Will it work?
greetings
I am a translator!
120.000, 2 days left. What happens if you get say 135.000? The crowdfunding will get extended for a few days, right? Anyway congratulations and good luck, the sum you managed to collect is not a trivial achievement, well done!
46 hours left and it's already at 128,160. I am optimistic that it will succeed.
EDIT: 31 hours left and it's at $137,945. At this rate it will succeed.
I pledged for a Libre Tea Computer Card and a Micro Desktop Housing back in June (didn't have an account on this forum then). I spread the word to my friends and family but none of them really have any extra money.
Thank you for all the hard work you've done toward making a freer world. People like you are essential---everyone who's contributed in any way. This whole forum deserves many rounds of applause and a massive toast to the EOMA68 folks, the Trisquel people, and all the other free software contributors!
All the best.
Peace&rage.
heya cryptobilly, no problem - down the road this will be lower-cost, plus do bear in mind that there will be a 2nd-hand market pretty shortly (especially when in 6-12 months or whenever it is) when we have newer EOMA68 Computer Cards. subscribe to the arm-netbook mailing list for now and i'll set up appropriate mailing lists / forums from there, okay?
really appreciate your support, and mostly for your recognition that i've merely provided people with the opportunity - you all grasped it and made it happen.
Congrats on this success! I will certainly subscribe. Thank you. I also just wanted to say that I think your prices are perfectly reasonable. I just happen to live in an overwhelmingly poor area and most of the folks I'm close to don't have a lot of money for extras. Just didn't want you to think I was criticizing the pricing. Far from it. Again, congrats.
The campaign jumped *4%* between 44 and 39 hours to go. If it can keep up this rate of growth, It'll be done with 24 hours to go!
I don't want to jinx it, but...
FUTURE OF COMPUTING, HERE WE COME!
Yep, not much left to go!
yeah it's now only $750 to go, with 22 hours remaining. i think it'll be fine. hooraaay :)
If you modify the chip pocket https://getchip.com/pages/pocketchip, can you use the computer card together with the chip pocket hardware? Thanks.
ah now that's a very interesting question. i forgot to add that onto the future roadmap. i'll have to look at what interfaces they're using: ironically it's conceivable that an adapter would be more than the cost of the pocketchip!! so we just have to see. one thing i _can_ do because they've done open designs is turn that into an EOMA68-R8 Computer Card. the BOM would be something like $10 which is pretty amazing. and yes i would love to adapt the pocketchip.... there is so much to do it's overwhelming.
I have considered two options. One is, you buy pocket chip and somehow gets the computer card connected. The other, you 3d print a cabinet rear part that is made for the computer card. It should have a speaker which the pocket chip has not.
Great you got funded. I thought it would stay at 50000,-. This approach was what thinkpenguin at one point was hinting at.
Several people are complaining about the keyboard on the pocketchip.
Could be the pocketchip is not a suitable option.
28 hours left and 95%... I can smell the tension.
The goal was just reached a couple minutes ago! Congratulations! :D
I am a member!
:) I'm so ecstatic. I'm so glad that some people have finally got some understanding of why I've been saying this project is so important. It's not the specs that matter. It's the control it gives to the user and provides solid ground to stand on for doing higher end 100% free systems down the road. Is the first card a bit slower? Yes, but its still usable and it's now much easier and will be much faster to produce faster quad core systems that are free. With this base it's going to be a lot more cost effective to do what we need to do and that's critical to making progress (and we wouldn't make that progress if we didn't get the costs down). Short of this we'd be going backwards. All the hidden proprietary pieces were a big blight and we can finally begin to remove them in a more permanent way. We are not stuck on outdated CPUs (A20 is a little old, but the estimated time to quad core might be eight months, not several years) or forced to decide between a proprietary firmware for some 802.11ac chip and a less desirable external USB wifi adapter that sticks out. There is still a lot of work to be done here and elsewhere (802.11ac), but we've hit a major turning point, for 100% free software computers. Hopefully in the coming years we can push into new territory and secure the release of 802.11ac code, fight back against the FCC's terrorizing of free software users, and continue to make progress in other areas. There are certainly things we've not even begun to solve. Like 3D accelerated graphics (we're taking one step back here on the A20, but two or three steps forward in other areas).
I am a translator!
157,460 buckaroonies !!!
yeeeehaaawwwww
congratulations Chris and luke
I remember the Lima driver that was being developed. What happened to that?
101% funded!
Luke and Chris:
Congratulations, gentlemen, on a job well-done.
I'm very happy to see that the campaign was a success. I harshly could believe it would be until the last few fays. So, congratulations to all people that made it possible! And especially to the "core" "team" that put so much energy in this project. You really deserve to see the campaign being a success!
Thanks A LOT for all you've done!
Its success was unavoidable
Yep let's admit it, we knew it from the beginning but we needed to make drama for the sake of thrill, I suppose.
Congratulations everyone for such outstanding victory, should we celebrate this somehow?
So happy this was funded. Here's to many more successes!
Size of computer card
If there had been a much bigger sized card form factor already on the market, would it had been preferable to utilise it instead?
I ask because in an adequate time frame I find it not likely that duo core fast x86 computer cards can be made that small, same goes for the 5w limitation.
Allwinner a64 the same as r40?
' 200 lines of code for libdram'.
Are they encrypted or why can they not be provided?
I am a member!
The R40 is a Quad Core 32 bit AllWinner CPU to be released. It's a pin for pin upgrade to the A20 from what I understand. There is a 64 BIT CPU that might be an option with more ram and everything. Though these all have Mali400 graphics. There may be a possibility of a future CPU with all of this and free graphics, but it's only on the road map for discussion and would take a bit of work on our part to make happen. Depending on what happens here we might have to travel down another road. All of this stuff is being evaluated, but none of it is easily done. It will probably be years away. The good news is we're closer to succeeding now than ever before. There is momentum here which is new. Before it was a struggle with no realistic chance of success [of making progress moving forward].
the (completely independent team) inside allwinner which has nothing to do with any of the other teams, is stating "we want to maintain control over our hardware, we [naively] believe that this is the best way to do it".
the fact that they're pissing *everybody* off and forcing people to waste their time reverse-engineering the unencrypted libdram.a seems to have entirely escaped their attention.
In another post I turned down that phoning arm to get them to release software would make arm do that. If a department in allwinner can release the software and you got its phone number and email, people should contact it.
Can you put the computer card into a x86 notebook with no mainboard? Can you connect display, keyboard, touchpad, speakers, battery and powersupply? How do you connect the devices? Thank you.
I am a member!
No, it's a new standard, so you would need a housing that says "EOMA68 compatible", unless you are operating the computer card independently of any housing, in which case you need the cable kit. Right now only two housings are being manufactured initially (in the coming months). One is a desktop housing and one is a laptop housing. Other housings are planned for the future as well as other EOMA68 compliant computer cards. When it says EOMA68 compliant it means it'll work with any EOMA68 compliant housings. Anybody can design a computer card or housing around EOMA68 [though they'll need to comply with the standard and get certification to make said claim].
>cable kit
Is it for sale?
There are no compatibility obstacles about a notebook's display, keyboard, touchpad, speakers, battery and powersupply?
Can I buy a pc card port and get it to work if I solder wires to the pc card port?
- Anmelden oder Registrieren um Kommentare zu schreiben