Slashdot Mirror
Open Source Graphic Card Project Seeks Experts
An anonymous reader writes "Could this dream of many open source developers and users finally happen? A 100% open sourced graphic card with 3D support? Proper 3D card support for OpenBSD, NetBSD and other minority operating systems? A company named Tech Source will try to make it happen. You can download the preliminary specs for the card here (pdf). The project, though a commercial one, wants to become a true community project and encourages experts and everyone who have good ideas to add to the development process to join the mailing list. You can also sign a petition and tell how much you would be willing to pay for the final product."
Does it count if it's more than 2 weeks old?
I've kind of waited for this for years.
In theory other companies might steal the design and build and sell the card on their own, but if the design is community-owned, then that actually works to lower prices...
Anonymous Cow
Building a good open 2D card? Mabye... I doubt it's really feasible, but have at it. Chase that dream.
But a 3D card? You are going to make a card to run the latest Quake and Doom? Or even release back of the games? Do you realize how much time, how many thousands of man hours go into these cards? The dollar amount for the simulators, the fabs to make the prototypes, etc
This could however, make a great teaching tool.
I take it back... if the card can target elementary 3D and stellar 2D, it could (in a few years) be THE card to own for a commodity Linux box. Target your audience carefully and don't get caught up in the IdSoftware upgrade cycle! :)
Agile Artisans
It makes me sick.
...until we have open source DRAM.
This is a really great idea, but it will probably never work, a mailing list will bring way to many points of view.
Really what a project like this needs is the developer to shut out the open source community, until the project is done. If linus had made a large project out of the original kernel, I seriously doubt if it would have ever been completed. This should be kept simple, and then open sourced, only once there is a good code base to build from.
Graphics cards are so expensive these days when they are first released, and I can't imagine this every becoming very profitable or worthwhile if they are only going after a very small niche market. Maybe by having us do all of the work for them, they won't see any labor costs...whatever, as much as I hate to be negative about this, I don't see it getting past vapor.
I can understand that this card will never compete with ATI and nvidia which raises the question, is there any reason why ATI can't open source their old graphic cards, such as their 7000 series. Surely that technology is no longer critical to their lead. Sure many of those cards aren't being sold any more, but there are still plenty around and this may open up a niche market so they can produce some as a low-cost device.
----
Go canucks, habs, and sens!
If you RTFA, you'll see that there's been some activity since the last Slashdot post. The designers want to try to raise some awareness.
If you want to do 3D work, use an operating system that has been designed for it.
OpenBSD has been designed to be a border network system. NetBSD has been designed for portability and targeted mainly in embedded systems. Sure, you can use them as desktops (and ive used openbsd as a desktop for 5 odd years now), but if you want to do serious 3d work or play games, you use something more suited to the task.
oh my god! ive got a hammer! where are those nails....
Hardly a dupe, since the project has risen from speculation to preliminary specs and a petition.
Do you see what I did there?
If you'd read-up on this subject, you'd have seen that these folk *do* know their hardware.
They are also not being overly ambitious. While they expect to be able to develop a card which has 3D accelleration for desktop applications, they make no bold claims about gaming.
Indeed, this card is being designed as the ideal desktop-card for open-source systems with open-source drivers and firmware. Any gaming performance, while unlikely, should be treated as a bonus.
I have already pledged my intention to buy one of these cards just out of curiosity.
Ripping an new rectum in the fabric of spacetime.
May I suggest you use a new type of interconnect? My video card was kind of hard to push into place.
And the screws that hold it in place should be square heads.
I think the company would make a ton of money just making these as a reference platform and selling them to University students looking for a way to program their own GPU on the cheap for research purposes. Heck, Xilinx should do it themselves, and give all these students exposure to Xilinx parts (and their crappy design software) before they even find out who Altera is.
This project looks interesting. I'd sign on to help out, but this gets dangerously close to what my Day Job is, and I don't think my management would smile on my participation...
It's like saying:
"No, it's impossible to build a replacement for Microsoft Office. Do you realize how much time, how many thousands of man hours went into this software?"
But there you go, Open Office is doing pretty well.
If anything, development of a good "open-source" 3D card could be hampered by patents.
tech source makes graphics cards for sun microsystems computers, i've got a raptor in one of my ultrasparc10's. I'm sure they have some fabrication experience, just visit their website, they've got quite a few products.
Wgat sense does this make. There are some people (not me) that might pay up to $500 for the newest ATI or Nvidia cards. But they do that with the knowledge that the hottest 3D applications will take advantage of them. More importantly, that is the price they might pay for those cards today. It's well known that in six months those cards might be worth half that, in a year perhaps around $100. How can anyone say how much you would be willing to pay for the final product when by that time it might not even compete with the $100 cards?
I'm an American. I love this country and the freedoms that we used to have.
Well as mentioned over on OSNews. There's the always issue of patents. The vorbis people had to deal with them. These people will as well. I suggest looking through the archives for suggestions already discussed. e.g. DSPs.
I recommended them buying their way in by obtaining the patents to the Tseng 2D chip, and the PowerVR Kyro 3D chip and building from their.
The other way is doing some truely innovative work (basically reinventing 2D and 3D graphics).
Dude, come back when you got pixel shaders.
Yeah, it COULD happen. But it will also be crap. Does anyone really think a company could simply start competing with nVidia or ATI on features and power?! Heck, 3dfx couldn't do it. Matrox essentially gave up. And what about Virge?! Dare I even mention bitboys?!
Come one folks, let's get real.
If someone says he and his monkey have nothing to hide, they almost certainly do.
cheap labor conservatives - they want to keep you hungry enough to be thankful for minimum wage.
For 99% of users, this could be a great card. If it does great 2D, and can do good 3D (especially features like those used in Apple's Quartz, or Project Looking Glass) it would work more than well enough. Lets face it, for a large number of applications, a GeForce (origional) quality 3D would be MORE than enough for most anything many people would do. And if the graphics are localized into a small area (say a little 200x200 area of a window), then even such a card would be able to render very nice looking graphics (just like a "slow" card could run Doom 3 looking great at such a low resolution).
I'm with you. For a quality, commodity card this could be great. Plus, with the FPGA, not only could be hack the DRIVERS, you could hack the FIRMWARE! Think! You could buy the card, and write software to take the burden off the CPU for decoding MPEG2 or 4. You could even (with a little kernel help) swap firmware on the fly so you could have that video decoding, and then enter a command (or press a button on your desktop) to have the 3D firmware put in. When you're done, go back to video decoding acceleration.
Hell, make it run SETI in the background at super fast speed when just using 2D (like using nVidia cards to do scientific calculations on the GPU).
These things could be a LOT of fun to mess around with. I think I just sold myself on one ;)
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
while not exactly a GF6800, it does use a FPGA, which may lend itself to some interesting modifications.
I can see it now: custom logic patches to change the core for extra performance on your favorite game...
The difference between spam and poop is that you don't have to dig through septic tanks looking for real food. -- Me
""No, it's impossible to build a replacement for Microsoft Office. Do you realize how much time, how many thousands of man hours went into this software?""
Repeat after me. Hardware is not software. Software is not hardware
Overestimating is not any better than underestimating.
"If anything, development of a good "open-source" 3D card could be hampered by patents."
I've said as much elsewere. The vorbis people have shown that patents can be dealt with. However graphics is considerably more complex.
What if a large company got behind this. They would say, we will release the specs and let you open source them, everything chip design, reference board. But, we are the only manufacturer that is allowed to produce these boards that run this open source. Would you work on it? Do you think other people would How long would it take to get to production? What kind of premium would you pay for it?
"brxref
It's better to have a finished product that meets a limited set of goals than an over-engineered design that never gets properly implemented...
If that FPGA is still in there when the card is completed then there's going to be alot more this card will be capable of than just 2d/3d. 3d accelerator cards for games showed up around 7 years ago. We're starting to take advantage of other processor intensive things now. Where's my physics/ai accelerator? 2d and 3d graphics will make the card usable now, but where's the next step?
Of course, there are going to be more concerns than what cool stuff you COULD eventually use the card for. Like if it's too expensive and won't sell.
I'll leave the speculations about the price/performance vis-a-vis ATI's and Nvidia's cards to others. There is something else that worries me. The management behind this engineer gives the go-ahead based on positive techie feedback. They actually go ahead and build the things and techies actually buy it and improve the drivers. The likely subpar performance is overlooked because it is still the fastest video they've ever had on their Powerpc and Sparc boxes. ....only maybe it won't be based on an FPGA and it isn't going so open this time.
Now that a little bit of money is rolling in, it is time to start the ball rolling on the OpenGPU2.
Do we have any idea if this botique manufacturer won't jerk the football up at the last second?
From the petition site:
We endorse the Make you visible to TS new open 3d videocard Petition to .
What does that mean?
Comment removed based on user account deletion
I am glancing at the specs and I have a couple thoughts.
The first is that these are respectable specs - providing you don't want to to any gaming.
I think that is a really important caveat. I know that every once in a while people get all excited because the usual suspects port there games to Linux - you know ID and Blizzard come to mind.
It is a good thing that these two companies do this, but it is a bad thing that there are really only two companies that do this with anything approaching reliability.
Thing is... a card with these specs, especially considering that it is a year if not more away from reality will never cut it for any sort of gaming. You are going to produce a card with 3D support that doesn't have the muscle to handle any 3d games that are produced.
If you are fine with that then there is nothing wrong with those specs. This card will be able to handle email, porn and movies as well as anything ATI produces.
My 2nd thought is a bit more practicle.
Actually there may not be anything practicle about it. Might just be wishful thinking really.
What about 3DFX? What about OPENGL?
Between the two things isn't half the work already done?
I know it might seem insane - nuts even, but back in the day 3dFX had some very respectible hardware. They didn't fail cause there stuff was poop, they failed cause they underestimated nVidia (which in turn underestimated ATI). The hardware is still out there, the code is still out there. It just isn't being utilized.
Would there be anything wrong with utilizing these old resources to achieve this goal?
I can't wait for these to come out so I can put one in my Indrema... Imagine how many FPS I'll get in Duke Nukem Forever!
You see? You see? Your stupid minds! Stupid! Stupid!
I couldn't find any details of the project on Tech Source's web site, so I don't know what their motivations are. But if they hope to corner the Linux market, designing it via committee and then open sourcing the final hardware design won't solve the real problem.
Open source the API -- that's all anybody needs.
I can see it coming, a microsoft gfx card, sigh.
There is no sig
The proposal itself does not look like it is going over well. Look at the comments on the Tech's proposal site and you will see flames about the card not holding up to the standards of gaming such that ATI and nVidia do. If they are able to find programmers and developers that are superb at their jobs, maybe even ex-ATI or ex-nVidians, the card has a great chance at surviving among its rivals. Furthermore, ATI never had good software to begin with [drivers, etc] compared to their hardware. Opensource has also always seemed to prevail in one aspect or another in comparison to closed source. the development is a great idea, and even if it does not take off right away, it is a step in the right direction...
_
Free 27" Sony WEGA TV
Make sure you can "beowulf" them together. The GPUs are not going to be powerful so there needs to a way to make it more powerful.
Comment removed based on user account deletion
Uhm, sure this would cause headaches, but couldn't you just compute the fixed inversion of the divisor outside the fpga and use it?
Ie, you need x/n for your perspective interpolation, so you instead calculate 1/n outside the fpga and do x*(1/n).
Of course, if n is not a constant, or a small group of constants, then external calulations of its value will slow things down dramatically.
Also, why not just use the 'Microblaze' softcpu for the spartan3? This includes a mapped instruction todo a divide fairly quickly on the fpga.
man is machine
Osho
FPGAs are also slower than ASICs. This, and the cost, are the reasons why commercial manufacturers use ASICs. You may have a great design, but if it is limited by the performance of your FPGA you lose.
FPGAs are designed to be universal, and to do that they feature programmable interconnects. But the number of those interconnects is limited, and many FPGA designs are thus constrained. You may have plenty of gates left and no way to get to them... With ASICs this is not a problem because if you need a wider bus you build it there, on your own silicon. In FPGAs the busses are already there, and you can't add more.
Yet another concern is tools. Xilinx, for example, offers a free download of some bare minimum tools. They work OK if you are making a door lock with RS-232 control. But they fail miserably, to the point of being unusable, on a complex design - which this one is. Better tools, such as Synplify, will cost you your yearly salary. How many developers have access to that kind of tools? And once you switch to some specific tool you are committed.
Finally, there is a problem with skills of developers. There are many s/w developers who are very good with C/C++. But not that many are good with Verilog (and its wickedly evil predecessor, VHDL :-) Hardware design is very, very different from software design. And you can't debug it, you only can simulate it. Simulation tools, such as ModelSim, are absolutely not free on the level that you need for this design.
To summarize, this project can be done, but not by a bazaarful of people but a small, dedicated band of wizards who locked themselves up in a small cathedral. Even if these wizards release their works, none of mere mortals will be even able to open their files, since the tools to do that are not free.
And besides, why would any sane person, who is not burdened with FOSS thoughts, want to buy such a card even for $100? This cash buys you a decent entry-level Quadro, and if anyone suggests that this design can beat Quadro I won't believe that...
And if anyone wants a real entry-level card, then it can be had (Vanta TNT2, for example) for $10 in any bargain bin, at many places. Beat that first.
without dividers, perspective interpolation is going to be pretty tough
Perhaps there exists a cheap ASIC divider/trig unit (a 487?) that they can use as a coprocessor...
The way I see it, most of the cost of the latest ATi or nVidia cards is to cover R+D expenses. The fact that the price drops drastically in a year or two is evidence of this.
The advantage of an open source hardware project isn't just that you have documentation for the hardware and can therefore write drivers for it. The real advantage is the same advantage that open source software gives you; namely that you can hack on it and make it better.
Imagine an open source video chip project that you could send design patches to in the same way that you can send patches to the Linux kernel. There could be simulation software to run tests with, and if you wanted some reference chips, you could download a snapshot spec and take it to a fabricator. In fact, there's a business opportunity right there. You could take orders and print chips on a regular release cycle, say twice a year. Of course, I didn't RTFA, but might this not be what this company is proposing?
Sure, it might be a bit expensive now to have chips printed, but if there is a demand for this kind of service, the price will drop and the options will multiply. Eventually, you might be able to buy a kit at Radio Shack that will burn chips equivalent to today's high end graphics chips. And when that happens, there will be this open source (GPL?) chip spec waiting for you to burn, and there will be a driver ready for it when the HURD is finally released.
A republic cannot succeed till it contains a certain body of men imbued with the principles of justice and honour.
"Parts of them contain licensed copyrighted code from companies like sgi, which will not permit their stuff to be released to be public."
The test of that argument will come when SGI goes out of business, and Nvidia buys the remainders.
A lot of Nvidians were former SGI employees.
If it was a communist graphics card you would be forced to pay for one...
I don't think anyone will force you to buy this card.
You are a dumb ass.
I was scouring this thread looking for someone else to say this because I knew couldn't be the only one to realize it.
I have never understood this project. If they want to start with something at least equivalent to a five year old SGI graphics pipeline abd build from there, then I'd say go for it. But the specs on this card don't look any better than the stuff you get right OOTB with an intel chipset (which, after sufferng with this goddamned nvidia system for too long now, is the reason I'll not be buying another AMD system).
So is the whole point of this card just to pick up the slack for AMD?
I've heard that 3D cards of today are exponentially increasing in number of transistors. It's been said that the problem of displaying 3D is "embarrassingly parallel". Hence, the performance of these cards far outstrip the CPU for parallel processing.
Some of the thoughts expressed by experts are that 3D cards may become general purpose parallel computing cards.
If it weren't for bottlenecks in the AGP bus, it would be possible to use 3D cards of today for more general purpose computing (I'm fuzzy on what the actual hold ups are here...timing issues?).
There have been Slashdot discussions about using the graphics card for audio processing, because audio is usually less than a 32 bit stream. The problem is that audio and often general purpose computing have "real time" requirements.
Also, make sure your open source card supports ARB_fragment!
I think they won't be able to compete with an FPGA-based design: ultimately, those chips are too much work to program and they don't use their transistors all that efficiently when you already know you are mostly going to be doing stuff with numbers.
If, instead, they can come up with a high performance vector or SIMD machine which also has graphics output at a decent price, then they might really have something that people would snap up, not only for doing killer 3D graphics on, but also for doing more general computation.
A tiny software company who put it together "on their lap" literally, without any billion dollar investments a-la Microsoft. Never underestimate the power of a small group of highly educated, passionate individuals. I bet most of the work that goes into cutting-edge graphics chips is done by a team of 5-10 people. Can such a team be put together outside NVidia/ATI? You bet!
Think the majority of computer users - non-geeks. :
Think what the majority of those machines are used for
1) Internet browsing at home/office
2) Multimedia at home
3) Office work, presentations
4) Small time and educational games in schools
Now, if we had an open source card that could take care of all the overhead processing that today's desktop eye candy requires and then some, you've just got yourself a solution to this problem:
"I'm your average joe blow. I do not know what XFree86 is. I do not know what a driver is and I don't want to. I want it to work, I want it to work well and I want it to look good."
You've just cleared away 90% of that market's concerns about a video card that just works. Since it's open source, it's cheap. Since it's open source, it'll be used in universities. Since it's open source, there will be an expanding knowledge base for it.
Forget high end games. That will come in time. Target desktop eye-candy - great looking shadows, transparency, 3D Desktops, god knows what else. Target an interface BETTER than the Mac. Heck, TARGET SVG implementation! Use this to give our community artists free reign with their themes. I think there's some VERY serious potential here.
I think this is a gold mine that might actually help with better Linux adoption.
Find a job you like and you will never work a day in your life.
Because what you said is stupid. Five to ten people who know what they're doing is enough to put something like this together in two years time, tops. Granted, their product won't give you GeForce 6800 performance in its first incarnation, but these days it has become a lot easier to design custom logic. Besides, they're essentially "standing on the shoulders of the giants", so they already know what _not_ to do.
... how all the comments are "Linux this" and "Linux that", when the article submitter obviously went out of his way to not mention Linux.
Marginal Change and Improvement (MC&I) it the path of evolution and it isnt' required that they get it perfect the first time out. There will get much support in the non-western world and good support every where. The first cards might lack features and key functions but give it a few years and let everything build upon that which came before it..
.1 or .4?
I mean did you ever see the early releases of Linux? or any other open source project at release point
http://www.hawknest.com/
I've been rather unhappy with Nvidia and Ati offerings in the way of drivers so I'd happily buy this card. I'm looking to change to a new card in the next six months so if its timing is right it will be good for me. :) :)
Also I would be willing to pay a premium you would expect from R&D in this field plus for the effort. I wish this project well and hope it comes to completion.
I ate your fish.
I think it would be more practical to design such a "card" based on a previously developed chip which could be "bought".
Call me a pragmatist.
Learn to spell, American retard.
the hardware industry is based on obsolescence. A company like Nvidia wants to have their -current- and -in production- card to be as fast as possible. Once it's not in production anymore, they don't care. If they released the drivers and (expecially) the specs to the card, someone(s) could improve the driver, and thus make the card perform better, or add a useful property, thus making it more attractive and thus hurting sales of the current card. An opensource driver and specs would also mean that obscure/niche OSs (BeOS, neutrino, skyOS etc.) could be supported by the older card, thus making it moore attractive and....
This, in addition to the very god point made by the poster above.
Sigged!
This is what 3dfx is now, cheap voodoo3s are laying all around you. They do opengl and the uber glide. Write up some drivers and they'll be just as useful as anything else if not more so.
Sorry, but drivers have an absolute limitation of the hardware they are interfacing with. They cannot make it into something it's not. If the hardware only has X level capability where the current standard is 10*X, a driver cannot solve that.
That's the problem with this idea. You are limited by the capability of real hardware. HArdware design is much harder than software design when you talk performance. It's difficult to squeeze performance out of silicon. That's really what makes AMD, Intel, IBM, etc so good. It's not that they can design fast chips, most EE people can do that in VLSI. It's that they can design fast chips and then actually produce them.
Also hardware costs real money to produce. There is an actual cost associated with every card. With open source software, it's ok if it never makes a dime, so long as those that make the software are content to donate their time for nothing. That's not the case with hardware, even if all the design costs are donated, each card takes real resources, and thus costs money, to produce.
I'm just not seeing what this is supposed to offer.
I'm really doubtful that the specs listed can
be achieved in a Spartan-3 FPGA, particularly
the DDR at 200 Mhz should be a real trick.
Anyone who's looked at the monkey business
required to get 133/266 working would red flag
this one from the start. I'm also very suspicious
of the AGP 4X number...
-- All that's left of me, is slight insanity, whats on the right, I don't know. -- Bob Mould
Dear "Melted".
The following list of posters await one of your "Your Stupid" posts.
Pyte (140350)
Jeff DeMaagd (2015)
justins (80659)
osho_gg (652984)
tftp (111690)
Please be prompt with your reply.
"Five to ten people who know what they're doing is enough to put something like this together in two years time, tops. "
Another 'Dear Stupid' has indirectly addressed that issue.
"Granted, their product won't give you GeForce 6800 performance in its first incarnation, but these days it has become a lot easier to design custom logic."
Designing is only part of the problem. Economics is one of the many other issues. Addressed by a fellow 'Dear Stupid'.
"Besides, they're essentially "standing on the shoulders of the giants", so they already know what _not_ to do."
Unless some of those "Giants" are helpful patent holders. Knowing what not to do is going to be a rather confined field.{1}
{1} Something I've noticed that hasn't been addressed by them, so far.
You know, with HP and IBM saying they're backing Linux, and HP selling linux laptops, I'd say they should try get help from there. If they're going to succeed, they need a vendor anyway, and what better way than HP's linux laptops? Adapt the chips to both desktop and laptop formats, get a high rate sales like HP and everyone will be happy: the linux community because it'll finally happen, the company because they're making money, and the customers because they've got 100% Linux support on their hardware, which to me is the best price/value to find right now!
---- I am certain of only one thing : I know nothing else.
ATI already supposed specs for their R2xx cards. So everything up to a ATI 9200 has accelerated 3d support under X.org using the standard radeon driver. You won't get speeds as fast as the ATI drivers and some things like texture compression aren't supported due to patents but it gives good performance for something like chromium b.s.u and tux racer.
Check out gatos.sourceforge.net for info on the open source video input/output support for ati cards.
"When you sit with a nice girl for two hours, it seems like two minutes. When you sit on a hot stove for two minutes, it
Maybe you didn't notice it, but the GPU is a Xilinx FPGA. If users will get a change to write/customize their own FPGA code, this may result in some cool home-brew multimedia algorithms. Not to mention that tech companies will be able to use it to optimize products like video players, etc.
Tech Source is the company that supplies cards for our Solaris workstations. The driver quality is pretty decent, but we are only doing 2d. I would guess that they are fully capable of doing a good 3d card though. My only qustion would be price, as I think we paid $300 for our last Tech Source card, and it was a 64meg card.
"He's lost in a 'floyd hole"
I'm sure the serious possible contributors won't spend a lot of time on this unless they can be assured that their contributions are freely available in the future.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
The project, though a commercial one, wants to become a true community project and encourages experts and everyone who have good ideas to add to the development process
Hehe, I always knew that there was a better alternative to outsourcing the manual labour to cheaper countries (like ATI & Nvidia do). All you have to do is convince people it's good for the whole of mankind, and you'll even get the R&D for free. What's next, the open-source Nike equivalent? Beats the PR-hassle you get with child-labour any day!
All jokes aside, I would encourage the project, as long as it does not follow the current gaming hype. The power of ATI/Nvidia is that they work closely with gaming companies, and even bundle their products and research. That is a huge competitive advantage over any newcomer! (Industry) Power comes in great numbers... So it would be a good idea if the OS gfx-card would target current OS applications and games and collaborate with them. That could stimulate potential spin-offs of further games and apps designed for this card and create a whole new industry.
Btw. is there any hardware which is currently being developed in an OS environment... and successful?
How can you compare software development with hardware development? You need lots of money to create hardware, and if it doesn't sell because your tech is expensive and under-perform, you won't have money to improve on it. I don't think TS is sitting on mountains of cash, so if this project fails, it's likely to be the end of this ideal.
Open Source software also has the price and the replication advantage, for both costs $0 in most cases.
Clearspeed http://www.clearspeed.com/ is just coming to market with their CSX600 'application accelerator' processor.
It has 96 execution elements, 96 ops simultaneously for your data. Sounds like ideal for graphics processing. Power consumption 5 watts, 50 GFlops of computing power.
And they make PCI-X cards for PC systems. You can have several cards in one system for compounded processing power. Now, all this monster would need is the graphics output parts and drivers. They even have a full development kit for both Windows and Linux. The card's programmed in C.
Perhaps the PC's of the future would have two CPU's, one linear general purpose CPU (current x86 based) for program code and system management and one massively parallel CPU for tasks better suited for it. If there's no one true road to happiness, make it two then.
Might want to consider setting up a site for people to register their interest and potential orders, not just how much you would pay for but actually get the orders.
I don't remember if it was successful, but Sony has done this in the past. I know it failed once due to (I believe) a weblogic crash due to too many orders or weak system.
If the website is mentioned every time a story appears on slashdot or some other site, you can continue to accumulate and update information. If you make transparent the financials behind it, people may rush in to get you over the threshold of a precalculated breakeven point (including reasonable profit of course).
Personally I am in the market for a graphics card in the next 6 months. I am planning on getting the best I can afford at the time, and am curious what this project might offer to sway me. Sure performance is not likely to beat the top of the line of the other competitors at the same price point.. at least that is what one would guess. Maybe not true? Well, the FPGA looks really cool.
Consider that the fastest supercomputer in the world is the GRAPE-6 (GRAvity PipE) built on FPGAs for simulation of gravitational interactions (of globular clusters, etc.).
I was thinking it might be closer to something insanely great if you go for the multiple channels now for example. Maybe if you ask about that on your site you'll get people to agree. (How much more would it cost? etc.).
Also I don't know what the FPGA would promise, presumably quick firmware updates from the net of course. Could part of it be used for another purpose, or is that too difficult? Could an additional FPGA be turned into a chip that runs linux (use it on a PC) or perhaps be flashed with the results of another project (I'd love to have a Perl chip.. make it and they will come?) Could another chip or expanded memory provide say a video wall controller with edge blending for multiple screens in realtime? This kind of thing alone might sell enough to make it useful. What do commercial image processors have that this couldn't?
I just saw a sexy video switching fabric thingy here
I am curious about what exact "X.org eye candy" this would enable. I am guessing some of: "Direct Link for this comment Brilliant, and about time By Bryan Kagnime (IP: ---.polarnet.ca) - Posted on 2004-11-28 08:23:43 I don't really care so much for the 3d gaming aspect, distribute with the card an opensource operating system like Slackware with some 2d desktop eyecandy (translucency/transparency/openGL) and I'll buy a card for everyone I know with a comp. This'll show users *what* linux is all about, distrobuting a superior product and opening the market share for innovators." ?
One post on osnews mentioned realtime encoding/decoding of video streams, and though I am not sure this would not still impact the rest of the machine considering the design, that sounds neat!
128MB is enough to hold a couple frames of 20 times the resolution of a 1024x768 screen and still have over 30 MB left over. What if it included support for edge/corner blending and warping for a video wall? Is it conceivable that this could take the output from a fast consumer card and provide 2D warping and other effects for displays using multiple projected patches? Consider what it is good at. How about talking over the network or other bus to other oss graphics cards for multiple projector support.
If some nonvolatile memory was included, the card could remember a video wall wallpaper and open window/document information, or keep some megapixel images or something else always available. Would this be useful, say for quick startup or as a backup for important memories?
How about selling with an external patchbay that can take many video sources and provi
Based on their own specs this card will only support PCI and AGP. Furthermore, they have no mention of any support for programmable shaders (pixel or vertex). That means this will be an "older" card than a R100 or TNT card.
Developers who think they will learn how modern graphics cards work by contributing to this project are simply misled.
Ozo
Even though this is scheduled for a year in the future, I don't think standard TVs will be gone away by then, and good TV-out support is something absent in ATI/NVidia video cards. S-Video is missing from the PDF spec.
The absolute #1 focus for this card (if they hope to get people to pay more than $30 for it) needs to be fully reprogramable by mere mortals. It would be absolutely wonderful to get a general-purpose FGPA in a computer. People pay more than $100 for crypto cards, video capture cards, etc because hardware is so much better at those tasks. This would wipe the floor with them, because you could program in a new codec or cipher.
Even if it didn't have any video-output at all, I'd still pay $100+ for a PCI card version. Once video encoding apps are optimized to send the processing that's hardest on the CPU to the FGPA instead, I expect we'll see huge increases in encoding speed. That, BTW, also leads to much more complex codecs (MPEG-6 anyone?) that reduce filesize/bitrate significantly.
Besides that, I would also like to see a bit of effort in making sure it works on non-x86 hardware. Since this company makes video cards for SPARC systems, I that surely would not be difficult for them to handle.
If this thing actually sees the light of day, it will completely change what a videocard is. This also strikes me as a potentially piviotal moment in computer hardware. Perhaps, a few years from now, the biggest graphics card maker will have a museum wing dedicated to remember how it all started back in 2004. Yeah, I know it's a stretch, but this really does have that potential.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
"Just because you can't seem to see value in this doesn't mean that someone else can't."
Oh I see the value. However as an engineer, I also see things you apparently are more than willing to bury your head in the sand about.
Neither you, melted, nor the company have addressed the issue of patents. Also hardware isn't software, and all the wishing in the world will not change that fact.
I'm an expert at 3D card design, but if you want my help, you're going to have to pay me. My time is too valuable to give away for free.
I should add to my comments.
Someone suggested the SUN MAJC GPU which is another possability.
Basically the problem is that the community doesn't have a GPU under it's terms. Several ways have been suggested. Buy, or donate preexisting hardware. e.g. the staroffice/blender way. Build completely from scratch with existing ideas, and hardware as reference. e.g. the Mozilla way.
One way is to simply find as much hardware already under our terms (TV Tuners and MJPEG for example) and incorperate that as part of the design. The overall design will not be as efficient as say a from scratch ASIC design (in a number of areas), but it will be better than nothing.
Bootstrapping however is needed to hold people's rather fickle attention, especially in the face of cheap competition. Also one of the makers out there may pull a "harmony" which we don't want.
I could think that it might be fun if we had all the specs neccesary for anybody to compile their own firmware with gcc, and load it into the card in a safe manner.
TRANSLATION OF TRANSLATION: All the fruit of your labor are belong to us.
TRANSLATION OF TRANSLATION OF TRANSLATION: You've been 0wned, l00zer.
"Most M$ Office users use only a fraction of the "features" available in M$ Office. "
Kind of like the "You only use 10% of your brain" argument. Both are false. Which fraction is being used?
"So tell me again why an Open Source 3D video card project can never make it?"
No one is saying "never". However we are doing the one thing that zealotry apparently is unable to do. Provide a reality check. Here's one "fact" about OSS that will be harder to bring to this project. Software is easier because everyone (potentially) can be a participant (It's written into the GPL). The same isn't true of hardware. From the skills, and knowledge needed, to the actual creation of hardware. And there's so many obstacles. This project is more for ideologists, for that's the one's carrying this all the way through.
I'm not really looking for the latest and greatest accelerated 3D in a Linux video card, and I don't think there's all that many people who need it.
That being said, what I would like to see in a card of this type is built in TV-out capability, either through coax, RCA composite, or S-video.
I'd also like to see a dual-head card that supports multiple monitors.
If this project can deliver a card that supports these features and is easier to set up in Linux than the old nvidia geforce 2 I'm using now for TV-out (on a Myth box), or the Matrox G450 for dual monitors (desktop system), I'll definitely be in line to buy it.
A 3D accelerator is basically a vector processor, right?
And there are "off-the-shelf" (albeit closed source) vector processors available from various companies, right?
So you can still build an "open card" - a vector processor, a few DSPs, a RAMDAC and a bus interface linked together aren't going to be nearly as difficult to do as an entire graphics accelerator chip.
Then you take mesa, and really, really optimise it to run on your vector processor. You would then have really fast OpenGL support. I don't know how this would work out in practice, but it sounds a helluvalot simpler than what they're proposing.
http://www.icculus.org/manticore/
Manticore already exists for some time and it is also what they call Open Hardware. If they could work together, this could result into a good implementation for a Linux/Un*x hardware design.
F/OSS & IT Consultant
No one has to write an OpenGL driver from scratch. You just start with MESA and start offloading stuff to hardware as much as you can. It's not a great route to a great system, but it's a straigh forward route to something that works and is feature complete.
Well I've looked through their site, and a couple things stood out.
1) There product line is basically 2D, and that's it.
2) They already have a card with Linux support. Why reinvent the wheel?
3) There products are for the high-end business/government space.
So we have a company with no 3D experience, or economic experience in the middle to low end consumer space.
They have their work cut out for them. 2D is a solved problem, and has been for over two decades.*
3D is were everyone's headed and that's much, much harder to pull off in an acceptable way.
*solved enough that some are complaining that it's being neglected by the 3D companies.
I really doubt any true expert could join them, as the real experts are working for the card manufacturers, and signed NDAs...
The only ones that won't maybe are working for ILM, Pixar, etc, and those are equally jealous.
Unless someone can say there are some heavyweight video hardware guy in some university or so, I'm guess the effort will start from the very scratch.
Got Pike?
It's probably more worthwhile to get VIA to document in detail how to talk to the S3 graphics engine in their midrange motherboard chipsets.
People are probably right that this isn't going to produce a 6800 Ultra graphics card rendering 100 FPS in CounterStrike, but they will learn a lot about what really is needed by their customer(s). Learn the level of complexity and revise much of their design in the second revision. I don't see this project producing a graphics card tomorrow, but it could start putting pressure on these manufacturers. Maybe they will realize if they don't give us what we want then in the future they are going to have to deal with a cheap graphics card with an open design. I would venture to say that one of the most costly components of a PC is the graphics card if you are talking high end performance, and this is a way to start chipping away at that cost.
I'm not a fan of this idea. I think it will be lots of talk, with little-to-no production. I could well be proven wrong, but more cooks doesn't make a better soup.
I would contrast this with the development of the Linux Kernel or Mozilla due to the highly specific domain knowledge required. The people who are deep into chip design are few and far between compared to the number of software coders out there. This much smaller communitee is further hobbled because it could seriously be seen as a conflict of interest for a person working at Intel, NVIDIA, Matrox, ATI or whoever to contribute anything... and if they did contribute, it could bring up contamination issues that could mire the project in legal problems.
In short, I think this idea will not fly as proposed. I'd be interested to be proven wrong, but I'm not holding my breath.
"I will trust Google to 'do no evil' until the founders no longer run it." Hello Alphabet.
And that's after a slashdotting.
*sigh*
I guess the world just isn't ready for ventures like this.
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
...they better make their chip open source like the Leon2 SPARC is. That is a successful opensource chip project.
Your CPU is not doing anything else, at least do something.
The concept sounds great, but this card looks to be incredibly slow. What's the good of an opensource 3D card if you can't use it to run any modern games?
As an AC noted, hardware design is not softwatre design and what applies to one doesn't apply to the other. OSS has the opertunity to be just as fast, or faster, or just as featured, etc on a given platform. Why? Well it's all just bits in the machine. You can design whatever you like, you have the same constraints the commercial people do.
Not the case with hardware. Here you have to actually have something physical built, in this case, a dedicated graphics processor. Well, this company isn't talking about designing one and then having it fabbed. Instead they are going to have a card with an FPGA on it, and program that. FPGAs are chips that you can literally reprogram, from software. It's pretty cool, you design the circuts, and then send it to the chip which implements them. Great for prototyping.
Now of course one might wonder why these aren't used everywhere. I mean how great would it be to have a reporgrammable chip for everything. Ya well the thing is a FPGA is MUCH larger and MUCH more expensive than an ASIC for any given transistor count. If you can do it for $1 with an ASIC it'll probably take a $100 FPGA to do it as well.
So the problem with this card will be twofold:
1) Whatever price level the card is at, it'll be far inferior to a normal card of the same price. The bigger the FPGA, the worse the disparity will get.
2) Even if they put the biggest, baddest, FPGA they could get (which would be in the thousands of dollars) it still won't even come close to the low end of current hardware.
That's the problem here. In software, you are running on the same platform as the commercial software. If they can do X on PC Y, you can also do X on that PC. However with hardware, the hardware is your limiting factor. If your hardware has X transistors and your competitor has 100*X transistors, they can just do more than you. Your design can be the most brilliant ever, it'll still be slow be comparison. And if their chip costs less than yours, well then you are really sunk.
It's cool as a toy, but as for any real performance, not even going to happen. The Xiling FPGA they are using has about 1.5 million transistors. A GeForce 6800 ultra has 222 million transistors. Also the ASIC runs much faster, is more customized, and gets more done with a given amount of gates.
There's just not a real comparison.
Two things:
s way, you can leverage university students and hobbyists who want to make other types of cores or even single board computers since there's onboard RAM.
You're not going to get AGP 4x nor 200mhz DDR on a Spartan-3. If you do, you're going to have spent so much time laying out the gates by hand that you might as well have just went to work at McDonalds and bought yourself a Virtex4 with that money. And even then, you won't have a easy time laying out the rest of the design (the actual 3d and 2d processing cores). Whew...and then RAM? I forgot how much the Spartan-3 has in off chip IOs... but I doubt it's THAT many.
Next. Make the board double as a prototyping platform for people who don't have access to cheap FPGA boards. Check out "http://www.fpga4fun.com/board_dragon.html".
Thi
The reason is very simple: To make a copy of software so that one more user can use it - you just have to download it. To make a copy of hardware so that one more user can use it - you have to actually manufacture a piece of hardware.
---
Mod me down, you fucking twits. Go ahead. I dare you.
(I read with sigs off.)
The iCandy of Mac OS X's Quartz graphics layer is important because it sells. Companies with big fabs will produce only what sells; otherwise, they're not doing their duty to ensure maximum return on investment.
This is somewhat like somebody trying to get people to work on an open-source version of DOS. Sure, you get your freedom of the free software, but who would want to use DOS?
You'd be surprised as to how many devices embed FreeDOS, and how many firmware update discs include it. In addition, some users of video game emulators, EPROM programmers, and other time-sensitive applications still swear by DOS, as the application runs in what is effectively kernel mode and doesn't get randomly interrupted.
It would appear that GeForce SLI is architecturally very little like Voodoo2 SLI. Unlike Voodoo2 SLI, which interleaved scanlines from two video cards to increase fill rate, GeForce SLI interleaves whole frames from two video cards to increase fill rate and T&L.
Doesn't NVIDIA already own patents on the basic methods of teaming GPUs through the various technologies that it has branded SLI?
From the specs, that's roughly 1998 technology in graphics boards - triangle fill, but no "geometry engine", i.e. no 4x4 transformation matrix multiplier.
Heck, it's pre-1994 technology. Even the original PlayStation had a transformation and lighting unit called GTE (Geometry Transformation Engine, nothing to do with the company now known as Verizon).
If this card gets produced and sells reasonably well, ATI and Nvidia will be able to kill it easily by open-sourcing their own drivers, even just for their older cards.
That said, at the end of the day, the project would have benefitted the community.
Chuck Norris: Socialism == a thousand years of darkness.
They have experince with TV-tuners, why not include a quality TV-tuner with v4l2 drivers and a HQ TV-out.
:)
Then they might actually make some money on it