Slashdot Mirror
Open Source Graphics Card Available For Advance Orders
mollyhackit writes "The Open Graphics Project, which we've been following since it first started looking for experts four years ago, has just announced that the OGD1 is available for preorder now. The design features 2 DVI, 256MB RAM, PCI-X, and a Xilinx Spartan-3 FPGA along with a nonvolatile FPGA for programming on boot. FPGAs are reprogrammable hardware which means the graphics card can be optimized for specific tasks and execute them faster than a general purpose CPU. The card could be programmed for certain codecs to speed up encoding or decoding. An open hardware design means potential for better driver support. Of course you could always use the FPGA for something else... say crypto cracking."
The SPARTAN 3 is a hobbyist FPGA. Cheap, and a lot of gates, but slow. A Virtex 4 would've been nicer :)
But does it take a beowulf cluster of these play GLQUake?
What, me worry?
They're probably going to own 100% of the high-price videocard market with that.
... does it run Linux?
At $1500.00 each, I don't think the adopters are going to be on slashdot! Or at least not the ones who I know!
"My immediate reaction is "WTF? What kind of moron doesn't make things 64-bit safe to begin with?" Linus
Nice
That's PCI-X, not PCI-E. The rest of the stats are also a retro-blast as well.
I'm not sure what kind of architectures you could really test with this thing. It has slower memory on it than is on my motherboard. I honestly believe you could write software renderers faster than this thing.
Done with slashdot, done with nerds, getting a life.
Since I'm a programmer, not a HW guy, what does these chips get me, the equivalent of a Voodoo3? I mean, dual DVI and stuff, very good and all, it doesn't sound like they have any real software base to put w/ those cards yet, so that sounds like a lot of very low-level programming. Sounds like it will be a while before anything is written for the hardware, hell it sounds like it might be a while before you get any actual hardware!
Save $100, only cost $1400. That amount it better be able run Doom at 10,000 fps, now I can say I have the fastest fps possible.
Freedom of Speech only include discussion that are approved by the RIAA, MPAA and DMCA.
Pci and pci-x is dieing
And then have it run Linux (or some other free OS)? I think that'd be pretty neat.
Having recently taken a graduate class where I had to write my own shaders for OpenGL, it was neat to play with the video card on that level; however most cards are quite limited with what is open API.
This card, while too expensive for me, might spur some interesting projects - cypto stuff and Ray tracing come to mind. I hope someone does something great with this.
At this stage of their development.
This is not a finished product by any stretch of the imagination. These are prototypes. Back in the day prototypes were wirewrapped nightmares and they cost a lot more than $1500!
This card isn't particularly useful on its own, but if you want to write graphics card drivers - It sounds like the bees knees. You don't have to go out and buy 10 different cards, cause you can just emulate the suckers on the FPGA - write and verify your driver and move on.
Prediction: The real iPhone killer is going to be sex robots from Japan. Think about it.
crypto cracking
you might be getting ripped off if you're paying $1500 for a Spartan-3 board.
I guess they don't really have the board volume to get low prices. But If you want a graphics card for $1500 that's probably less functional than an NVidia commodity card, I'm not gonna stop you.
OTOH, If you're interested in FPGA programming and a novice at it, you'll want to get a MUCH MUCH MUCH cheaper Spartan board (like 50 to 150). See http://digilentinc.com/ for good starter boards.
If you're serious about FPGA programming (or just willing to pay $1500 to $3000) you will definitely want to get a board with a Virtex or Stratix on board:
http://www.xilinx.com/products/devkits/HW-V5-ML501-UNI-G.htm
If you want to have it on PCIx:
http://www.xilinx.com/products/devkits/HW-V5-ML555-G.htm
You can also get FPGAs socketted for AMD's Hypertransport bus and Intel's FSB:
http://xtremedatainc.com/ (Altera FPGAs)
http://drccomputer.com/ (Xilinx FPGAs)
http://nallatech.com/
http://celoxica.com/
(some of these vendors also sell PCI solutions)
FPGA programming environments still mostly suck. it's a market impeded by proprietary standards and a whole lot of NP-Hard algorithms. We're working on it...
How are you going to emulate video cards that are undocumented enough to not even have existing open source drivers?
A drivers developper usually develops the same driver for the same family of cards. When he needs more than 1 card, he usually needs the complete range of the same family of cards. A RadeonHD developper might need to have both entrypoint cards (say, HD2400) and high-end models (HD3870X2). Same goes for a nouveau developper and nVidia cards.
Spartan 3 doesn't seem to have a huge amount of cells. So, even if you could cram entry-level cards designs into it, that won't be usefull for developpers because they don't need all the entry-level cards of all different brands (which could be fitted inside the FPGA), but a whole range of card of the same maker (of which the bigger models won't fit inside the FPGA).
What this is interesting for, is developing drivers for a system that DOESN'T exist yet. At least the developer can simulate the entry-level card and develop some proof-of-concept drivers which might work on the final product and might serve as starting point to biuld drivers for the bigger variants. All that while at the same time engineer modify the chip design. This enabling communication and exchange of idea between the hardware and software designer.
And that's what actually this board is done for : to have a platform on which designer could build the real final card, and which will help the develop and test these designs before they are made, even is they can only test the "low-entry" ones.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Nice links there. I was wondering if you have an idea on where to get bargain basement priced DSP boards for audio. Mono 24 bit I/O is enough.
I don't have the kind of cash they're asking for, for a graphics specific FPGA. If they could tailor the board towards the FGPA market in general, I'm sure they'd find people interested in more than just it's rendering capability (me!!).
:D). If the card can do offline rendering efficiently enough to experiment with discrete pipelines (more gates = more fp precision!) I'd be a happy graphics geek.
I'm concerned about the shelf-life after I'm done tinkering.
I'd like an I2C bus, a few led connectors, and some magic so that I can connect a general purpose daughter board the FPGA's address bus (ie: implement USB, LAN, audio support that way). Every FPGA should be able to run as a Tanenbaum CPU by law!
As far as rendering goes I can't see an FPGA being as fast as an ASIC - propagation delay is going to hammer it, and syncing will be a bitch - but I'm still interested in what it can do offline (assuming I can get a vesa console
Good luck!
Matt
I've got a 12-processor Sun E4500 that I want to put some graphics boards in and use as a workstation, and I've been having an annoying time finding anything that fits in a PCI slot, has proper open source drivers, and has dual DVI-D outputs. The closest I seem to be able to get is the Matrox G550, but that can only do up to 1280x1024 for DVI-D. This looked perfect, even if I'd have to spend my free time for the next n months writing Verilog for it, until I noticed this.
That's right, you need a closed Windows-only tool to synthesize and download logic for the FPGA. Bleh. :(
... does it run Crysis well???
Fifty watts per channel, baby cakes.
At least, it doesn't appear to me that it would. The product page states that the thing is sold blank. Unprogrammed. Meaning it's not gonna run *anything* till someone programs the thing. Once someone codes it up to run OpenGL/Direct3D decently, maybe it could run 3D games OK; kinda hard to tell. The hardware in it appears to be top-notch, in terms of lots of high-speed ram. Can anyone give us any idea what kind of performance that FPGA can give?
It looks like, basically, this thing is a $1400 prototype that OEM's could use as the basis for a consumer video card.
Can someone out there who knows more about hardware design and fab than I do tell me - once someone has come up with decent programming for an FPGA, can non-programmable, cheaper, maybe even faster, chips be fabbed? I assume that is generally how the design process works - start with an expensive, programmable chip, get the firmware correct, then mass produce non-programmable chips that are much cheaper?
I'd love to see a graphics processor that could be programmed to create graphics that look like classical hand-drawn animation. I think you'd need to do some curve-fitting in X,Y,Z and T in order to achieve that. We all know there are cel-shaders and vector renderers that can render 2D stills that look authentic, but that's still a far cry from animating something that looks like a Disney classic, or like anime. Fitting 3D polys in X, Y, Z into curves in X and Y may be trivial, but figuring out how to turn data from X, Y, Z, T into X, Y, T is the real challenge for non-photorealistic cartoon rendering.
I'm more of a video card user. But I could try to learn what it takes to program this card. I'd just want to be sure that what I want to do is feasible. The actual designers might understand if it is or not.
Much of my computer use, including almost all programming, is done in text mode with Linux virtual console semantics, which work better and faster than terminal emulation under X does (for people that are used to it having done it this way since Linux came out, and on other systems before that). The trouble is, modern video cards have not kept up with text mode because most people don't use it. While they can still do it in basic VGA modes, they cannot do more advanced levels of text mode. Some cards, such as the Matrox Millennium G450 AGP, do OK up to a point, but have limits (maximum number of video scan lines in text mode is 1024, and a limitation on pixel clock).
What I want in a video card is one that can do text mode and still also do graphics mode, both in a pixel geometry at least as much as 2560x1600 (which the OGD1 lists). I don't need a lot of the other features for this usage case that other video cards focus on, such as 3D rendering.
Maybe this card can do some new text mode advances, with the right FPGA programming, such as 16-bit character modes (with 16-bit attributes) and not be limited to 8 or 9 pixel character width. Maybe it can also cache all the text buffers in its 256MB of RAM. Assuming 256k bytes per buffer, that's up to 1024 virtual consoles (but a lot of memory would need to be taken away to use for font storage). Another possibility would be overlaying graphics and text mode together.
What I don't know are two things: 1: how easy it is to program this card ... and 2: how easy it would be to do what I want to do. I can't know the 2nd without learning the programming, unless someone that already understands the programming for this card and understand what it is I want to do can figure that out and tell me.
Another idea might be way beyond this card until a GPU is available. That idea is to run an X server right inside the card. Then a simple driver in Linux/BSD could pass the X connection streams into the card, and a process can do network listens for remote clients.
now we need to go OSS in diesel cars
"If I had the money to spend I'd help support them because I'd love to see them get the money to build a truly open video card that could compete with it's modern rivals."
This makes me wonder if someone could setup a foundation of some sort to act similar to Public Radio/Television. I don't have the money to build or run a radio or TV station, but when 15000 people donate 50 or 100 bucks, suddenly the money is there to run the radio station. The problem with this, however, is that public radio and television are fairly well developed at this point. People immediately get benefit for the money they donate (in fact, benefit before they donate). Whereas with hardware design like this, there's no guarantee that any decent, full consumer product will ever be produced (that is, it's speculative), and even if it is, the best you as sponsor can hope for is that you get a discount when the product is released.
Heres another (cheaper) open source graphics card project
http://wacco.mveas.com/
There is no way in hell these people can compete with ATI/NVIDIA. Have you ever been to NVIDIA? Do you have *any* idea how many really smart people they have working on these problems 60 hours a week?
This project would be so much better off reverse engineering Cuda to make an open source driver than trying to make their own graphics chip. Hell, even Intel is having a very hard time getting a high-end graphics chip to work, and they've got so many more resources than this project.
Open source software works because anyone can hack on it and produce comparable stuff with zero initial investment. Hardware does not work that way. There is just way too much of an initial investment required. Even with FPGAs it's too expensive, and you're way too far behind to start with.
These people are idiots to think they can succeed here unless one of them has a 90nm fab in his or her backyard. (Sorry -- this is qualitatively different than trying to write your own OS, which is done all the time in undergrad classes.)
A Formula 1 car might not even be street legal, let alone comfortable for anything other than racing.
Right tool for the job, can't be said enough. Even CPUs have things like math coprocessors. It's no surprise that even if a video card technically can run Linux, it might not be very good at it. (A fair example: PS2 Linux.)
Don't thank God, thank a doctor!
I am quite intrigued by this offer. I've heard of this project before and was interested when I first heard of it. Now that they are ready to offer the hardware, I find myself seriously considering purchasing one. My question is this: Given that I am an experienced software developer (C, C++, Java, Perl, etc, etc, etc) that has spent most of his career on business and financial systems (though I have done some device driver development and near-real-time processing systems [Currently working on control systems and UI for Radiation Monitors for ports of entry]) and that I've always had a lot of interest in graphics programming, what could I do with something like this? In other words, if I've never done any FPGA programming and not much 3-D graphics programming (though I am familiar with matrix algebra and such and have read the OpenGL Bible cover-to-cover a couple of times), but, could I have FUN with this thing? Anyone with any experience with this type of thing have any opinion? Look forward to people's ideas and recommendations.
Over-the-top Response Guy! Giving "Over-the-Top Responses" since 1970.
There are literally thousands of FPGA prototyping boards on the market, many of which cost much less than this one. So while you could use this for other things, I can't imagine why anyone would spend the extra money unless they wanted to use the video specific features like the dual DVI interface. Furthermore, the purpose of the project is to develop an open source video card, and this card was created as a tool for those developers to experiment with.
So, it was created to prototype a video card, and it's only practical uses are real-time video (output) processing, thus it is a video card.
This sounds way more practical than the OpenGraphics thing. $1500 on top of having to find a PCI-X board? No thanks.
...hosted by Xilinx even!
http://git.xilinx.com/cgi-bin/gitweb.cgi?p=linux-2.6-xlnx.git;a=summary
-- PGP keyID: 0x4C95994D
there is a reason that fpga's are not normally used in these types of applications. they are slow and get quite pricey. the ones my company puts into certain military hardware cost $2000 each
No you can't. If you had the info to emulate you could write a driver, and probably a lot easier.
This is a research & development tool for designing hardware or for use in highly specialized applications.
Whether Project VGA is "way more practical" depends on what you're trying to do. The OGD1 was intended as a tool for developing a graphics card with a high-performance 3D rendering pipeline. The FPGA won't compete with the ASICs from ATI and Nvidia, but is perfectly suitable for a prototype to demonstrate that the logic is functional. Eventually that logic will be put into an ASIC, and used to make graphics cards that are cost-competitive.
how did you get modded up? Your comment is insanely stupid, an emulator requires you to already have intimate knowledge of the graphics card you are trying to emulate, they are also infinitely more difficult and time consuming to write than a driver, not to mention how are you going to adequately test these drivers without the cards int he first place, not to mention none of the market leaders in graphics cards are likely to publish enough info for you to write a good emulator in a timely fashion.
You know I just had to ask!!! :-)
My ism, it's full of beliefs.
While Project VGA does look very interesting (and much more practical than this particular project), I have to point out that PCI-X is almost always inherently compatible with PCI, and vice-versa. In this case, the manufacturer specifically states that it will work in a PCI slot. If you're having trouble finding a PCI motherboard, then maybe you're doing it wrong.
Screw the rules, I have green hair!
You're mis-understanding at least a part of the target audience.
One of the MAJOR problems with Linux desktops for a long time has been the reliance on proprietary drivers for any decent performance out of video cards. This card may not be the "fastest gamer card" out there when it arrives, but it will allow tuning of the drivers to card hardware in ways not possible to the open source world currently. Personally, I would be happy with enough acceleration potential to make 3d graphs/Blender style scenes render smoothly - I'll sacrifice the ability to play anything more complex than bzflag at high fps as long as applications for Real Work and nifty desktop features work well.
While $1500 is not something I can afford right now (and I'm not the target market anyway at this stage) I would definitely pay a significant markup to obtain a consumer graphics card that is open. Indeed, it would be extremely interesting to use such a card and other freely available hardware designs to create a completely open hardware system.
The open source video card is such a wonderful idea. Bringing video cards into the open source movement would be a huge leap forward into making open source a viable option for all users. Why sully the whole thing by making it look like it is some sort of hacking tool? Time to grow up and join society. Open source can be the salvation of the computing society, which should NOT include hackers. By the way, I hate that term. A REAL hacker is not someone who commits crimes with computers, and it is a shame that term has been co-opted by people who want to break the law, or who are just unwilling to join us in the civilized world.
Open Source: Eroding the Digital Divide
One of these in an OpenSPARC machine?
Oh yes... the future is now
Just bought a NVIDIA GeForce 9600GT with 512 MB for 100$ and have it running under 64Bit Ubuntu. If I consider this while reading the price of the open source graphics card this really puts things into perspective. I understand that this card is not only a graphics card but an all-around programmable FPGA and I would really consider one for 150$ but not for ten times the price.
If you really want totally open source drivers they'll be available for ATI cards in the not too distant future.
This graphics card when it finishes won't be "open source" because you won't be able to change it, it might have open specifications, and it might have a good relationship with the open source community, but the open source community is just as bad at maintaining a relationship with hardware vendors as hardware vendors are at maintaining a relationship with open source.
If you're willing to pay $1500 for your ideology that's your call, but I'd rather pay $500 and get a card that's substantially faster, and is actually programmed to do something other than diagnose itself and I don't really give a rats if the drivers are open or closed source.
If this works as well as most Linux Discos, gods help my grandmother who will have to plot her only graphics thought a command line before she can play solitaire.
I've followed ProjectVGA for awhile, and while it's fairly simple, it's got a major advantage in that it's PCI. It's probably the cheapest PCI/FPGA dev board in existence.
The main caveat is that because it's a 2-layer board, signal integrity might not work out at high operating frequencies. Also, since the idea was to make a VGA-compatible video card, the FPGA chosen doesn't have a lot of logic, compared to to the OGD1. The part is 10 times smaller.
But you could still write something useful that would fit in 400K gates, I reckon. At least I'm planning to look into it. Maybe an open-source sound card.
Just to let you know (and because I'm a pedant ;) ) it's spelled "dearth", and it means "lack of" or "not large enough supply of".
:)
Basically, you said exactly the opposite of what you were trying to say
It's official. Most of you are morons.
I don't think this is meant to be a GPU, at least not this card. I think this is meant to be what the GPU is made on. If that's the case, then this is really cool.
Coming from the software side here is my idea of what this could mean: if the OS has problems with the GPU card, the GPU's specs are open and the driver could be matched all the way down to the bare metal. If the GPU doesn't do what is needed, the GPU can be redesigned on this FPGA card. If the tools used to work with the FPGA card are troublesome, the tools can be programmed better. If the FPGA doesn't do what is needed, the FPGA card can be redesigned.
I think it's pretty cool that a card can be completely open from before the design of the card. Sure, the price tag is a bit steep, but with time, I wouldn't be surprised if this project—or another like it—becomes a platform of choice in making video cards.
I think you guys are forgetting about how money works in the grant funded/ public university/ bureaucracy world. Colleges buy burned DVDs of documentaries filmed with a Flip for hundreds of dollars. If you can catch the attention of the right professor or department head, you can charge whatever you want.
No, that is not with this FPGA is about.
In fact, the original implementation of the device was a graphics card, but you do not get that when you order it.
You get a piece of hardware for prototyping, essentially, for all sorts of applications.
You COULD use it to develop a new type of graphics hardware core, then use it to send off to be manufactured in a different process technology that is much cheaper in bulk quantities.
But the article headline says this is some sort of open source graphic card that is suitable for the masses....
Which is not the case.
-Hack
Free seems to always be, by far, the most expensive alternative.
Just ask Munich how great their Lunix deployment went. So many millions of Euros spent on "free"... and it's still not working.
... the obligatory Beowulf cluster of these?
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
Yeah, I think that's the sound of GP going down a large well...
Your mind is clear / The things that you fear / Will fade with how much you / Believe what you hear
How can anybody think this is going to play 3D games with only 62,208 logic elements (equivalent to about 30,000 transistors)?
Not going to happen. Not even close.
It'll do some "rotating cube" demos, sure. At a stretch it might even accelerate a 3D desktop.
A full OpenGL or Direct3D pipeline though? Not a chance. The original GeForce was about as minimal as you can get to be considered a full pipeline for fixed-function 3D APIs (no shaders) and it had 125,000,000 transistors.
ObCite: http://www.tomshardware.com/news/Nvidia-GT200-geforce,5441.html
No sig today...
My hobby is women. I spend $4k per year on 3, whereas this $1.5k hobby would realistically have more results, without the clap. The ladies got to go!
62,208 logic elements is equivalent to 3-4 million gates, or 12-16 million transistors. This card is an FPGA development card. It is intended for *testing* the rendering pipeline. It won't have as many functional units as the eventual ASIC, but it will demonstrate that the functional units and rendering pipeline work correctly.
Ok, but even so .. the first GeForce has 125 million transistors. You're well short of that.
And that's without counting waste. Your "logic elements" will waste a lot of gates (according to a quick google as much as 50%) compared to real transistors.
What I mean is that I'm sure you can make a working graphics card in theory (We'll ignore the fact that it took four years just to get a development board made) but the people in this thread are talking about competing with ATI/NVIDIA.
Not gonna happen.
No sig today...
The FPGA card is NOT intended to compete with ATI and Nvidia. My understanding is that even the eventual ASIC version isn't intended to compete with the high-end cards.
Yeah, I worked with the original SGI machines. Flat shading, no texture mapping, no depth buffer...
If everybody's clear that this will only do very basic 3D graphics then fine. Reading this thread though, that doesn't seem to be the case...
No sig today...
Project VGA is an interesting toy: it's a bit like buying a project kit to learn how some basic piece of technology works. It provides an environment adequate to reproduce functionality in extremely-low-end video cards (VGA/SVGA w/o acceleration) that is already 100% supported through free drivers on just about every video card.
The OGD1 is something else entirely: it provides a development environment adequate to develop fast 3D rendering using a simple, but fast architecture that the OGP has designed. It also will provide a platform to test new video architecture concepts, which can then be tested in actual use, instead of on a simulator.
It's true that their architecture won't (quite) compete with NVidia or ATI -- mainly in the area of fancy shaders. But for engineering and scientific uses, it will be just as fast and run much cooler. When an ASIC version becomes available, it will be very cost competitive in a niche market.
Until then, the OGD1 is a very large hobbyist (or moderate commercial) FPGA prototyping board with fast D/A and DVI ports, which make it somewhat unique.
Project VGA is much cheaper, but it doesn't provide any of those capabilities. It should also be appreciated, that these two projects are working together. Project VGA cards will probably be used to test the VGA bios for the OpenGraphics architecture.
Actually, the cool thing is that the PCI controller on the OGD1 is an open source design, implemented on the smaller XP10 FPGA on the board, created by the OpenGraphics project.
A small win perhaps, but they didn't have to pay any licensing fees for the PCI or agree to any limitations on production.
I imagine they'll have implemented a newer standard by the time a consumer ASIC-based card comes out of this. Or if not, they'll use something more widely available.
Meanwhile, PCI-X does what is needed for this project, and without proprietary code. Since open hardware is the point of this project, this is consistent with OpenGraphics' goals.
That's _really_ expensive, and it doesn't include a decent FPGA or PCIe. There are several other FPGA-based boards that supply much more, at or around the same price.
For a little more than 2x the price, I can have a 4 lane PCIe board with full HD (DVI or HDMI included) support and a much better FPGA (Cyclone 3C120).
I think it would have been a better use of time to work from an existing FPGA development kit, rather than re-invent what didn't need inventing. The hardware that goes into developing FPGA kits is _not_ closed source. In most cases, the FPGA vendors want you to use those board designs as starting points for your own.
I'll be looking at the IP, for this board, and re-implementing in a (much better) off-the-shelf development kit.
"...The smart and lazy ones I make my commanders." - Erwin Rommel