Slashdot Mirror
Free Software Friendly Graphics Card?
An anonymous reader writes "There's an interesting discussion on KernelTrap with a hardware company that is talking about developing a 'free software friendly' graphics card. The idea is to fully disclose and document all register interfaces including the BIOS, providing Linux and BSD users with a fully supported video card. The hardware engineer proposing the idea summarizes his viewpoint saying, 'the whole issue comes down to this: This is technically feasible. Should we do it?'"
Does your company have to divulge any proprietary secrets in order to leave everything open for this card? If so, is that okay or does that do them harm?
http://www.busyweather.com/
Yes.
thisnukes4u.net
Can it be cool and still be royalty-free? Or are you going to get shut down by the big boys for stepping in their patents?
How about a Free Software Friendly Audio Card to go along with it?
I don't know about others, but I've had *way* more trouble getting audio to work on my linux boxes than I've ever had configuring video.
What? Create a functional and supportable video card that is platform agnostic and will just work? The problem is, it is too logical. Unfortunately, it won't work in todays economic environment. Unless you are screwing over your competitors, your customers, or your employees, you can't make a buck.
I like the idea. My only thought is, are they going to have enough pull to make this happen? Graphics cards are much more than just throwing a few hundred million transistors on a chip. You have to worry about pipeline architecture, parallel texturing units, and (most importantly) well optimized driver software.
Can this company create a card that's competitive? And if they can, will they get pushed out of business through patent litigation?
Javascript + Nintendo DSi = DSiCade
They may have the best drivers for their card in two years, but I don't see how they can compete with Nvidia/ATI even with opensource drivers
open hardware? free [as in free speech not as in free beer] hardware?
SHOULD we do it?
/etc. ,BR>
Yes.
WILL they do it?
No.
~~~~~~
It's a "trade secret thing. nVidia doesn't want ATI to know what they are planing / doing so they can make their buck...
I'm not a prophet or a stone-age man,
I'm just a mortal with potential of a super man.
I'm in!
It is an interesting idea. Open source hardware, sounds good to me.
Why bother?
I think it's very unlikely that anyone starting from scratch will be able to compete with nVidia or ATi on performance, and there aren't all that many geeks who care about hardware openness enough to give up the value.
So, I predict that it will be expensive and low-volume, and (sadly) will eventually fail.
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
While this won't become mainstream, I think it is a really cool idea.
If nothing more that just to see what's involved in all the development. I mean, I'm an MS EE student, and I can't even tell you all the details in how to build a video card. I mean, I could on a large perspective, but I'd be curious on learning details, and just sort of exploring around the design for myself.
But, I guess I'm geeky like that.
both for the company that does it and the linux community. The company that does it will immediately have all the people building linux boxes flocking to them-- the people building games machines will still go with whatever's fastest, but people who just want everything to work will be happy to know that at least with this one card, they'll never have to wait for drivers when the kernel changes, never have to worry about a buggy video driver, their system will just be that much more solid. So that's a nice little boost in sales. Maybe not a huge boost, but compared to the amount of work involved in opening the specs up, it's a great cost/benefit sort of thing.
It would also be fantastic for the linux community because the existence of such a card-- and the preferential treatment the card would receive-- would put pressure on all the other cardmakers to follow suit, or at least tighten up their linux support.
It's serving a small market(right now), requires thousands of man hours of design and testing, requires expensive fabrication equipment(too expensive for this company probably), and is unnecessary because current video cards work fine under Linux. At least well enough that spending $500 to buy a mediocre card by a small company is out of the question. And yes, it would most likely cost that much. With little demand, high development costs and high fabrication costs, it will be that expensive.
Hurricane Ivan: A 17th century prison collapsed. All of the inmates escaped.
Should we do it?
In March 1923 the British mountaineer George Leigh Mallory was asked why he wanted to climb Mount Everest, and replied, 'Because it's there'. Good enough for me. It's a noble idea and technically feasible. Wish I could help...
Until now, open source software has proven to be able to scare M$. Why can't open source hardware scare competitors of it's field? Obviously it's not the same but hopefully, if they all planned it well, and by the article it shows that they got a nice idea, I'm sure a project such as this would get sufficent support to progress.
I'd be willing to pre-order 5 at $50/each.
Be sure to have PCI in addition to AGP.
Concentrate on making a product that works at displaying things on a screen rather than making a product that pumps billions of polygons per parsec or something. You'll lose all the customers who are just interested in penis extensions but look very attractive to people who would just as soon have a computer that just works and go play games on an xbox or something.
Its going to cost $500 and have the same performance as a $10 S3/Trident/Generic chipset card. Good luck to all 6 people who plan on buying one.
Only the State obtains its revenue by coercion. - Murray Rothbard
For those who didn't RTFA, the guy is talking about an FPGA, not an ASIC. Reasonable 3D is a pipe dream. OTOH, everything would be open spec: BIOS, card layout, and everything. As an FPGA it would be completely reprogramable.
I think it'd be great to hack around on, but considering the price of perhaps $100 I don't see this selling in quantity.
Seriously. nVidia already has kick-ass hardware and the best drivers available under Linux, plus one of the best, if not the best, installer for Linux that I've ever used. It would probably take less effort to convince them to open up completely than to create a new card.
Do you have ESP?
I have ordered some FPGAs for my company that cost $4000 each -- yes, that's four thousand dollars per each single chip. Now, I don't know how much you can pay if you want the largest, fastest FPGA, but saying "in an FGPA" gives one no indication of the end price of the proposed board.
The real "Libtards" are the Libertarians!
It isn't like audio cards are solving quantum cryptography or pushing the envelope of technology or something. They're playing audio! With ALSA linux finally has a lovely audio architecture, but it doesn't count for much since the drivers to make it work are such a fucking pain.
The nature of audio cards and what they do hasn't really changed in 10 years. Why is it that the software support for these things is so immature even when compared to crap like video cards that are bleeding-edge marvels of technology?
I'd be willing to pre-order a graphics card that fully documented it's specs and cooperated with the Linux community for my desktop. The problem is that many companies aren't prepared for such a thing, and don't have a way to take your money. So, helps us out... Where do we pre-buy one?
Sean
Nvidia are extremely open-OS friendly. Their driver itself may not be opensource but they have excellent driver and developer support under Linux and BSD, and the graphics card market is so cutthroat that if they were to divulge driver secrets, it would be suicide. Keep in mind that one of Nvidia's huge strengths has always been their driver technology, which is miles beyond ATI.
I have always been extremely happy with Nvidia and at this point I see no reason to buy any other make of card.
Yes, it would be nice if they could opensource more of their technology but I can't see that happening. I think they've bent over backwards to provide support to Linux, more than any other competitive graphics card company.
Intolerance for ambiguity is the mark of the authoritarian personality.
In such cases I hate speculation. If it *can* be done and a person/company/organisation wants to put up the initial capital to get it going then it should be done.
If it isn't done then there will always be the painful 'what if' forever after.
Personally I'd pay up to $250USD for the type of card they're suggesting if it had fast (ie: usable with UT2004) 3D OpenGL acceleration.
Ripping an new rectum in the fabric of spacetime.
So please do it. I know some Linux users take pride in their amazing ability to get some piece of not-really-supported hardware to function, and in fact there are whole companies which provide installation of Linux on unsupported laptops as their business, but this is not fun and is a waste of time.
When can I buy it?
It's been a while since I bought a video card (my ATI Xpert 2000 still works fine), but aren't Matrox's cards well supported by free drivers?
Seems that most people here didn't bother to read the article. (Big surprise.)
This is a 2D only card. He would not try to compete with BigBadVideoCardVendor. He knows that development of a competitive 3D card is out of the question for now. But you have to start somewhere.
Unlike an opensource software project, an "opensource" hardware project can't "show me the code" in order to gain legitimacy and gather developer attention. He's looking to see if there is real interest so that he can make a case to his boss. He seems to understand the risks involved, and I hope he can make it work.
What are Matrox and VIA doing these days?
Help fight continental drift.
If there is a market for such a card, it should be low end, as cheap as possible.
You will never compete with ATI/nVidia, and they are ignoring the low-end ($40) market. Other low-end manufacturers (S3, I'm looking at you) have absymal Linux support.
Forget about 3D. The number of people who 1. use 3D on Linux, and 2. don't buy the latest ATI/nVidia, is too small.
Cater to large-scale specialty installs, multi-head installs in schools, etc.
"I assumed blithely that there were no elves out there in the darkness"
That would rock the house considurably.
GRAPHIC MANUFACTURERS SELL GRAPHICS CARDS, NOT DRIVERS.
Open source drivers are a great inducement to purchase a card.
Even if the card is slower then others and slightly more expensive, I would still buy it. If it's very much slower and very much more expensive then it would be a issue.
Ok what I am about to say will only make sense is you understand what ISA's are.
Get out of your mind the ISA slot and the x86 ISA is teh suck. PowerPC has a ISA, for example.
It's a standard way on which software is ment to interact with hardware in your computer.
For example you first created the 386. Most of what the software ran on was raw hardware. However the modern pentium4's and Althons are VERY much DRASTICLY different from the original 386 cpu.
Lots more REAL registers. Lots of extensions, SSE, MMX, so on and so forth.
Why then are they able to run programs and even DOS OSes designed from the i386?
BECAUSE THE ABSTRACTION NEEDED TO FIT INTO THE ISA STANDARDS IS BUILT INTO THE HARDWARE.
So what we need for video cards is a ISA for them. Like the VESA standards, but for hardware 3d acceleration.
Something built around OpenGL, because it's open standards and universally accepted, unlike DirectX which is NOT just for 3-d but for input, sound and all sorts of other stuff and is only specific to one vendor operating on only one platform.
Think about it.
Video cards are mini miniture computers.
They have a micro proccessor.
They have RAM.
They have a BIOS.
So on and so forth.
So why not build the drivers for the video cards like you build a OS?
And why not build a opensource OS for it built around a Open ISA standard for OpenGL capable video cards?
Maybe a GDOS? Graphical Driver OS?
That way you have a choice. You have a generic OpenGL capable drivers that will run only any compatable video card irregardless of make or model. The GDOS would be something exceedingly simple. It only has one purpose, take care of OpenGL instructions from software running on it's Parent OS and transform it into instructions to be ran on the hardware itself.
Then people like Nvidia and ATI could take that Free G-DOS and add extensions to it for their own private optimized rendering stuff that sits outside the normal OpenGL standards. Propriatory ways of rendering Anti-Aliased text for example.
If they don't want to release their secrets to propriatory bits of software they dont' have to.
But if you don't want to run the propriatory software you still have full standards-compliant OpenGL drivers. If they are a bit slower, then so what? I'd rather have slightly slower Open source, open standards, drivers then slightly faster closed source drivers anyday.
I care more about the stability of my system then anything else.
Then when the OpenGL standards are upgraded, or you need a new generation of ISA to get rid of the cruft it would be simple, since you only dealing with a single-tasking, single-purpose, specialized peice of hardware. Backware compatability would be taken care of by allowing older cards to render in Software (Mesa) the bits that they can't render in hardware due to their oldness.
The OS would be kept independant of it. The kernel would be kept out of it. The G-DOS could be in it's own memory space or even in userspace (since with displays your only dealing with one user at a time)
G-DOS 1.0 cards
G-DOS 2.0 cards
So on and so forth. With in this framework their would be very much room for performance growth. It would reduce User's suffering, increase stability, and increase ease of debugging and testing.
And if some companies don't want to join in with the standards, along with everybody else. Then dinosaurs realy do go instinct, you know. But I don't see that happening. After all companies like ATI and Nvidia already do belong to open standards groups like OpenGL.
because they offer drivers for Linux. I'll have to admit though that if there were a truly "open" card out there that would allow me to play the games I play I'd go for whatever worked. I'm not religiously behind Nvidia.
for those of us that read the article, we see that the entire nVidia/ATi argument is practically moot. the developer explicity says that the card will be primarily 2D because his employer won't give enough funding to produce an ASIC. Thus, they're using an FPGA and will only really be able to implement a 2D core.
[move
A nice useful mid range card at a competitive price. I'll take 2, To start with and more later. We are starting to sell dual boot systems at the white box store where I'm a tech and sales type. We have sold a few in the last 2 months, some Fedora, some Suse, and one Mandrake. A nice mid-range card supported on Linux and Win XP would be perfect. Just make it a bit cheaper than the Radeon 9600, with similar performance and I'll be able to sell the hell out of them. One of the biggest complaints amongst Linux users is support for video and audio.
Professional Politicians are not the solution, they ARE the problem.
nVidia doesn't own everything that they sell. They cross-license existing technologies from other companies who are keen to keep getting royalties checks from nVidia. nVidia couldn't open that if they wanted to, and judging from their attitudes in general towards open source I bet they would want to if doing so wasn't such a massively bad business move. They would have to rewrite everything from scratch with a cadre of lawyers watching every move to ensure that they didn't step on the patents of their partners, all to help 15 people sleep better at night because they didn't give into evil capitalism.
Just a reminder, drivers are a cost for video card manufacturers. They sell a card and have to bury the cost of driver development and maintenance into the cost of the card. Open sourcing driver development lets a card manufacturer profit from the hardware while the community develops drivers for them and they get good karma to boot. This would be a fairly inexpensive/low risk way for a low-end (PCI only it appears at the moment) card manufacturer to get their "foot in the door".
Only the "big boys" (Nvidia and ATI) have anything to lose by open sourcing their cards. People would actually see to what extent they fudge their cards and drivers for benchmarks.
They that can give up essential liberty to obtain a little temporary safety deserve neither safety nor liberty.
Ben
It may be technically feasible, but what about financially? The interesting thing about open-source consumers is that they're mostly talk, but when it comes down to actually buying all of they stuff that they claim to want for Linux, they don't vote with their dollars. Just look at the failure of Lokigames to make a profit, not to mention id's big profile attempt to push Linux by doing a simultaneous Linux/Mac/Windows release of Quake III - sales of Linux Quake III were abysmal.
Expecting geeks to pony up a few hundred bucks for an open-source video card that has little if any chance of competing with ATI/Nvidia on speed seems pretty unlikely.
This will cleate many new jobs in China.
The only reason that I would want a graphics card with open specs is so there could be an opensource driver for 3d acceleration. Every major card works without issue in 2d mode. It's the 3D acceleration support that's flaky!
So unless they make a card that can play doom3 on linux as fast as some of the new nvidia cards they're out of luck.
What's the point of using a second rate graphics card just because the drivers are open? I'm better off buying a cheap Geforce, using the open 2d drivers and still having the option to reboot into windows and play games.
An open, documented piece of graphics hardware has tremendous commercial potential, and would see considerable use in price-sensitive markets. Such a design would succeed not because geeks buy FPGAs and burn the design onto it, but rather because chipset vendors and embedded systems designers could simply use the "open, standard" video card implementation and avoid reinventing the wheel.
Once the 2D core has been proven commercially, the companies that use it will be interested in adding features such as 3D acceleration. Then we'll see the combination of volunteer and professional collaboration we're so familiar with in the F/OSS world.
Please come up with something original not "FreeGeForce" or "OpenRadeon."
This guy is way out there
yes
just because your a schizophrenic doesn't mean people arn't really out to get you
It really should be something that laptop vendors can use.
What I would rather see, instead of a card, particularly, would be a design for a generic register interface that any vendor can implement. Each vendor can provide as much optimization as their market will bear. The creator of the spec would have first-mover advantage, but eventually everybody would have to support it (in addition to whatever else they had). Then, any new laptop would work with at least the generic driver. I know VESA was an attempt at something similar, but it was at the wrong level and too weenie.
Maybe there is already an interface in use in some "obsolete" card that could be lifted wholesale, and then cleaned up and modernized. It seems a suitable subject for an ECMA standard.
no
I wasn't sure whether to read this as "Free, Software-friendly, Graphics card," or "Free-software friendly graphics card".
Probably the Sam Adams isn't helping. Time for bed.
Download my free songs!
While we're at it, they could throw in decent tvout to this card. Just a simple framebuffer with ability to sync to vert refresh, double/triple buffering. And independent from main display. It doesn't even has to have backend scaller, processors are fast enough to scale DivX to 720x576 in real time.
Hell, I would pay up to $50 for a simple PCI card with low resolution (enough for PAL/NTSC), tvout, vert sync, double/triple buffering and good support in mplayer (so it means completelly open specs).
Robert
PS No, there are no cards on the market in the price range of up to $200 that match all those specs.
PPS No, dxr3 doesn't count, one has to compress video to mpeg1/mpeg2 in order to play it on this card, which results in lower quality. And because of this it eats too much cpu, as well as there are constant problems with a/v sync.
Bastard Operator From 193.219.28.162
"Unless you are screwing over your competitors, your customers, or your employees, you can't make a buck."
Well that explains the porn industry.
It's quite simple,
Cost to copy a 500k tarball:
Cost to fab a graphics card: $Hundreds
Bascilly one of the things that makes free software popular is that it's so easy to make thousands of copies of it. Hardware would be much harder to take off as it requires the consumer to actually fork out money to try it.
I've been through two Nvidia motherboards now and I don't see any value in them. Before I had an S3 motherboard with open source drivers and everything worked great. I paid extra for one of those boards with the nvidia audio chip on it and THE DAMN THING DON'T WORK. I'm still using the open source 810 drivers because Nvidia's own drivers don't work with the thing. And no one can tell me what's wrong because - guess what - no one can see the source to find the problem.
The S3 isn't really "free" either - the guy just managed to connive enough info from S3 to make drivers but the major distros won't use them because of potential legal issues. If there were an open source friendly chipset that could be bought on relatively inexpensive motherboards I'd not only have one, I'd be selling them.
Because back in Mesa 3.1 beta 2, the developers are dumb enough to turn the LGPL licence into XFree to appeal to the then-incompetent XFree developers. As a result, no video hardware makers want to pitch in, fearing that competitors have edge over them by not releasing improvements. This decision cost everyone dearly. They should have told XFree developers to get lost, and stick to the LGPL, then we won't have to pay $50 for a 'decent' driver that should have been there at the first place!!!
Close source solutions, simply to prevent exploding levels of complexity, necessarily impose constraints on what hardware designers can do. The more complex the software, the more expensive the maintenance, the lower the profits.
With Open Source and Free Software, there's no such constraint. If necessary, coders can always #ifdef the relevent sections, and distros can always provide the alternative binaries. As such, you can modify such software to make best use of the hardware, rather than limit the hardware to maximise the software's profit margins.
On that basis, it should be very easy to build a graphics card which implements in hardware some simple, commonly-used graphics functions or libraries that should logically be in the graphics card anyway. The FOSS drivers would look much like the FOSS drivers for any other card, except that those portions would just be stubs, with the hardware doing the work.
Doing that gives you: Differentiation (always a good thing, if you want to sell a product), improved performance (again, that usually doesn't hurt) and a fairly solid market niche, which ROTM graphics cards don't have. More importantly than all those, though, is that such a card would work faster under Linux or a *BSD than it could under Windows. You can always patch X11 or an Open Source kernel to not do something. I'm not convinced Microsoft could still do that with Windows, even if they wanted to. Too many things depend on too many other things happening a specific way, and in software, to make it trivial to cater for a graphics card that does more than other graphics cards, rather than less.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
That's what my nvidia card combined with suprnova gives me. a free software friendly graphics card.
if they open everything, and the card is able to play Doom 3 reasonably well, I will devote myself to the company exclusively. I would name my second child after the person responsible for getting the company to divulge all the information.
this is the one thing that every linux user has wished for. and damnit, someone needs to finally make the decision. others will have to follow, or lose market share.
This is doable, if you're willing to compromise on not being a "complete" graphics solution, IE, not trying to implement hardware accelerated everything.
I'd suggest simply a dumb frame buffer, and doing everything in software. Then your solution is simply a memory controller, plus a CRT controller and an output graphics DAC. Really, if you implemented a VGA controller from the early 90s sans the backware compatible bits for MDA,CGA, EGA, etc, and opened up the line/pixel resolution you'd be there.
If you make the goals modest enough, this could probably be done with a field-programmable gate array, an external graphics DAC, and some RAM. The only tricky part is external DACs are hard to come by these days and aren't cheap. (Prior to their integration in current-era parts, they were down in the $1 range, but since they're integrated now, the only stuff that's commonly available are insanely high end workstation DACs.)
Back in the mid-late 90s, right as 2D acceleration was hitting its peak and 3D acceleration was emerging, there were netlists of VGA designs for sale for as little as $500. So designs of this level aren't hard.
If you're thinking even full 2D acceleration, it gets much harder. and if you're thinking 3D acceleration in an "open" project that would be competitive with even the slowest Nvidia/ATI parts... you're on drugs.
One alternative would be to approach an existing vendor about opening up an "old" product. However, getting a fresh production run of an old product wouldn't be cheap -- you're basically talking a million to get masks made, initial wafer lots, etc. Hence the FPGA suggestion, since that's commonly available hardware that doesn't require any manufacturing specific to the design.
However, some manufacturers may still be in low volume production with a suitable product. Someone who used to be small player in the PC graphics market years ago but isn't now would be a candidate -- perhaps a Silicon Motion or an Avance Logic (part of Realtek now, I think, though I doubt the video products are still active, tho the audio parts are) could be persuaded to open up a part that's in sustained low-volume production.
Seriously, though, if you can't offer significant volumes -- the minimum probably being on the order of 10-20K/quarter, and that's VERY small in this business -- don't expect to get much help from existing vendors.
It's possible to build an extremely-low-functionality graphics card with open source software. It's so easy, that it's already been done. The 'nv' driver gives decent 2D performance from very cheap NVidia cards, and it ships with all modern Linux distributions. What's the peoblem?
What's not possible is to build competitive 3D cards. NVidia spent a few billion dollars making the NV40 chips. They're currently finalizing design on the NV60 range of chips, two generations in the future. What's truly amazing is that NVidia has found it in their interest to make extremely competitive Linux drivers for these chips/boards, and they give them away. And they "just work". While I deeply appreciate these Linux drivers, I cannot for the life of me understand the business case for spending a few to ten million dollars to write them.
Thad Beier
I love Mondays. On a Monday, anything is possible.
Yes it definitely makes sense to develop such an FPGA-based board and letting FOSS crowd to hack the FPGA code similary to the software. FPGA are really powerful toys these days and can easily compete with ASIC devices. Total reconfigurability extends it's lifetime a lot - I know that the cameras http://developers.slashdot.org/article.pl?sid=02/1 2/04/1526226&tid=8&tid=106 are shipped now with much improved "hardware" without any actual changes on the boards.
Are there any graphic cards that natively support rendering of Ogg Theora? What about wavelet BBC Dirac that is still in alpha stage and many specs are not finalized? Can _any_ of the high end cards do that?
The FPGA-based with the open HDL code can. And will support bleeding-edge just emerging tasks, formats designed after the hardware is manufactured.
Concerns about the licensing and patents? Yes those are nasty things. But have you ever downloaded any applications or drivers to your Linux box separately from the distro you've got? Because of the "licensing issues"?
That can work for reconfigurable FPGA too. Still I prefer to avoid those - our next cameras (FPGA code is under development) will support Ogg Theora, not MPEG*
And we will be happy to buy such cards and ship them with every our camera.
"Seriously. nVidia already has kick-ass hardware and the best drivers available under Linux..."
Look through the Linux Nvidia forums, then tell me they're the best.
Isn't that basically what you get under windows? Less support related hassles, and ease in driver installation?
FWIW, at home, I buy a nice Nvidia card for my windows box, then when the time comes it migrates to my linux box, or my windows box becomes my main linux box.
I'd really like to see this open source card happen, it would be very cool, but honestly I would be much farther back in line to buy one.
In a business situation for servers and desktops times hundred(s), it could make life in a mostly linux work environment easier. The pain in the ass part is always about installing the Nvidia drivers for 3D support, otherwise the current modes/defaults are sufficient. So this card would have to roll with sufficiently good 3D performance and be cost justifiable for budgets vs the existing status quo.
Why do the current video card vendors feel the need to have their own custom hardware interface anyway? They all have to ultimately provide OpenGL or DirectX drivers anyway, why not just implement OpenGL or DirectX on the video card's BIOS?
Remember back in the day when the VGA cards first came out and how you had to custom program for each video card? Then the VESA standard came out and made things much simpler. I ask again, why not do the same thing for hardware accelerated 2d and 3d cards using existing standards like OpenGL or DirectX?
The would still protect their proprietary GPU design, while making video drivers trivially simple at the OS level as well as platform independent. Need to update the "drivers"? BIOS flash...
Or is there some compelling reason AGAINST doing that that I'm missing?
...It can compete with the latest GeRadeon TiFX 9 million.
OpenGL 2.0 would be a must, D3D acceleration also. Audio would NOT be a bad idea - I'm sick of Creative.
But seriously - a while back a company tried to produce a video card that was more of a daughterboard that supported new stuff....
How about a card where you can change out/upgrade it's "core", add your own ddr2/3 memory, and maybe even expansion stuffs like VIVO/tuning? A truely generic GPU'd card that's a standard could become it's own long-lived fad.
Getting marketshare is something else.
You'd have to hit it BIG with the gamer community. And the card itself has to (somewhat) mirror the abilities of the FireGL/Quadro for more consideration. Open drivers are just a boon. This needs to be a sub $250 card, preferably obtainable for under $150 to really strike gold.
Why can't all fpga/microcontroller manufacturers just release free optimizing compilers???
Thin-Clients is the name of the game - particularly linux-based thin clients (terminal clients).
Those are usually bought by big companies that use MS-TerminalServer or Citrix. So the graphic cards inside those are purchased by the thousands.
Means that if you manufacture a good(stable) and cheap 2D card and sell it to the thin client manufacturer - you make lots of money.
The fact that the card is "OpenSource friendly" will be liked by all office PC and terminal client manufacturers. Usually they don't get this level of technical data from card makers.
regards, Boris Ratner.
I currently own a couple of nvidia cards. I enjoy that NVIDIA is providing 3d accelleration for my installed software. What this Free Software Friendly board is capable of is minimal. It's essentially an ancient 2d acceleration. 3d support is off the table. I can find that elsewhere; I think there's a few OSS drivers that do that with proprietary cards. Perhaps they can't work on obscure platforms. I don't work with obscure platforms regularly, thats why they're obscure!
From a ROI perspective, you have to convince me there's some improvement over the status quo. I couldn't care less about the source. I know that 3d graphics are among the most alien software topics to developers. Its difficult, especially when you're mixing it with low level programming in a performance sensative environment. Not providing 3d means I'll look for a second card. More likely, I'll be looking at a different card that offers more functionality, even on Linux, at 50 dollars, than this can offer at 100.
Simply put, an free-software friendly board lacks a community to push it forward, and I don't see it treading water among the highly competitive graphics card market. If you want this to sell, you need to identify and explicitly cater to your niche market. Promote it as a learning tool, and grease the community wheels. Just putting it out there and expecting the world to recognize its value won't net you much.
I Browse at +4 Flamebait
Open Source Sysadmin
My answer is an unqualified "No, do not go forward, there is no market." The market for 2d cards is totally dead not because people don't use systems just for 2d work, but because there are so many that can be gotten very cheaply, and driver support for all platforms is great. I can easily get a 2d card that is fully supported under Windows, Linux, BSD, BeOS, etc, etc. It's getting a 3d card that is likewise supported.
So I don't see any gain here, espically since it's likely to be more expensive. You aren't going to see many people except hard core OSS zealots use a 2d only card simply because the architecture is open.
This is not a good business plan. If you're going to make hardware, don't make your specification distribution policy be your driving motivator. What I mean is...
If you really think there's room for another video card manufacturer, then by all means jump in the game, and by all means release your specs freely - god bless you. But if the only competitive advantage you anticipate is the way you release your info, I don't see it.
Surely there're more creative and needed areas into which R&D funds can go. Making yet another low-end video card whose main appeal is to a very niche market which may "like it" but by no means "needs it" due to its free information policy seems wasteful.
Just my thoughts.
Ecce Europa - Web Design for Business
If you intend to make no dollars, then no, it will not be worth it to anyone for the project will die in infancy. Spinning boards is not cheap and most established hardware developers will not even contemplate your design when they can have nVidia or Intel take the brunt of the risk. Screwed up logic I know. But in practice when the developers product fails thyey fail, but never the less they hold on to the idea that if they have a big name provider then they are somehow insulated from the decision of what unit to go with. Too risky (I know, either way the product fails) I guess.
Unless you are willing to promote said video chip as a product, you shouldnt spend your time at it.
On the brighter side, if you did promote your open chipset, you would find some small amount of customers ready to purchase.. nay.. demand your chipset. It could be a way to break into the market.
Dont bother unless your willing to make a bussiness of it. Otherwise the endevour will be mostly pointless.
I think you underestimate just how much I just dont care.
Now we'll need virus scanning of the hardwares software.
------
insert sig here,here, and here
I read the article.
It is definately about 3d.
(3) How do you feel about the choice of neglecting 3D performance as a priority? How important is 3D performance? In what cases is it not? (4) How much extra would you be willing to pay for excellent 3D performance?
they would neglect 3d right now.
I think that's a bad idea, X.org is moving to 3-d only...
I would be OK for mobile and embedded situations were gaming is not expected. Ultraportable inexpensive laptops. Handhelds. Cars. Koisks machines. Media players.
that sort of thing.
For normal sized laptops, and desktops and such you definately need 3-d within the next few years. I expect it would be a REQUIREMENT. 2-d performance is falling by the wayside. The future is using 3-d api for 2-d acceleration.
Maybe make a 3-d only card? (who knows..)
I'd be whilling to pay 300 dollars for a decently performing card with good 3-d capabilities.
With the PCI Express slots, I'd probably buy a couple for a multimonitor setup if they were 170-250 dollars.
DVI connections on cards, no fancy-ness needed. No TV out is needed, no dual outputs is needed.
A specialized TV-out card would be nice for my PVR setup. Maybe even a specialized HDTV out card.
(We are talking about a card that won't be released for a couple years now, right?)
I personally think it would be a great idea! And though it would be difficult, it's far from impossible: One friend of mine is making a 3d robotic vision system using two black and white cameras and some FPGAs and PIC microcontrollers, and he's eventually going to try to make a neural network interated circuit (he has sponsers to pay for the masive costs of that) with back propagation and such for object regonition! My point is, he's doing that alone and it's a massive task, so surely a group of people could make a 3d accelerated open standard video card! For one, commercial video cards limit themselves by clock cycles and don't do too much in parralel, however, a singal FPGA could fit many small floating point processing units (talking about... thousands) on a singal chip, and they could all run in parralel (and each can run 500 MHZ fanless, so add a fan and heat sink and it could clock even faster). So it could have plenty of power to 3d accelerate! In fact realtime raytracing could even be don't by clustering a reasonable number of FPGAs, however that's an excessive number of FPGAs for a singal video card.
I would suspect that a good deal of driver development is already outsourced.
This could definitely be economically viable for the manufacturer, but less than great for the graphics development industry (well, then again, they'll still have the defense and embedded industries).
How good will it be for the consumer? Will there still be a reference driver?
Obviously the project would NOT be competing with nVidia and ATI or Intel or even SIS on the battlefront that those companies are used to dealing with. You do not have to be the fastest in terms of fps or even cheapest to be purchased. You have to be reliable and fixable. We are mostly a PC shop with about equal parts Linux and win32 with some sun machines thrown in. Almost non of this machinery needs to play doom3 as its just windowing and such and in many cases, just text. Yet we buy ATI cards just the same. We need them to be reliable and compatable with our chosen OSs. It saves a whole lot of money that way.
You see the point yes?
I think you underestimate just how much I just dont care.
Can you program Transmeta cores to do OpenGL instead
of usual macro-ops? If so then can an 8-way
Transmeta blade be used as an off-the-shelf fully
open graphics card (yes, hideously expensive).
If this were released I would buy one the moment it was released.
I don't care very much about 3D acceleration; as long as it accelerates a little OpenGL it's OK. The important part is 2D. The 2D performance of most cards (well, of the Radeon and GeForce cards) sucks. I mean, why is it that I can't even get 15fps in Doom II using prboom without OpenGL at 640x480 when I have a Radeon 9100 (AGP 4x) and a dual AthlonMP system? Why does GeForce (the music visualization plugin) get 10fps when the window is 800x600? Especially when the processor isn't even being taxed by it?
Hardware RENDER support, Xv (colorspace conversion and decent scaling), XvMC, the ability to process a large number of pixels being thrown at the card (e.g. when running a music visualization program like GeForce, or when running an old game like non-OpenGL Doom), and fast gemetric primitives. That's what matters.
HAL 7000, fewer features than the HAL 9000, but just as homicidal!
Just use this instead of giving away proprietary info: http://www.opencores.org/projects.cgi/web/vga_lcd/ overview
The intel integrated graphics chips already cover this ground. Midrange performance, open specs, DRI friendly. Put a recent Intel graphics chip on an AGP card with dedicated memory, and I'll buy one in an instant.
If they have patents on key aspects of this card, and are not infringing on anyone else's IP, then it's in their best interests to publish the specs and sell more cards. They have everything to gain and nothing to lose.
If their card uses trade secrets which, if discovered, would hurt them, that's a good economic reason not to disclose.
If, as is very likely the case, their trade secrets infringe on patents or patents-pending that they are not aware of, that's a VERY good reason to not invite unwanted attention.
There's a dirty little secret in many industries, and I suspect video technology is no different:
You can violate patents left and right as long as nobody suspects you are doing it AND you don't tell the world you are doing it.
My recommendation, sadly, is to publish "functional specs" so the OSS drivers can be made to WORK but without the nice competitive bells and whistles, and publish binary drivers for the good stuff.
File defensive patents on anything that's reasonable to patent, and after 18 months* or so when all the patent-pending stuff is published - including your own patent applications - go back and make sure you aren't violating anyone else's patents. If you are, negotiate. Once negotiations are done, publish the full specs.
*It's my understanding that as of a few years ago, patents are published about 18 months after initial filing in the USA. Can anyone confirm this?
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
My Radeon driver (on Debian) works pretty well and I imagine its author had limited access to the hardware specs.
So I hope that with open specifications, open source developers will create an awsome driver. The open source community can work together with the hardware provider to create a product that both sides are happy with.
I wouldn't be suprised if the open source drivers eventually surpassed the proprietary ones in performance and quality. Best of luck! I'll buy one.
You won't compete performance-wise with high-end consumer 3D cards. You won't compete budget-wise with low-end consumer 3D cards. You're going to have crappy Windows drivers unless you wave money in front of your developers to work on that uninteresting part and _maintain_ it, _and_ shell out for the Microsoft developer packages for this sort of thing.
Your revenue? Linux geeks who are patriotic enough to pay for a product with less bang for the buck than a standard commercial card, and who will take the promise of eventually-less-buggy drivers some time in the future as being more valuable than a buggy but adequate and fast 3D driver now.
The thread makes mention of hardware cost control, though they're having serious trouble making that competitive (hard to beat quantity-millions for bulk rates). However, Alan Cox's message highlights a serious problem - you have a lack of programmers for cards that specs are already known for.
The only realistic solution I can think of is to pay coders to produce a minimum adequate driver implementation for the new card (or heck, even one of the old ones). Making a decent accelerated 2D driverfor an experimental card is a few person-months of work, if memory serves. Making a decent accelerated 3D driver is a few person-years of work. The budget for this is within reason, but still has to come from somewhere. As there isn't a deeply pressing need for this being felt by most people (see previous point), the pace of volunteer development will be slow (as is shown by Alan Cox's comments about current driver projects).
I'm not suggesting taking this outside the open source community. I'm suggesting paying open source people enough that they can do this as their day job, and have _incentive_ to do this as opposed to some other interesting project.
This is the Big Problem. I can't stress it enough. Any easy way of implementing _anything_ to do with 2D graphics cards was patented a decade ago or more. Any easy way of implementing any basic 3D was patented more recently, but is still patented. Even crawling through the patent database to look for implementations that were missed will take a lot of time, and cost a significant amount of money (you need experts on graphics algorithms and VLSI design to do this, and patent lawyers to back them up; see previous point about volunteer time vs. needed schedule).
Big graphics companies solve this by doing the requisite grunt work, and aggressively patenting everything they can think of as a defensive measure. The standard way of solving patent conflicts is bitter litigation followed by cross-licensing relevant patent portfolios from each other (we've seen this in other parts of the hardware world often enough too). A low-budget open source card project won't be able to afford either of these, and both will eventually become necesary (someone will claim you're infringing no matter how clear it is you aren't, because it's in their best interests to make the claim).
In summary, the only way that I can see an open source graphics architecture being developed is if the community and donors scrape together several hundred thousand to a few million in startup capital to fund hardware and software development, and to deploy lawyers. A side benefit is that you might even be able to afford chip spins if you're on the high side of the funding scale, though it'd probably be more wise to divert the funds to multi-platform driver support and patent portfolio instead.
Variant options that come to mind:
The catch is th
There is no way that this company can come out with thier first graphics chipset and have it be in any way reasonable right out of the gate.
Instead, it would be MUCH smarter to just wait until a chipset company goes bankrupt, and then raise money and buy a slightly older chipset, and all of it's documentation outright. This has the advantage of the millions of pre-existing cards using that or a derived chipset already in production/use. Possibly a company might even be willing to donate/sell at a cut rate a much older chipset. For example, I don't think it is really necessary for a chip like the s3 virge to remain propriatary for whatever company owns it today.
If it were fully released as open source then medium sized card manufacturers like Asus or MSI etc, who don't own the chipset tech for the cards they currently make, could get the free chipset almost immediatly put into production at a modern fab plant, which with minor modifications would make it several orders of magnitude faster. Since it is an older design it uses on die real-estate somewhat efficiantly (or at least uses a modest amount of area).
What about patents? I can't find where the legal aspect has been discussed. While technically feasible it doesn't mean jack if it's not legally feasible.
Anyone got any info regarding what would be possible without any patent licensing?
Closed-source drivers may be good enough for ordinary users, but for a hobbyist doing much low-level development work, a piece of mysterious code (especially in the kernel) can become a significant hurdle. It also means that I can't have my kernel too different from a Nvidia-tested one (their current wrappers can accommodate ), which is sometimes needed for some advanced bleeding-edge feature. Sure, they kept up with the 2.6 release relatively well, but who knows about future releases? Or what if I used a patch from IBM to have some feature not normally needed by gamers? With source at least I can do something, which is a warm fuzzy feeling even if I may not have the time for it.
Good idea. Graphics cards in particular. Lot of good thoughts on the kerneltrap page as well. Upgradeable card and good integral drivers, yes.
Dumb non developer question.
As CPU clusters have proven that for some applications they can be a very good and cost effective solution, as opposed to jut whopper iron, is it possible or has it been done with the video cards? Is it possible to have a cluster of cards that work together effectively? Or would it be too slow, they couldn't talk to each other fast enough?
Here's an off-the-cuff guess as to the parts cost for one board (I'm sure I have most of the prices wrong):
- FPGA $30
- PCB $5
- DAC $10
- DVI transmitter $10
- RAM $20
- Assembly $??
- Development cost $??
- Profit $??
That should actually read:Like one of the posters on the list said, if you can come out with an open and really good 3D implementation and platform, then people would probably flock to you. Sure, dirty dealing is part of business, but the momentum of a good product is hard to kill.
I bootleg Fizzy Lifting Drinks.
Damn, there isn't a single good comment in this whole discussion... Does anyone here realize what a huge difference a fully-open videocard would make?
Yes, you'd have working drivers, which is valuable, but barely worth noting. The big deal will be the more advanced features.
HDTV is developing pretty well, and even if you can't get HDTV broadcasts, there's plenty of HD material on the internet. Unfortunately, most computers aren't fast enough to play 1080 material in any format, and I'd bet there's a few that can't handle 720 video encoded with MPEG-4, WMV, etc. The real answer is to have hardware decoding... MPEG-1/2 are all that we see now, and even that is pretty rare under Linux. I happen to be lucky on that front, but xvmc doesn't allow you to deinterlace before it's displayed, so it's fairly useless at this point.
When you have all the specs for the FPGA, you can just download the latest upgrade, and have full-fledged MPEG4/Theora/WMV decoding on the same videocard, meaning a 100MHz PC could playback HD-DVDs perfectly. No doubt Tivo would be equally as interested in the features of this card.
Even if you don't have a videocard powerful enough to decode your favorite codec, you'll still get serious gains from it being open. If you check-out mplayer's vidix drivers, you'll see that you can get serious performance improvements if the developers have the docs for the card. It's hard to explain what a HUGE performance boost you would get from having a fully-open card.
Plus, FPGA programming is getting a bit of attention lately. It wouldn't be hard to imagine companies setting up clusters of computers, and filling every available PCI slot with this graphics card, and using the cards to do most of the calculations. Remember the PS2 cluster? Imagine the processing power of that, but on steroids.
In addition, think of all the groups trying to setup display-walls, with multiple monitors. Being able to do that much easier with this card could make it a big seller, if nothing else...
As someone who has setup several Unix machines for multimedia, I think there would be a big market for this, even if it costs, say, $60, and has no 3D support. If you think you need 3D support everywhere, you're probably mistaken. If you're running anything other than x86 (or maybe MacOS on PPC) you've got practically no options for hardware-accelerated 3D anyhow. So, putting a 2D card in there, instead of wasting money on a new Radeon, makes everything work better, and you loose nothing.
Personally, I have only 2 suggestions.
1. Make it as cheap as possible, while still being fully functional. If it sells for $30 (maybe after a few months) I'll buy dozens of them myself!
2. Include as many output options as you can. I use S-Video a lot, but very few have interlaced TV-output support. DVI is important for those with LCDs. Composite looks like the next standard for HDTV output, and that could turn into lots of sales (especially if your card costs less than ATI's Radeon/HDTV adapter!). I've heard lots of cards don't work with HDTV well because they can't output an interlaced signal at HDTV resolutions.
Dual-head support would be very nice, at least if you can include dual overlay support with it. Then you only need 2 cards for a 4-head Linux system.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Linux and FreeBSD, both i386 only is not good enough. What if I am running NetBSD on a Mac? Open source drivers are the only option, plain and simple. And nvidia and ati don't have to write them, they just have to give up the specs to let other people write drivers. This wouldn't even comprimise their precious IP either.
So "we" need a card, but "they" have to pony up the resources and infrastructure to make it happen? If "you" see the need and the utility, figure out some way to contribute (if possible, besides buying lots of their cards). Here's a question: Could an open source video processor be designed purely around OpenGL and still be viable in this age - i.e. with many games being developed around DirectX9, would an OpenGL-only design be good enough?
Less is more.
1) The card also manages to be a top-performer, and can rival Nvidia and ATI in 3d performance for games, at least price-for-price. If someone has the choice of a card that can play their games in Linux (or on a dualboot system) decently, for the same price as a card that is significantly lower in performance, they'll probably pick the one that performs.
2) The card is dirt cheap, and any performance inadequacies can be overlooked (provided it at least does basic OGL stuff).
3) They somehow manage to get it to be used by large laptop manufacturers (are you listening, IBM?), so that Linux laptops can reliably suspend, etc. (which isn't currently the case).
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
That was one thing that made the 8 bit computers and also Amigas so good was the openess of the hardware designs. I remember combing through hardware books to look for points in the systems to play with or alter. There isn't any reason why hardware should be closed unless it's a security product. I think this would bring about a new computer revolution starting in the open source community.
*It's not what you can do for the Dark Side but what the Dark Side can do for you!*
My first reaction was that this is really cool.
The second reactions was, NO 3D?!
xorg will soon need hardware 3D for best performance.
Most home users want some 3D.
Every card I've ever tried had 2D support and the extra price benefit from widespread use.
I just don't get it now.
BTW, why was this on kerneltrap instead of being an xorg discussion?
If tyranny and oppression come to this land, it will be in the guise of fighting a foreign enemy. - James Madison
Might I also add that I personally don't think there's a market for an entry 3D card maker right now, and that option 2) as stated above is fairly unlikely to happen, particularly when you can get a nvidia card for $50 that can play new games about as well as a console can.
I'd say the future market for graphics hardware in linux is on laptops, and that you'd be much better off doing that, if you could figure out how to get into the market.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
It'd be inside my next computer, and inside every computer I advise buying from now on.
I am a free Software man, as in GPL, and having half-functional video cards bothers me a lot!
Our government is changing to Free Software, and we can't get a 100% Free working system due to video driver issues - it is just a shame on video card manufacturers.
-><- no
Video card manufacturers are once again screwing everyone. Their costs for video cards include Produciton and research costs, distribution costs, retailer markup and marketting. And even with all that they cost more than that estimate.
Obviously not every video card can be top of the pile but gaming is only one application of video cards, some people buy matrox simply for their multymonitor support.
Unfortunatly these guys are trying to fit in between onboard video and low end video cards.
And that gap will be filled as soon as the major players realize it's there and open some drivers.
Also their old cards have massive open source support.
I personally find this very interesting because my Linux box had it's Video Card go out. My main box has a ATI card so even an upgrade card move won't solve my problems! I really need a new card! And it's going to have to be an old nvidia card because ATI doesn't have the support I need.
In fact it will probably not have the highest performance/price ratio just because of driver support. I will spend as little as I can to get driver support because I can't imagine getting good price/driver/performance on linux from a graphics card.
These guys might be on to something. I imagine it's really really hard to break into the graphics game right now.
Yes, I would support them IF the performance was at least comparable to some of ATi/nVidia mid-high range cards (9800 Pro for example). I wouldn't support them if it performed like a TNT2.
"With Microsoft, you get Windows. With Linux, you get the full house" - unknown
This is the idea.
Right now ATI and NVIDIA and any other 3d card manufacturer out there feels that it is not in their best interest to reveal these specs. This doesn't make sense because:
1. If a company wants to reverse engineer another company's BIOS it can do it.
2. If a company wants to reverse engineer another company's driver it can do it.
So the reason for hiding these specs is security through obscurity. Not a good reason.
Let's give them a good reason to open source their software. Let's use Money!
So we (as in the Open Source Community) hold a BIOS freedom drive. People put money towards freeing the software of a video card. Any card manufacture can participate. Cards are choosen by eople voting with their dollars as to which two (2) cards they want free. This allows overlap and more compitition between manufactures. The first manufacturer to release the source code under the GPL get's the kitty. The other manufacture get's nothing.
Remember Blender, (the 3d modeling program), when it was opensourced for 100K. The company couldn't make a profit from the software, and decided to get a bit of their money back. They got it back in 3 months. Blender is a little side project compared to video cards. I'm fairly certain Video cards can get a wee bit more money.
GO FOR IT!
A path that could be very fruitful is to design a video card to be used in a TiVo-like device. In particular, in addition to the good suggestions involving the Render and Damage extensions, a 2D-only card should do some hardware accelerating of IDCT and motion compensation, so that i.e. DVD's and MPEG-4 files can be played with a very minimal CPU. Work with systems integrators that are willing to put MythTV on a silent fanless system with a pcHDTV card and your video card/chip. This could be a good way to go for smaller but demonstrated market, where the part is easier to design than a 3D-nvidia-ati competing beast. Actually doing the video and TV on the same part is a good idea, if it can be done, since these machines are usually space and PCI-slot constrained.
I do not think, out of the gate on a small budget is reasonable or feasable to get a 3D part. It would be better to start small, and plan some features for the second generation. For funding, take pre-orders. Oh and hype the shit out of it, on slashdot.
Secondly, how feasable is it to put a cheap off-the-shelf CPU on the part to handle the 3D workload. Certainly that's faster and cheaper than a FPGA. CPU's with MMX or Altivec instructions can be had in the 1-2 GHz range for < $50.
-- Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 1=0.
I recently built myself a new Linux based system. The hardest part was finding the right video card. I wanted a card like that support mpeg2 (encoding and decoding) and perhaps a TV tuner. No card was available with Linux support. I could care less about 3d since I never play games. I finally compromised for a low cost nVidia card for the mpeg2 and what a pain its been given their so/so prioritary drivers: - In Mandrake 10 after install the card and their drivers caused KDE application to crash. I traced it down to a problem with GLX. The semi-secret option "NvAGP" in the XF86 config fixed the problem. Option "NvAgp" "0" ... disables AGP support
Option "NvAgp" "1" ... use NVAGP, if possible
Option "NvAgp" "2" ... use AGPGART, if possible
Option "NvAGP" "3" ... try AGPGART;
8 hrs wasted!
- Next,mplayer did not recognize the mpeg2 hardware. The very reason I picked the card. I recompiled mplayer with all the correct flags and finally got the hardware recognized. Another 6 hrs wasted.
I been using Linux since the 0.9 Kernel and can not remember having as much difficulty. Yes, I am a developer so I have always had an advantage using Linux. Since I also develop on Windows I realize how helpless one is when problems occur with prioritary software. You'd be luck if the tech support person actually owns a computer or speak your language. Granted nVidia may install properly on your system but if you have problem your our of luck. Now wait a few years an lets see if nVidia is going to keep on supporting you.
It time someone else got into the market, drop the 3d heat generators chips from their cards and sold a quality product for the non gamers. My bet is their are more non gamer in the PC world by a few orders of magnitude than gamers. I am all for an open source video card with good documentation!
is it compatable with windows?! I know i know... but theres still some games that unfortunately work best in windows and directx :P
1) There are many old graphics cards that are fully documented. These cards are just fine for 2D stuff for 95% of Linux users, excluding the few 3D gamers, OpenGL developers, and people who run insane screen resolutions.
.13 micron will run you a few million dollars, then more testing...
2) It's just not feasible to implement a cutting edge 3D graphics processor -- it takes tens of thousands of man hours to just get a working + tested Verilog description and a layout. Then a fab spin at
So in conclusion, we already have 100% documented 2D cards that are "good enough" like the S3 ViRGE, and it's pretty much impossible for open source developers to produce a decent 3D card.
To those who find binary drivers too distasteful, but still want to find the best card for their needs: the DRI project has a very useful catalogue of supported cards to aid you in your selection.
For example, there documentation on ATi's cards: http://dri.sourceforge.net/cgi-bin/moin.cgi/ATI
I recently picked up a ATi FireGL 8800 because of information on dri.sf.net and have never been happier with a card. It doesn't perform as well as the FireGL X1-128 that has now moved to my secondary workstation, but not having to deal with ATi's drivers and having the 8800 work equally well in FreeBSD and Linux is more than enough of a reward.
I have also used nVidia's binary drivers (with a card that has moved to a media box), but they are not ideal either. I will say performance is remarkable however.
I suppose my graphics needs have become more modest, however, and others may have more pragmatic concerns.
Matrox? Don't count on it. They've just "recently" gone from being one of the best supported video card makers, both 2D and 3D, in linux, to one of the absolut worst after they switched to a closed source model of providing their own drivers.
"Worst?", I hear you say, "How can that be? My ATI doesn't work great either!". Well, consider this; It's been almost a year since their last driver was released. It doesn't support Linux 2.6 yet. People are trying to patch things up, but it's a losing battle. It doesn't support SMP either, which means that any P4/HT users are out of luck. And I'm not just talking about not actively enhancing the drivers for SMP, no, it will outright crash and bring the the whole computer down with it if you som much as think of starting an OpenGL application. Oh, and while we're at it, there is of course no support at all for AMD64.
Quite frankly, Matrox has remained so apathetic to the Linux crowd that I'm now convinced that they tricked us all just to get our money, and deep down inside they just hate us.
Slagborr
In servers and such. I'm working at a place right now that likes to build their own servers for whatever reason, and right now they're all equipped with Radeon 9200 cards (I don't know why). If this card were open, they wouldn't need to worry about writing the driver, so they should be able to make a decent, stable card for _cheap_! Sounds about right for server use. This might also be handy for MythTV users.
Never eat more than you can lift -- Miss Piggy
If you just need the 80x25 text display it is very easy to implement in hardware. I did one recently using about 1% of the resources of the FPGA. And this is something with colored text and everything. I even borrowed the font bitmap used under Linux console mode so you can't even tell the difference.
My long term goal is it use the remaining 99% of FPGA resource to build a custom cpu and port uclinux onto it. Then add a keyboard and you have a computer on one chip.
You're swimming a bit upstream, because both NVidia & ATi have adequate (albeit closed) X drivers. So your real distinctiveness is for non-X graphics like cybercafes (noX browsers) or home theater PCs.
Or maybe games?!? Are NV&ATi forbidden from writing DirectX for Linux? OpenGL may be nice, but porting is always a b!tch.
I have an nvidia card, I run Linux exclusively, It displays my desktop fine and I can play Doom3/UT/NWN just fine with it.
It cost me very little and works fine.
I seriously doubt their are enough people that *really* care that much to fork out a premium to get something that doesnt work as well as existing products.
I have a few Linux boxes in my house, and I can sympathize with the limitations of the current crop of GPUs on Linux. But, the way I see it, there are only a few possibilities here:
1) A company that is not Intel, nVidia or ATI designs this chip. Modern graphics chips are VERY complex, and VERY difficult to design. The industry went from 30+ companies less than 10 years ago, to only 3 today**. Most of the architects ended up at the 3 remaining companies. So, there's problem 1: lack of unemployed graphics chip designers.
Assuming problem 1 can't be solved, that doesn't mean that a GPU can't be designed. A simple chip can be designed, but it would have trouble even keeping up with Intel's integrated chipsets. So there's problem 2: not powerful enough to do much of anything.
If the chip were to ever turn a profit, lawsuits would be flying from Intel, ATI, nVidia, and probably anyone else with any graphics IP (Creative anyone?). The point being that it would be virtually impossible to create a chip that doesn't infringe on existing patents. Say what you will about patent law, but that's the way it is. There's problem 3: lawyers.
2) nVidia or ATI designs this chip. It would be considered only if there were a large enough market. Unfortunately, even if every Linux user were to buy one, the market there is extremely small. Here's the overly simplified equation to determine the minimum price of the chip on store shelves:
And it's a moving target, in that the more it costs, the fewer potential customers.For argument's sake, lets say that it were profitable to do this. Since both nVidia and ATI have an interest in not letting the other know of any hardware tricks, the cards would be slow and crippled. Just like option 1, this chip would not be powerful enough to do much (probably along the lines of a GeForce 256, or Radeon 7200).
3) Intel designs the chip. Don't laugh. This is the most plausable solution so far, but still just as unlikely. Intel only makes integrated GPUs, and they have the largest market share in that segment. They compete with nVidia and ATI by pricing their chips extremely low, not on performance. Their graphics chips are treated like swag just thrown in to sell their chipsets. If it looked like Intel could sell a good amount of chipsets by doing this, I think they'd consider it. But it comes down to market share again.
In any case, just like option 1, this chip would not be powerful enough to do much of anything (see the pattern yet?)
4) A group of independent engineers/academics/hobbyists create the chip. Only possible if some philanthropist donates a ton of money to the cause. Assuming everyone works for free, the necessary tools alone will cost close to $1M, and the cost to create the actual ASIC is insane. And don't think that you'll get it right on the first try...or the second...or the third...
Or you could just put it on a really big FPGA. See option 1 for extremely optimistic performance characteristics.
I apologize for raining on /.'s parade. The real--not living in a fantasy world--solution to the GPU problem on Linux is not getting open-source hardware/drivers. It is to keep increasing Linux's non-server market share. It's a cliche, but money talks, and the IHVs are more than willing to listen.
**yes, S3, Matrox, et al are alive, but control only a very small sliver of the market.
Back in the day, when IBM and Apple were going at it in the home computer industry, IBM decided to allow other companies to copy their hardware. I am assuming they must have released a significant amount of their propretiary secrets to let this happen. If so, it seems that video card companies could *perhaps* follow a similar principle. Suppose nvidia released their specs -- whose to say their business model would crash? I find it at the very least a speculative argument either way. Please expand on this or refute, slashdotters (i.e., find a google search about IBM's business strategy against Apple to support / deny my claims :)
"Recursive bipartite matching"- try it!
The first company to come up with a graphics card good enough to play the games I play (e.g. Halo/Halo CE, Tron 2.0, C&C Renegade, Rollercoaster Tycoon 3D) AND which has windows drivers good enough for the aformentioned games (Open Source would be nice but I
suspect you cant do that on windows because you need to sign NDAs to get at some of the OpenGL/Direct3D hooks/devkits) AND which has Open Source linux drivers good enough for WINE and whatever else will get my $ (especially if it can do all that nice programmable shader stuff in OpenGL on both linux and windows so I can have a play with it)
As for issues relating to development (i.e. the issue with not being able to make an ASIC due to costs), just find a DSP or some other kind of CPU which is geared for the math involved in doing this stuf (there must be a CPU thats fast enough, cheap enough and has the right features somewhere) and then write the whole card as (closed source) firmware for this CPU. Then, there would be a defined (and open) interface to this firmware.
I get a deep gut rage when my Linux box that used to run for hundreds of days at a time freezes in the %$##@! Nvidia driver because a screensaver came on.
It might be more worthwhile to work on better relationships between Linux developers and Via. Via sells a large fraction of the motherboard chipsets (if it's not Intel, it's probably Via) and, as a commodity part manufacturer, doesn't have a strong business interest in a proprietary interface.
If Via can be brought on board (assuming it isn't already) that provides more leverage for dealing with other vendors, like nVidia.
I just plug my card in, and it works.
I run the driver installation package, I get drivers for GLX and the card itself.
Then I configure X (unless already having done so), and start it.
And then I enjoy a perfectly well-working card.
If you're having trouble, are you using some crappy framebuffer hacks? Don't. Get X.
I've been using Linux since the late 90's and I do remember, back in the day, support for almost any hardware could be a real pain. But that was years ago.
For video I believe *both* Nvidia and ATI are supported. Personally I've been using Nvidia cards the whole time, XFree86/Xorg goes so far as to include a 2D video driver but NVidia makes their closed driver available and updates it about as frequently as their Win32/64 counterparts. For that matter I'm working right now on a Mandrake 10 amd64 system with the 64 bit drivers installed and working great.
With sound I've used Soundblaster for ages (since I finally upgraded my Monster Sound Aureal based card) and I've never had trouble with getting sound to work. And for that matter, its been years since I've actually had to do anything myself to get it working (god bless hardware autodetection!).
The only screwy issue I can think of is pc builders should probably take the trouble to preinstall the Nvidia drivers (I'd hope they do) so Linux new comers don't have to jump right in (if you've ever done it, you know its not hard, but its a new way of doing things..so..:)).
Quack, quack.
Not knee-jerk activism. The article was a person that claimed to be from a bussiness, asking a bussiness question, and that's the kind of answer I gave. Sorry, but it's the truth. If you think that there will be serious sales of a 2d graphics card that costs more than many 3d cards, you are kidding yourself. As a project for fun, this would be really cool and I'd say good idea. As an actual idea to make money, it's crap because it won't.
It's like say you came and told me you wanted to design a web browser, but only one that did HTML 3, no CSS or JavaScript or layers and so on, just simple HTML, as a personal project. I'd say cool, and great luck, sounds like a great way to learn, and might develop to something cool. If a company asked me should they spend money to develop that becasue they'd like to sell it, I'd say never, because there are better browsers out there for free.
Also hardware isn't like software on this level. You don't just go throw a graphics card together from parts at RadioShack. Even if you have the full specs, you need a company to produce it and that is expensive. It's not like precision chip fabs are easy to build or cheap to operate.
Oh and yes, ATi and nVidia WILL survive, so long as they make money. If they don't, it'll be because someone else is. They also aren't teh only two, S3 and Matrox are two other major players, and there's probably a few more I forgot about.
Your legacy architectore argument is stupid too. If you build something on cheap, commodity hardware with no support contract, you've no right to bitch when it fails. YOu replace it with more cheap hardware. If the system must be kept running for decades on end on the orignal platform, you pay for a system that does that. At work we have an IBM 390 mainframe that is about 15 years old. IBM still supports it and maintains it, and we pay them for a yearly contract to do so.
Get off it, an open graphics card of this nature solves nothing worth solving. It's not a serious competitor so they aren't going to sell or force a change. It's a neat idea, but only if implemented on hardware with an actual chance of having a market.
Isn't that a bit like a pretty girl walking up to a geek and saying, "It's like this: technically, I'll fuck your brains out. Should we do it?"
I mean, unless she has an STD, the answer is pretty damn obvious.
My Greasemonkey scripts for Digg &
Make an X.org accelerator - There are a lot of people who dont care much for pushing polygons, but would love to have a fast, high quality grahics card that intergrates with X.org or XFree86 and works without hassle.
Support multi-head operation with robust Xinerama support, good colour calibration etc. and provide hardware acceleration for compositing, video4linux overlays, SDL hardware blitting, X primitives, Freetype font renderers, DirectFB acceleration - this card could form the heart of every low-cost or embedded linux system sold in existing or emerging markets round the world, and provide significantly better 2D desktop acceleration than ATI or NVidia, who seem to put 100% of their efforts into appeasing the Doom3 players.
Even if its not a match 3D-wise to a Geforce FX6800, it wouldn't be hard to do a better job of supporting Linux APIs than 90% of the manufacturers out there.
I gots ta ding a ding dang my dang a long ling long
I would definately buy this. Graphic drivers is probably the hardest part to configure when setting up a linux box. If its open source/ this should solve that problem. Also make sure it runs cool. No fan too. This would be perfect for my servers. Now I buy Matrox G550 for my servers/ but its getting difficult to find and not many other cards come with no fan.
From the web page:
...free!
Concider the pocket calculator. Every low end pocket calculator is now made in China and the wholesale cost must be $5 or less each; yet these calculators have all the functions that a $100 calculator had just 10 years ago.
The same process will happen to graphics and video cards. The chinese will license and tweek older card designs and flood the market with $10 cards that are "good enough" As they will be running very low cost operations there will be no desire to spend money developing and supporting in-house drivers.
If the Open Source community wants to lobby any company, start with these chinese companies. They will be open to any method of reducing costs.
Instead of having OTHERS making an open-spec graphic cards for the opensource community, let me throw couple of questions to all:
Knowing what we know now about nvidia and ati graphic cards, what they do and how fast they do their stuffs, is it possible to somehow invent a decent graphic card from scratch ?
If the above question is yes, then, how much you think it would cost ? Including R&D, finding ways around patents, prototying, taping, manufacturing, and so on, until the thing is on the store shelf ?
Community-force have done miracles, Linux is just one of them. But so far, this happen in the software arena, not hardware. I do understand that it's a totally different arena altogether, that's why I throw out the feasibility question and the cost question.
Someone did some calculation and figured out that Linux worth 500 Million or so. Now, let us figure out the rough worth of an open-source/spec piece of essential hardware.
Thanks !
Muchas Gracias, Señor Edward Snowden !
You'd be surprised how much of a market does exist for computers which aren't a l33t 3D gaming machine. Think: corporate market.
For example, there are more servers sold with an ancient 2D ATI Rage graphics chip on-board, than gaming machines with a GeForce 6800 Ultra. Or, heck, until very recently Sun still sold workstations with a renamed ATI Rage PCI card in them.
The problem however is that
1. that's a bulk low-profit margin market. It's not about selling marked up boxed graphics card, it's about selling bulk chips at mere cents above the manufacturing cost.
2. it's already being savaged by integrated graphics. OEMs already operate on single digit profit margins. It's increasingly hard to justify even the extra traces or mobo space for an extra graphics chip, when for 2D any integrated graphics are already good enough.
3. Competing for merits -- _any_ merits, including OSS drivers -- in the 2D market it's gonna be a major feat. For 2D _all_ current chips are supported well enough by F/OSS drivers.
A polar bear is a cartesian bear after a coordinate transform.
I have implemented a video card in FPGA using a sim. I've done extensive testing and come to several conclusions.
1) You can't (even with EXPENSIVE) FPGA chips build a graphics pipeline that performs more than just moderately well.
2) FPGA chips lack the performance to make good use of high end RAM products
3) Simple VGA compatibility is not a major issue, can be done in very few gates (mouse cursor acceleration is the exception). Higher resolutions with a VESA interface are even easier since VESA typically is operated as just a frame buffer device. Also, VESA generally runs without palette mapping. 15-bit per pixel minimum can be forced
4) Acceleration becomes an issue, rectangle clipping is easy, region clipping is not (lots more gates and registers required). Alpha blending is not too hard on a range, but intelligent alpha blending is more problematic. YUV support is sort of easy, but still, in combination with region clipping becomes much more complex.
I can go on for a while, but I believe that my point is being made. You can't make a competative video card in a FPGA chip.
That being said, I've also investigated using large scale DSP chips in combination with FPGA and the performance gain was substantial. Still roughly 50% of a comparibly clocked card from ATI or NVidia, but much better.
I think that for the purely acedemic value, this is a great project. I also think this guy could sell at least a few thousand of them to hobbiest.
Now if you want to make something really cool, get a Motorola G4 CPU (heard they're cheap these days), mount them on a PCI or PCI Express card, then provide a HIGH SPEED bridge between the host machine and the card. Do this for a low cost and I think you'll have the entire "Brew your own supercomputer" community as well as the "I want to run Mac OS X on my PC" community banging at your door. I've seen another card that does this but comes with 4 CPUs and costs over $10000. Do it with one CPU and leave that board open.
They won't need to supply drivers.
Okay, Free Software community aside (we already know our answers and reasons) - why should a hardware manufacturer hire a team of experienced coders to write drivers for every platform out there? They can save quite a lot on just releasing a "bare bones" driver i.e. for Windows and providing only some Q/A and help (AND full documentation) and have Linux, BSD etc drivers written for free? :) They make their sales from hardware anyway, drivers are and always (or at least for a very long time) have been available for free and as long as new features in drivers are coming, they just boost card sales.
So... why pay when you can have the same thing for free?
And as to "opening up the design" - add some fast layer of indirection or something alike, just obscure the hardware a -tiny- bit and you're safe - competition won't steal your hardware design - hackers need the API hooks and specs, not internal plans.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
when you said big boys Nvidia and ATI... sort of made me giggle. they are the leaders of high end graphics cards, but you are missing out the real big leader (in terms of market share), intel pretty much owns graphics end :D
the point of all of this, if i'm reading the discussion thread right (on lkml) is that a 2d and maybe even 3d card could be had for about $150. i would be willing to buy a card with good support and natively open source drivers for that price.
"but money is the God of Algiers & Mahomet their prophet." - Rich. O'Bryen June 8th 1786
Why don't more people get into trying to reverse engineer the nvidia/ati drivers, just catch the calls sent to the GL and see how the driver translates those functions to actual algorithms sent the GPU?
This seems much simpler than development of a 3D card from scratch. At least if it does become reverse engineered then nvidia/ati will either have to change their implementation (meaning hardware) and or go ahead and provide more information since the information will already be out there.
...is Radeon on chip R200 (8500, 9000, 9100, 9200).
I use Radeon 9100, and I can:
- watch movies in mplayer (xv)
- use framebuffer
- play Doom3
- use lots of other OpenGL apps (like Blender)
- avoid any binary drivers (everything is open source)
If your primary need is free software system, then R200 is best choice for you. If you are going to buy nVidia then stop talking crap how "free" you want to be.
Ok, so we now start considering developement of specific, OSS community centric, graphical cards. Why? What exactly does it resolve?
My Matrox G400 worked quite nicely when I checked it last time. Of course, aside from two days of X configuration and some X jokes by themselves.
I understand that everybody feels that there is some serious problems with graphical layer on Linux, but WHY THE HELL is anybody refusing to admit that all of this could be easily resolved by removing some 20 years old legacy layer from the system.
And as regards to the graphical card - thank you, but this will be a total flap. medium 2d performance, nonexistant 3d performance and crappy support under other OS'es.
Think:
- Patents
- Time To Market
- Delivery channels
- Volumes and coressponding price per unit
In short, stop wasting your time trying to cure horse with a broken leg by reseating the jockey. FIX THE X! Kill it. Basta. Deleted. Removed. No more X. No more forks. Just remove it and replace with something monolithic, closely tied to the kernel and fucking working. Finito?
Graphics cards have become the single greatest hassle when installing Linux. Many of them don't even work correctly in frame buffer or VESA mode anymore, and if they do, they are slow.
I think that we are in for a major change in graphics hardware, going from more proprietary, special-purpose hardware to basically vectorized general-purpose hardware. If you keep the card more general purpose, you will probably at least get lots of orders from universities and research labs working on new ways of doing 3D graphics and using 3D graphics cards for compute-intensive applications.
On the other hand, Linux hardware vendors and users would probably also like to have a cheap, low-end card that just works with every OSS and supports commonly used functions. So, something with good 2D acceleration (both bitblit and Postscript models) and some cheap 3D support would serve those needs. And such a card could also become popular for Windows if it accelerates Windows desktop functions well and (in contrast to all the proprietary trash that's out for Windows) has a simple, clean, and hence reliable, driver.
A couple of points, though. First, it's probably the high-end open 3D card that would pay the bills, at least initially, and it would be sold to a niche market. But a high degree of programmability and flexibility would be its selling point. Second, sadly, decent Windows support would probably also be important for it to sell well because many people still want to have the option of booting into Windows.
I think this could be a good way for a smaller graphics card company to get a steady revenue stream because, while the market is small in relative terms, it is probably a decent size in absolute terms and you'd have it largely to yourself.
Oh, another thing that would be important would be good marketing: banner ads on OSS sites, getting the drivers into X implementations, making sure the major distributions include suport, etc.
I'm curious what data you've used to come to such an absolute conclusion about the spending habits of open source users ?
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
I know nothing about how much money, time and resources are required to design a graphics card or layout an FPGA. However, my uneducated initial reaction is that this would be very tough to pull off financially. I'm afraid that most of the posts on /. and KernelTrap aren't particularly enlightening either. The variations on "if it's cheap and plays Doom III", and "I'll buy one, maybe two, then another one in 3 years" really aren't helping to make a business decision.
My view, is if there were ways to play off the FPGA and the openness angles to find some niche markets that could foot the development bill and provide some manufacturing volume, there might be a way to pull this off. Later, as the design improves, the bill of materials goes down with volume, and Moore's Law helps performance, the design will make more sense for more and more mainstream markets.
What niches? Well, I don't know. A semi-cheap, semi-standard way to get an FPGA must be useful for someone. Maybe include special video processing functions that enable dirt cheap embedded CPU's to do motion detection? Maybe an all purpose, programable co-processor for HPC apps that happens to render a GUI as well?
Where does the cross platformness that a fully open source driver and known design have the most benefit? Heavy industrial tools? Science? Interactive TV? Security? My bet is on some sort of embedded market.
Then there are creative financing options that might be useful, things like preorders, sponsorships, bounties. Maybe not starting with a fully open license, i.e. split out the driver core like nVidia, but have a contract with someone like the FSF to open source the complete driver when a certain volume has been achieved.
Anyway, I've probably done enough uneducated rambling...
... there is no way in hell this would be successful. The market for this is too small- basically Richard Stallman and non-x86 BSD and Linux boxes. Plus you are competing with 2D-only drivers on cheap Nvidia cards.
In Soviet America the banks rob you!
Should it be done? Absolutely yes. I'd love to spend my money on a product I can use completely with free software.
Getting people to make ethical choices begins with offering them. Need another justification? Look at it from the customer's business perspective: this is something they can protect their investment in by leveraging what the community writes for it to do (or what the owners are willing to write for themselves).
I thought everyone's gripe is that ATi & nVidia don't open source their video drivers... because they are better than the existing open source drivers.
But don't we already have debian-free-qualifying GNU-hippy-pleasing drivers that run all ATi & nVidia cards? Don't we have drivers for every integrated graphics chipset?
Why would anyone care about the GNUness of the manufacturer's driver... if that manufacturer's driver were not better than the existing FS/OSS driver? If this guy builds his FPGA 2D card, won't all the FS/OSS people still run their ATI Rage 128 with their open source driver?
Idogeddit. Don't we already have open source drivers?
There are no trails. There are no trees out here.
The idea is that there would be an "operating system" in firmware on the video card--or perhaps a beefed-up BIOS. Video card makers would handle the differing raw hardware interface with this "GDOS"and from the PCs perspetive the interface would be the same..so the main computer OS could use one driver for whatever GPU is used.
cool idea...might be a good compromise for ATI and nVIDIA. They should agree on a standard "GDOS" specification and write implementations for their cards firmware, which would replace the pre-compiled, closed-source driver. They could then leave it up to Microsoft, Linux developers, etc. to write a single, standard, OpenGL-capable driver--basically a 3D version of a standard VESA driver.
Putting the "secret sauce" in the firmware means they really don't have to talk X86 if they don't want to-it could be a lot of special, proprietary microcode if need be. It also wouldn't be platform-specific so they could concentrate on making great video cards instead of writing drivers (which most often they are crap at doing anyways).
It is well known that various drivers and OSes with the same card result in significantly different performace. If a driver is buggy or slow, then they can yell at Microsoft or Linux developers or whatever to fix it...right now they have to shoulder the responsibility.
Are there any ATI drivers yet which don't crash Xorg 6.8? I can't find a single one, and I've tried every driver I can obtain.
I wouldn't mind knowing, as running Xorg 6.7 feels dirty when 6.8 is out and has so many more features.
Karma: It's all a bunch of tree-huggin' hippy crap!
Even if it is not the fastest, 100% X11 compatibility, open source drivers and good signal quality (don't forget the DVI port for TFTs) could give it the same status Matrox had for a long time in the Windows world:
Not the fastest, but a really good card for office use and fast _enough_ for last year's games.
C - the footgun of programming languages
How come this has not turned into a Slashdot poll?
I believe a lot of the Slashdot reades run Linux/BSD or something like it and would benefit from a graphics card that 'Just Works (TM)'.
And, should they make it modular enough with empty sockets for additional processors + multiple SODIMM sockets (would 4 SODIMMS make 4*64bits = 256bit?) for dual/quad memory channels... then one could buy entry-level card with one cheap processor + 1 64Mb 400Mhz DDR SODIMM, then upgrade it later if needed.
Preserve old classics: copy your collection onto all hard drives.
The reason video cards are so expensive is because they contain massively complex graphics processing units to do 3D calculations.
PCI modem, video tuner and sound cards on the other hand can be bought for a tiny fraction of the price, even ones produced by backroom outfits.
If this video card was high quality but 2D & DVI only, then it could be produced fairly cheaply, and would suit environments which want a cheap 2D video card that is assured a smooth ride with Linux desktops.
Companies rolling out thousands of Linux desktops would be friendly to it. Businesses would buy it.
He's making an oblique reference to the sony VAIO line. +3 will do, just so it isn't hidn.
I guess today is a passable day to die.
If I said that I was going to build an operating system in my free time, with the help of a few strangers on the internet who will be working only for "a vauge, but warm and fuzzy feeling", you would all probably laugh. Than if I added that it would be of a high enough quality that it would compete with Windows and Apple, and that I would be GIVING IT AWAY! You would all politely cough and send for the people with the straight jacket and the butterfly net.
But the fact is that it happened, despite all of the reasons, facts, and very loud voices that all screamed in unision, it won't work, it's a waste of time, and a proprietary solution is always better.
You can't go wrong with open source. Hardware, software, whatever. Someones going to buy it.
There's a lot of people here saying "noone cares about 2D". I don't know what 3D web browser they're using to make these comments, but I'd certainly like my 2D web browser to render anti-aliased text a bit faster.
Here's my requirements list:
- Proper stability. I used to think NVIDIA was bad at this, but then I tried ATI. Am I the only one who gets the impression that if I looked at an ATI chip through a microscope, I'd see a bunch of microscopic duct tape?
- Shit-hot anti-aliased text rendering.
- XVideo extension (colourspace conversion and scaling)
- DVI out. Why are DVI cards more expensive than analogue cards? Thinking about this makes my head hurt.
- apt-gettable drivers.
I'd actually buy if all these features were still alpha quality, as long as I was convinced that they were going to be properly supported down the road.It's interesting that even closed-source proprietary Windows drivers have riced-up unofficial forks. Think about that. People want to tweak their drivers so much that they're prepared to do illegal binary patching to do it. The demand for tweakable video cards is clearly there, but its an open question whether you can tap into it.
I think it's worth trying just for the excitement. It could start a paradigm shift in the graphics industry. Or it could flop dead. Sign me up!
fish and pipes
I believe that capital is the difference between software and hardware development. For those interested in writing software, a computer is probably available and thanks to GNU and others, no more money needs to be spent. Find a good tutorial and get coding.
However, there's a little more to hardware development. You may be able to perform the design stage without spending any more money, but when you actually need to start making silicon... where's the money going to come from? Unlike compiling code, making chips and PCBs costs money. By the standards of most individuals, a lot of money. As much as I like the sound of this idea, I don't think anybody will be willing to risk their money on this.
Of course, some generous donor could come along and help, but I think that's somewhat doubtful.
-ReK
md5sum -c reality.md5
reality: FAILED
md5sum: WARNING: 1 of 1 computed checksum did NOT match
All performance issues asside, but I don't think that such a card, even if fully open sourced could really work better than the a regular nVidia one with closed sourced drivers, which is really extremly easy to install compared to a lot of Open Source stuff out there. After all I went for a Geforce card instead of a ATI 9200 (for which the specs are available to the DRI developers) because of exactly this issue. I don't care to much about OpenSource if that still makes it a hell to get it working.
That said, I don't think the advantage for other architectures would be that huge, ie. in Macs one doesn't have much of a choice for a graphic card, but simply gets what is inside and who in their right mind would replace some ATI 9700 or whatever against a slower, probally not by MacOSX supported 'OpenSource' graphic card?
Overall, yes, having a graphic card with fully OpenSource drivers would be nice to have, however nvidias Linux support so far is great and I don't think that an inverior OpenSource card would really help all that much. Its probally easy to just ask ATI or nVidia how much money they want to open the specs and then just start a little 'call for donations', however even then people would need to develop an open source driver themself, which will probally never get better then the default ones supplied by the manufactors.
The Tao of math: The numbers you can count are not the real numbers.
I'd most definitely buy a fully open-specced graphics card, even if it cost a little extra. Better that than a tainted kernel. Although to my mind the question should not be "should someone make an open spec card>", but "shouldn't everyone be obliged to make their hardware open spec?"
If I am the rightful owner of a graphics card {say}, then Common Law says that nothing concerning that graphics card is a secret from me -- and nobody can stop me using whatever techniques are at my disposal for discovering what I have a right to know {which is, basically, every true fact in the universe}. They can bind me to keep secret what I discover; but, as long as I own the card, it is not secret from me. Nor can it possibly be secret from anyone else who owns an identical card.
Proper enforcement of this law is what we need. {It never really mattered up until now, because nobody ever envisaged technology getting as complicated as it has become today; you could figure out how something like a mechanical clock works just by looking at it, hardly needing any instrument more sophisticated than a magnifying glass.} Anyone who owns an nVidia or ATI card already has the right to know how to program it -- they shouldn't have to fight for that.
And spare me the bitching about revealing things to competitors. Are nVidia really so naïve as to think ATI don't reverse-engineer their cards, and vice-versa? Come on, guys.
Je fume. Tu fumes. Nous fûmes!
It has no branches. No loops.
But that card is not exactly easy to get anymore. Would I pay extra for a card that just fucking works thank you very much with the bloody latest freaking kernel and no damn binary crap? Hell yeah. My gaming machine is a windows machine and it has pretty much the latest Ati in it bought at the full price at launch.
As smarter people then me have pointed out on the forum were the question was asked it would at least have to be an excellent 2D card capable of handling video and blending and all the other nice things needed for a desktop with ease. No point in replacing my G400 with an older card.
Multihead support could be a nice bonus.
I would definitly pay for a card that works with source embedded in the kernel that I could just download the latest bleeding edge living dangerous "Backups? BACKUPS??? We don't need no stinking backups!" give me the alpha code kernel and it would work. Or blow up but at least no bloody binary crap.
Would it sell enough? Graphics cards for desktop are extremely cheap. I seen plenty of people buying S4(?) cards for a couple of bucks and consider that expensive. Then again both nvidia and ati and perhaps even matrox seem to make a living out of selling graphics cards wich cost more then the CPU. If the 2D performance is hot enough so that you can have a nice enlightenment/kde/gnome desktop with all the whistles playing multiple video streams scaled and corrected then that card would be extremely intresting.
But perhaps you need to talk not to us but to those who make desktops. After all it is the Dells and IBM's and HP's who field the support calls for mallfunctioning video drivers. Would they be willing to include a card in their office desktops, non-gaming home systems where they knew what was happening inside the video drivers?
Would the likes of Apple perhaps be intrested in a video card that they could write their own drivers for?
One deal like that would be worth far more then a million /. nerds saying they would buy it.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
Silicon Motion and Fujitsu happen to make video chips with full documentation.
SMI -> http://www.siliconmotion.com/en/sm731.htm
Fujitsu -> http://www.fme.gsdc.de/macrofam/mb86296c.htm
You can download the databooks from the sites above. Maybe someone could make a board with one of these parts. The fujitsu chip has hardware T&L as well and is probably better suited to small embedded systems.
The main thing that really annoys me though, is that in an otherwised fast and streamlined boot process, Nvidia forces me stare at a massive NVIDIA!!!!!! spash screen for about five seconds every time I run X on my laptop.
I really don't understand why this is necessary. It's not as if the Windows drivers that Nvidia provides force the same irritation on Windows users.
On the assumption that they'll make an effort to keep their closed source linux drivers up to date and efficient, I'd probably be willing to tolerate their policy considering everything that you've just stated. (In principle I still dislike that I can only use a very limited set of operating systems that they choose to support themselves.) I just can't understand the need for nVidia to incorporate that splash screen.
I also have an nvidia card which runs on an exclusively-Linux PC. Granted in my case it's because I've had the card for 4 years since the machine was a Linux box.
Performance-wise I'm very happy with the proprietary drivers. Having said that, though, it makes me very reluctant to buy a new card.
I'd prefer proper open drivers for a card. Partially for the very practical reason that there'd be a higher chance of it being supported on newer or older kernels, or for at least a while should the manufacturer fold - or even just lose interest in supplying linux drivers.
Also I might be reluctant to spend money, but I'm more likely to spend money of hardware than software. So I'd gladly spend a bit more on a card from a company that's actilvely supporting the F/OSS community. Plus if i'm spending the cash on a new card, I'd rather like to be paying towards getting the drivers I need rather than the Windows drivers I no longer have a use for.
I don't game too much, but I do like the occasional blast on Stepmania or NWN. Plus am rather addicted to 3D screensavers. So I certainly wouldn't want to risk going back to 3D-crippled drivers.
So yeah. As long as it at least had passable 3D support I'd gladly pay extra for a card which had/encouraged open drivers.
Tiggs
"120 chars should be enough for everyone..."
... looking for some investors helping to produce a graphics card 'owned' by the community? :-))
The open source community could develop the hardware design themselves in a public forum. Of course, we'd need Linus or RMS or some other great guy from the Linux world to lead/represent such a project. Maybe IBM could support that???
Maybe it is finally time to make a progression from software desgin to software AND hardware design, isn't it? Linux hardware would be the last link to ultimate Linux superiority. *g*
I would definitely buy a Linux notebook whose hardware has been developed by the open source/Linux community.
There should be a very publicy known website that shows which hardware is Linux compatiable, has Linux drivers, etc.
It doesn't matter how much better the OS is over Windows, it's like Linux is Steven Hawkings and Windows is Shaq... Sure Steve is smart and all, but he just can't get a 2 pointer...
What causes me to ALWAYS go back to windows? Games and shitty hardware support for Linux. I would pay the extra money to get a supported hardware, cause I could save money in the long run.
Right now, no Linux OS can measure up to Windows. It's only business going to Linux, not at home users. Mac is going to soak up the people frustrated with windows, because OsX has Unix, and has games, and everything else...
Linux has the business world convinced, now it's time for the hardware vendors. Get a publicy run site that points out what hardware to get for consumers, that could get something goin...
Why is no one mentioning the "Manticore" open source GFX hardware design? :/
Check it out at http://www.icculus.org/manticore/.
Too bad that this project seems inactive.
Most people here are assuming this company is going to develop a new graphics chip, which would be a huge undertaking.
(Observe BitBoys. Heck, why don't they release their Verilog source for the Glaze3D^W Avalanche^W Axe to the masses? They lost their practically speaking one and only chance of making high-end GPUs, so it doesn't matter any more.)
But what if these guys are going to "merely" build a card around an existing GPU (Radeon or GeForce, or Matrox or S3 or XGI or hell 3DLabs), with their own drivers?
Remember how good the community developed 3dfx drivers (WickedGL and others) were, back when the cards were still around? It's not so impossible for a company to make their own drivers from scratch -- especially if they can leverage the open source community via a well managed, well presented, clearly focused project.
I'm predicting this company has some talent in add-on card design, and some talent in driver writing, and this combo could well be enough to produce a lovely prefectly modern graphics card.
Okay, off to RTFA now...
Why not petition ATI etc. to publish the full specs for an older chip (e.g. Rage 128 Pro).
A couple of years ago I spent weeks of spare time hacking this chip! (I had access to the register spec.) Managed to write a full BIOS for this card that would initialise it from scratch - PLL, SDRAM etc..
I think that with clever programming, this chip could easily accelerate all Render acceleration functions. It would certainly perform better than ANY FPGA implementation.
There are a *lot* of hidden registers not listed in the Register Reference - take a look at the Windows driver DLLs with a hex editor - some of the register names are listed.
The gfx. engine is very flexible - you can even 'misprogram' it and use 2D drawing ops with the 3D setup engine and vice versa.
The (hidden) triangle rasterizer initial/step registers can be used with the SCALE function to draw alpha blended rectangles IIRC.
There's also a simple microcoded processor that can be reprogrammed to implement different CCE command formats.
There is also an IDCT engine, this was not documented in the register spec though. There was at one time an XFree86 header file with the locations of the registers. Search IDCT_RUNS on google.
BTW from experimentation, the IDCT works together with the motion comp. accelerator, set the IDCT bit in the MC_SRC_x register and write to the IDCT_RUNS and LEVELS registers (in the right order of course). It will cause the MC to fetch pixels from the IDCT instead unit.
"the person who only cares about "good enough", not "awesome" performance"
Wel I'm one of those people. I almost NEVER use the 3D part of the card, because I'm doing mostly desktop-oriented work. I suppose I drop into Quake2 every now and then, but overall I just need a 2D card.
It's funny to see all these Linux users complain about the latest/greatest cards not having drivers. The xfree/xorg folks have to write the drivers themselves, so it just takes time for them to figure out the cards on their own. I have a RADEON 7500 and it's got _FULL_ 2D and 3D hardware acceleration without binary drivers. The performance of the card is more than enough for my and most user's needs, even games like UT and Quake3 play beautifully on it.
My guess would be that this company, if successful, would produce a card that might be fully supported, but the performance would be less than that of the reverse-engineered xorg drivers. What's the point of making a fully-supported card if the design and fabrication resources will only yield a card that performs worse than the competition?
I'll stick with my generation-behing-cutting-edge ATI cards, thank you very much.
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
The linux/*BSD community is always more than happy to pay about half what it costs to make something. So unless you're cranking out tens of thousands of these things a month, you may as well not bother. The linux/*BSD community isn't big enough to support a flow of tens of thousands a month without the windows market there to help.
I wish these guys well; but don't predict much happiness.
I was dismayed when ATI stopped releasing information to the DRI developers. I bought a number of Radeon 9200's expressly because they were the last card that could run opengl applications reasonably well with all free software drivers.
I like free software drivers. They create the possibility that talented driver programmers will do interesting things that aren't on the big video card copmanies' radar, adn they keep the vidcard manufacturers honest. Free software is sometimes the rearguard of innovation, it's true; but if the rearguard never shows up, the frontline never moves.
microsoftword.mp3 - it doesn't care that they're not words...
This is *not* a Video card with an FPGA, this is an FPGA-card with Video-out capabilities; that is: a customary programmable FPGA board at the price of a mid-range Graphic card. The VGA-core (2D or 3D doesn't matter a bit) is really nothing more than a demo application and probably the first thing which gets replaced when this baby is put to the use it is meant for: video- and signal processing, I/O+control applications and custom hardware acceleration (crypto, compression, ECC/FEC, you name it ...).
See that I can plug in a camera and those video pattern-recognition tasks which now require custom built DSP- or FPGA-boards can suddenly be run on commodity hardware. Attach a micropohone, and we're talking about next generation PC speech recognition. Add some spare I/O-pins and get 1us response times for your hard-realtime applications by "hard-wiring" critical IRQ-handlers on the FPGA.
The potential for bringing FPGAs to the PC mass market is HUGE and even more so in the context of OpenSource. Once supported by the OS, resource demanding programs might ship with their own FPGA acceleration modules. This product might have the potential of becoming the standard for this new technology.
LFS'er can conjure up some weird 'batches' *indeed* :)
but ... I found it *verry* odd that an LFS'er doesn't know about the `uname -r`--thingie ... unless he's a wannabe-LFS'er :-)
I don't claim I know more than I know, and if you know you know more than I know, then by all means, let me know.
Don't build just a card, build a standalone Gigabit networked box.
Build a small sized motherboard with Processor and inbuilt Graphics, Sound, USB and Network. Use a AMD64 or some other processor with hyper transport like bus, and connect the bus as direct as possible to the Graphics, Sound and Network etc subsystems.
Bypassing the AGP/PCI[-X] bus will deliver better performance and avoid a few patent claims.
Don't bother with any IDE/SATA/SCSI Drive, PCI bus or Non USB[1.1/2] legacy device support. Have it boot off flashbios and PXE/Etherboot or via a USB device.
Have two * one Gigabit Ethernet network ports as standard, with one port able to pass though packets from the other, making daisy chaining possible.
Have it capable of adding up to 4gig of standard DDR2 memory.
For display output include DVI plus RGB/S-Video/AV Video PLUS the ability to send the digital video stream out over Ethernet packets.
The latter ability gives the video the ability to create virtual displays, Include the ability to receive and mix in Ethernet packet video from other boxes.
For audio output , include optical output plus 5.1 channel sound. Include the same ability to in/output stream out over Ethernet packets.
It can function as a business graphic remote LTSP/X terminal,
A networked high performance media center/VOIP for HD Displays,
or with enough memory. a diskless workstation,
or with good enough OpenGL performance, an X-Box killing Games Console,
or as a node for a Beowulf multimedia system.
I'll add that Matrox cards beginning with the G550 series will very likely not work with your LCD flat panel in DVI mode and Xfree86/Xorg. That's been the case for a few years.
I suspect Matrox dropped the Linux market because it was returning a profit, not because they wanted to "trick" people. (That's juvenile.)
OTOH, not all of us care about 3D and games. Before anything else, I want speed and beautiful text and images.
-- Slashdot: When Public Access TV Says "No"
My suggestion: why not slash the development cost dramatically by letting the open source community develop it (or continue to refine it from a simplest starting point that provides only a basic video display, so that you can at least sell a working product)?
In other words: just provide a hardware shell, period. Sure, without any proprietary IP, at some point it will become a commodity item that competitors can reproduce at will. But that will take some time, and no competitor will be interested anyway until the volume becomes significant. Until that happens, the hardware shell can be sold at a premium, but probably for a lesser premium than if the cost of developing the Verilog code had to be recouped. And even if it is cloned, by being first you will still have the advantage of the trust that goes along with brand-name recognition.
I don't know anyone gulible enough to do tha... No actually I know many people who would be that gulible, i just don't want them to be presented with the oppretunity to do something so foolish. How long would it take a company to create something like this 4-5 years? Do you really want to spend over $100 for a vauge idea? The chances of it ending up as a decent card are slim to none. I'd be really impressed if they came out with something on par with a geforce 2 in 4 years.
Well.. maybe. Or Maybe not. But Definitely not sort of.
And i meant free as in freedom.
Has anyone tried feeling out XGI and SiS?
These companies are trying to find a market for their parts.
Perhaps they might consider totally opening up specs and assisting in open driver development for their hardware in exchange for having a fully dedicated revenue stream from open source and multi platform systems.
I know that XGI is having big problems trying to penetrate into a market that both ATI and Nvidia are so very strong in.
I am surprised nobody else has mentioned
/.
http://www.opencores.org
Truly open hardware (the source code for it is downloadable and simulatable - just contact your local silicon foundry to have a batch manufactured)
As you will see there is already a project for a VGA/LCD driver, it is PRODUCTION/STABLE so there already is a commercialy viable free (as in speach) open controller chip which has already been ASIC proven.
You can download the Verilog code and inspect it alter it etc... We are not just talking the registers but the complete internal workings to the gate level in a format suitable for mask generation.
I have to say I'm surprised by the ammount of negative comments towards the concept of openIP - it seems the suits have taken over
Think of the government contracts. Think of signing a deal with IBM or SuSE. Think of the possibilities of future ideas that could be implemented with these cards. It is ideas like this that can fizzle, or they can turn into one of those snowball ideas that turn into great ones. Who would have thought that releasing the source code to the Linux kernel would have gotten us to where we are today?
My beliefs do not require that you agree with them.
The other thing is that Matrox is boring. It may not do Doom 3, but is is really, really stable. ATI and Nvidia don't offer crap, but state of the art complex 3D drivers are not what you want to draw some graphs and reliably show tables.
See my journal, I write things there
I'm deeply skeptical that this can be done for anything other than 2D operations - or exceedingly simple/slow 3D. You aren't going to be playing Doom3 on this card.
The design is based around FPGAS's (Field Programmable Gate Arrays) - which have a much lower density than full ASIC's.
A modern graphics chip has about a dozen full function floating point processors on it (for implementing shaders) and a large RAM cache (so that you get good performance from textures) and the highest possible bandwidth into it's main memory. I don't think you can come even close to that with FPGA's.
So, the best this activity can do is to come up with a card that does 2D graphics and does 3D only using the CPU and Mesa. That'll suck for anything that uses OpenGL rendering. But for 2D graphics, the majority of the drivers are ALREADY OpenSourced - 2D just isn't that difficult and the interfaces are pretty much standardised.
www.sjbaker.org
As an excercise, suppose a community is formed, and does a useful R&D, and copylefts it (e.g.: GPL).
Maybe VIA, or some other low-cost competitor would implement the design, benefiting from zero R&D cost, and from a potentially big buying force (free software users).
VIA wins, because they can sell good quality products with slashed costs, and no licenses, plus they get a good image ("OS" is a nice marketing term nowadays)
We win, because we can get more powerful, and standard video hardware.
I believe this kind of deal is the most feasible, and I don't think it's a bad scenario.
The difference is that you need a 'seed crystal' to get started.
If the company funds enough of the drivers to get them working decently and into mainline Linux, they'll have cards that will be used and loved for years.
If they incrementally improve the hardware design so that improving the drivers increases the performance on newer fully open hardware, they'll have a dynasty of completely open cards.
I very much want OpenGL on my SMP desktop, and I tend to use strange setups/configs that nVidia just can't handle. I would be happy to fix the problems in the nVidia code, if only I could.
If TechSource gives me a fully open and documented platform with decent hardware 3D support for less than 300 euro, and even the most basic of OpenGL drivers, I will buy it, because I can improve it and fix it myself.
Shae Erisson - ScannedInAvian.com
I currently use Windows to do digital video editing, however I'm in the process of trying to put together a "build" that is Linux-based and composed of Open Source software.
One of the major snags for me is trying to determine what video/graphics card to go with. Alot of the work I do doesn't really take specific advantage of the graphics processors (I'm currently using an old Matrox card if that gives you any idea), but in choosing a new card it would be very useful if it could offload some of the filter/effects processing (or even MPEG crunching for DVD's) to the card's coprocessor.
Unfortunately finding this out has been difficult. Trying to figure out which cards are supported by Linux, and then on top of that figuring out which ones would actually be useful for my work (as opposed to playing games) has been even more difficult.
If a company came out with a card that said it was 100% Linux compatible, it would at least answer the first question, and open the possibility of answering the second; in other words, I would buy one.
second society
>
>You might. Most Open Software users would not.
>
Allow me to correct your typo:
You might. Most Open Software users I know would not.
The world is bigger than you and your circle of friends.
Thanks for thinking of us, whatever the outcome!
As to the "should we do it" question: assuming that one is not legally restrained (by third-party licensors, etc.) from Revealing All, I'd say it boils down to this: how much money do you make selling graphics card drivers? I thought so.
I have an ATI AIW 128 (16 MB) and it seems not very Open Source friendly. If I play Quake3 in WinXP, my computer never craches. In case of Linux (Fedora2) and X.org the crash takes place in every second or third attempt to play Q3. Fortunately I have reiserfs on my Linux partition, so I don't need to fsck after every crash. And I haven't succeeded to find out the cause.
The card is going to be backwards... Period.
The closed-source nature of the ATI and NV drivers is partly due to the manufacturers' desire for secrecy, but also largely in part because some of the technologies used in the card are patented and the patent covers parts of the driver. In at least one such case, the patent is NOT owned by either company and the license they have prohibits them from including it in open-source drivers.
This is why when UT2K3 came out, it would only run on NVidia cards under Linux. At that point, ATI cards had a pretty good open-source driver that supported almost all features of the card *except* for S3 texture compression, which both ATI and NVidia had licensed but didn't own the patents for. NVidia's binary drivers supported it, the open-source ATI drivers couldn't. Now we have binary drivers for ATI cards too.
I would love a high-performance video card with open-source drivers, but it's going to take a lot more than one interested/open-source-friendly manufacturer to provide that. For the really high-performance cards, you have to deal with a nasty web of cross-licensed patents.
retrorocket.o not found, launch anyway?
IMO, the key is to not focus on leading benchmarks but making the card feature-complete and easy to integrate into X Windows. Raw performance usually just isn't that big of a deal (any decent 3D card made in the last five to seven years is adequate for many people's work).
This means:
- a genuine and complete OpenGL implementation (most but not all needs to be in hardware)
- tested and easily installed drivers for the newest and previous major release of X Windows for Linux and at least one of the BSDs.
- an ability to drive respectably high resolutions at faster than 60Hz.
- focusing on visual quality over speed (antialiasing that doesn't suck would be nice)
Make a usably fast card that produces a very nice display output that also works nicely with X Windows, and you can have a very nice slice of the market. Give it basic modes for dealing with Windows' default drivers, and you could probably make a profit just off of the Linux/BSD crowd who are happy to finally have a company that cares. Get it packaged in those Wal-Mart/Linspire/Xandros/JDS PCs, too.
-- "Makes Little Debbie look like a pile of puke!" - Moe Szyslak
How about using DSP's instead of FPGA's. DSP's have lots of adders and multipliers on the die, which you would need lots of for doing matrix operations which are the majority of 3D operations ?
If you can produce a multi-head, 2-d card, thats fast, fully supported on Linux at strange portrait resolutions. You can sell them for upwards of $1k apiece, and I would thank you for the oppertunity to buy them. Then I could stop buying these:
_ So lutions/Dome/Dome_C3_Grayscale.html
f x4 50.htm
http://www.planar.com/Products/Medical_Displays
(the card, I'd still buy the monitors, there pretty awsome for medical displays )
and these
http://www.techsource.com/products/datasheets/g
Those cards are not very fast and don't have lots of memory on them. This niche market could support a company quite well whose real goal is to make open sourced hardware/software cards.
AC
I think they are just cutting costs at our expense. I remember buying a video card, video capture daughterboard, and an integrated PCI TV tuner card as a bundle from them, when such a thing cost upwards of $300. This was a big investment for me at the time, but I was itching to edit video and DV wasn't big yet. These cards had excellent capture quality with hardware compression, so I invested big time in the whole Matrox system. Win2k came out a month later and they decided not to support it. Win2k was a leap ahead of 98, so many ditched the hardware.
Hundreds of calls and emails went to Matrox, especially on the part of people simply trying to get them to release specs on a single chip (MGA-VC064SFB-C) so linux drivers could be written. The 4 non-matrox chips on the board, including the Zoran capture chip - had publicly available datasheets. This wasn't a big part that was missing, but was necessary. Two open source projects started, and made progress, but halted or decided not to support the older generation of hardware, due to lack of specifications! Matrox even ignored a petition with nearly 400 signatures!
Other people were willing to do the work that would have given real value to the Matrox name - as a maker of rock solid hardware that continued to be a workhorse, even years after purchase. All they had to do was release some specs - the hardware was impossible to beat in price/performance at the time and the linux community would looked to Matrox with appreciation because a solid capture board that did hardware based compression wasn't easy to find or afford for linux. An asinine move if I'd ever seen one.
This was before they released open source drivers for a newer generation of cards, and it showed their true character before they cancelled open source drivers for the Parhelia line.
Simply put, don't buy closed source Matrox. They have a history of cutting development early for closed source drivers and not releasing specs!
I'm tired of waiting for nVidia to catch up with kernel changes, and tired of getting drivers that don't support all the features of the hardware that I buy (or broken drivers, like the ATI driver that I could always count on freezing my wife's computer in the midst of a spiffy 3D screensaver every so often).
I have five PClones, and would cheerfully buy graphics hardware for them from a company that takes Linux seriously by providing or allowing full-featured, open source drivers.
This is insane! A company is wondering whether it should develop a free software friendly graphics card, fully disclose and document all register interfaces including the BIOS, and provide Linux and BSD users with a fully supported video card... A company which basically asks whether it should make our dreams come true!
I've read the story on the front page, I got excited, really excited. I made a caffee and started to read Slashdot comments expecting to see similar excitement about the fact that finally a company making graphics card will do exactly what we all have been asking for years.
But what does the Slashdot crowd say? Let me quote the Score:5 replies, the most representative and valuable voice of the Slashdot Community; advocates of Free Software, Open Source, Freedom and Libery; users of Linux and BSD; programmers and hackers; experts and activists; visioners and innovators; in other words, the very essence of modern intelligentsia:
Sincerely,
Pan Tarhei Hosé, PhD.
"Homo sum et cogito ergo odi profanum vulgus et libido."
Actually, the PC was about as non-proprietary as IBM ever got in those days. The machine itself was an almost direct clone of the Apple II, cobbled together in what was then record time for IBM using a few off-the-shelf Intel chips and quite a bit of common 74-series TTL gates. The only thing proprietary about it was the BIOS, and even then, it wasn't like it was secret (IBM published a listing of it in their tech manuals, just like Apple did then) -- they just refused to license it to anyone.
For video cards, it's not quite as simple. You have hardware that's several orders of magnitude more complex than the old PC, stuff that may or may not be covered by patents and (more importantly) trade secrets, bits of code that were licensed under NDAs, and all sorts of other pitfalls and gotchas. Nvidia in particular is pretty paranoid when it comes to this, for reasons I'm not quite sure of, and it'll take some doing on their part to convince them otherwise.
-lee
Binary-only drivers might be the only current
solution to NDA woes. The real problem is,
kernel driver support is spotty for less well
utilized linux or bsd distributions. And
bleeding edge video card manufacturers would
rather commit sepiku (sic) than reveal how they
have achieved their performance superiority.
High performance OpenGL hardware based video
cards might be best supported (based upon open
standards), but are extremely pricey.
Short of a F/OSS-based video card development
effort (that can create an economy through
high volume production, I don't see any real
solution.
The feature I specifically shopped for - the MCP-T audio adapter (and that's not easy to find on a mini ATX motherboard) - does NOT work with Nvidia's drivers. And, since they won't release information on how it works, there are no oss drivers.
I've had two nvidia motherboards. I was just fine with the S3 but I bought into all that "we have to show them there's a demand for support" blah blah blah and wasted a couple hundred bucks on these things. They work, but I'm no better off than before - in fact, I'm worse off, because the audio section on this motherboard sucks worse than the AC97 sound on the S3 board. How's that for irony?
The theory and economics of open sourcing hardware drivers is really well covered by Eric S. Raymond in his now famous book the Cathedral and the Bazaar. In particular, The Magic Cauldron section deals with the theory and economics of exactly this sort of thing and follows up with an afterword that explains Why Closing a Driver Loses Its Vendor Money.
Signatures are a waste of bandwi (buffering...)
I agree with your points - this is only a good thing.
Unfortunatly, the "slashdot community" has changed. It isn't primarily "free as in freedom" advocates anymore, I think it is more "free as in beer" advocates.
As one indicator, it seems that the majority of posters on this topic care most about 3D performance in the context of game playing under Linux. I.O.W, I think the "slashdot community", at least those commenting on this story, are gamers who've decided to have a go at running Linux. They just don't understand the principles behind the FOSS movement, and why open source, and specifically in this case, at least open specifications are important.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
As other posters mentioned this would be fantastic if it was mainly a very hefty FPGA system that happens to have direct video output. I want one to be my "perl chip" and the other to accelerate astronomical simulations. Another (okay I'll buy 3 ??) could provide additional memory and filters with oss software that provides cinematic transitions or other support that would let linux offload to the fast board work like Apple's new Core stuff. Personally I'd also like to have one with several outputs for projectors, maybe with some nice edge blending for extra points. Would also be very nice if they could add some camera inputs and some dsps for some structured light projectors / sensors I want. There are tons of things that would be useful without necessarily implementing all of OpenGL.
I suggest that they:
The first version will be slower than the big budget video cards for the average off-the-shelf fragfest, no question about that. But, the hacking potential is hard to pass up, and it doesn't become obsolete that quickly because you can reprogram it into accelerating any number of things besides video.
Eventually, an entire set of firmware-loadable functions will develop around this, and seeing as the FPGA can be reprogrammed on the fly, it may be possible to make an adaptive driver that "swaps out" an infrequently used set in favor of another set that is repeated more often. Thus, video playback, for instance, may start out a bit choppy on a slow machine but will smooth out after a few frames. I don't know if this will obviate freezing the design in ASIC at some point, but it is worth a try.
All performance issues asside, but I don't think that such a card, even if fully open sourced could really work better than the a regular nVidia one with closed sourced drivers, which is really extremly easy to install compared to a lot of Open Source stuff out there.
This is about Free Software, people deserve to have their freedom. Even if the card was slower than all the cards in the market, but had complete documentation, I would still value it higher than the cards for which docs are non-existant.
My radeon 8500 didn't need any "extremely easy to install drivers" because they were included in the XFree86.
I second that. In the lab where I work, all of the dual-head systems run Matrox cards. The people doing atmospheric modeling don't give a rat's ass about 3D performance. (Yet almost every video card review seems to be written by a 13-year-old with the only emphasis being frame rates in silly games.)
?..
I don't know about Matrox, but
Nearly 2.5 Million Active Sites running FreeBSD (June 2004)
Excerpt:
"[FreeBSD] has a secured a strong foothold with the hosting community and continues to grow, gaining over a million hostnames and half a million active sites since July 2003."
To sum it up, FreeBSD is growing pretty fast.
Since we are in the Linux section: Of course Linux is growing *faster*. So what? Due to the radical differences between the GPL and BSD license, Linux and *BSDs will always be complementary, by attracting different investors and appealing to different users.
The guy who proposed this was not talking about a high-performance card - or a card with any 3D acceleration.
;-)
Remember the S3 Trio32 and Trio64? It sounds like it'll be something like that - though presumably with more video RAM, a faster DAC or DVI, and better 2D acceleration. Maybe more like a FreeG450
Why not simply use the `nv' driver and not use a 3D screensaver? Reliability is excellent. If you need 3D, the card this guy was talking about wouldn't do you any good anyway.
On the other hand, I use the `nvidia' driver on my home PC and find it 100% dead reliable as well, so I guess results vary.
--
Craig Ringer