Slashdot Mirror
S3 Tries to Get Back Into PC Graphics
mikemuch writes "ExtremeTech has a review of S3's attempt to get some traction in the lower-end graphics card market, the Chrome S27. Though its specs look great--256MB memory, 700MHz core clock rate, 1.4GHz memory clock, and 22.4 GB/sec memory throughput, it still manages to underperform similarly priced video cards from the red and green graphics companies."
Only, if they start to provide drivers for the Open source community.
My Averatec has a unichrome and am having difficulty getting it to
work *well* with anything other than the X vesa driver. No DRI, etc.
Help out S3!
I will buy it regardless of worse performance. Otherwise, I'll still stick with Matrox.
Another company to ignore.
Supply full GPL/BSD licensed source code to the X.org and kernel.org for inclusion in mainline. That will trigger a lot of positive support.
Besides, I don't really see a downside, because who, besides free software lovers, would be motivated to buy something non-nvidia and non-ati at this point?
Cheers
Simon
I remember the day when my PC was finally faster than the processor on the Virge, but boy, Descent looked kickass in special 'S3' mode. Of course that was also 1996.
Go S3!
I want to delete my account but Slashdot doesn't allow it.
If this card was around the 65 dollar range, I could see putting it on the shelf here at our little shop, but it falls too short for serious consideration. The open source driver issue... that's a good point. It would do fine for our Linux guys around here if it had the right drivers.
MadOgre.com
I think it is the initial drivers that make or break a new entry to the video market. You can have the best hardware in the world, but if you release it with bad drivers, you're pretty much doomed. You might claim a little back with one or two revisions, but most people won't care at that point.
:(
*hugs his Volari V8* Poor thing never really made it.
With the advent of xgl, compiz etc. we really NEED a decent 3D card with open drivers on Linux. I couldn't care less about gaming but xgl sure as hell looks awesome! I don't need a full-blown NVIDIA or ATI card for that. Open your drivers S3 and I promise you I'll be buying at least 6 of these cards as they become available.
Learn from the mistakes of others. There isn't enough time to make them all yourself.
> Besides, I don't really see a downside, because who, besides free software lovers, would be motivated to buy something non-nvidia and non-ati at this point?
People who do not play high-performance games might not want to pay $100-$600 for a graphics card. Joe User is far more interested in multimedia playback than 3D graphics. Intel's sells their embedded graphics cards for $7, and they are the biggest seller of graphics cards. Plus, they have open source drivers. There is plenty of room in the low-end for S3, although they have a lot of work ahead of them if they want to compete with Nvidia and ATI.
---- "XML is like violence. If it doesn't fix the problem, you aren't using enough."
This is what, the third or fourth time S3 has tried to 'make a comeback' in the graphics market? It always seems to happen the same way. Big price features in a small price card! While the hardware might be ok, and compete on the low end.. The drivers never seem to catch up. This is what the big players have. An established codebase for drivers that gets tweaked each time a new chip comes out.
Predictions for new S3-super-thinamagig:
1. Early previews from hardware review sites- "Shows promise!" "Should compete at entry level" "Good for casual gamers." Drivers will be buggy.
2. Card released many months after initial previews. What was mid grade is now low end, and card doesn't look so hot against current competitors. Drivers still buggy. S3 promises bug fixes and performance improvements.
3. Several off brand Taiwanese manufactures will make cards featureing new S3 chip. Cards will quickly be relegated to bargain basement prices in retail and online shops. Mobile versions of chips will be found in cheap low-end laptops and versions of the core will be seen integrated in to via chipsets for cheap onboard video. Drivers still buggy.
4. S3 continues product line and no longer updates drivers. (Drivers still buggy.)
With any luck S3 will do better than their previous attempts, but they've got a lot to prove. In all likelihood, this will go the way of the S3 savage, S3 chrome, trident cyberblade, XGI volari, powervr2, and powervr KYRO.
First they compare a $115 card to cards costing $125 and $129. Then the price drops to $99 and they 'stand by their review' against those more capable boards because they didn't pan it for performance, but for basic flaws? Uh huh. That would be because SLI mode doesn't work? What sort of idiot would buy a $99 card for SLI work? Ok, AA doesn't appear to work for GL, that is bad but will almost certainly get fixed in the drivers pretty soon.
It looks like S3 is trying something interesting, throw high speed but dumb hardware at the problem of 3D instead of trying to put more compute power than a P4 on a board. But they are going to discover that the drivers are a big part of the equation, it was clear that their drivers probably what was holding their scores down on several of the tests. Since they obviously don't have a lot invested in them yet perhaps they are the ones we should be pushing to support open source. Despite what that PR moron at Nvidia said I suspect the Open Source crowd could whip those drivers into shape in short order, Use the right license (MPL or BSD) and they could roll those improvements back to Windows and carry the fight to ATI and Nvidia.
I know I'd certainly switch from ATI Radeon 9250 (most current 3D with Open drivers) to this new S3 tech if it had an open driver.
Democrat delenda est
Stop nurgling up this place, kid
The GPL2 implies that you allow people use your patents, or else!
I'm still trying to figure out what people mean by 'social skills' here.
i've been trying to keep track of video card comparisons, and rank the cards since the 5200. since neither i nor my friends can buy $300 cards, i kept the list to the lower end of the spectrum. what is interesting is where the current generation of integrated graphics on the motherboard compare to which cards.
**best price/performance**
nVid 7600 GT ($210)
ATI X1600 XT ($170)
nVid 6600 GT ($140)
**best price/performance**
the faster at top:
ATI X800 Pro ~$250 ($150 refurb)
ATI 9950 ultra (N/A)
nVid 6800 LE/XT (LE=slower)($150,$300)
ATI 9800 XT(~$185)
ATI X700 PRO($125,135)
nVid 5900U/5950 Ultra($250)
ATI 9800 PRO(~$130)
=ATI 9700 pro
=ATI 9800 ($90??)
=nVid 5900/5950
ATI 9700 ($110)
nVid 6600 ($100)???
nVid 5800 ultra
(3GHz)
nVid 5700 Ultra (N/A)
ATI X1300 PRO($105)
ATI X700 (not pro)
ATI 9500 Pro ($95 used)
(yes it beats 9600pro!)
=nVid 5600 Ultra
=ATI 9600 pro/XT ($100)
=ATI X600 PRO/XT ($100)
nVid 5800
ATI 9800 SE(128 bit)
nVid 5700/5750
nVid 6200 non-tc (under $100!)
=nVid 5600
=ATI 9500/9550/9600
ATI X300 non-Hypmem???
nVid 5700 LE (MINE)
nVid GF4 Ti 4600
nVid 5200 ULTRA
nVid 5600 XT (XT=lower)
ATI 9600 SE
this last group of expansion cards is equal to the current generation of integrated onboard graphics
***very slow***
nVid 5200/5500
nVid PCX 5300
nVid 6200 Turbocache
ATI 9200 SE
ATI X300 SE Hypermemory
current generation of integrated graphics chipsets:
-- Intel GMA950
-- nVidia 6100/6150
-- ATI xpress 200
i disable sigs
3dfx is coming out with Voodoo6!
troll
/troll
It was durning the mid 90s in one of my first jobs that I got wind of this. The company was called Chipworks and what they did was breakdown microchip in order to diagnose IP violations. Some of the things you see under a microscope on these chips are just amazing. Clowns, Trains, little designs lots of crazy stuff. I remember being shocked at the time when I found out that the manufacturer of the chip under my lens was S3 and that we were investigating them for IP violations.
Does anyone know how that turned out? I left the company before I found out and haven't read anything about it. Also, any bets THIS time. Once a cheater, always a cheater is a truism that might apply here.
In which Mac computer would you put this -budget- graphics card?
The only one you CAN put a new graphics card in is the high-end $2000+ Power Macs.
Developing Mac drivers for this would be a waste of money. There's no market.
However, I do sympathize with linux users who want quality drivers for all types of graphics hardware. I doubt, though, that NVIDIA or ATI will ever release open-source drivers for linux. I think they can and should take the desktop linux market seriously and release high-quality, closed drivers, even if it affects the OSS purity of the linux operating system.
For decent article reviewing some of these issues, see this.
Is it even possible to compete on price in the low end of 3D graphics? They way it works now, older ATI and NVIDIA models get priced insanely low. Not because they are that much cheaper to make, but because their bigger brothers have pushed them out. But if a card/company starts at the low end, how can it compete?
This assumes that the cheapness of older models is subsidized by the profits from the newer. Is this the case?
-matthew
"THERE IS NO JUSTICE, THERE IS ONLY ME." -Death
I had one of those Diamond Stealth 64 video cards. ;) It used a S3 968 chipset. Does anyone remember using VESA? SciTech Display Doctor? :) I remember having to use this utility for games and demos. Oy!
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Is it
* the hardware? Not able to make the hardware?
* the drivers? Not able to utilize the hardware?
* bad management? Not able to do a project like this?
* bad implementation? Good ideas on paper, unusable in real life?
* All of the above.
Could someone with a lot more understanding of how highend graphic hardware than me give an explanation? As long as I can remember S3 has tried to compete but have failed.
Thomas S. Iversen
"throwing High speed but dumb hardware" is something new and interesting now just because s3 cant pull anything sophisticated out of their ass? Quick reminder: ALL Ati and Nvidea cards are brute force renderers. Look for tile based renderes or the old talisman concept to see something different.
... sorry, can i have what you smoke?
"trying to put more compute power than a P4 on a board"
What? If they are high speed, how do they _not_ have more computing power than a p4? And how is having that much power a bad thing?
And about the price issue:cry me a river. Till this cards will actually be in the shops, it will be 20$ more expensive while those other cards will ahve dropped below 100$, if the last 2 generations of "s3 fucks up their comeback" is any indication.
And "open source guys whipping the drivers in shape in short order"
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
I thought S3 was bought out a few years ago by one of the big graphics card players. So what's this???
Or else in this case is that the what 1000 people that might buy this card because it has open source drivers?
Yea that will scare them.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
This is exactly why NVIDIA and ATI keep their drivers closed-source.
If you look at S3's product, you see a device that has great hardware specs (looks great on paper) but fails to deliver because of buggy/incomplete drivers. S3 isn't alone in this - XGI faces similar problems.
The truth is that a lot of the performance of modern GPUs comes not from the hardware but from the drivers which supply it with data. NVIDIA and ATI keep their drivers closed-source because they don't want a company like S3 to benefit from their software - NV and ATI love the fact that everyone else has buggy, slow, incomplete 3D drivers, and that's the way they want to keep it.
I still HAVE my old Diamond Stealth graphics card. I don't remember which one, but I have the daughter board that plugs into it for a whopping FOUR MEGABYTES of VRAM!
Boy, do I remember both of those... ah, the days of DOS gaming...
Is a stable interface in Linux so that the graphics card companies can write closed source drivers that don't need to be updated with every minor kernel revision. The problem is that graphics drivers contain proprietayr, licensed code. There's no real way around it if they want to support all the features. Even OpenGL itself must be licensed. Well, they can't just go and relicense the code and open it up, even if they want to.
So this is a situation where Linux needs to make a concession, if they want better support. This attitude of "open source always!" needs to give way to an attitude of choice. One where you provide all the tools necessary to do open distribution, and open distribution of your own tools, but the option to use closed source for those that want to.
If you don't want that, fair enough, but then you can't be too angry when the graphics companies won't accomadate you and your rather small marketshare. If you won't be accomadating to them, don't look at them to be accomidating to you.
Oh yeah, I remember that option. I never upgraded my RAM. Man, we didn't need 3D cards back then!!
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Intel is. They have the low end graphics market virtually locked. The integrated Intel accelerators are very, very cheap (it's like $10 difference for a board with grpahics) and they are quite capable. They aren't even remotely high end by any means, but they get thw job done. So to compete, you've got to do better than that. Well then, as you noted, there's older nVidia and ATi cards for cheap.
It was the Virtual Reality Graphics Engine.
http://wiki.duskglow.com/tiki-index.php?page=Open- Graphics
IIRC they are shipping FPGA PCI cards and you can download the chipset image. The plan of course is to sell PCI ASIC's for $150 or so. They have a pledge page where you can give them an idea of how many cards they can sell for a first run.
Before I part with'em: two pennies weigh ~4.996+/-0.014g, have a zinc core, and the face of Lincoln. You can keep 'em.
Or else in this case is that the what 1000 people that might buy this card because it has open source drivers?
No, the "or else" is a legal threat put forth in the GPL 2 when someone breaks the GPL by enforcing his patents in the attempt to circumvent it.Yea that will scare them.
And the reason that I just call it "or else" rather than telling you what the GPL 2 says is that you'll need to work it out yourself by reading the license, because it is really hard to tell how such a situation would work out.
Moreover your post is pretty dysfunctional because I am just pointing out that if they once make the drivers free source, then they might have foregone on their patents forever, so the statement of the GP is no good.
I'm still trying to figure out what people mean by 'social skills' here.
I was simply pointing out that this and many other video drivers will never be completely open sourced because of issues like this. S3 is still in business because of royalties on it's patents.
Linux really has got to develop a stable binary driver interface so that closed source drivers are not such a pain.
The current state of affairs has not forced a single vendor to open source a driver but it has caused many users a lot of pain when upgrading their system.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
Somehow the notion of S3 (or god forbid SIS) climbing back into the gaming video card market is like a slasher film monster coming back to life.
Have they ever worked well for gaming (and no, solitaire doesn't count)?
Mod flamebait, it's obviously designed to provoke controversy. As others have stated, you've gotta start somewhere. With a company that has so little marketshare to begin with, they go with the largest share of the pie first, and that is the Windows market. Course a little nudge nudge wink wink could do wonders with getting them to write Linnox drivers.
Do not downmod posts "overrated" simply because you disagree with them.
What's rather funny, is to go back to older games like Doom 2... when I think of Doom 2, I think "It had decent graphics." Then install it and run in... and the first words out of my mouth are "Oh god, how did I EVER PLAY THIS?". It's easy to forget how far things have come.
Amazing. They managed to start out with all specs at least equal when not better and STILL managed to get mediocre performance at best.
When will S3 learn. They don't DO 3D. They have have had a bad reputation when it comes to actual 3D acceleration since the old days and it has only gone downhill since then. It's time to give up and stop burning money on this when they could be dedicating their technology to something useful instead of screwing over a few uneducated users here and there.
(awhile back)
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
All they have to do is release the docs for their hardware. Open source developers will write their own (superior) drivers. Quit making up bullshit excuses for corporations to keep cheating us. I bought the fucking hardware. It is mine. I own it. I should be allowed to use it. Provide the docs so I can write my own drivers for any OS I want.
I remember buying a Stealth 64 DRAM VLB card /w 2MB of DRAM, simply because of the fact that it had a VBE 2.0 BIOS, so I could avoid having to use SciSoft's crap.
hello dear sirs my name is jamesh i are india (bihar) can u guide me install red had linux 9?
Please don't.
Yup. That is why I use this to help the graphics. ;)
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Please contact S3 and VIA.
Contact information here :
http://www.s3graphics.com/en/company/contact.jsp
But realise that when you take a "My way or the highway" attitude, you are being as much of a prick as MS is. You are also making it harder for you platform to ever go mainstream. Most users don't want open, they want working, and they don't care how that's acomplished. They do not want to spend 2 hours compiling a new compiler only to spend 3 more building dependancies for an app (which is something I've had to do with OSS). They want a nice clicky installer that will just drop binaries on their system. They don't want to have to futz with new drivers all the time and endure rhetoric about the hardware makers needing to "open up their code", they just want their pretty graphics.
I'm not going to tell you that you can't demand everything be OSS, I'm going to tell you that you then can't throw a fit when most of the world doesn't comply with your demands. Linux on the desktop is still vastly in the minority. That means you don't get to dictate policy. If you tell the hardware makers "You must do this or you can't support us" don't be supprised if their answer is by and large "Ok, we won't support you."
Now nVidia and ATi have decided they will support Linux, but on their terms. They aren't going to open their source, they can't, they'd be sued. So, that means you need to deal with the pitfalls of that, the incompatibilities, the slow releases, etc. If you don't like that, you need to be willing to compramise. You cannot have everything you want in life.
Also you need to recognise that some of us do not want everything to be open. I don't believe that will be some glorious revolution, I don't believe openness in everything is the way to go. You are free to disagree, but that's my stance, and I'll continue to support companies that choose not to go OSS. You don't have to like it, but there's nothing you can do about it.
I personally want to see Linux grow because while I'm not an "OSS everywhere" person I do think it has value, and I think MS needs real competition. However I believe for it to grow in the desktop market to any sort of credible level, concesions have to be made. You have to change the OS to make it more consumer friendly and yes, that means supporting closed drivers among other things.
..their drivers and mysteriously the next generations of ATI and Nvidia had very similar characteristics, they could sue the hell out of them and get the courts to force them to look at the code. Ya, can of worms, but that's all we got to fight back against the closed source guys-and who wants to speculate that they DON'T go around lifting open source stuff, generally speaking? I bet it happens *daily* all over the planet. What industry association or police force is actually vetting their closed source code to see if it has snippets of GPL code (or other open licenses) in there illegally? I'd bet a year's pay that if you kicked over big corps closed sopurce rocks you would find a TON of decent code hijacked from the FOSS community, because they know they can get away with it!
Yugo built old Fiat designs in old Fiat factories that were purchased and moved. The Soviets also licensed some of the same designs. The truly ugly Soviet cars were local designs. The worst of British auto manufacturing has been carefully preserved in India. You thought globalization was new, or that socialism was immune to it? The problem is too much power in too few hands. Centralized power leads to poor performance, no matter what economic or political system it masquerades under.
No, I would not think that they would go with the largest share of the pie. That share is taken.
The place to start is with niches which are less contested, and grow those niches. Tackle the larger market segments only after refining the product and organization against the weaker competition.
The Japanese auto makers did not enter the US market with full-sized sedans. They worked their way up from economy cars, which Detroit viewed as unprofitable. They built dealer networks, brand recognition, and customer tastes in the segment with the weakest competition. Only after they were strong there did they move up to larger cars. Same pattern with pickups: small, then mid-sized, and full-sized (the largest market segment) came last. Small bulldozers, now every size.
I'm not saying that they should ignore Windows. But they should put some effort into open drivers, where ATI and Nvidia are weak. Intel seems to grasp this. But there's a real lack of low cost, modest performance graphics choices for open systems using AMD processors.
On my 233 Mhz Cyrix with integrated SIS 5597/98 onboard video, using 32 megs of my shared 160 megs of RAM, played Unreal Tournament at medium detail at 640x480 (good for that day) at around 35 FPS - no dynamic lighting. That's Direct3D - OpenGL was not compatible with the integrated video.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
I don't think that's a good solution. It'd be an adquate half-measure, but would sacrifice: ... and it'd make the kernel a heck of a lot harder to debug when issues _do_ come up. Ugh.
... and I'm not convinced that can't be avoided.
... but would be concerned about making it easier to explore the underlying workings of the card (if I felt those were an important secret) through the now-exposed driver/hw interface, and would still be concerned about legal issues. If I were S3, I'd seriously think about developing new, clean drivers and opening them up ... or just publishing *full* hardware specs and seeing what happened. S3 long ago licensed out its few scraps of interesting tech, and they're not going to catch up to NVidia and ATi on R&D, so they need another angle.
- flexibility for the kernel team
- non-x86 users' choices
I'd be all for it, except that accepting that option now would lock everyone into it almost indefinitely - and I *don't* want to be stuck with that situation down the track. Of course, the current option sucks even more, so I'll admit to being somewhat ambivalent.
I do think you may be confusing the types of licensing involved here. Many of the required licenses are patent licenses, and don't have to conflict with open source distribution. I do expect that most of these drivers will include 3rd-party licensed libraries and components as well, so I don't personally see any way a graphics card mfgr could "open" existing drivers. I don't see why an underdog with nothing much to lose like S3 can't do it as a play for market share. Let's face it - they don't have much to protect. Their technology is behind, verging on crap, and they don't need to fear ATi or NVidia poaching ideas. The only likely threat is IP-related ("Hey, now that we can see your source we can see that you're using $patent-technology-x, pay up)
Were I NVidia, there's no chance I'd ever open my drivers. They're a huge asset to NVidia, and probably contain lots of juicy sue-able goodness. If I were ATi, I'd be more inclined to think about it - since the ATi drivers are *complete* *crap*
...Linux only.
Another person to ignore.
It's OK Bender, there's no such thing as 2.
I'm no longer entirely convinced that open source has a huge benefit for video card drivers.
Okay, yes, it's probably not a bad idea, but I read an article from some guy at Precision Insight about this.
The argument was that video card drivers are among the most complex of drivers. It's kind of like Mozilla -- there is a significant barrier to just dropping in and hacking away -- so you don't accrue programmers left and right.
Granted, I'd still buy any card that has good open-source driver support and is more modern than my Radeon 9250, but I'm part of a limited set of people.
Open source works very well for a number of things, but there are definitely systems that it's less effective at:
* Systems that require a significant understanding of a large deal of code before commits can be made. People get scared away from Open Office and Mozilla. Emacs has done very well, but even though it's a sizeable codebase, it's also extremely modular on a by-feature basis. You don't need to learn much to be able to write a new useful feature or understand how a particular feature works.
Anything that requires non-general knowledge has a significant cost in an open-source package. Custom code that replaces standard code (string classes, replacement data structure code), knowledge of the software package's threading model, strange conventions, you name it. Having a codebase that is easy to drop into and start working is always nice, but it is especially important for open source projects.
* Systems that cannot be used by the developer. Okay, this probably isn't an issue for graphics card drivers. Really, really esoteric stuff or things that few programmers have interest in (like simplified interfaces for existing software) don't tend to attract many volunteer developers.
Games with limited replay value are another example. Successful open-source games, almost without exception, have very high replayability and make relatively little use of elements like plot twists, eye candy, or anything else where the enjoyment only comes through once or a few times. Such games tend to either be multiplayer or involve a significant random element. If games can't be played over and over by the developers, the developer doesn't get much enjoyment out of playing them and drifts away.
Any program relying on (nontrivial) preemptive multithreading will be buggy.
Maybe the reason why NVidia are against Xgl is because it uses 3Dprimitives to do all its' work, whereas AIGLX puts 3D with X 2D primitives. Xgl being widespread would mean that the nv driver is worthless.
NVidia, how about telling us who, how and why you can't tell us how your card works?
I've owned 3 S3 cards, a Virge DX, Savage 3D, and a Savage4 (which I still own two of though neither are in service). What I've learned about S3 is that their first few driver revisions are atrocious. Historically, S3's drivers usually take a few revisions to get to the point where they're useable, and by the time S3 stops supporting the product they're great drivers. M
y last straw was when I bought my Diamond Stealth S540. $129 for a 32MB card was a steal at the time and it was one of the few 2D/3D solutions that worked properly with my Alladin V Socket7 board. The driver that shipped with the card couldn't play the just released Quake 3 Arena Test (my Real3D i740 and Permedia2 cards could play it flawlessly). AnandTech gave owners a ray of hope during a round-up by stating that engineering drivers included a working OpenGL ICD that worked with Q3ATest. Performance with the Savage4 got better with each driver release and I was completely blown away when I finally got a chance to run Unreal Tournament using the special textures under S3Metal. My next card was a Geforce2MX and I haven't looked at S3 ever since, especially when I could use the Geforce to play Q3A under Linux.
I tried using the Savage4 with TrueSpace3 and Ray Dream Studio only to find artifacts and garbage being left all over the screen. I got in touch with one of the OpenGL driver engineers and, to his credit, he took my input and screenshots to try to fix the driver.
Anyways, it seems S3 is still up to their old tricks. Make huge claims about their products, then release them knowing they don't live up to their claims.
Fifty watts per channel, baby cakes.
I'm not as interested in how it does against NVidia and ATI. To me, it seems more like competition for on board graphics. And if that's the case, how does it do again GMA900?
If the problem for S3 is the drivers, then the open source community can help S3 write drivers for Linux, and both sides will benefit from it. Then S3 will have a product comparable with the others and open source will have one more set of quality video cards to play with.
The above can be achieved by S3 offering mock-up environments of their hardware that do not reveal the algorithms. This is a practice followed quite often in the military where 3rd parties are called to supply software for 'secret' hardware: the contractor puts out 'dummy' specifications, and the subcontractor codes around these specs; then the contractor takes the source code and applies it to the real specifications. As long as there is interface consistency, the problems are minimal.
>Though its specs look great--256MB memory, 700MHz core clock rate, 1.4GHz memory clock,
Great ? That's better than the specs on the machine I'm typing on (Slot 1 600 Mhz Athlon, 192 Mb RAM, 16 Mb PCI graphics card, 12 Gb harddrive)
Kids today... Pah !
They make lots of stuff.
* EPIA small/home theatre CPUs and mini-ITX boards
* AMD and Intel chipsets (K8Tx00, PT8x0, CX700, etc.)
* basically any southbridge not made by NVidia or Intel
* Envy24 audio chips (lots of cards, from Hercules to Terratec to M-Audio)
* VIA Velocity GBE and Rhine 10/100E ethernet chips (used in many, many desktop cards)
* A metric fuckton of small desktop switch integrated platforms... and a bevy of wireless MACs
Plus the S3 video chipsets are used in a whole lot of video-on-board motherboards even if the north/southbridges AREN'T from VIA. Basically if your server doesn't have an ATI Rage-XL or Intel IGP on it, then its got an S3. Tons of HP and IBM rackmount systems have them.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
My understanding was that Unichrome was a VIA chipset. I happen to have one in one of my little Epia machines, and the DRI etc do in fact work (either with the provided drivers from VIA's website, or from the openchrome project). They are, of course, a bit buggy, and some of the newer-gen chipsets still have issues (we have some at work) which are only currently rectified in the X.org CVS builds.
S3 has always been a company run by complete retards. They never bothered to appeal to the high-end gamer until it was way too late, and by that time they had fallen critically behind.
ATI and Nvidia learned the hard way: appeal to the entire market, low to high-end. Use the innovations gained in the latest high-end designs on the next-generation mid-range. Thus, your designs are constantly fresh, and games are constantly using new features, which means customers actually have a reason to buy your new products. At the same time, your products are always in the spotlight, so even those gamers not willing to shell out for top-shelf may purchase one of your low-end offerings.
Take a look at S3. For about two years they dicked around with the ViRGE, never doing anything to attract the hardcore gamer, just selling a lower-midrange "Free-D" solution with crappy drivers.
Then they came out with the Savage 3D, and looked like they got a jump on the industry with the first graphics chip on 0.25 micron process. But then the truth came out: the yields and speeds available on the early 0.25 micron process were lackluster, the drivers still sucked, and the card was limited by a maximum 8MB onboard and a pitiful 64-bit memory bus.
S3 continued meandering, even a year later. The Savage 4 (two texture pipes) looked competitive to the TNT2 / G400 / Voodoo 3, but then they tacked-on that same repulsive 64-bit memory bus, and the drivers STILL HAD SERIOUS VISUAL QUALITY BUGS!
So, S3 finally "got it", and offered the Savage 2000, a 2 pixel pipe, 4 texture unit card said to be clocked even higher than the GeForce 256. It offered a real 128-bit bus, and T&L. Unfortunately, it was too late to save S3. By then, they were purchaed by Diamond Multimedia, and the writing was on the wall. The Savage 2000 was late, didn't have functional T&L, and didn't outperform the competition.
There was some talk of releasing an updated part with functional T&L and DDR, but then Nvidia released the GeForce 2 GTS, and a few months later ATI released the Radeon, and everyone knew S3 was toast.
Via purchased S3 because Via needed an integrated video chipset in this day and age. Via has improved (slowly) on the assets they acquired from S3, but they will never catch up with ATI and Nvidia. They still market standalone cards with their Chrome series of chips because having cards on the market improves brand awareness and makes a few bucks, and new features in the standalone cards can trickle down to the integrated chipsets (where the real money is for Via), like the K8M890.
They key here is not "best", just "good enough." Via doesn't have to beat ATI or Nvidia, they just have to offer a better integrated graphics solution than Intel, something that runs Vista acceptably.
Man is the animal that laughs.
And occasionally whores for Karma.
The grandparent post contains the information that I wanted when I first saw the article. I still purchase Radeon 8500's when I build a desktop machine, because they are the fastest cards that just work with my open source distros.
If there were another company that offered cards that were faster and just worked, I would buy those exclusively. The same would be true, I'm sure, of Linux vendors.
But as the poster said, look elsewhere.