Slashdot Mirror
Half-Life 2, ATI, NVIDIA, and a Sack of Cash
Latent IT writes "If you're into games, and unless you've been living under a rock for the past few days, you've heard a bit of a rumble from Valve on the relative quality of ATI vs. NVIDIA cards. Starting with articles like this one (previously reported), Valve told the world that the ATI 9800 Pro was nearly three times faster in some cases than the formerly competitive NVIDIA offering, the 5900 Ultra. Curiously, this happened at an ATI sponsored event, "Shader Day". But the story hasn't stopped there. NVidia released this response, essentially claiming that their new drivers, that were available to Valve at the time of their press conference, would make for vast, legitimate performance improvements. An interview with Massive, the creators of the Aquamark 3d benchmark, seems to confirm this opinion - that the NV3x chipset wasn't designed around any certain API very well, and the drivers are critical in achieving good performance. Anandtech writes here about the restrictions Valve placed on what benchmarks could be run. However, the key to this whole story may be this: an article, which I haven't seen get much coverage in all this, seems to make everything a little clearer - Valve stated that their OEM bundling deal with ATI came from the fact that ATI's cards were so superior, and that they were "performance enthusiasts". However, if the Inquirer is to be believed, the bundling deal was a result of an outright auction, on what will probably be the most popular game of the year. Which year that might be, is another issue altogether. Whatever happened to just making hardware, and making games?"
Apparently, they can't even if they wanted to. There was a comment in the previous thread on HL2/ATI.
& ci d=6930186
http://games.slashdot.org/comments.pl?sid=78019
Raw power, eh. That must be why the Radeon 9700 Pro, with a GPU clock of 325MHz, was equivalent to the GeForce FX 5800 Ultra, with a GPU clock of 500MHz. The Radeon 9700 Pro was so focused on raw power that it put out a whopping fifty-seven watts of heat to the 5800 Ultra's mere eighty! The 5800 Ultra had a far more sophisticated cooling system, of course, which consisted of a copper heatsink that stole a PCI slot and a banshee-like fan. Now that's what I call finesse! Do your homework before you post :P
Parent is on crack. Absolutely nothing in your post is true. ATI has had a superior gpu architecture since the radeon 9700 pro hit the market a year ago.
The Nvidia FX series has been plagued with problems from the get go, with Nvidia resorting to a
massive pr blitz and outright cheating in their drivers to compete with ATI.
Parent post is truly laughable and shows an ignorance of what has transpired over the last year in the video card industry.
you do realise that Nvidia was found to be cheating in those 3dmark results right?
/ at i_vs_nvidia/dx9_desktop/001.htm
they inserted static clipping planes and swaped shaders out
take a look at this article using the latest build of Halo PC and take note of the developer comments at the top
i think you will find your wrong
http://www.gamersdepot.com/hardware/video_cards
also tomb raider angel of darkness being a shite game is irrelevent in this matter
its one of the only DX9 programs out at the moment and so is useful as a benchmark for DX9 performance
Having not used ati's cards under linux--don't own one--I can't say exactly what their support or performance is like. However, I can say that I strongly disagree, and feel that nVidias offering in linux arena has improved tremendously in the past year. They are now releasing windows-equivalent versions, offering an easy to use installer, and run a massive forum for users. Can ATI say the same? All that said, however, platform support is only a marginal issue to the real question of performance. An important question, but still not the issue of this article. As for which one performs best, until I see a proven independent third party (which most hardware review sites are not), I will base my decision on what seems to be the consensus: nVidia rocks for the money! ATI is good, but from what I hear, no quite as good. THat last part, though, has no grounding whatsoever in statistically (or otherwise) proven facts, because I simply do not have them (and from what I can tell, neither does anyone else, even ATI & nVidia)!
"We don't know what we are doing, but we are doing it very carefully,..." Wherry, R.J. Personnel Psychology (1995)
ATI has just released official XFree86 4.3 drivers.
The driver even handled an upgrade to Kernel 2.6 without flinching. NVidia AGPGART support doesn't have to be hacked in any more either, it would seem.
No more mucking around with the FireGL drivers from the German branch of ATI.
ATI bought the guys who did the chip for the Gamecube, and they did clean paper DX9 design for ATI. ATI went from being a year behind NVidia (DX8 generation) to being a year ahead.
In the R300, ATI decided to do all their calculations in 24 bit floating point: essentially a pure next-gen chip. The NVidia Geforce FX design was based on their DX8 chips, which were far and away industry leaders in fixed-point calculations; NVidia didn't figure that floating-point performance would be very important this generation and tacked it on. What they ended up with was a chip that had a high transistor count, was very good at legacy, fixed-point operations but could not keep up with ATI in floating point. Even then (about a year ago) NVidia's chip might have been competitive but they had process problems that made the chip clock slower than expected and about 9 months late.
ATI's superiority in floating point shaders has been demonstrated by various benchmarks (including some open-source benchmarks, which are the only ones I really take seriously) time and again. NVidia can only be competitive this generation when they 'tweak' their drivers for particular benchmarks. These tweaks sometimes consist of rewriting floating-point shaders to use their legacy fixed-point functionality, and on some occasions of even using pre-generated shadow models to replace the dynamically generated models of benchmarks that run over a known scene.
NVidia's NV3x generation seems weak, compared to ATI, and very weak unless game coders ignore API standards and write custom shaders that do as much as possible in NVidia's legacy hardware. Of course, by historical standards NVidia's NV3x isn't weak at all--they blow away all their competitors and ATI's pre-R300 products. It's just that the design choices made by ATI's new designers allowed them to leapfrog a generation.
Word.
As Valve has said. If the driver isn't an official release, it's not appropriate to bench with it. NVidia may turn their driver around, but given the industry's history it's a reasonable expectation. Valve knows and has stated a number of times in a couple of recent interviews that the majority of their customers are NVidia users. Hence the great deal of time optimising of their code.
The auction is just good business.
The title of this post says it all, really.
The videogame industry ALREADY generates more revenue than the movie industry (at least in the u.s).
Seems to me Nvidia has a crap card and they have been covering it up for a while now. 1. Bad Future Mark results Nvidia: We stopped participating a while ago, thats a ATI benchmark. 2. Poor Tomb Raider Performance. Nvidia: Who cares. 3. Poor HL2 Performance. Nvidia: You should of used our 50.xx drivers that don't render fog, and aren't out yet. Someone posted this picture. I think it says it all... http://myweb.cableone.net/jrose/Jeremy/HL2.jpg Apoptosis
Half-Life 2 uses the Source engine, which Valve has been building from scratch for the past 5 years.
ASCII stupid question, get a stupid ANSI
If you look at the FX architecture, it has a serious problem.
It can't run "true" DX9 spec games worth crap.
Why?
Because to save die space, nVIDIA engineers decided it'd be best to use 32 bit FP units, compared to ATi's more numerous 24 bit FP units. DX9 specs call for 24 bit precision computations, which is the ATi native precision (which can then be mapped to 16 or extended to 32 bit precision, if asked for) whereas the FX which has to operate in 32, 16, or 12(?) bit modes basically loses half its registers (or more, if you are comparing to 12 bit registers) because it must run in 32 bit mode to be compliant.
End result? Less high speed registers on the FX part, more swapping from ram and less FP computational power to go around.
And this is only a simple example. I believe it has been noted that that Carmack eluded to many ugly optimizations in using lower precision math or proprietary shader paths he had to make to the D3 engine for the benefit of the FX not sucking utterly in terms of performance. It isn't really a playable DX9 part, all in all.
If valve says they spent serious time working for the Geforce codepath (and indeed, it is quite a bit faster in hyrbid mode, but now they are making it well known that it isn't running "true" dx9, which it the truth. It should also be noted that this hybrid mode is what the D3 benchmark was run in which offered the nVIDIA part such stellar performance, specifically noted by Carmack.) then they probably did so. Either that or they would have mentioned nothing.
Drop the "it must be corporate scandal" bit. If you read some of the specs and dev notes you will note that they more or less universally have their gripes in getting DX9 performance out of the FX part.
ATi runs IMMENSELY faster on the default path(you know, the one they're SUPPOSED to use) Valve spent 5x as long optimizing the nvidia specific path than the default path. Man, I'm SO mad at ATi actually using an API like it's supposed to be used, and stuff.
They would have to result into a 50% gain in HL2 in order for the FX5900 Turbo to catch up to the Radeon 9800 Pro, not 100%. The graph with the customized nVidia code path has 40fps vs 60 fps. Although, of course, the nVidia path is lower quality, since the 5900 doesn't do 24bit precision.
Also, I wouldn't call it a CLEAR lead in Doom 3. The nVidia scores 20% higher on medium quality, but the Radeon takes the lead on high quality. Again, nVidia calls driver problem.
Myself, I will be upgrading for Christmas, when I will know for sure which one works best, and how the drivers are. This is also the time when the FX6000 Super Mega Turbo and Radeon10K Elite Pro Plus Plus(Or whatever) push the prices down on the "older" cards ;)
ASCII stupid question, get a stupid ANSI
Shows how much you know. OpenGL was relatively static for a long time. In fact, I wrote an article about this on OSOpinion awhile ago. But the ARB seems to have gotten its ass in gear, and as a result, OpenGL is managing to keep pace with DirectX in the programmable hardware department. When OpenGL 2.0 comes out, OpenGL will take a leap forward. Also, if you actually take a look at the APIs, you'll see that there are very few platform compatibility issues. Both APIs are pretty much self-contained, so OpenGL's progress really isn't affected by its cross platform nature.
A deep unwavering belief is a sure sign you're missing something...
From driverheaven.net:
From 3dgpu.com
Now we have independent verification of Valve's problem statement of the beta Nvidia drivers for Half-Life 2.
Looks like ATI is the way to go this year.
Please don't spread lies about a monthly fee, really.
There will most likely never be a Mac version. Valve has stated many a time that it is simply not part of their plans to port to Mac or Linux (although they don't mind WineX at all, and in fact fixed their anti-cheat when it was found to be incompatable with Wine).
What about "was static..." do you not understand? The pace of OpenGL progression prior to 1.4 was pretty glacial. It always had full support for graphics card features, but you had to use vendor-specific extensions. More recently, 1.4 and 1.5 came out in quick succession, with support for shader technology.
A deep unwavering belief is a sure sign you're missing something...