Final Words
Both of our custom benchmarks show ATI cards leading without anisotropic filtering and antialiasing enabled, with NVIDIA taking over when the options are enabled. We didn't see much improvement from the new SM3.0 path in our benchmarks either. Of course, it just so happened that we chose a level that didn't really benefit from the new features the first time we recorded a demo. And, with the mangoriver benchmark, we were looking for a level to benchmark that didn't follow the style of benchmarks that NVIDIA provided us with in order to add perspective.Even some of the benchmarks with which NVIDIA supplied us showed that the new rendering path in FarCry isn't a magic bullet that increases performance across the board through the entire game.
Image quality of both SM2.0 paths are on par with eachother, and the SM3.0 path on NVIDIA hardware shows negligable differences. The very slight variations are most likely just small fluctuations between the mathematical output of a single pass and a multipass lighting shader. The difference is honestly so tiny that you can't call either rendering lower quality from a visual standpoint. We will still try to learn what exactly causes the differences we noticed from CryTek.
The main point that the performance numbers make is not that SM3.0 has a speed advantage over SM2.0 (as even the opposite may be true), but that single pass per-pixel lighting models can significantly reduce the impact of adding an ever increasing number of lights to a scene.
It remains to be seen whether or not SM3.0 offer a significant reduction in complexity for developers attempting to implement this advanced functionality in their engines, as that will be where the battle surrounding SM3.0 will be won or lost.
UPDATE: CryTek has pointed out that the new lighting implimentation is essentially the same but uses branching in the pixel shader to accomplish what needed to be done in multiple shaders under the PS2.0 path. This indicates that the conditional rendering feature of SM3.0 is actually faster than using multiple shaders (which gives NVIDIA 6 series cards a performance advantage when multiple shaders would have been required).
Facebook
Twitter
RSS
36 Comments
View All Comments
DerekWilson - Friday, July 2, 2004 - link
Thanks Pete, we'll be setting AA and AF in the benchmark batch file from now on ... We've updated the site to reflect the fact that the first run of numbers had NV 4xAA set in the control panel (which means it was off in the game).We appologize for the problem, and these new numbers show an acurate picture of the NV vs. ATI playing field.
Again, we are very sorry for the mistake.
Bonesdad - Friday, July 2, 2004 - link
Wait till you see the numbers for NV's 6800 Ultra Extreme with Cheese!!!Pete - Friday, July 2, 2004 - link
Derek, was AA on for the nV cards? Apparently nV's latest drivers change behavior once again, to require AA to be set in-game, rather than via CP (which does nothing).Perhaps you could avoid this mess of ever-changing AA settings by using AA+AF for comparison screens? It'd also have the added benefit of showing the games in a more positive light. :)
joeyd - Friday, July 2, 2004 - link
gordon151 - Friday, July 2, 2004 - link
pio!pio! x-bit labs tested the difference between performance with the 1.2 and 1.1 patch on the NV3x (5900 Ultra) and well it wasn't pretty. NV3x actually saw a rather big performance drop using the new patch. I dunno if nVidia is gonna do anything about this since they seem to be turning a blind eye to the NV3x line with respect to future optimizations.DerekWilson - Friday, July 2, 2004 - link
trilinear optimizations are onanisotripic filtering optimizations are off
AA has less noticable benefit as resolution increases, but nearly vertical and nearly horizontal lines are still obvious in games with high contrast scenes.
kmmatney - Friday, July 2, 2004 - link
Do you really need AA on when running at 1600 x 1200, as in these these benchmarks? Just wondering if its much of a benefit at this high of a resolution. I never go past 1024 x 768, so I wouldn't know.pio!pio! - Friday, July 2, 2004 - link
So how about just the performance jump from FarCry 1.1 to 1.2 w/o using these high end shaders? (Ie for the previous generation Geforce 5900 crowd and lower)AnnoyedGrunt - Friday, July 2, 2004 - link
Does that mean trilinear optimizations on, or trilinear filtering on?Thanks,
D'oh!
DerekWilson - Friday, July 2, 2004 - link
we used driver default:trilinear on
anisotropic off