facebook rss twitter

Review: ATI CATALYST 5.1 Performance Analysis

by David Ross on 18 January 2005, 00:00

Tags: AMD (NYSE:AMD), ATi Technologies (NYSE:AMD)

Quick Link: HEXUS.net/qa6k

Add to My Vault: x

Far Cry, UT 2004, Sims 2, Xpand Rally

We now move on to our brief performance examination based around our own set of HEXUS benchmarking titles to see if there are improvements of note. All of the benchmarks you see here will be run at 1024x768, 1280x1024 and 1600x1200, first without AA and AF and then with 4x anti-aliasing and 8x anisotropic filtering.

I'll begin by mentioning three titles that showed no change when using CATALYST 5.1, either for better for worse. First up was Unreal Tournament 2004, which is already largely CPU-limited and thus gains little from video card driver improvements these days, and so it proved here. Then there was Far Cry - We found with our CATALYST 4.12 comparison that performance seemed to be basically optimal here, and so it has proved again, with no fluctuations in benchmarks on our 'Research' timedemo. Finally, The Sims 2 retained the steady framerates we've seem since CATALYST 4.11, so no changes there either. Obviously, there is little point in showing charts which... Well... Chart nothing, so I've therefore omitted them from the article.

Which brings us on to the titles which did show some changes in performance - First up, Xpand Rally, using our own custom recorded replay, played back using FRAPS.


We start without anti-aliasing and anisotropic filtering, which shows some almost imperceptible changes for the better - Mostly in line with the margin of error, but the result at 1600x1200 suggests that there is indeed a very small improvement, although admittedly nothing to jump from the rooftops about.


It's a similar story with AA and AF switched on in-game, as framerates rise slightly with the use of CATALYST 5.1, most notably at 1280x1024.