Originally posted by: black_zion They're not so much trolls as they are those who are blind and will not see with eyes wide shut, to mix a nonexistent metaphor. There are articles out there about everything on the reputable sites except one which actually explains what video drivers are and how they work, as well as the impact a game's code has on performance. Every time AMD and nVidia release new Catalyst and Forceware driver sets they all copy the release notes and maybe run a couple of benchmarks with the old and new sets, which is what causes people like these to think that AMD and nVidia have everything to do with the performance of a game after it is released, even though many developers have to release patches which increase performance and fix issues which never should have been present in the first place. Drivers do play A part in performance, especially as far as multi-GPU performance is concerned, but the game's programming is the major determining factor.
Ok. Let's start over.
Why do you think it's more to do with the game code (if not completely) more than the AMD drivers?
And then, why do you think the game runs flawlessly on nvidia but not AMD?
We can ignore the multi-gpu performance issues because we both know this is a driver issue and it is present on both AMD and Nvidia setups in rfactor 2 but this is not the issue being raised. Based on the single gpu perforamnce alone, AMD cards are performaming at less than half of their respective nvidia counterparts.
If i recall correctly, it is either one or two key graphical settings in game (i think enviromental reflections was one of them) that when set on anything other than disabled, causes the performance issues we are seeing on AMD cards.
Are you suggesting that the code handling these settings could be somehow biased towards nvidia hardware?
I can only take the word of the developers at this point in believing that they have done all that they can from their end until AMD steps up with new and specific game driver support. Whilst i have no proof the devs are right/wrong or whether they're not telling the truth (which i think is highly unlikey because this issue can only lose them sales as rf2 has a demo and a paid copy can be refunded), it's very common to find new drivers released for new games. Sometimes the performance gains from new drivers are very small (bordering insignicant and on the rare occasion worsens performance) and sometimes they are game changing....i'm interested in why you think this is not likely the case with amd cards on rf2?
I can understand your reservations to an extent....i mean seeing that everything on full, the GTX 770 is performing a ridiculous 2.5x better than a HD 7970 (one of the worst case comparisions from the compiled benchmark results), when in any other game they should be on par with one another.
The benchmark graphic even shows something is clearly wrong with updated amd drivers in rf2. The HD 7970 should be performing on average 45% better than a HD 6970 (varies between 20% to 75% with different games) yet a HD 6970 on older Catalyst 12.11 beta drivers is besting a HD 7970 on newer Catalyst 13.11 beta drivers by 48%. In relative terms, the HD 6970 is performing 214% better by going back to older drivers. Ok, perhaps a bad example as their beta drivers but if amd beta drivers are anything like nvidia beta drivers, they are almost always as good as their respectively made WHQL predecessors and successors and i've never encountered one that has taken a step back anywhere near as big as a 50% reduction in performance with newer drivers.
I'd honestly like to hear your thoughts and/or anyone else's on the above.
02:07 PM by