Any video card benchmarks using results from modern games?

[Naruto]

I am seeing lots of benchmarks of games I don't play, so I really don't have any idea what is the best card to upgrade to for my gaming tastes. I know the games such as quake 3 are a standard and UT2k3 is becoming standard. I say that more than half of the benchmark games or tools used I don't play or use. Are there any benchmarks using Call of Duty? Homeworld 2? C&C Generals? Warcraft 3? Jedi Academy? Medal of Honor? Battlefield 1942? Thanks.

 

[modedepe]

Check out some of AT's fall video articles here. They benched quite a few games, including some of the ones you mentioned.

 

[Matthias99]

The problem is that not *one* of those games has any sort of built-in benchmarking (some, like BF1942, have FPS displays, but no timedemo capabilities that I'm aware of), and so testing with them requires using an external program like FRAPS, which is, to be frank, a huge pain in the butt. Some sites are doing it, but it takes a *long* time to run lots of tests at different resolutions, etc., and the results are harder to compare directly. Quake3 and UT2K3 both have very good demo recording/playback/benchmarking capabilities, and it's easy to reproduce and verify the results.

 

[jiffylube1024]

Originally posted by: Matthias99
The problem is that not *one* of those games has any sort of built-in benchmarking (some, like BF1942, have FPS displays, but no timedemo capabilities that I'm aware of), and so testing with them requires using an external program like FRAPS, which is, to be frank, a huge pain in the butt. Some sites are doing it, but it takes a *long* time to run lots of tests at different resolutions, etc., and the results are harder to compare directly. Quake3 and UT2K3 both have very good demo recording/playback/benchmarking capabilities, and it's easy to reproduce and verify the results.

And those games are also both heavily optimized for by both ATI and Intel. It is actually a lot better to benchmark on games that aren't as high profile (or as commonly benchmarked) to get an idea of overall performance. As for Quake3, UT2k3, 3dmark 2k1 and 2k3, those games have been optimized for time and time again and aren't that accurate for a "big picture" view on performance.

The GeForce4 Ti cards are incredible on UT2k3, for example, but on other modern games they can be a bit lacking.

 

[Bateluer]

Originally posted by: jiffylube1024

Originally posted by: Matthias99
The problem is that not *one* of those games has any sort of built-in benchmarking (some, like BF1942, have FPS displays, but no timedemo capabilities that I'm aware of), and so testing with them requires using an external program like FRAPS, which is, to be frank, a huge pain in the butt. Some sites are doing it, but it takes a *long* time to run lots of tests at different resolutions, etc., and the results are harder to compare directly. Quake3 and UT2K3 both have very good demo recording/playback/benchmarking capabilities, and it's easy to reproduce and verify the results.

And those games are also both heavily optimized for by both ATI and Intel. It is actually a lot better to benchmark on games that aren't as high profile (or as commonly benchmarked) to get an idea of overall performance. As for Quake3, UT2k3, 3dmark 2k1 and 2k3, those games have been optimized for time and time again and aren't that accurate for a "big picture" view on performance.

The GeForce4 Ti cards are incredible on UT2k3, for example, but on other modern games they can be a bit lacking.

Q3 is optimized for by everybody, its been out so long, but still Nvidia and Intel parts run it faster than ATI and Intel.

Anandtech benches Homeworld 2, C&C Generals, Jedi Academy, NWN, and a few others on their last couple of vidcard benches.