- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]jaquith[/nom]While I appreciate the benchmarks, a couple didn't make sense until I noted 'beta drivers'...[/citation]
Beta drivers is a concern for you? In years of testing I can confidently say they rarely have any perceptible impact. When they do, which is very rare, they represent where the WHQLs are going. I can't imagine why that would be a problem, what's your beef with betas?
[citation][nom]jaquith[/nom]...and no defined benchmark just a walk through on (1) one level so ±4 FPS or larger variances are common with these type testing methods. What's your variance observation and how many times did you repeat each run or did you just do it once?[/citation]
It's a precisely defined run with almost no variance. With multiple (5x) tests the difference is less than 1.5 FPS, usually within 1 FPS. It's an extremely reliable method and I use a long (over 2 minute) repeatable sequence of actual in-game stealth play, so the variables are minimal. The jungle area seems to be the hardest on frame rates so it's an ideal worst-case-scenario from what I've tested.
[citation][nom]jaquith[/nom]What's your thoughts on the game itself? Have you played multiplayer? Any help is appreciated.Example BF3's campaign is bad, but multiplayer is great.[/citation]
All that's covered on page one, please read before commenting.
Beta drivers is a concern for you? In years of testing I can confidently say they rarely have any perceptible impact. When they do, which is very rare, they represent where the WHQLs are going. I can't imagine why that would be a problem, what's your beef with betas?
[citation][nom]jaquith[/nom]...and no defined benchmark just a walk through on (1) one level so ±4 FPS or larger variances are common with these type testing methods. What's your variance observation and how many times did you repeat each run or did you just do it once?[/citation]
It's a precisely defined run with almost no variance. With multiple (5x) tests the difference is less than 1.5 FPS, usually within 1 FPS. It's an extremely reliable method and I use a long (over 2 minute) repeatable sequence of actual in-game stealth play, so the variables are minimal. The jungle area seems to be the hardest on frame rates so it's an ideal worst-case-scenario from what I've tested.
[citation][nom]jaquith[/nom]What's your thoughts on the game itself? Have you played multiplayer? Any help is appreciated.Example BF3's campaign is bad, but multiplayer is great.[/citation]
All that's covered on page one, please read before commenting.
Facebook
Twitter
Instagram