microlite

Those numbers sure sound sweet gdogg I'm really looking forward to public release Power to gdogg's RL and OS contentions!

The problem is that all games do not support DirectInput or have a bad implementation of it. For USB the mouse polling rate can be as important as the frame rate in FPS games.

@gdogg Hehe, you guessed right These are small and easy to use tools, I'd really appreciate it if they're tested hope they don't cause any problems, they're important for FPS games! Download - Anir's Windows Mouse Acceleration Fix disables capability of erroneously enabled acceleration Download - SweetLow's USB Mouse Rate Adjuster setups polling rates for HID compliant USB mice Download - Tscherwitschke's Mouserate Checker checks PS/2 and USB mouserates in realtime

The total size of the installation doesn't matter much to me, it's the unnecessary services running and the RAM wasted that matters. I don't think average fps numbers will change much generally, to prove the advantages better I think framerate graphs would be needed. There would not be so many background processes running that could disturb the game and there would be more RAM available for it. The advantages could be faster loading and more stable performance. The SWAT 4 numbers posted earlier in this thread suggests that Vsync was in the play. Guessing it was enabled at first with the monitor at 100Hz (it couldn't maintain 100fps so it dropped to 50fps to stay in sync). Next it looks like Vsync was disabled and the fps was limited by software (ingame fps limit set to 85). Vsync should be disabled for benchmarking and enabled for playing. In-game fps limiters should be disabled in both cases (or set to 0 or 999).