the idea is having a mouse rate thats equals fps * integer which in my experience results in the smoothest mouse movement. And since i play cs mainly where we use 100 fps unlike quake with 125fps (where i could use 125x2 250hz) i hope you see my point... I know you're gonna tell me - use 500hz; but the problem is the mice i use, g1 and mx518 aren't stable on 500hz so you see the problem. I know xai is suppose to have exactrate or something where you can change mouse rate in steps of 1hz(not sure dont have a xai), so i was wondering is something like that possible with other mice editing the hidusb.sys or some other method.
Having recently debugged a Xai exactrate problem with MouseMovementRecorder (included in the MarkC accel fix), I can say that the Xai exactrate is not much use really.
You set the Xia driver to record mouse movement at any arbitrary frequency, and it does so, BUT it can only report those movements to Windows at the usual USB polling multiples of 1ms, same as any other variable report rate mouse.
If the Xai exactrate is higher than the USB report rate that your PC can handle, then sometimes each USB report/poll gets sent TWO Xai recorded movements, which does nothing except waste time in the drivers (and confuse MouseMovementRecorder).
It is no better than SetPoint (or HIDUSBF) setting the report rate to 1000ms.
WM_INPUT, DirectInput, WM_MOUSEMOVE, I would say this Xai 'problem' affects all of the above, because it happens at the USB level before any input even gets to any mouse driver.
Would you people stop it with the mouserate checker.... you don't even know how that program works, or how usb polling rate works. Does it feel unstable in-game? If no you are perfectly fine.