View Single Post
  #9  
Old 06-28-2022, 08:47 PM
mg is offline mg
Registered User

Join Date: Jun 2022
Posts: 27

Just saw your edit,

Quote:
I suspect, for some reason, under certain conditions, the game fails to deactivate mouse acceleration, but only to the right.
Per my previous message I think we should be able to invalidate that hypothesis at least (I don't use mouse acceleration in Windows and we can prove the game doesn't use raw input, yet I see the problem anyway).

Thanks for noting that just alt-tabbing can impact this: I was never satisfied that it was about scene complexity (especially since I could never observe any actual variability in the frame time).

Definitely an odd one!

Quote:
Tests on the up-down axis should be considered.
Oh indeed, I'm already quite sure it's impacted as well. A quick test is to rapidly flick the mouse left and right without mouse acceleration enabled many times, then return it to its starting point. By all logic, assuming a sensor of a high-enough quality, the character should be aiming in the original direction exactly. I can do this test successfully in non-Novalogic games. Yet, quite often in Novalogic games, the character ends up aiming further down at the end. If I wiggle the mouse that way for long enough, the viewpoint will eventually point straight down (or as low as the game will allow). For now I'm assuming that it's the same issue in that some slight movement up or down is not restored the other way on subsequent movement because of these skipped inputs we're investigating.
Reply With Quote