|
|
#1 | |
|
Registered User
Join Date: Jan 2009
Posts: 23
|
Just tried driver 304.37 on Debian "Squeeze" 32bit - GeForce 210
Everything worked well except I always like to launch glxgears as a tool to test all is well (like not too low FPS), I never use it as a real tool for precision testing, just for basic checking that acceleration is working ok, but I did notice the FPS were like really low this time: With previous 295.71 I get like 16700 Frames per each 5 seconds With newer 304.37 I got like 300 Frames per every 5 seconds. So I went back to previous version. Anyway, as I said, glxgears should never be considered an optimal tool for measuring FPS but for sure something must be wrong right? Thanks! |
|
|
|
|
|
|
#2 | |
|
Registered User
Join Date: Feb 2008
Posts: 163
|
This explains why you have less fps with 304.37
http://www.nvnews.net/vbulletin/showthread.php?t=176815 Quote:
300 frames divided by 5 seconds = 60 fps which = the screen refresh rate (this means sync-to-vblank is on).
__________________
leigh123linux |
|
|
|
|
|
|
#3 |
|
Registered User
Join Date: Jun 2006
Posts: 678
|
google: vsync
|
|
|
|
|
|
#4 | |
|
Registered User
Join Date: Sep 2010
Posts: 6
|
You will get the same "slow" FPS rate if in 295.71, you start the NVidia X server settings application, and then under the OpenGL setting you thick the Sync to VBlank option. What I'm trying to prove to you is that as far as this is concerned, there is no actual bug in the 304.37.
Last edited by pvelkovski; 08-17-12 at 08:45 PM. Reason: spelling error |
|
|
|
|
|
|
#5 |
|
Registered User
Join Date: Mar 2012
Posts: 31
|
|
|
|
|
|
|
#6 | |
|
Registered User
Join Date: Mar 2012
Posts: 31
|
To all this.
Does it have to do with the TrueBuffer activated? |
|
|
|
|
![]() |
| Thread Tools | |
|
|