When conducting a Directx test with a frame cap there is a time dilation issue.
When a frame cap is applied the white screen stays only over a number of green light flickers/readings from the OSLTT device.
If the timer of the Directx mode white screen flash is made over a number of frames it can extend if a frame cap is applied.
Say you had a white screen flash for one second @1000FPS, a 200FPS cap would make it 5 seconds.
I would like to suggest a warm up screen/period where the FPS can normalize and be counted so the white screen duration can be calculated to the FPS in the warm up period.
Currently the FPS can take some time to get to any fps target so a warm up period will work well for this and avoid users starting the test before the target FPS is reached.
Or some other method of calculating how long to show a white screen for during a test that wont overlap and of course low latency.
There is a work around of extenting the auto clicker or doing it manually but that would take up a lot of time.
Say a 0.5 auto clicker of 50 tests which would be 50 seconds with 1000fps, if the fps was 200 the test would be 250 seconds.
The utility of frame capping a low load Directx app is quite useful, you can play with varying FPS targets with various Nvidia control panel settings to identify latency such as Low latency Vsync.
When conducting a Directx test with a frame cap there is a time dilation issue. When a frame cap is applied the white screen stays only over a number of green light flickers/readings from the OSLTT device. If the timer of the Directx mode white screen flash is made over a number of frames it can extend if a frame cap is applied. Say you had a white screen flash for one second @1000FPS, a 200FPS cap would make it 5 seconds.
I would like to suggest a warm up screen/period where the FPS can normalize and be counted so the white screen duration can be calculated to the FPS in the warm up period. Currently the FPS can take some time to get to any fps target so a warm up period will work well for this and avoid users starting the test before the target FPS is reached. Or some other method of calculating how long to show a white screen for during a test that wont overlap and of course low latency.
There is a work around of extenting the auto clicker or doing it manually but that would take up a lot of time. Say a 0.5 auto clicker of 50 tests which would be 50 seconds with 1000fps, if the fps was 200 the test would be 250 seconds.
The utility of frame capping a low load Directx app is quite useful, you can play with varying FPS targets with various Nvidia control panel settings to identify latency such as Low latency Vsync.