Originally Posted by peteer01
Taking a photo of an LCD or any other monitor is hardly an exact science. I've taken close to 100 photos of LCD projectors and my computer's LCD screen to test the image lag of various projectors, and it's possible to end up with a 1/120th of a second photo that is one frame crystal clear, and a 1/500th of a second photo that shows two different frames equally, giving the impression of strong blur. Don't assume that a 1/320th image is how every 1/320th of a second appears.
Bad timing is accounted for while taking the shot. Actually you can use this to your advantage in order to know more about your display
.Here is an example:
Lets take the example of an LCD with switching time of 4ms running @60FPs (16ms per frame) and no blanking time. I ll assume switching starts at the begining of every frame. Frame 1 starts @time 0.
0->4ms (the pannel is switching from previous frame to frame 1)
4ms->16ms (Frame 1 is on)
16ms->20ms((the pannel is switching from frame 1 to frame 2)
20ms->32ms(Frame 2 is on)
Now if you shoot with a speed 1/320, the shutter will remain open for 3ms. If this time window occurs during the intervals(7->13ms, 23->29ms) you WON'T
get any blurr and ((6+6)/32=37) thus 37% of your shot should contain no blur if manufacturer advertised spec are correct!
In the case of single ship DLP with negligible switching time (in micro seconds) and blanking time, the numbers comes out around 62.5%.
With very fast shutter speed and a lot of shots, it is very easy to reach conclusive results about panel response. I took 25 shots with my LCD and Guess how many clean shot(ie no blurr) I had with my LCD with a shutter speed of 1/320s????? NONE!
and that was conclusive for me to know my LCD was short in response speed. With the 11s2 DLP @1/1000s I had 9/10 shots without any blur and 7/10 shot with a shutter of 1/320s. Hence I decided to post one with no blurr since the remaining 3/10 were due to shot bad timing(overlap with frame transition)
SAH measurments on the other hand, requires much more sophisticated tools and methods to get accurate results. My method was in no way intended to be an industry benchmark for accurately measuring SAH
. Yet it was enough for me to illustrate what I see with my eyes and assess different display tech.