I have worked for a few years on developing computerized vision tests for research purposes and did not realize just how difficult it can be to guarantee the ac

I have worked for a few years on developing computerized vision tests for research purposes and did not realize just how difficult it can be to guarantee the accuracy of the stimulus on different monitors. There is so much that can effect the output image that it's a nightmare. Often the Edid info monitors return is missing data or has marketing values listed instead of actually values. (Screen size of 21" vs 20.5"). There can be differences based on whether the monitor is plugged in via dvi, dp, hdmi whether you have nvidia,amd,intel gpu's. Is scaling being done on the GPU or display or both? How is windows color profiles effecting your image? How are font settings effecting your image? Are you running fullscreen or windowed cause that matters. What if timing is critical to your stimulus? Is your software aware when displays are added and removed ? I originally wanted the software to support arbitrary displays and gpus but the problem of showing actually the light you want to on a monitor is way harder then I ever would have imagined.