Display Latency Comparisons with DIY Arduino tool

[SeppNel]

Hello, this is my first post here.

I recently build a latency tester with an Arduino and a light sensor. It seems to work well, but now I want to start comparing the results to other systems and I have some concerns.

First, because this is measuring complete system latency (Arduino + PC + Display) if I want to compare just the display latency I would need to use the same PC, otherwise the latencies would not be comparable.

My question is that, if I enable VSYNC (at the same refresh rate) it should basycally nullify the effect of the PC latency, at least if the PC can update at the refresh rate, making the Arduino + PC part equal and allowing me to compare display latency.
Is this correct?

Second, Im quite satisfied with the results im getting, wich if someone could doble check that every at least seems correct here are my results on a cheap 60hz Monitor:

VSYNC Off = 47.04 ms AVG
VSYNC On = 83.24 ms AVG

But it would be nice to be more accurate. Im thinking that if I know the latency of the arduino part, I could subtract that amount to get a better aproximation to real world. Im guessing the Arduino latency would be mostly constant. Does that make sense? And if it does, does someone know how could I measure it?

[RealNC]

Second, Im quite satisfied with the results im getting, wich if someone could doble check that every at least seems correct here are my results on a cheap 60hz Monitor:

VSYNC Off = 47.04 ms AVG
VSYNC On = 83.24 ms AVG

Without an FPS cap, this looks just about right to me.