Timing issues about short stimuli

Zhenzhen · November 2019

Hi，

In my experiment, I want to present a very short stimulus (16/32/48 ms), followed by a mask. To test whether the actual presentation time is 16/32/48 ms, I used canvas.show() to log the timestamp of both the target and the mask. I calculated the presentation time by subtracting the target_timestamp from the mask_timestamp. On my own computer(refresh rate 59Hz), the calculated presentation time was about 18,34,50 ms respectively. When I tested the same program in the laboratory(refresh rate 120Hz), the calculated presentation time have a delay of about 39ms under the xpyriment backend ( like 54/70/86ms respectively) and a delay of about 28 ms under the psychopy backend. I am very confused about this problem, and I don’t know the cause of this problem. I would appreciate it if you can give me some clues to solve this problem. Looking forward to hearing from you!

eduard · November 2019

HI,

the images you uploaded cannot be displayed (I think something is wrong with the forum, @sebastiaan do you know anything about it?)

Regarding the timing, could it be that the lab monitor is also larger, so that more pixels have to be flipped on each frame (I'd be surprised, honestly). Do you have the chance to run the experiment with your pc, but the lab monitor? I am not terribly familiar with graphic cards, monitors and alike, but maybe the problem originates there. It is unlikely that it is related to you experiment. To be really sure, you can try this timing check, described here: https://osdoc.cogsci.nl/3.2/manual/timing/

Sorry & Good luck

Eduard

Zhenzhen · November 2019

Hi Eduard,

Thank you for your reply. I also considered that this may originates from the difference of hardware？Because the time intervals between two canvas without sleep is 28 ms in the laboratory while the refresh rate is 120Hz. However, this time interval is 16 ms on my own computer as the refresh rate is 59 Hz.

Zhenzhen

sebastiaan · November 2019

Hi Zhenzhen,

It would be easiest if you upload the experiment here so we can see what might cause the delay. It's probably in part due to the interaction between the hardware and OpenSesame, but even so there might be ways to optimize it.

(I think something is wrong with the forum, @sebastiaan do you know anything about it?)

Ugh, yes, some attachments got lost during the recent forum struggles.

Cheers!

Sebastiaan

Zhenzhen · November 2019

Hi Sebastiaan,

I really want to figure out the cause of this problem. Thank you very much for your time and help！

Sequential.osexp

sebastiaan · November 2019

Hi Zhenzhen,

There's nothing wrong in your experiment, as far as I can see, that would unnecessarily slow things down. And for me it works fine. This suggests that there is an unfortunate interaction between the video card of the lab computer and OpenSesame (or rather OpenGL, which is the software that PsychoPy and Expyriment both use for display presentation). I would take this up with the lab managers to see if you can tweak some settings on the lab computers to fix this. The problem is not, in any case, with the experiment.

Cheers!

Sebastiaan

Zhenzhen · November 2019

Hi Sebastiaan,

Thank you for your help!

Cheers,

Zhenzhen

Howdy, Stranger!

Categories

Timing issues about short stimuli

Comments

Howdy, Stranger!

Quick Links

Categories

Timing issues about short stimuli

Comments