Variable Durations for same Script file

[JSalmon]

Hi,

I have an interesting problem that I was wondering if you could shed some light on. I'm running an experiment where I'm trying to show a picture for 20ms. Because the refresh rate, this translates into 24-26ms (in the log file). However, and this is the problem -- every once in awhile (one in four participants) this 20ms instead gets converted to what must be three refresh rates and all those trials come up as 35-37ms.

Fortunately it is consistent (once it switches to 35 ms it stays for all trials for that participant). But it IS a problem. Why would the same DirectRT script file give me 25ms sometimes and 35ms (consistently) other times?

To my knowledge no other settings have been changing on the computers to mediate this. This SEEMS to be a DirectRT thing/problem. Anyone encountered this before?

It seems like a problem with the script file interpreter. Sometimes it's interpreted as 2 screen refreshes/frames, and sometimes as 3!!

Regards,
Josh Salmon

[jarvis24]

It sounds like you are working with a 12-13ms refresh rate. If so, the your displays will either be on for say, 12, 24, or 36ms. If there is a lag in your display hardware you can compensate by subtracting off ms from the requested values. In your case though you are getting 24 sometimes and then 36's. I can't say for sure what is going on, but one thing I would try is changing the requested value to 15-17ms. High enough so that DirectRT will not try for 12 and low enough that a few ms of display hardware lag will still make 24ms possible. It's unusual to have to do this, but you can, at least, count on the log file data to give you the actual display times so you can play around with stuff like this.

See also: http://www.empirisoft.com/Support/showthread.php?t=55