timing delays in Matlab 5.2 under Mac OS 9.2 ??

Hi,

We had just finished an fmri scanning session, and we noticed
some time delays in our experiments. Using one experiment
that we did as an example, the retinotopic mapping, we had a
wedge of 45 deg, jumping at 45 deg to the next position, and it
stays at one position in the circle for 4 secs (therefore, it has 8
positions to bypass before completing one circle).

So, the wedge should take about 32 secs to complete one circle,
and since we are repeating this 8 times, then the whole
experiment should take about 256 secs.

However, this is not the case. With the above values, we got
more than 256 secs overall during a trial run without the fmri. So
we calibrated our timing to 3.87 secs instead of 4 secs for the
duration when the wedge stays at one position. It worked fine
during this fake trial.

However, during the real run and with this new value, we still got
more than 259 (!!) secs overall, and each circle lasts slightly
longer than 32 secs. We are using the function 'getsecs' to get
the timings from the experiment, and no other programs were
running at the same time.

We are using Matlab 5.2 under Mac OS 9.2 on a G4. I saw on the
website that there were some bugs found for Matlab 6.5 under
OSX, but none for Matlab 5 with Mac 9.

I am not sure what is going on, because this program does not
involve any human response time... we are just trying to get the
time length of the displaying of the stimulus on screen.

If anyone has any take, suggestion or comment on this issue,
please do let me know...

Thank you very much!

Tian
Ophthalmology Research Lab
University of British Columbia
It's impossible to diagnose your problem without seeing your code.
Please post it to the list or send it to me.

keith

On May 27, 2004, at 7:34 PM, wtianlim2003 wrote:

> Hi,
>
> We had just finished an fmri scanning session, and we noticed
> some time delays in our experiments. Using one experiment
> that we did as an example, the retinotopic mapping, we had a
> wedge of 45 deg, jumping at 45 deg to the next position, and it
> stays at one position in the circle for 4 secs (therefore, it has 8
> positions to bypass before completing one circle).
>
> So, the wedge should take about 32 secs to complete one circle,
> and since we are repeating this 8 times, then the whole
> experiment should take about 256 secs.
>
> However, this is not the case. With the above values, we got
> more than 256 secs overall during a trial run without the fmri. So
> we calibrated our timing to 3.87 secs instead of 4 secs for the
> duration when the wedge stays at one position. It worked fine
> during this fake trial.
>
> However, during the real run and with this new value, we still got
> more than 259 (!!) secs overall, and each circle lasts slightly
> longer than 32 secs. We are using the function 'getsecs' to get
> the timings from the experiment, and no other programs were
> running at the same time.
>
> We are using Matlab 5.2 under Mac OS 9.2 on a G4. I saw on the
> website that there were some bugs found for Matlab 6.5 under
> OSX, but none for Matlab 5 with Mac 9.
>
> I am not sure what is going on, because this program does not
> involve any human response time... we are just trying to get the
> time length of the displaying of the stimulus on screen.
>
> If anyone has any take, suggestion or comment on this issue,
> please do let me know...
>
> Thank you very much!
>
> Tian
> Ophthalmology Research Lab
> University of British Columbia
>
>
>
>
> http://psychtoolbox.org
> POST a message to: psychtoolbox@yahoogroups.com
> UNSUBSCRIBE by sending a blank message to:
> psychtoolbox-unsubscribe@yahoogroups.com
> Yahoo! Groups Links
>
>
>
>
>
>
-----
Keith Schneider
Department of Psychology
Green Hall
Princeton University
Princeton, NJ 08544
(609) 258-1480