| Author |
 Topic  |
|
|
jamie.morris
BIAC Alum
59 Posts |
 Posted - Dec 20 2004 : 2:51:50 PM
|
The last 3 times I have scanned a subject on the 4T I have had a problem with Cigal crashing during the middle of the experiment. Is there a known conflict that I am missing here? I am running Cigal locally with all stimuli in it's appropriate folder on the D drive.
|
|
|
diaz
BIAC Alum
USA
212 Posts |
Posted - Dec 20 2004 : 2:58:00 PM
|
I have had a problem where the screen turns white or black (I'm not sure if this is what you mean by crashing). When this happens it has always been the screen saver coming on - so I just turned it off via the control panel.
Does Cigal give you any kind of error message when it crashes?
Michele |
Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center |
 |
|
|
jamie.morris
BIAC Alum
59 Posts |
Posted - Dec 20 2004 : 3:00:24 PM
|
I have seen the screen saver error and taken care of that. However, this is simply an error where almost all of my stimuli load (there are a lot of them) and then Cigal "unexpectedly had to quit."
|
|
|
|
melissa.slavin
BIAC Alum
Australia
44 Posts |
Posted - Dec 20 2004 : 3:00:58 PM
|
I've had a problem with Cigal crashing after 2 of our 3 runs for a long time. I assumed it was something to do with memory, and always restart Cigal after run 2. I'm sure Jim can answer this more specifically. By the way, I always restart the computer before starting at all - perhaps superstitious, but has resulted in less problems than in the past.
Melissa |
|
|
|
francis.favorini
Forum Admin
USA
618 Posts |
Posted - Dec 20 2004 : 3:34:03 PM
|
Folks,
Which computer(s) (Donders, Mach, Kulpe, Darwin) does this occur on? What dates did it occur? Does it always crash on the same file in the same place? I.e., is it repeatable or is it random?
-Francis
P.S. I believe Jim is away starting today.
|
IT Director, Brain Imaging and Analysis Center |
|
|
|
melissa.slavin
BIAC Alum
Australia
44 Posts |
Posted - Dec 20 2004 : 4:31:32 PM
|
| For me, it is on Darwin (that's the one on the right, right?). |
|
|
|
jamie.morris
BIAC Alum
59 Posts |
Posted - Dec 20 2004 : 5:04:57 PM
|
| I've only run from Darwin (computer on the right). It normally happens once after the first couple of runs. I rebooted before this afternoon's subject and it hasn't happened yet. |
|
|
|
jamie.morris
BIAC Alum
59 Posts |
Posted - Dec 28 2004 : 1:05:59 PM
|
| The same problem is occurring with Mach (the computer on the right at the 1.5T). Kevin Pelphrey and I had Cigal crash from an "unknown error" at the beginning of run 2. Kevin reports that this has happened on several occasions. |
|
|
|
diaz
BIAC Alum
USA
212 Posts |
Posted - Jan 05 2005 : 4:47:14 PM
|
Cigal crashed again several times during a run today. This occurred on Darwin (the 4T computer on the right). Each time an error box appeared (an unexplained or unexpected error occurred) and cigal had to be closed. I had previously successfully ran all of the files (on the same computer - so it does not appear to be linked to a particular file).
Michele |
Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center |
|
|
|
jim.voyvodic
BIAC Faculty
138 Posts |
Posted - Jan 05 2005 : 10:49:00 PM
|
I am still in South America and can't really help, but this problem doesn't sound like anything I've seen before. I will need more information (of the sort Francis was asking about). I gather that the error message that appears is a Windows error box, and not a CIGAL error? Does it only happen during a Showplay paradigm? Does the computer crash? Most importantly, is it reproducible? and if so, send me the paradigm (PPF) files that can produce the problem.
I will look into this next week when back in the northern hemisphere
(while my tan quickly fades away).
Jim
PS Happy new year!
|
|
|
|
jamie.morris
BIAC Alum
59 Posts |
Posted - Jan 06 2005 : 3:45:41 PM
|
The problem occured on the 1.5T today. It is always a Windows dialogue box that says that Cigal.exe had to shut down for unknow reasons. It does not crash the computer, but does remove the Start Menu Bar from the bottom of the screen. It once again happened after the third run.
|
|
|
|
jim.voyvodic
BIAC Faculty
138 Posts |
Posted - Jan 13 2005 : 12:04:07 PM
|
I believe that the problem(s) discussed in this topic are all caused because CIGAL did not have enough memory allocated to load all the stimulus images. Although memory is cleared after each run, it gets fragmented over time so that if you are using close to the maximum available memory initially you will run out of space after many runs.
The fix was simple because the CIGAL start-up file (sysstart.imp) has a line that allocates memory, but it was being capped at 200Mb (mostly for historical reasons). We have removed the cap, so now CIGAL will allocate 95% of the computer's available memory. On the scanner XP computers this will give you 600-800 Mb for loading images (assuming you aren't running too many other programs at the same time.)
To see how much memory is allocated click on the "About CIGAL" entry in the File menu. You will see something like:
...
High memory: 480.0 Mb allocated, 99.9 Mb free, 30.0 Mb available
This would mean that your data are currently using all but 99.9 Mb of the 480 Mb allocated to CIGAL, and the computer has 30 Mb that you probably could have allocated (but didn't).
Having already increased CIGAL's memory allocation on all 4 scanner PCs and Gall/Programs, this memory overflow problem shouldn't occur again with any current paradigms.
In the next ShowPlay release we will make sure that you get an error message if you try to use too much memory. The recent problems probably generated a "memory overflow" error message while loading images, but then didn't stop, so the program crashed when it tried to access memory that hadn't been allocated properly.
Jim |
|
|
|
diaz
BIAC Alum
USA
212 Posts |
Posted - Jan 14 2005 : 2:08:02 PM
|
| I ran 2 subjects yesterday and ran into a collection of error messages with Cigal during both runs. They always occurred after I had selected "RUN" from the drop down message (immediately after selecting the paradigm file). They either occurred immediately or before the stimuli had completely loaded. The errors were either 1) a generic windows box stating "cigal.exe has encountered a problem and needs to close . . ." 2)(Cigal) error 42 stack(internal memory space) overflow and 3) (Cigal) error 89 stack memory allocation list corrupted. I received all 3 errors multiple times and each time cigal had to be restarted. Once it was restarted, it loaded and ran the same file without incident. I was using the computer on the left at the 4T. |
Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center |
Edited by - diaz on Jan 14 2005 2:18:25 PM |
|
|
|
Dichter
BIAC Faculty
190 Posts |
Posted - Jan 15 2005 : 11:04:08 AM
|
Cigal froze in the middle of a run during a scan today at the 4T. I was using Darwin (the 4T computer on the right). There was no error message; the program simply froze on an image. I quit out of Cigal and ran the same paradigm again, and it proceeded without freezing. I had previously successfully ran the file on the same computer - so it does not appear to be linked to a particular file.
Additionally, the screens on both 4T computers intermittently went black throughout the scan, but the subject reported that his image through the goggles from Darwin was uninterrupted.
-Gabriel |
Gabriel S. Dichter, PhD
UNC Departments of Psychiatry & Psychology
http://www.can.unc.edu/ |
|
|
| |
Topic |
|
Cigal Issue with Darwin on 4T
