crash when going through tutorial

[Thorsteinn]

I am a new user with brainStorm and just started going through the tutorial. In the "Importing MEG data" section the import data window crashes when I check the box for removing DC offset.

I'm running Matlab 7.3 on Unix and using the lates BrainStorm version.

I am also experiencing trouble importing RAW data, I suspect it has something to do with the fact that I am working on a Sun computer with big-endian representation and not litle-endian as is the standard for Intel processors? Has anyone else runing Unix (Sun) had this kind of trouble?

[Sylvain]

Hi, thanks for reporting on your troubles and welcome to BrainStorm !

May I ask what is the original data format you are using ?

As for the raw data format and DC offset issues, I need to know more about the error message you're getting.

We have had no experience with Sun systems so far but maybe some other users have. If troubles are persistent, I'd suggest you ship me a bit of your data by mail or ftp.

[Thorsteinn]

Thank you for the response Sylvain.
I'll try to be more specific with the errors I get. Both errors occurr when following the tutorials supplied on the website, so at the moment I am just using the demo data supplied.

(i) In the "importing MEG data" tutorial the "Importing data from..." window crashes (closes down) when I check the "Remove DC offset" box.

(ii) In next tutorial section, "Importing RAW data", when using the Phantom Database the data is not read in correctly (no data gets loaded). Following is the printout from the "Messages" window after trying to import the data:

Converting sensor 1 with string "Stimulation input" to "OTHER"
You have loaded 182 Sensor Channels, and you have loaded
a data matrix that is 0 rows by 0 columns.
The number of rows in the data matrix must be equivalent to
the number of channels.
You have loaded a total 182 Sensor channels, including
possible references,
BUT, you have loaded 0 rows of data.
The number of rows in the spatial temporal data matrix must
match the number of channels.
Please add/subtract in rows in the data matrix, including
reference data, even if it is a row of zeros.
See also HELP_DATA_CHANNEL, HELP_DATA_RAW, HELP_DATA_DATA

For both of the bugs listed above there are no messages (errors and/or warnings) in the Matlab command window.

[Sylvain]

Hi again - thanks for the details

(i) In the "importing MEG data" tutorial the "Importing data from..." window crashes (closes down) when I check the "Remove DC offset" box.

Do not click this box before you select the dataset you want to import once you have selected a format you want to import from, from the pulldown menu (import data from...)

(ii) In next tutorial section, "Importing RAW data", when using the Phantom Database the data is not read in correctly (no data gets loaded). Following is the printout from the "Messages" window after trying to import the data:

It seems the channel information is properly loaded - not the data array. Which file do you select as a raw data file ?

[Thorsteinn]

Hi again - thanks for the details

Do not click this box before you select the dataset you want to import once you have selected a format you want to import from, from the pulldown menu (import data from...)

Still crashes.


It seems the channel information is properly loaded - not the data array. Which file do you select as a raw data file ?

I use
.../brainstorm/Phantom/Sample_Raw/Mixed_data_Trial1.raw


Since I am only working on this Sun station temporarily and no one out there seems to be running BrainStorm on Sun machines I am planing on continuing my work next week when I get my permanant computer (Windows/Linux).

But thanks for the quick responses Sylvain

Cheers - Thorsteinn

[Sylvain]

OK - I cannot replicate this crash on my XP machine or a Linux station. please keep me posted when you have your new install running under XP/Linux.

From our experience, this part should be runnign just swimingly.

Cheers!

[Thorsteinn]

Hi,

I am now running on XP. The bug marked (i) in the post below still happens, I just skip removing the DC bias to avoid it! But the error marked (ii) is gone.

In the "Importing MRI Data" tutorial I have had some trouble. Following the tutorial exactly an error occured when importing the "somMRI.mri" file. The error was:
Error using ==> set
Bad value for figure property: 'Visible'.

I traced it to line 255 in the m-file:
...brainstorm/Toolbox/bst_mriviewer.m
Changing the line from:
fig = openfig(mfilename,'reuse');
to
fig = openfig(mfilename,'reuse','visible');
got rid of the error.
Maybee the bug is related to the Matlab version I'm running? Maybee someone can replicate this error?

I am running Matlab 7.3.0 (R2006b) and BrainStorm 2.2.21 (05 Dec-2006).

Cheers - Thorsteinn

[Sylvain]

Looks like it's something happening with the newest version of Matlab. If you can go back to 2006a, would be perfect, otherwise you might encounter other annoying things like that until we upgrade BrainStorm to Matlab's latest tricks.

But please keep us posted if you get stuck somewhere as generally things can be fixed with minor tweaks.

Cheers