Back to the main page.
Bug 1197 - Problem with reading Neuroscan 4.5 format file
| Status | CLOSED WONTFIX |
| Reported | 2011-11-28 16:01:00 +0100 |
| Modified | 2019-08-10 11:46:22 +0200 |
| Product: | FieldTrip |
| Component: | fileio |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P2 major |
| Assigned to: | Vladimir Litvak |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Vladimir Litvak - 2011-11-28 16:01:36 +0100
Created attachment 193 Example cnt file See below. I tried to convert the attached example file and the number of channels is actually correct. It's just that every other channels is practically flat (actually switches between two values with a small difference). Vladimir ---------------------------------------- Hello I got your contact info from colleagues at the Cyclotron research centre in Liege, and am hoping you can help with a problem we have encountered in trying to read Neuroscan .cnt files in SPM8 eeg. Our colleague Caroline Schnakers has been importing .cnt files without problems using Neuroscan 4.2. We are using Neurocan 4.5 and are encountering problems in converting the files. When opening the file only every second channel appears in SPM. Thank you very much Marianne Løvstad -- Marianne Løvstad Psykologspesialist/Ph.d. student Sunnaas sykehus tlf: + 47 93 45 20 03
Robert Oostenveld - 2011-11-28 17:09:09 +0100
this has the characteristics of a 16 versus 32 bit problem, which has been an issue with neuroscan files before.
Vladimir Litvak - 2011-11-28 21:06:55 +0100
(In reply to comment #1) Oh yes, it's that again. I didn't think so at first because I thought that at the time we got complete rubbish with the wrong setting but yes, nothing new here.
Vladimir Litvak - 2011-11-28 21:08:08 +0100
(In reply to comment #1) Oh yes, it's that again. I didn't think so at first because I thought that at the time we got complete rubbish with the wrong setting but yes, nothing new here.
Boris Reuderink - 2011-12-06 13:17:30 +0100
Created attachment 198 Script to reproduce bug. Please rename contributed .cnt file to bug1197.cnt.
Boris Reuderink - 2011-12-06 13:19:17 +0100
Confirmed in SVN revision 4938.
Vladimir Litvak - 2011-12-06 13:22:10 +0100
This bug has been known for a long time. The reason for it is that there is no way (at least that we know of) to determine whether the data file is 16-bit or 32-bit. This needs to be specified manually by the user. Perhaps it's worth checking whether there is now any code around that can solve it, but if not that's just how it is.
Boris Reuderink - 2012-01-03 12:23:27 +0100
Assigned to Vladimir since he confirmed that it is a known bug.
Boris Reuderink - 2012-01-03 12:23:49 +0100
Changed status to resolved:wontfix.