BCI2000, Brain Vision Recorder, MATLAB Cogent Toolbox issues

[AnDi]

Hi,

I am having problems with my setup.

I am using the Brain Vision Recorder, which sends the signals to the BCI2000. I am using the RDA Client as the source module for the BCI2000.
For the presentation of my stimuli I am using the Cogent Toolbox from MATLAB. The problem is as follows: I start the Brain Vision Recorder and BCI2000, they work fine. But as soon as I start the stimuli presentation the following errors occurs:

RDAClientADC::Process: RDA block numbers not in sequence.
RDAClientADC::Process: Lost connection to VisionRecorder software.

Then the BCI2000 crashes. I have heard of similar problems, if the CPUs are totally overloaded, so I checked. This does not seem to be the case.

Thank you very much for any help or suggestions.
Anita

[mellinger]

Anita,

the errors occur when BCI2000 roundtrip time (signal processing plus stimulus presentation time) becomes longer than 40ms, which is the time interval for data blocks from the Brain Vision Recorder. When BCI2000 is blocked for more than this, it cannot receive data packets from the Recorder, and thus data is lost, which is the cause of the error message. This has nothing to do with CPU load.

Are you using Matlab's Cogent toolbox from BCI2000's Matlab Signal Processing module? If so, I guess that the toolbox is blocking BCI2000 during some operation, e.g. to synchronize with the display's vertical blank.

Best regards,
Juergen

[AnDi]

Hello Jürgen,

I am using the cogent completely separate from the BCI2000, although I am sending markers through the BCI2000 Connector.

Thank you for the help,
Anita

[mellinger]

Anita,
are you using a large number of visual stimuli, or large stimulus files, such that image data might reside in virtual memory? This could be the cause for BCI2000's timing problems.

Juergen

[AnDi]

Hello Jürgen,

I am presenting a new stimuli every 500ms - 1s, although I tried just opening a small, blank screen and even this caused the connection and thus the BCI to crash. Worst case scenario, I have thought about just running the stimuli on a separate computer. Would be nice if I could solve the problem a different way, though.

Best,
Anita

[mellinger]

Anita,

sorry that I could not help you with this problem. Have you found a solution in the meantime?

Best,
Juergen

[AnDi]

Hello Jürgen,

I have not found a solution yet, but I will keep trying. If I don't come up with one, then I will change the set up.

Thank you very much!
Anita