bci2000.org BBS

I downloaded the latest stable source code (version 4230) via the SVN versioning software. Compilation in 32 bit mode with visual studio 2008 is flawless. (Included all component modules in the build). However on starting up the BCI2000 Launcher and choosing Biosemi2, DummySignalProcessing and Feedback demo, with appropriate parameters, I get the error:

Preflight::Biosemi2ADC Could not connect ...

Note that I am using the Biosemi the 64 bit USB driver for the Biosemi ADC, downloadable from the biosemi website. I thought this may be an issue with incompatible drivers (64 bit biosemi USB driver). I then tried the build on Visual studio 10 in 64 bit mode after installing the Qt libraries and following the instructions at:
http://www.bci2000.org/wiki/index.php/P ... _for_Win64.
But then the biosemi module does not compile.

Has anyone encountered this problem with windows 7? Is it possible at all, currently, to use biosemi with BCI200 on 64 bit WIndows 7 ?

Thank you.
Sandeep.

PS: A also tried the LABVIEW_DLL.dll that comes with the latest Actiview software (v7.01) on biosemi's website. But the error is the same.

Hi,

I do not think you can use a 32 bit app and a 64-bit driver. Have you tried to use their 32-bit driver?

Gerv

That didn't occur to me since I was using Actiview 7+ which is required for windows 7 and which requires their 64 bit driver. But yes I should try it out since if won't really need Actiview if I use BCI2000. I was going to fall back to windows XP, but I should try this first. Thank you.

Regards,
Sandeep.

I confirm that 32bit drivers work on Win7. I have a 64bit machine, but I was unable to connect to to BioSemi even through Actiview before I started using 32bit driver which fixed all of the problems.
Note that there are quite a few of different 32bit dll drivers circulating around so I had to try out two - three before it started working.

Thanks for the confirmation.

I am able to use Actiview 7.01 (the so called "vari" one - featuring a scrollbar that mitigates the fixed resolution issue) on windows 64 bit 7 with the 64 bit USB drivers from Biosemi's website. Versions of Actiview prior to 7.0 will not work on 64 bit windows 7. Check the downloads section in Biosemi's website.

Regards,
Sandeep.

On windows XP, with the 32 bit USB driver, we ran the BCI2000 launcher with:
Biosem2 + DummySignalProcessing + FeedbackDemo
along with the parameters:
SourceCh: 16
SampleBlockSize: 16
SamplingRate: 256
SourceChOffset: 0.03125 (x 16)
ChannelNames A1 A2 ... A16
EEGChList: 1 2 3 ... 16
TriggerChList: [Empty]
TransmitChList: 1 2 3 ... 16

with other parameters being left at their defaults. The following (sole) exact error is obtained:
DataIOFilter::Preflight: Biosemi2ADC::Preflight: Could not connect to device.
The error is almost instantaneous.

We notice that there are parameters such as:
SignalSourcePort, SignalSourceIP, ConnectorInputAddress ... etc.
Not sure what these mean, but do they have to be filled in?

Any ideas?
Thank you.

Regards,
Sandeep.

I had the same error so I followed the instruction in the readme file by the plugin creator:

If errors regarding “Cannot connect to the device” or “No Cable or Power” occur, please replace /prog/Labview_DLL.dll with the one provided from BioSemi (http://www.biosemi.com/download/USB%20d ... 032bit.zip)

//Petar

Much obliged for that info. Did it and now am able to connect to the ADC for the very first time! Thanks again. Now to get the stimulus presentation codes synced with the signal. I'll post progress here.

Best Regards,
Sandeep.

Update:
The above post was for windows XP.
Still trying to install the 32 bit USB drivers on windows 7. Having some trouble doing this.

pjercic wrote:I confirm that 32bit drivers work on Win7.
...
Note that there are quite a few of different 32bit dll drivers circulating around so I had to try out two - three before it started working.

Just noticed something. Did you mean the 32 bit USB drivers or were you talking about the LabView_DLL.dll files ?
I'm not able to install USB driver Windows (32bit) from Biosemi's website. This is the only place I know of to get the 32 bit USB drivers. Are there any others or was there a misunderstanding here.

Best Regards,
Sandeep.

I finally got the windows 7 system to work.
I used the following:
* Uninstalled the 32 bit USB drivers and reinstalled the original 64 bit USB drivers from the Biosemi link USB driver Windows (64bit)
* Replaced the LabView_DLL.dll file in the BCI2000 build with the LabView_DLL.dll from the Actiview 7.01 folder from http://www.biosemi.com/download/LatestA ... 1-Vari.zip. This is a beta version for windows 7 64 bit.

I just connected electrodes A1, A2, CMS and DRL with gel and for the moment, just to check that there is some sort of signal. The other electrodes A3 - A16 are just plugged in with no gel.

There does appear to be some sort of signal if I use the "Auto-scale" option in the signal window, but it is at 200000 muV and very small. So I'm not sure if I'm getting any real signal. I see numbers with a variation of about 20 from A1 and A2 while the variation in A3 - A6 is of the order of just 1 or 2 units.
If I use any other scaling (10 - 3000 muV), I don't see any signal at all. So I don't follow what is going on. Shouldn't I be seeing some signal at very low voltages?

But at least the Biosemi2 module seems to be working. The question is what signals am I getting.

Best Regards,
Sandeep.

A signal of 200000 muV is very large rather than very small.

Use the Highpass option in the display context menu in order to suppress large voltage offsets.

Rather than connecting un-gelled electrodes to inputs, connect these inputs to ground using either cables, or putting all the electrodes into a glass of undestilled water, together with a ground electrode.

Regards,
Juergen

Yes, that is what is confusing. Using AutoScale (200000 muV) I can see a (very) small signal. But at a scaling level anwhere from 10 to 3000 muV, I see nothing at all - just a flat line.

Best Regards,
Sandeep

Right.
Use the Highpass option in the display context menu in order to suppress large voltage offsets.

-Juergen

Thanks. Works beautifully now.

I guess this means that there were some very large amplitude low frequency oscillations which were simply "off the scale". ?
[edit: OK you said voltage offsets - which counts as a 0 frequency oscillation.]

Now to figure out the Stimulus presentation.

Best Regards,
Sandeep.