Problems while measuring on Windows 7, but not Windows 10.

zfmengzfmeng Japan
edited July 2019 in OpenBCI_GUI
Hi~

I did several tests these days with Cyton board.
But there were some tesing problems occuring (powering off and on) while measuring, as the picture shows.


When I start data stream, the signal stream is not consequently but unstable.
Many interrupts appeared just like the picture shows. I didn't find the causes.

Besides, I downloaded the OpenBCI GUI 4.1.2 on the windows 7.
It seems not run properly, and could not collect data.
Is it related with the operate system ?
Now I run the GUI on Windows 10.

Very looking forward to hearing from you.
Thank you very much indeed.

Comments

  • retiututretiutut Louisiana, USA
    edited June 2019


    System Requirements

    Hardware

    1.6 GHz or faster processor 2 GB of RAM 400 MB of hard drive space (minimum)

    Platforms

    OpenBCI GUI has been tested on the following platforms:

    • OS X 10.8.5 or later
    • Windows 8.1 and 10 (32-bit and 64-bit)
    • Linux Ubuntu Desktop 18

    OpenGL acceleration is required.



  • wjcroftwjcroft Mount Shasta, CA
    edited June 2019
    The GUI runs fine on Windows 7 (64 bit) as well. Although staff may not test it there on every release. I have Window 7 64 here. Many commercial and educational labs are still running Windows 7. My guess is that there are more Windows 7 and 10 installations than Windows 8 out in the field. On the other hand 32 bit Windows 7 systems are getting much more rare since their base hardware is essentially obsolete.



  • zfmengzfmeng Japan

    Thank you for your reply.

    I checked with my computer. 
    There's no problem with the hardware.
    It's 1.8GHz and 2.2GHz.
    The RAM is 16GB (15.8GB available), and 16GB (16GB available).

    Is the measuring problem related with the movements of the subjects ?

    Thank you very much indeed.
  • wjcroftwjcroft Mount Shasta, CA
    I don't understand which OS are you using? Does it work on 10 and not 7?

    All EEG will show artifacts when the subject is moving, talking, blinking, etc.
  • zfmengzfmeng Japan

    Thank you for your reply. Sorry for my poor explaining.

    Now I'm using Windows 10 with 1.8GHz, and 16GB (15.8GB available), but there are always problems appearing like sudden restart while testing.

    So I want to change to another computer which is Windows 7 with 2.2 GHz, and 16GB (16GB available), but it couldnot measure.

    Thank you.
  • wjcroftwjcroft Mount Shasta, CA
    Has this Cyton EVER worked? For example when you first received it. Did it work on ANY computer?
  • wjcroftwjcroft Mount Shasta, CA
    What happens when you perform all the steps of the tutorial?

  • zfmengzfmeng Japan
    Thank you for your patience.

    Yes. I followed the steps as the link.
    It could work well on my computer (Windows 10, 1.8GHz, 16GB RAM).
    I guess maybe the electrodes didn't connect stably all the time while testing so that resart current appeared.

    Thank you very much.
  • wjcroftwjcroft Mount Shasta, CA
    If it is not working as well on Windows 7 as Windows 10:

    * be sure you have the latest FTDI driver on W 7
    * try running the GUI from Processing IDE, instead of the standalone
    * be sure to adjust FTDI device settings to avoid buffering delays
    * on W 7, ensure the machine is relatively lightly loaded before starting the GUI. If you have many background processes using CPU, it could interfere with the realtime processing that the GUI and Hub need to do.
    * go through the tutorial with the W 7 machine and report here what is failing.

    Regards,

  • wjcroftwjcroft Mount Shasta, CA
    As Richard mentioned, your Windows 7 machine must support OpenGL. Sometimes older laptops have obsolete OpenGL drivers. If you are getting a blank or black screen with the GUI, visit your manufacturer's Support page and download the latest graphics chip driver(s).
  • zfmengzfmeng Japan
    Thank you.

    I'd like to try as you explained.


  • zfmengzfmeng Japan
    Hi~

    There is still problem during measuring, suddern powering off and on during measuring process.
     I cannot find the reason.


    Thank you.

  • retiututretiutut Louisiana, USA
    edited July 2019
    @zfmeng I don't see a problem with the photo you have shared. If an error is happening, it will show up in the Console Log.

    You can access the Console Log by clicking the button in the top right of the OpenBCI_GUI. Then, you can share the text file here in the forums.
  • zfmengzfmeng Japan
    Thank you.

    The normal measuring page is like this.

    However, it changed to zero like restart suddenly. Is that poor contact ?
  • retiututretiutut Louisiana, USA
    edited July 2019
    Side by side comparison:
    https://imgur.com/a/tBtSO9K

    First thing I notice is that there are no artifacts in channel 4, which seems to be ECG data, as I see PQRST.

    There are some artifacts in channels 1, 2, 6, and 7. @wjcroft What do you think/see?

    The channels are not turning off an on, or losing connection. The time series data "Auto" adjusts the Vertical Scale based on the data.
  • wjcroftwjcroft Mount Shasta, CA
    The first screenshot (channels 1,2,6,7) shows high frequency noise bursts. Since you have bandpass 5-50 filter, my guess is that this noise is even higher than mains frequencies. The regular ripples in the FFT looks like some kind of harmonics.

    Be sure to position the subject and cyton, cables, away from any potential sources of EMF energy.
  • zfmengzfmeng Japan
    edited July 2019
    Thank you for your explaination.

    About the EMF energy, does the mobile phone or computer matter ?

    And how about this problem like suddern restart which appeard in my experiment all the time ?

    Thank you very much indeed.


  • wjcroftwjcroft Mount Shasta, CA
    Richard @retiutut may comment, but I think your 'restart' is combination of two things: (1) a packet was lost, possibly due to EMF interference. (2) the GUI DSP bandpass filter goes wonky with the supposed "all zeros" sample inserted in the stream. I think Richard may have a fix for this already that substitutes an interpolated sample for the missing one.

    More importantly, you want to reduce radio interference. Either by changing your Cyton radio channel to another one, and/or moving away from possible EMF sources (Bluetooth, Wifi, mains power supplies, etc.) Another hack you can use is put the dongle on a usb extension cord, then position the dongle itself as closely as you can to the Cyton.
  • wjcroftwjcroft Mount Shasta, CA
    The recording file does not contain this glitch, only a missing sample. By turning off the GUI filtering, the glitch should disappear.
  • zfmengzfmeng Japan
    Thank you very much. I understand.

    How to change the Cyton radio channel to another one ? 
  • wjcroftwjcroft Mount Shasta, CA
    Click the right arrow,

  • zfmengzfmeng Japan
    Does it mean to reset the devices ?
  • wjcroftwjcroft Mount Shasta, CA
    Read the section on Optional Settings. You can change the channel to whatever you like.

    It's probably doubtful that channel changing will solve your lost packet issue. Try the usb extension cable idea.
Sign In or Register to comment.

© Vanilla Keystone Theme 2026