Calendar Latest Topics
 
 
 


Reply
  Author   Comment   Page 2 of 2      Prev   1   2
Jaewon

Administrator
Registered:
Posts: 971
Reply with quote  #26 
Hi niemeyej,

I remember the issue. That is why I asked the version number of your toolbox.

I examined the toolbox package (May 19, 2016 version) included in the GitHub ML and found a problem. The toolbox is not as I distributed and was repackaged before being added to the repository. During the process, the daqmex files of the previous version were included, without being updated. Unfortunately I don't manage the GitHub repository and cannot fix it from there. I re-link the May 19 version, so please download it and update your toolbox.

ftp://helix.nih.gov/lsn/monkeylogic/NIMH_daqtoolbox_(May-19-2016).zip

I removed the package from FTP because I thought the GitHub package was updated and I don't distribute the toolbox as a separate package any more. A newer version is included in the current NIMH MonkeyLogic, but the data structure has changed slightly, so the GitHub ML needs some modifications to work with the latest NIMH DAQ Toolbox.
0
niemeyej

Junior Member
Registered:
Posts: 18
Reply with quote  #27 
Ok, that solved the errors.  However, now I'm getting cycle rates between 60-70Hz.  I tried R2014a, 2013a, and 2013b, and the rates are similarly low for each of those matlab versions.  I saw another thread in which someone determined that R2014 might have some issue, but that user said that the problem disappeared when they switched to a different matlab version.

Is this another known issue?

Just to recap: I downloaded the latest version of monkeylogic linked in github, then merged over your above-linked NIMH-daqtoolbox.  Everything seems functional now, but just slow.
0
Jaewon

Administrator
Registered:
Posts: 971
Reply with quote  #28 
You may want to check the following thread, although there isn't a solution posted.

http://forums.monkeylogic.org/post/touch-screens-7491343?&trail=100

I provided NIMH DAQ toolbox but I don't manage the GitHub version. Maybe you can test my NIMH MonkeyLogic and see how it does, although I am still working on it for a performance improvement.
0
niemeyej

Junior Member
Registered:
Posts: 18
Reply with quote  #29 
Ok I've tried our your NIMH Monkeylogic with several matlab versions now.  The performance is a lot better, but still below 1kHz.  My cycle rates are coming out to about 700-750Hz.  This happened with 32-bit versions of R2014a, 2013b, 2013a, 2012a, and 64-bit R2013b (64-bit was actually about 670Hz).

I'm not really sure what to try next


edit: attaching image of from Monkeylogic Latency test with NIMH-monkeylogic
R2013b_64bit_latencyTest_NIMH_ML.JPG 
0
Jaewon

Administrator
Registered:
Posts: 971
Reply with quote  #30 
Can you show me your timing file? I can't find the bottleneck, unless I run the script myself. Thanks.
0
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.