[open] RuntimeError: Unknown Command

ckoops

--- UPDATED 8/5/15 ---

We are running OpenSesame 2.9.6 on a Windows 7 PC and are setting it up to interface with an Eyelink II tracker. We're still in the early stage of setting up the system, and when running OpenSesame's sample Eyelink experiment, we get this error:

Unexpected error
line: 230
exception message: Unknown Command
exception type: RuntimeError

Traceback (also in debug window):
  File "dist\libqtopensesame\misc\process.py", line 139, in run
  File "dist\libopensesame\experiment.py", line 347, in run
  File "dist\libopensesame\sequence.py", line 47, in run
  File "C:\Program Files (x86)\OpenSesame\plugins\pygaze_init\pygaze_init.py", line 198, in run
  File "dist\pygaze\eyetracker.py", line 72, in __init__
  File "dist\pygaze\_eyetracker\libeyelink.py", line 176, in __init__
  File "dist\pygaze\_eyetracker\libeyelink.py", line 230, in send_command
RuntimeError: Unknown Command

Do you have any advice on what we could try next to troubleshoot the connection?

We have the Eyelink software running on the display PC which is connected via Ethernet to our PC running OpenSesame. The connection works, as far as we have been able to see. On the OpenSesame PC, we've installed PyLink 27 as explained here: http://osdoc.cogsci.nl/2.9/devices/eyelink/#installing-pylink

Chris

eduard

Hey,

Which "eyelink experiment" do you mean? The "Pygaze eye tracking template"?
Does the error also occur, when you set the the tracking mode to one of the dummies in the pygaze_init item?

Eduard

ckoops

Hi Eduard,

thanks, yes, we're using the "Pygaze eye tracking template".

and also yes, we don't get the error when we use either the simple dummy or the advanced dummy. We've been using the EyeLink one. That's when we get the Runtime error message.

Ps- We updated to v. 2.9.7 today, but the error persists

sebastiaan

Hi guys,

It's probably the force-drift correct option, which sends a command that only the EyeLink 1000 recognizes. This issue is already listed:

• https://github.com/esdalmaijer/PyGaze/issues/36

For now, what you could do is simply disable this option, and you should be good to go.

Cheers!
Sebastiaan

ckoops

Hi Sebastiaan,

thanks for pointing us to the earlier thread. Yes, our error message is identical to the one discussed there. We've tried it again, this time with 'Force drift correction (for EyeLink 1000)' disabled, but the error remains exactly the same.

Do you have any other thoughts on how we can troubleshoot this?

-Chris

sebastiaan

Hi Chris,

Nevertheless, it's that error--I can tell by the stacktrace in your first post. What seems to be going on is that this option is enabled regardless of whether the box is ticked. A bug on top of a bug, in other words. What happens if you edit the script, and set eyelink_force_drift_correct to 0 manually:

set eyelink_force_drift_correct 0

Make sure you don't use the controls to change this afterwards, because it will set this variable back to 'no'.

Cheers!
Sebastiaan

ckoops

Hi Sebastian,

with these steps, we were able to get the experiment to run once, right after starting up the whole system and OpenSesame, but only once. After that, it's back to the same error, even though the script line still has "0" (and of course the box is unchecked).

Any ideas on how we could make it work permanently?

Thanks -Chris

sebastiaan

Hi Chris,

As soon as the plug-in controls get applied, the script is regenerated and your changes to the script are lost. So what you could do for now is change the script, save the experiment, and then run it using opensesamerun (which won't change anything in the script):

• http://osdoc.cogsci.nl/usage/opensesamerun/

This is a pretty annoying problem for EyeLink II users. (And you guys have it difficult enough, with the painful helmet and everything.) I'll commit a small fix to PyGaze next week and upload a repack of OpenSesame 2.9.7 (i.e. the same version of OpenSesame repackaged with an updated PyGaze).

Cheers,
Sebastiaan