> Could be i'm missing something. I've never used an Eyelink, so despite being involved in implementation of the callback mechanism, i don't know how a real Eyelink actually behaves wrt. the callback. I only know how our "Eyelink simulation" test code behaves (EyelinkGetTrackerImageDemo run in dummy mode, with the Eyelink('TestSuite') in use).
>
> The reason for introduction of the callback was to allow reception and display of eye images from the trackers camera, something only possible via the callback and it was the most requested feature at that time.
>
> I assumed that the callback gets called very frequently to update the eye camera image, so the callback would get called often and regularly enough to do things like animated calibration targets? But maybe this doesn't happen or not in all setup modes?
Yes, this is what I had experienced. On our production machine with Eyelink connected the callback code is only called every 100ms (or possibly every 10th frame as there is some very small jitter?) if the camera image is NOT displayed (what is the default during experiments in our case; and about every 50ms with more jitter if the camera image is displayed). This rate (10Hz) was too low for smooth animation.
> The calibration target drawing routine is not reached in PsychEyelinkdDispatchCallback if the Eyelink only queries for key press and a few other cases, as that doesn't set needsupdate = 1.
Yes, I have measured timing before the needsupdate/return part and also tried setting needsupdate=1 for key press checks in the switch part.
I assume callback rate could be changed in the Eyelink c/mex code. However, there is actually no need as we have a working setup using the old code path.
Best,
Andreas
> That's something one could check, if the callback gets called too infrequently or if it just return; s prematurely in some cases.
>
> Anyway, if it works old style and you don't need the video feed that's also fine.
> -mario
>
>
>
>
>
>
>>> -mario
>>>
>>> --- In psychtoolbox@yahoogroups.com, Andreas Widmann <widmann@> wrote:
>>>>
>>>> Hi,
>>>>
>>>> We were running into the same problem drawing animated calibration targets.
>>>>
>>>> Yes, by default EyelinkDoTrackerSetup calls the callback function el.callback (usually PsychEyelinkDispatchCallback), which has its own local EyelinkDrawCalibrationTarget function.
>>>>
>>>> You can either empty the el.callback = [] field before calling EyelinkDoTrackerSetup using the old codepath (see lines 26 to 41 in EyelinkDoTrackerSetup) or adjust the local EyelinkDrawCalibrationTarget function at the end of PsychEyelinkDispatchCallback.m to your needs. The latter will not work with animated targets but should work for static targets.
>>>>
>>>> Hope this helps,
>>>> Andreas
>>>>
>>>> Am 28.08.2013 um 18:04 schrieb C Ludwig <c.ludwig@>:
>>>>
>>>>>
>>>>>
>>>>> Hello,
>>>>>
>>>>> I have just upgraded my lab PC, switched to Linux and the latest and greatest version of PTB 3.0.11.
>>>>>
>>>>> I always draw my own calibration target for use with the Eyelink system (I prefer a '+' over the standard oval). As usual, I have modified EyelinkDrawCalTarget.m and when I call this function directly it does, indeed, draw a + as intended. However, the standard calibration procedure that you enter with EyelinkDoTrackerSetup persists in drawing the oval calibration target. So now I'm thinking that the Eyelink calibration routine no longer relies on EyelinkDrawCalTarget in order to draw the targets.
>>>>>
>>>>> Is this correct? If so, how do I change the nature of the calibration target?
>>>>>
>>>>> Many thanks,
>>>>> Cas
>>>>>
>>>>> --
>>>>> Casimir Ludwig
>>>>> University of Bristol
>>>>> School of Experimental Psychology
>>>>> 12a Priory Road
>>>>> Bristol BS8 1TU
>>>>> UK
>>>>> Tel. +44 (0)117 3317251
>>>>> Fax +44 (0)117 9288588
>>>>> Skype pscjhl
>>>>> http://casimir.psy.bris.ac.uk
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>>
>>> ------------------------------------
>>>
>>> Post your message to: psychtoolbox@yahoogroups.com
>>> Please indicate OS9, OSX, or WIN version, and include your full name.
>>> Denis Pelli, David Brainard, and Allen Ingling.
>>> http://psychtoolbox.org
>>> Yahoo! Groups Links
>>>
>>>
>>>
>>
>
>
>
>
> ------------------------------------
>
> Post your message to: psychtoolbox@yahoogroups.com
> Please indicate OS9, OSX, or WIN version, and include your full name.
> Denis Pelli, David Brainard, and Allen Ingling.
> http://psychtoolbox.org
> Yahoo! Groups Links
>
>
>