# Calibration Issues **URL:** \ **Category:** Support\ **Created:** [April 2, 2018, 8:45pm UTC](https://mworks.discourse.group/t/calibration-issues/340 "2018-04-02T20:45:16Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![cstawarz](https://yyz1.discourse-cdn.com/flex031/user_avatar/mworks.discourse.group/cstawarz/32/3_2.png) [@cstawarz](https://mworks.discourse.group/u/cstawarz)\ **Post date:** [April 2, 2018, 8:45pm UTC](https://mworks.discourse.group/t/calibration-issues/340/1 "2018-04-02T20:45:16Z")
Hi Chris, I believe Simon may have emailed you about this some time ago, but I’m having issues with getting a tight calibration with the calibration protocol. Consistently, the top left, middle, and bottom right calibration points are off by about 2-3 degrees in a consistent direction for each point for every calibration the monkey completes. This has made it so that I can’t lower the fixation window below 5 degrees - otherwise the monkey gets consistent errors and becomes frustrated. All of the other points are spot-on, which leads me to believe the monkey is in fact looking at the points and it is not an issue with the eye tracker. This issue isn’t fixed by implementing a clear calibration or stopping and restarting the protocol. I very much need to be able to get a fixation window of 2-3 degrees because I am working in V1. I’m not sure what could be causing this. Any thoughts? Thank you, Yvonne ---
**Author:** ![cstawarz](https://yyz1.discourse-cdn.com/flex031/user_avatar/mworks.discourse.group/cstawarz/32/3_2.png) [@cstawarz](https://mworks.discourse.group/u/cstawarz)\ **Post date:** [April 3, 2018, 2:09pm UTC](https://mworks.discourse.group/t/calibration-issues/340/2 "2018-04-03T14:09:25Z")
Hi Yvonne, Can you send me an [event file](https://mworks.github.io/documentation/0.8/guide/running.html#opening-the-event-file) from one of your calibration runs? It would be helpful to see the actual numbers that the calibrator is producing. Also, have you tried using the [linear eye calibrator](https://mworks.github.io/documentation/0.8/components/linear_eye_calibrator.html) (as opposed to the [standard eye calibrator](https://mworks.github.io/documentation/0.8/components/eye_calibrator.html), which uses second-order fit functions)? That might be more appropriate for calibrating digital EyeLink data. Thanks, Chris ---
**Author:** ![cstawarz](https://yyz1.discourse-cdn.com/flex031/user_avatar/mworks.discourse.group/cstawarz/32/3_2.png) [@cstawarz](https://mworks.discourse.group/u/cstawarz)\ **Post date:** [April 3, 2018, 5:50pm UTC](https://mworks.discourse.group/t/calibration-issues/340/3 "2018-04-03T17:50:36Z")
Hi Chris, Switching to the linear eye calibrator seemed to fix the issue! I wasn’t aware it might be better for digital EyeLink data. Thank you for suggesting it. Best, Yvonne ---
**Author:** ![cstawarz](https://yyz1.discourse-cdn.com/flex031/user_avatar/mworks.discourse.group/cstawarz/32/3_2.png) [@cstawarz](https://mworks.discourse.group/u/cstawarz)\ **Post date:** [April 4, 2018, 2:44pm UTC](https://mworks.discourse.group/t/calibration-issues/340/4 "2018-04-04T14:44:14Z")
Hi Yvonne, > Switching to the linear eye calibrator seemed to fix the issue! I wasn’t > aware it might be better for digital EyeLink data. It was just a guess, based on my assumption that the transformation from EyeLink “pupil” coordinates to MWorks eye coordinates is linear. It’s not clear to me why a second-order fit wouldn’t also work. In any case, I’m glad that switching solved your problem! Chris ---
**Author:** ![cstawarz](https://yyz1.discourse-cdn.com/flex031/user_avatar/mworks.discourse.group/cstawarz/32/3_2.png) [@cstawarz](https://mworks.discourse.group/u/cstawarz)\ **Post date:** [July 19, 2022, 9:48pm UTC](https://mworks.discourse.group/t/calibration-issues/340/5 "2022-07-19T21:48:03Z")