Leginon

Can Presets Manager get the super resolution mode more consistently? If so, you can set camera geometry there with "override preset" unchecked in the calibration nodes. See Creating_Presets_for_the_first_time

I can see from the code that calibration nodes use a different method to get cameras than nodes like presetsmanager but don't have a good reason for the alternative method to fail, still.

I attach here a python script to copy the calibrations from GatanK2Counting to GatanK2Super so that you can have the calibration even if you still cannot acquire binned super resolution image.

You will need to know the hostname for the K2 camera and the high tension in volts where the calibrations were performed. Run it in the same environment where you can run Leginon.

For example:

python copycal.py your_host 200000

The output looks something like this:

StageModelCalibrationData 6.20801792e-05
Uncomment in the script the line 'newdata.insert()' to insert to database

StageModelMagCalibrationData 38 4.11600335188e-08
Uncomment in the script the line 'newdata.insert()' to insert to database

StageModelMagCalibrationData 330 5.47185336369e-09
Uncomment in the script the line 'newdata.insert()' to insert to database

MatrixCalibrationData stage position 29000 -6.64085396084e-11
Uncomment in the script the line 'newdata.insert()' to insert to database

MatrixCalibrationData image shift 29000 4.82269015766e-11
Uncomment in the script the line 'newdata.insert()' to insert to database

MatrixCalibrationData defocus 29000 -2787561870.68
Uncomment in the script the line 'newdata.insert()' to insert to database

If you get some of these and no error is produced during the run, you can uncomment the line as indicate to insert the calibration in the database.

If you want to do the same copying to 'GatanK2Linear', then replace 'GatanK2Super' with it, and change pixelsize_scale to 1.

I was able to get the full formats for super resolution this morning using the preset manager but had a problem where the calibration failed. I was unable to follow up due to scheduling conflicts but will try at some point soon with a grid with better contrast.

There is the possibility that the full size super resolution is just too big to be practical for calibration because the need for doing correlation. Please try the copycal.py I attached above. No need to redo calibration when we know the ratio of the calibration should be.

On which computer should I run the script?

It should work on any of the computer you currently use because they all have the configuration for connection to database (sinedon.cfg) defined.

An updated version of copycal.py is now in the repository renamed copy_k2_super_cal.py . See Issue #3247 for usage

I looked for copy_k2_super_cal.py in the repository and can not find it.

http://emg.nysbc.org/projects/leginon/repository/changes/dbschema/copy_k2_super_cal.py