Leginon

Could you tell me if the bright points are at consistent positions or not ? And also also the typical count of the bright points ?

Yes the positions seem to be consistent (See image attached) As far as the average values they seem to vary as seen by a max of 2400 for one and a max of 500 for another dark image.

If I understand correctly, these are images taken in normal imaging mode but with column valve closed or no beam, not the Dark image you get in Correction Node, right ?

If so, this means you will want to add these points as bad pixels.

No these are the dark images taken in the correction node with our K2.

Please check the settings of Correction node to make sure that the right camera is selected. I have in the code to modify dark images to make them 0 so it should never show up like that.

The modification only take effect, however, if "Digital Camera" selection is on "GatanK2Super" not "None" (default before the camera class is initiated which takes a little time). It is one of those gui bug impossible to fix. The safest thing to do is to go into Correction Node and deselect and select the right camera.

I redid the dark reference making sure to select GatanK2Super and still get the non zero image.

Did you have DM 2.30 ? If not, make sure your dmsem.cfg corresponds.

On K2 computer what do you get when running this python script in python command window ?

from pyscope import dmsem
k = dmsem.GatanK2Super()
k.isDM231orUp()

The modification is only done for DM 2.31 and above.

We have 2.32

When I run it I get
False

Maybe there is some customized code at your site to keep 2.2 alive on the newer versions of Leginon.

Please give me the results of these. In the mean time, you can force a "return True" on isDM231orUp function if you are in a hurry.

from pyscope import dmsem
k = dmsem.GtanK2Super()
k.getDmsemConfig('dm','dm_version')
k.camera.GetDMVersion()
k.getDMVersion()

The three corresponds to different level of DM version number that we can pull out. First one from dmsem.cfg which you configure, second through SEMCCD.dll, and the third is the conversion result.

Ok I modified dmsem.cfg to the correct version and now get the proper dark image. Should I redo the brightimage also?

I changed dmsem.cfg before I got your last message. Do you still want the results of your script above?

There are a lot of changes after 2.31. Yes, I think that would be a good idea.

No. Since it is not a bug.

Results
k.getDmsemConfig('dm','dm_version')
2.32.888.0
k.camera.GetDMVersion()
40302
k.getDMVersion()
(40302, '2.3.2')

We are anticipating upgrading DM to 3 soon to be consistent with the F20. Will I need to change dmsem.cfg when I do that as well?

Yes, but the change in behavior is the same as 2.3.2. Thanks for the script results.