Attending:
- IMD
- Tom Fish
- Julio
- Mihaela Duta
- David
- Galvo scanner
Started with a discussion of Mihaela plan to implement a galvo scanner as camera module within Microscope.
Ian started saying that probably this is relatively simple but one critical factor is the pixel size. The infrastructure has no current knowledge of any meta data associated with the a camera device except image bit depth (8, or 16 bit) and a timestamp.
David suggested implementing a meta data dictionary, although exact contents are to be determined.
Apparently there is already python code that can run the galvo scanner and return an array of intensity data, this should be quite easy to interface to as a camera device.
- Documentation for devices
David mentioned that the documentation is rather thin as to what methods are required for each device class. Ian mentioned that it was in the abc.py file. David comment that this mixes required and default implementations of some classes, which is true.
- Timer code
Discussion moved on the the timer issues that crash cockpit on exit in different ways for different OS's. Ian said he thought that this was almost entirely fixed with the wx.Timer fixes in the gui/safeControls.py module. David says not as there is further code in the base device class addWatch function but no deleteWatch see issue microscope-cockpit/cockpit#828
There is also an outstanding issue in the lightPower handler with the _updater thread see issue microscope-cockpit/cockpit#865
There was some discussion about the multiple timer techniques used and it was suggested that we replace all variants with the wx.Timer as closely linked to the safecontrol approach as possible to unify the solutions.
There are also examples of wx.Timer use in the Aurox and boulderSLM device code that probably doesn't terminate properly but hard to test with no hardware available.
- Linkam Stage on Linux
David asked about the Linkham stage on Linux issue that has raised its head a few times. Apparently there is lots of code that is windows specific or even only works effectively by chance. The driver seems to work well on windows but the Linux version has lots of issues.
- Vortran laser and Thorlabs stage
Veronika is having problems with the implementation of the Vortran laser device and the thorlab stage she has. The thorlab stage had a serious bug that was fixed but only after months of delay. David doesn't have time to help her at present.
- Training notebooks
Julio is attending MiFiBio in the Autumn. He is thinking about setting up training examples in something like docker, or a binder instance. David shared a link tot he Napari example code to possibly include in a new training materials github repository.
- Stepper driven stage on Raspberry Pi
Julio has a colleague who is running a stepper motor driven stag via a raspberry Pi Hat. Ian asked for info about the Hat. In theory this should be relatively easy to add to microscope.
- ASI stage
Julio also wants to use an ASI RAMM stage that he hope to borrow from ASI. Ian says that he has a driver that mostly works but has some weird problems.
- No meeting in August so next meeting Sept 12th