Skip to content

Latest commit

 

History

History
113 lines (86 loc) · 4.8 KB

2022-12-13.md

File metadata and controls

113 lines (86 loc) · 4.8 KB
Raw

2022/12/13 - Microscope development meeting

Attendees

Attending:

  • Ian Dobbie
  • David Miguel Susano Pinto
  • Julio Mateos-Langerak
  • Danail Stoychev
  • Veronika Pfannenstill

Notes

  • Both Ian and Julio have been busy and not done any work.

  • David brought up Veronika has run out of digital output signals from her Red Pitaya. She has a BNC breakout box with 8 outputs, but needs access to the other 8 GPIO lines. Danny offered to help her connect them, they will connect afterwards to arrange this.

  • It was discussed that filterwheels hardware triggering do not work with the TriggerTargetMixin which expect triggers to come from a single line. It is common to use multiple lines to set a positiion number and an additional line to trigger the move. This is the state in the Cairn OptoSpin.

  • The Microscope website at https://python-microscope.org/ no longer redirects to the version at Micron. The source is https://github.com/python-microscope/python-microscope.org and the site is built and hosted by netlify. David is using his own personal account which has a limit on CPU time spent on building the website. While it is possible to have netlify automatically rebuild the website with each commit, because there is a monthly limit on CPU time spent on building the website, David will trigger it manually. Changes can be built and tested locally using Sphinx and make.

  • Cockpit does install on Python 3.11 because of wx (cockpit#834)

    The current version of wxPython does not install on Python 3.11. Because of that, Cockpit fails to install as well.

  • Need new microscope release (cockpit#835)

    Development version of Cockpit fails to install because it is dependent on the current development version of Microscope. While it is probably time to make a release of Microscope, this is not an issue when installing the last release of Cockpit. It means we need to make a new release of Microscope before a new release of Cockpit.

  • Julio has some updates for microscope and cockpit to enable pseudo global shuttering on sCMOS cameras. This triggers the camera in bulb mode and for exposures triggers the lasers, waits for the readouttime to allow the whole chip to be active and then stops the camera and laser exposure simultaneously.

  • David chimed in that he and Veronika was having trouble with bulb mode. Veronika has a Photometrics Kinetix camera.

  • Code style in cockpit (cockpit#649)

    It was decided to implement in cockpit the same formatting style as python-microscope (black and isort). This is already applied to the microscope source code. We will try and integrate Julio's current outstanding issues and then apply Black as a single commit across the whole of the cockpit source tree.

  • git pre-commit hooks to keep style in Microscope

    David and Julio have been using git pre-commit hooks to keep the style of the project. Julio has the hook running black automatically while David has it rejecting the commit unless it passes the black style check. David's option is because he often only commits part of the changes in a file.

    It was also discussed we could use Github Actions to run it on pull requests but we do not use Pull Requests and we do not want to have a bunch of extra commits that just change the style. There's not that many of us, so we will just tell/require that contributors just use something to ensure the style.

  • Danny has a number of outstanding updates. He expects to be able to spend more time on merging them in the coming months.

** Minor DM fixes for microscope that include the ability clear the queue and load a new pattern.

** Minor general fixes in cockpit, including defining camera ROI with the mouse in the camera window.

** Major changes in ao-tools, Danny will start pushing these changes to the ao-tools main branch in the new year.

  • Ian suggested that we move ao-tools to the same code style as Microscope and Cockpit.

  • David asked what exactly the camera cycle time referred to. Ian said this had been discussed at length previously and suggested that it should refer to the minimum time between camera triggers. Complicated by sCMOS camera can actually overlap exposures with the rolling shutter. This issue was previously discussed at microscope-cockpit/cockpit#522

  • Julio said his laser takes an appreciable delay in fall time of the laser. Ian commented that there is already a timing variable that should be applied to all digital trigger devices you just need to define it and it will add a delay with that particular device. Ian will raise an issue on cockpit github.

  • Ian mentioned running the Pi Pico as a timing device. Julio says he has a demo implementation reading an actiontable and flashing the Pico LED. There is already a cockpit issue for this.