Ensure FLI is not selected as primary camera for solve-field timeout problems

[lspitler]

No description provided.

[wtgee]

It might make sense for us to pick a primary camera permanently. Probably one in the center. The concept of "primary" is not as important for Huntsman so that might make it easier.

[lspitler]

Good idea @wtgee ! Funnily enough we'd likely move the FLIs eventually to the centre to minimise vingetting.

[wtgee]

The primary camera is mostly used for plate-solving so PANOPTES can do a simple closed-loop slew and to make a pretty picture. Not much use beyond that. We could strip out the plate-solving (or make it a separate background process) and this would no longer be an issue. We have vague plans to do that for PANOPTES anyway (especially my one at MQ as the Pi can't solve their either).

[AnthonyHorton]

Because of the wider field of view it makes sense to put the FLI cameras in the middle of the array, to reduce the change of vignetting by the dome shutter. @lee that's where they are already, I moved them last run! So if we don't want the FLI to be primary then probably need to pick one that's not in the centre.

I think Huntsman should have a defined primary camera, not so much for regular operation but to define the optical axis of the array for things like pointing model calibration and co-aligning the cameras.

Definitely think that we should make plate solving a background process as with Huntsman we're only really using it for the pretty picture, and it would be OK if that turns up a minute or two after the next exposure starts.

[wtgee]

Note that panoptes/POCS#612 will affect the primary camera setting some.

[wtgee]

Closing as stale.