First, ASCOM isn't compatible between Gemini I and Gemini II and I took lot of times because it looked so... At the end, I started from scratch. I took some time to find out that you need to set the mount as "Titan" in the Gemini and "Titan50" in ASCOM - now figue why... :-)
I was very surprised by the improvement in GOTO accuracy. With no pointing model, I am able to go from one target to another one within 40-60° and still have it in the small guiding field of view.
That's for the Gemini II installation but because of the three months downtime, I faced several (too many!) problems:
1/ WINDOWS (win10) was updated of course, it just never stops!
2/ when updating prism, AVIRA antivirus blocked update.exe software and I could'nt restore it so my best option was to get rid of Avira and use Defender instead... but so much time lost on this issue!
3/ With the current PRISM version, the OFFSET_TELESCOPE function doesn't work anymore, not sure why but it looks like a regression!
4/ ATIK drivers were updated and it now doesn't work in prism as ArtemisHSC.dll can't be found. I am able to use Ascom which may not be a bad thing as it better recognize each camera compare to direct driver interface. But the drawback is that now starting autoguiding from the script (STARTGUIDECONSIGNE) fails all the time and I can't figure out why. My only option for now is to have a manual interuption of the script to manually launch ans stop autoguiding so my observations are now less automated!
Anyway, I am now back to observation (I lost around 15 nights), but I have to wake up between each target for manual intervention... I hope to find a solution for this STARTGUIDECONSIGNE issue; one option is to replace the Atik Titan by another non Atik camera and see how it works...
Despite the better GOTO accuracy, the new system seems better in autoguiding and I have some surprising increase in light efficiency during the first nights I tested it. Almost 30% improvement in spectra SNR - I now need to reduce my exposure times to reach the SNR=100 target. Because it is less automated I am leaving the parameters as is but when I find a solution for the autoguiding, then it should increase the number of spectra per night.
