Last night I managed to quickly test auto-centring with the latest SGP beta 2.5.2.1 and the latest 10 Micron beta firmware 2.14.4 (just noticed a new update 2.14.6 this morning). All worked as expected.
As a recap: I had found with the previous SGP beta and my new W10 pc using 2.14.4 10 Micron firmare that SGP had froze and the 10 Micron firmware had reported 'lock clutches' and stopped tracking/slewing during auto-centring. Only a re-boot of the PC would free the software. All had worked fine with these versions on my W7 laptop before it died. Logs for this are on dropbox here (I hope I have the correct logs Ken/Jared):
https://dl.dropboxusercontent.com/u/32050952/sg_logfile_20160909192242.txt
https://dl.dropboxusercontent.com/u/32050952/sg_logfile_20160910230611.txt
https://dl.dropboxusercontent.com/u/32050952/sg_logfile_20160910230838.txt
Last night I slewed to my target, auto-focused and then did an auto-centre. All worked without freezing and was succesfully concluded and I then captured a sub before pausing/quiting the sequence and closing down with cloud. I do not think that the mount slewed after the plate solving position as my goto slew was accurate enough and within the tolerance (but I'm not absolutley sure, not paying sufficient attention to the position read out, sorry). Nevertheless, all worked as expected. Log here:
https://dl.dropboxusercontent.com/u/32050952/sg_logfile_20160930215326.txt
I will also post the mount log to 10 Micron as well as carrying out another test of this when I'm next out (possibly tonight if weather holds) but from a mount position outside of my auto-centring tolerance (20 px) to generate a noticeable slew.
My result is contrary to my expectation from reading Gunny's post however I hope the log posts help. I have a QSI ccd and Gunny a SBIG, W7 vs W10 pc but don't know if these differences are of any significance.
HTH
Barry