Hi Wade, peace.. just trying to understand. Like I said in my original post, it may be that I'm just seeking something that isn't really practical or possible. I have a full guiding setup, but when I get these posts using encoder mounts or my mount, showing incredibly flat controlled tracking, I'm trying to see if I can achieve this as well.
I see the same things in forums, on claims of incredibly precise polar alignment, and as I said in the previous post, I don't know how they can make such claims. I spent an hour one night polar drift alignment and had better alignment than usual, but it took an hour, and it still wasn't good enough for very long subs. Long subs for me, with a CMOS camera, are still under 10 minutes or less. I've done 20 minutes, but it's unusual. Most of my subs are 1 minute (luminosity). Although, having purchased some chroma 3nm filters, those seem to take some really long exposures to get a good amount of signal!
I measure drift in PHD2, using their drift alignment tool.. which takes forever to settle.. so it's a long process (for me). Usually when I'm doing polar alignment, I'm set up with camera, filter wheel, etc.. and want to do something that doesn't require me to take it apart and put in an eyepiece. And PHD2 also gives me an estimate of my polar alignment error, which is based on drift, not instantaneous bouncing around the mean which I understand is all atmosphere.
I've only done one model, a 35 point small model, and I think it took around 25 minutes (it's been a while since clear skies). And most of the plate solves were pretty fast. I noticed the meridian points, solved twice, and most others once, so it does more than 35 solves for 35 points. I use SGPro, which seems a bit slow to download files from my camera, which may make it slower than others. The results of the 35 point model did not allow 1 minute subs, and I had unacceptably elongated stars. But that's one data point, and there is a lot to learn.