I have had an Atlas EQ-G for almost 4 years. The first year I used it primarily for visual. After that I switched to imaging with an 8" RC. After well over one year of tinkering with gear mesh adjustments, guiding settings, PE correction via EQMOD, and as a last resort performing a Rowan belt mod; the mount has simply never worked consistently. The best guiding I ever achieved with the mount was close to 0.5" total RMS. Typically, it danced around from just below 1" to as much as 1.5" total RMS. Backlash was an issue and lessened with the belt mod, but continued as a problem with the mount. I did have stickiness in RA as well. I adjusted that by playing with the axis lock nut at the CW bar attach head at the expense of some play in the axis.
Ultimately, I gave up on the mount for imaging. It is sitting in my garage and has been unused for many months. Someday I'll rebuild it from scratch again and dedicate it to visual once my grandkids are old enough to go out at nights with me.
I ended up upgrading to a CEM120. My experience with that mount has overall been positive. As such, I recently I acquired a CEM40EC. Perhaps I had a lemon EQ-G, but the difference is night and day between guiding consistency of the iOptron mounts compared to the EQ-G.
As another data point, my brother has an EQ-G as well. He has only once ever gotten below 0.8" total RMS guiding error. His dances around from right around 1" to under 2". The good news for him is that he uses it at an image scale no less than 2.5"/px. I tend to be down around 1"/px or lower.
I recently got a replacement Atlas EQ-G from Orion as my previous one had significant backlash in the declination. I tried it out a few nights ago and found that the guiding performance of this new mount is significantly worse than my old mount (which usually stayed around .6"-.7" RMS). I've attached a guide log below; however, the TL;DR version is that the RMS average for the imaging session was 1.16" according to the log, but I would routinely come out and see that it was as high as 1.3"-1.5" even when pointing North (Iris Nebula). I'm just wondering if this is normal performance from the Atlas, or if there's some additional mechanical problem in the mount (to me it looks like there's significant declination backlash again and some "stickiness" in the RA). Any suggestions or comments would be greatly appreciated, and I'd be more than happy to provide any additional information!
Guide Log (Guiding assistant starts at 22:29:38 and Imaging starts at 23:30:32): https://openphdguidi...2_logs_kwRc.zip