This is fun to read and also a valuable contribution to preserving the historical details of how it was achieved. I especially appreciate his tone in approaching what had become a somewhat contentious subject:
> Hi, I was the animation supervisor on Rogue One, and as such I was intimately involved with the creation of Tarkin.
> I’ve decided to chime in for one purpose only, to clarify the process we used. I have no interest in trying to convince anyone to like the results more than they do, or to argue with anyone about how “real” our work looked in the film.
I'm one of those who enjoyed RO but also immediately noticed the CGI Tarkin being "off", despite the fact I'd not heard about it and didn't go in looking for it (I had heard something about CGI Leia though). It's helpful that the OP mentioned in the intro that many people never noticed it. Although CGI Tarkin clearly stood out to me, I'm a pretty serious SW fan (having seen the original when I was 12 and the entire opening trilogy many times since). So I'm unusually familiar with Peter Cushing's appearance and mannerisms on-screen in the SW universe.
Perhaps more significantly, I've also had a multi-decade career deeply involved in the creation and evolution of digital production tools and CGI as well as being a sometime professional (and, more often, hobbyist) film-maker. To be fair, once you start counting NAB and Siggraph trade shows you've attended by the dozen, it's reasonable to assume you probably can't see films or CGI the way most people do - and so I concede it's entirely possible CGI Tarkin was adequately executed for the majority of the intended audience.
However, I think that may miss the more important point that, whether CGI Tarkin in RO was "good enough" or not, doesn't much matter in the long run. We've always known creating perfectly photo-realistic CGI humans is extremely difficult, especially substituting CGI for a particular well-known human in a well-known live action context. It's pretty much the hardest CGI thing there is. Like most things in CGI, I'm pretty sure we'll eventually master it but at the time CGI Tarkin was done - it was wildly ambitious and, IMHO, very likely to fail. So the fact CGI Tarkin didn't abjectly fail and was, at worst, mildly distracting to critical eyes, is something the team that did it should be proud of and those of us with those critical eyes should, at the least, be tolerant of and, preferably, celebrate as a worthy historical milestone on the long path toward perfection.