You confirm that I am not as stupid as I first thought.
The software still is way too slow to be usable.
It's not about being stupid or things like that. I too experience occasional slowdowns when I'm pushing the application (like with stitching multiple-row, very wide panoramas, consuming lots of memory). But for less taxing assignments it performs well, even on limited hardware (although more capable hardware is going to give a better experience).
Speed has also a lot to do with what the application allows a user to achieve by clever implementation of operations (i.e. workflow benefits), not only raw processing power. As a Windows version user, I've only had a limited period of time to explore so far because the application only recently became available with a Windows version. But what I've seen, really is impressive.
I've been a Photoshop user for many (rough estimate some 20) years, and I've paid for many updates over those years. I consider myself a pretty well informed and experienced user of actually most of its functionality, unlike many users who only use a subset of the capabilities.
Comparing Affinity Photo's capabilities and application design choices is a refreshing experience. Is it perfect yet, no !, but it's on its way of becoming something awesome (and a lot already is).
Because I'm used to evaluating software (and have seen several good ideas perish), I've developed a sense for which are viable solutions and which are not. Affinity (Design and) Photo are! They are clearly made by people who come from the practice of making things happen because they need them themselves on a daily basis. And they are not yet at the stage of capitalizing on past investments, like e.g. Adobe. No, they are still eager, they are trying to please most of their (prospective) customers and they are pushing the limits.
It didn't take long for me to understand that they deserved backing, especially with the incredibly low price-of-entrance they ask for their software. Really a no-brainer. Again, the software is not perfect (but then what is?), but it has the potential of getting very close. I'll be submitting some suggestions for feature enhancements, not just for my personal benefit, but because they address fundamental benefits for all users.