Well, I wondered about that.
So I kept an eye on the CPU core temperatures. There is very little thermal inertia there. At the sampling rate m monitoring has, there were some spikes, but none really excessive, and as I mentioned above, the last crash was AFTER integration finished, and temps had dropped, when I went t save the image.
Next time it worked again, and I saved the image with no problem.
It probably IS the same problem. Mine is a newer CPU with 6 cores rather than 4.
Following The advice in the previous thread, I have turned off half of my CPU threads. :-(
Let's see how that goes.
I think I will open a support ticket with Apple, just to see what their response is.