Interestingly, I saw a 1973 video from the BBC that showed an animation of new grains forming in 0.8% carbon steel, and they described what seems to be well accepted by knife makers now - that normalization before quench is important (to get a clean set of new grains). Their animation showed the emergence of grains beginning around 750C (emergence starting at the boundaries of old grains).
The interesting thing to me is both that, as well as off-the-charts things people do with a forge to try to make up for the shortcomings of not having perfect heat control.
In the knife world (For anyone with the desire to buy an oven), it appears to be standard to normalize steel in an oven (instead of a forge) and then quench from there - so temperature and duration are well controlled. And then two tempers.
For the forgers, the forge anneal, hammering and triple quench seems to be an alternative.
(Of course, annealing in the forge and then hammering and just heating and quenching also works - it's surprising with the simpler steels how close the grain structure is from one to the next - and what it leaves me wondering is if I'm going to hammer out knives and chisels, am I really approaching anything needing this ideal treatment? I don't think I am, but I'll find out. I wouldn't be surprised to find that annealing a file and then giving it a forge soak and the a pair of quenches and a temper makes just as good of a chisel as hammering and being more careful....
...but the higher end knife makers who push angle and then do things like cut nails with their high hardness knives do show what's possible.
I'm less searching for perfection than guarding against downside. I don't want to come up with a routine and then find out it's less good/more coarse than just annealing and shaping the files and being more careful about temperature control.
I'm not interested in buying a heat treat oven at this point.
I'm headed for your link to read and see what I can glean from it.