Larrin,
I don't think forging stainless adds any benfit. Reduction in carbide size doesn't occur since carbide formation is accomplished during heat treating. Reduction in grain size is most effectively done with thermal cycling, and yes you can thermal cycle stainless, but with the CPM steels for instance, the grain is so small due to the CPM process, it's a waste of time to try to refine it more. This is all information I have received from metalurgists at Crucible, not my own conjecture.
If carbide size is reduced by heating up to 1950F to heat treat, than it would be reduced by forging, and yes, carbide size is reduced by forging; go ask Crucible again. I was just recently talking about thermal cycling stainless with Scott Devanna, it's not necessarily possible, he's reading up on it for me. I still have questions about a duplex grain (large grains forming with small ones) forming when heat treating multiple times, and whether this is also the case when normalizing/thermal cycling. And you're right, it's a waste of time to reduce the grain size of a CPM steel because the grain size is already way smaller than carbon steels ever get through thermal cycling and multiple quenching. And for conventional cast steels, most of them have a carbide size that is too large to get any reduction in grain size (grain size is dependent somewhat on carbide size). The only steels that have a potential for grain size reduction are conventional cast stainlesses that have a small carbide size such as 12C27, AEB-L, 12C27M, 13C26, etc.
And to fire one back at you, I still think 154CM and 440C benefit more from more reduction from forging than carbon steel, because of their large carbide size. Carbon steels carbides all dissolve at 1700F anyway, so there are rarely problems with carbide size or carbides forming in "strings" from rolling in one direction, which reduces toughness, but 154CM and 440C carbides do not break down as easily, and they form in "strings"., if any steel has a noticeable benefit from forging to shape, it would be 154CM, 440C, D2, etc. Also, the more they are reduced in size by forging, the more their large carbide size will be reduced. I'm not saying 154CM, 440C, or D2 are any better than simple carbon steels, only that they have more to improve on through forging than simple carbon steels.
Grains all recrystallize anyway, so forging to align the grain isn't much of an argument. You can normalize/thermal cycle and multiple quench a 52100 blade without every touching it to a hammer, and I guarantee you it would get just as small a grain size as if you forged it with low temperatures and then thermal cycled and triple quenched, and the blade that was not forged would perform just as well as one that was. Oh, and of course, there's no rule that says that you can't differentially heat treat if you don't forge.
We commonly make stainless damascus, and for somewhat proof of my point: we heat treated three blades, one out of D2, one out of 154CM, and one out of D2/154CM damascus. Obviously, if you have a ladder pattern of D2 and 154CM, each individual layer of D2 and 154CM has been reduced considerably, and there is at least some reduction in carbide size, and also grain size. When the carbide size is reduced, heat treatment is more effective; if you heat treat one blade of out 154CM, and one out of CPM-154, the CPM-154 will get harder every time. Anyway, the D2/154CM got harder with the same heat treatment than either the D2 or 154CM blade.