There is a whole thread on this from August 2019: "power triodes indirect vs direct heating".
Most of the science is considered.
Of course, there are scarcely any indirectly heated power triodes.
Russians made a number sort of “similar” regulators tubes: the low gain triodes - 6C19P, 6C41C, 6C47C, 6C33C, 6C18C, 12C42C. The 6C41C, 6C33C and the 12C42C are essentially the very same tubes but with different number of anodes, and consequentially with different dissipation power
However, in several places I heard different ratings of the first 6c41c, which should be in fact a "real" ordinary triode and by the logic of things should sound the best, from recommendations to ratings that can not be compared with the bigger brother 6c33c ...
I looked at data sheet curves of common DH and IDH triodes. The only one that impressed me with low curve rollover is 01A. Three other DHT renowned for fine sound, 10, 71A, and 45, are, IMO, nothing to write home about, if compared to 27 or 56.
I am not convinced that low rollover explains the sonic signature of DHTs.
An electron tube isn't a telephone. It won't tell you their best working point by the numbers in the datasheets. You have to hear it, have to test it.The curves are very similar, the difference lies in their performance.
Obviously there is going to be distortion in any device whose "curves" are not equally spaced straight lines. In the other thread, DF96 made the point that the reason for the superior linearity of 300B, 845, 211 and so on, is their size not the fact that they are directly heated.
That could be true but it mustn't.I don't use any small 9 pin tubes and haven't for years. Not that I'm against them, but I have found that the older and larger tubes just seem to sound better. The older types with octal, loctal, side contact and all the other larger old bases. I know this is a broad generalisation, but I'm just calling it as I hear it.