It has been a while since I have done much to mine. Good as a reminder as any :)
Does it make a significant difference to the amount of blocks they receive when out of the T20? Would it be more beneficial for example to spread a few more of the blocks to the T20-40, while still keeping the consensus T20?
I didn't do the maths, but 40 is very probable to still be able to run with a profit if nothing else is financed with it, witness nodes are relatively cheap. The difference from 20 to 21 is 80% though.
I'm always open to spreading the inflation wider, but I'm not sure creating a third "class" would be the answer.
Just wondering if it would encouragement more solid node operators.
The higher in the ranking, the more often a witness will be scheduled for block production. Therefore your idea for T20-40 to produce more blocks is already implemented.
should some more of the witness distribution from the T20 go toward the next 20 you think?
Yes, I'm in favor of lowering the gap between T20 and T21+