I wonder what you consider to be long time control

[Uri Blass]

VVLTC SPRT after good VLTC

is in one of the tests

I see that the VVLTC is only 60+0.6 when earlier VLTC was 240+2.4

I think it is wrong

I know that 60+0.6 is for 8 cores when 240+2.4 was only for one core but I consider 240+2.4 to be longer time control than 60+0.6

quality of play for 60+0.6 with 8 cores may be higher than 240+2.4 with 1 core but higher quality of play does not mean longer time control.

My opinion is that testing with many cores also in regression tests should be at 60+0.6 and not at faster time control because almost nobdy is using stockfish at faster time control and in any case testing with many cores is only a small minority of the tests

[Mikael Johnsson]

The metric is cpu time. 20+0.2 8th is 2.5x the amount of resources put into the search 

per game compared 60+0.6 1th. 60+0.6 8th is 2x VLTC