It depends on how fast you are. I think you’ll find that Frank Dick is quite pessimistic for 10.5 and faster, and quite optimistic for 12.0 and slower. I get the impression that he had a fairly small sample size in his data set and made incorrect assumptions over the range of possible sprint times. The error seems to be smaller if you average several sprint times (30,60,150,250).
I think you’ll find USATF to be much more accurate overall.