Fast 'n dirty vs slow 'n clean is a common
trade-off in a zillion
applications
And some have fast'n clean. If my src is "dirty", send me an example.
I wonder if this contest is misleading -- the default sinc width I use
was originally aimed at CLM-users; perhaps Eric's uses a different
default, so comparisons are of apples and oranges, so to speak.
I agree with Kjetil that linear resampling can be useful -- I based
"Leviathan" on that effect. Sometimes "clean" isn't the main
criterion.