Josep M Comajuncosas wrote:
> So when using a-rate delay times, is it possible to obtain delays lower than 1/sr ? or just
> up to 1/sr?

if you are using an interpolated delay (such as vdelay or deltapi) I think it is possible. 
However I intended 1/kr, when kr  is different than sr, not 1/sr
-- 
Gabriel Maldonado