[linux-audio-user] Advice needed re: latency tuning/optimization, esp. video

davidrclark at earthlink.net davidrclark at earthlink.net
Thu Feb 19 10:56:18 EST 2004


Thanks for your comment:

>It doesnt appear to - a friend of mine has been poking his to try and
>figure out whats going on. If you send a 48k impulse through it comes back
>with clear resampling symptoms.

As I'm sure you're well aware, this should not be the case --- at least
in theory (interpolation theory).  Now it is true that the emu10k1 does
32-bit integer math, not FP.  So perhaps this is the explanation.  I'm
unfamiliar with integer math interpolation and resampling; my
own resamplers are double-precision FP.  If there is any audible effect
of resampling of the same frequency, this should be a serious indictment
against the card and against whatever algorithms or math implementation
it's using.

Now back to 44.1 K vs. 48 K: The resampling problem you mentioned should 
only affect playback, not recording.  So I think that everything I said
previously still applies --- other than what we just discussed.

Thanks again; I'd be interested in hearing what resulted, even if by
private email.

More information about the linux-audio-user mailing list