From: Fred Maxwell <rockbox_at_anti-spam.org>
Date: Fri, 19 Mar 2004 20:21:24 -0500

BlueChip wrote:

> At 20:09 19/03/04, you wrote:
> >Jörg wrote (referring to levels above 93%):
> >
> >>I avoid it at all cost, because the distortion (clipping) is really bad.
> >
> >Wouldn't that have to do with the load being driven?
>
> No, you are thinking on the wrong side of the circuit ...You need to be
> thinking on the digital side.

So you are saying that the clipping exists in the digital realm (e.g.,
values exceeding 16 bits)?

Driving a pair of low impedance headphones will cause clipping faster than
driving high impedance. Clipping often occurs because of limitations to an
amp's power supply and lower impedance will suck the supply dry sooner. I'd
be interested to see if the clipping that occurs above 93% shows up driving
a high-impedance load like a line-level input. Neon John (a close personal
friend of mine, as you know ;) reports having to use levels as high as 96%
to 97% for good signal-to-noise ratio on his aviation muffs/headphones. It
seems unlikely that he would be turning it up that high if it lead to severe
clipping.

Regards,
  Fred Maxwell



_______________________________________________
http://cool.haxx.se/mailman/listinfo/rockbox
Received on 2004-03-20