[LAU] 48k vs. 44.1k

Len Ovens len at ovenwerks.net
Mon Jan 7 03:16:18 UTC 2013

On Sun, January 6, 2013 3:17 pm, David Olofson wrote:
> On Sunday 06 January 2013, at 04.38.01, "Len Ovens" <len at ovenwerks.net>
> wrote:
> [...]
>> I do wonder about samplers though. Are the samples mostly in 44.1k
>> if they come on a CD? Would that make them off-key when used? Would they
>> get rate-changed on the fly? Would that use more CPU? (questions,
>> questions, questions)
> [...]
> A sampler (software or hardware) is typically resampling everything on the
> fly
> anyway, modulating the pitch in response to pitch bend events, LFOs,
> envelopes
> etc. It's one of the most fundamental features of sampling as a synthesis
> method. Traditionally, an instrument would only have one or a few samples
> per
> octave, so resampling was required to even implement note pitch at all.

So as long as Jack is reporting the correct rate everything will be fine.
> These multi-gigabyte piano and orchestral sound disks may not make much
> use of
> that feature, but that's a special use case that came with live streaming
> from
> disk and machines with gigabytes of RAM. I'd think even samplers built
> specifically for this would still have the usual resampling features, and
> should work with any output sample rate - but of course, you're not
> getting
> any advantages from 48 kHz if you're still playing 44.1 kHz sample
> libraries
> of that kind.

With a sampler there would be no advantage, true. To only be using a
sampler though would generally be live use of sampler and other
softsynths. Though there are reports that the hardware is better optimized
for 48K than 44.1k so the output sound _may_ be improved. On the other
hand, if there is any recording being done and the sampler is being used
only as one or two tracks with many live inputs, the live inputs do
benefit and the overall sound quality _should_ be higher. We wouldn't
record the whole project at 12bits and 32k just because we are using a
vintage synth that uses 12 bits at 32k to generate it's sound, but we
might still use that synth because we like it's sound. (no I don't know if
there is a real synth that uses that combination :)

Len Ovens

More information about the Linux-audio-user mailing list