Recording at different bit depths/sample rates

[Heapdriver]

I want to experiment with different bit depths and sample rates,
to get a feeling for how the sound qualities vary. I have a
computer/studio setup with a Firewire 410 interface. I can do recordings
into my sequencer, but it defaults to 24 bit and I think the only other
choice is 16 bit.

I took a look at Audacity, but it seems to only support 32 bit, 24 bit,
and 16 bit.

Can anyone suggest a way for me to do this?

[Laurence Payne]

On Mon, 31 Dec 2007 04:21:53 -0700, Heapdriver >
wrote:

>I want to experiment with different bit depths and sample rates,
>to get a feeling for how the sound qualities vary. I have a
>computer/studio setup with a Firewire 410 interface. I can do recordings
>into my sequencer, but it defaults to 24 bit and I think the only other
>choice is 16 bit.
>
>I took a look at Audacity, but it seems to only support 32 bit, 24 bit,
>and 16 bit.
>
>Can anyone suggest a way for me to do this?


Your Firewire 410 is capable of sending the computer 16 or 24 bits at
sample rates up to 96KHz.

[Peter Larsen[_2_]]

Heapdriver wrote:

> I want to experiment with different bit depths and sample rates,
> to get a feeling for how the sound qualities vary. I have a
> computer/studio setup with a Firewire 410 interface. I can do
> recordings into my sequencer, but it defaults to 24 bit and I think
> the only other choice is 16 bit.

> I took a look at Audacity, but it seems to only support 32 bit, 24
> bit, and 16 bit.

> Can anyone suggest a way for me to do this?

You've had plenty comments, just supplementing with the fact that the 16 vs.
24 bits are counted downwards from 0 dB FS, and that the difference between
24 bit and 32 bit audio is that 32 bit audio has 8 more bits "on top". The
advantages of computing in this format are that cpu's do things in 32 bit
lumps anyway and that the headroom vastly reduces the risk of clipping
during processing and mixing.


Kind regards

Peter Larsen

[Laurence Payne]

On Tue, 1 Jan 2008 13:44:32 +0100, "Peter Larsen"
> wrote:

>You've had plenty comments, just supplementing with the fact that the 16 vs.
>24 bits are counted downwards from 0 dB FS, and that the difference between
>24 bit and 32 bit audio is that 32 bit audio has 8 more bits "on top". The
>advantages of computing in this format are that cpu's do things in 32 bit
>lumps anyway and that the headroom vastly reduces the risk of clipping
>during processing and mixing.

Sure, compute in a 32-bit space. In fact, if you're mixing on a
quality DAW program it doubtless does this anyway, without telling you
or giving you a choice!

[Norbert Hahn]

On Tue, 1 Jan 2008 13:44:32 +0100, Peter Larsen wrote:

>You've had plenty comments, just supplementing with the fact that the 16 vs.
>24 bits are counted downwards from 0 dB FS, and that the difference between
>24 bit and 32 bit audio is that 32 bit audio has 8 more bits "on top".

That's the case with 32 bit integer (fixed point) numbers for audio.

32 bit floating point is used for wav-files too, lots of software is
supporting ist. The floating point number seem to come in two different
kinds: 8 bit exponent and 24 bit mantissa and 16 bit exponent and
16 bit mantissa.
And I've even seen 24 bit floating numbers, 8 bit exponent and 16 bit
mantissa. But this format seem to be not common.

Norbert

[Peter Larsen[_2_]]

Norbert Hahn wrote:

> On Tue, 1 Jan 2008 13:44:32 +0100, Peter Larsen wrote:

>> You've had plenty comments, just supplementing with the fact that
>> the 16 vs. 24 bits are counted downwards from 0 dB FS, and that the
>> difference between 24 bit and 32 bit audio is that 32 bit audio has
>> 8 more bits "on top".

> That's the case with 32 bit integer (fixed point) numbers for audio.

> 32 bit floating point is used for wav-files too, lots of software is
> supporting ist. The floating point number seem to come in two
> different kinds: 8 bit exponent and 24 bit mantissa and 16 bit
> exponent and 16 bit mantissa.
> And I've even seen 24 bit floating numbers, 8 bit exponent and 16 bit
> mantissa. But this format seem to be not common.

Which is why I carefully did not translate the 8 bits to dB headroom ....
O;-)

> Norbert


Kind regards & Happy New Year!

Peter Larsen