[Jay[_2_]]

Hi.

I'm using Sonar 6 on a WinXP SP2 system, and a Gina3G 24-bit soundcard.

I have Sonar's engine set to 32 bits despite the soundcard being 24b because
: 1) That was the default installation setting, and 2) This option is about
internal processing, and I figured it had more to do with what my hardware
and OS can do than what my soundcard can.

The software's audio engine is using all the available bits this way, right?
It's not constantly degrading the sound because of the 24-bit limit of the
soundcard, is it? As long as the wav remains in the digital realm, the
limitations of the soundcard are irrelevant, right? So I'd have nothing to
gain switching the Sonar 6 engine to 24 bits, would I?

Assuming I've got that right :

1) Should dithering (during track bouncing, for instance) always be on in
the Audio / Advanced area?

2) Should dithering be on during exports, where all 50 tracks are bounced
together to a single wav located outside the program?

3) Am I doing myself a disservice exporting to 32b, or even having the
engine always set to 32b during everyday editing?

Thanks.

[Ethan Winer]

Jay,

I can tell you what I do, but others may (will) disagree.

I have Sonar's engine set to 32 bits

That's fine, and your understanding is correct.

1) Should dithering (during track bouncing, for instance) always be on in
the Audio / Advanced area?

I never use dithering because I'm not willing to give up even 1/100th of a
percent of my CPU for that. I suggest you try it both ways and see if you
can hear a difference. That's what I did and you can probably guess what I
concluded. :-)

3) Am I doing myself a disservice exporting to 32b, or even having the
engine always set to 32b during everyday editing?

I export as 16 bits because that's what goes onto CDs and it's more
efficient. And for tunes I'll convert to MP3 16 bits is more than good
enough there too.

--Ethan

[JSVice]

"Jay" wrote in message ...
Hi.

I'm using Sonar 6 on a WinXP SP2 system, and a Gina3G 24-bit soundcard.

I have Sonar's engine set to 32 bits despite the soundcard being 24b
because : 1) That was the default installation setting, and 2) This option
is about internal processing, and I figured it had more to do with what my
hardware and OS can do than what my soundcard can.

The software's audio engine is using all the available bits this way,
right? It's not constantly degrading the sound because of the 24-bit limit
of the soundcard, is it? As long as the wav remains in the digital realm,
the limitations of the soundcard are irrelevant, right? So I'd have
nothing to gain switching the Sonar 6 engine to 24 bits, would I?

Assuming I've got that right :

1) Should dithering (during track bouncing, for instance) always be on in
the Audio / Advanced area?
2) Should dithering be on during exports, where all 50 tracks are
bounced
together to a single wav located outside the program?

If you bouce the file and are then done with it, then probably yes. If
you're going to do some other processing later, then you may have the option
to add dither at that time (Waves L2/L3 for example). Some people prefer
the "sound" of one dither algorithm to another.


3) Am I doing myself a disservice exporting to 32b, or even having the
engine always set to 32b during everyday editing?

I don't know how the inner workings of the program are, but I would figure
that, if you record at 24 bits, then bouncing to a 32 bit file isn't going
to get you anything but some extra zeroes and wasted drive space. Maybe
someone who knows more about the engine can comment on that.

--
Thanks,
John


Thanks.

[Jay[_2_]]

"JSVice" wrote in message
...

I don't know how the inner workings of the program are, but I would figure
that, if you record at 24 bits, then bouncing to a 32 bit file isn't going
to get you anything but some extra zeroes and wasted drive space. Maybe
someone who knows more about the engine can comment on that.

I didn't mean bounce to 32b from a 24b project... the song itself and all
tracks within were processed with the program set to 32b while those tracks
were being edited and processed - before said export. So there wouldn't be
extra zeroes here, I don't think.

[JSVice]

"Jay" wrote in message ...
"JSVice" wrote in message
...

I don't know how the inner workings of the program are, but I would
figure that, if you record at 24 bits, then bouncing to a 32 bit file
isn't going to get you anything but some extra zeroes and wasted drive
space. Maybe someone who knows more about the engine can comment on
that.

I didn't mean bounce to 32b from a 24b project... the song itself and all
tracks within were processed with the program set to 32b while those
tracks were being edited and processed - before said export. So there
wouldn't be extra zeroes here, I don't think.

Ahh. Well, I suppose if you have the CPU power, it wouldn't hurt in the
least to keep everything at 32 bit internally. You should reduce artifacts
caused by rounding errors. All the math that gets done by the plugins and
other processing should be more precise at 32 bit. Whether the difference
is audible is the only question I suppose. Have you tried it both ways?
--
Thanks,
John

[dash10606]

There have been a number of discussions on the Sonar forum about this
topic. Here's a good 'sanctioned' explanation of how Sonar manages
bit depth, word depth, dither, etc.

http://forum.cakewalk.com/tm.asp?m=751063&mpage=3

As I understand it, your soundcard D/A 24-bit limitation *does* come
into play here, and the *optional* dither may be enabled to dither 24-
bit audio result of the 32-bit audio sent to your sound card. There
is a performance hit for that processing.

But that doesn't affect the files themselves - just adds some
processing as it exits the application and enters the audio realm.

Audio is passed around Sonar as 32-bit data (actually I think the new
Vista compatible version allows 64-bit word-length, too). When it
exits Sonar and is directed to the sound card or when it explicitly
rendered to a lower bit depth, you can *optionally* apply one of the
dithering algorithms. Remember, you may want to render without dither
as 32-bit audio or at 24-bits for someone else to mix or master the
track.

If you render your track at 16-bits you may want to dither at that
stage (or not, as Ethan prefers). If you render as 16 bits, you can
burn a CD without further bit reduction, so you probably want to wait
until that point. Another thing to remember is to dither the audio
data *once*, and to wait until the very last step of your processing -
i.e. when you reduce the bit depth for target media for the last time.

Dan Ash
White Plains, NY