Bouncing in session first vs. bouncing straight to disk...

[WATYF]

OK. A couple of friends of mine got some "official" Pro Tools training in New York some time back. One of the tid-bits of information that the instructor told them was that if your session is 48K (or above) you should always bounce your song into a stereo audio track (in the same session) and THEN bounce that track out at 44.1K. He said this was because if you bounce a 48K session straight out to 44.1, your plug-ins are "running" at 44.1 (the bounce sample rate) instead of 48 (the session sample rate).

So my friends took this to heart, and have been bouncing this way for some time... basically *doubling* their bounce time (once in the session and then again out to disk).

Anyway, I found out about their methodology a while ago and, as a computer nerd, it didn't make any sense to me, since I know that a sample rate conversion happens *after* the bounce (you have the option to Convert During Bounce or Convert After Bounce). What that means, in my little nerd brain, is that a temporary 48K WAV file is being created during the bounce, and then at the *end* of the bounce, it gets converted to 44.1K. So in that case, it doesn't make any sense that the plug-ins would be "running" at any sample rate other than 48K.

So I figured I'd test this out myself. I bounced 30 seconds of a 48K session various different ways. I bounced straight to 44.1 using Convert During Bounce... then straight to 44.1 using Convert After Bounce... then I bounced the song IN the session to a stereo audio track and directly converted that stereo track to 44.1... then I bounced IN the session to stereo track and bounced that stereo track out to 44.1

At that point, I had four different stereo files created four different ways... and every single file was the exact same size, down to the last byte. I then took all of them and brought them into a 44.1 session and then started inverting them and playing them in pairs. Every file that was inverted cancelled every other file out... which basically means that there is no difference, in even the slightest way, between any method of bouncing.


So my question is... has anyone else heard this "myth"... and why on earth is a certified Pro Tools instructor teaching people something that is such a HUGE waste of their productivity??


WATYF

[albee1952]

Well done on your personal experiment. I have to agree that it sounds like a big waste of time. I bounce using the Convert after Bounce choice as I assume it puts less load on the cpu and may avoid errors or a bounce that does not complete. As to your friends assertion that the plugins will run at 44.1K, I have doubts on this as I am betting that the sample rate conversion is the last thing to happen(at least using CONVERT AFTER BOUNCE).

[timebandit]

The only advantage I could see to that would be to have two copies of the mix - one at the higher sample rate and one at 44.1 for use on cd if desired.

I can't see how the plug-ins could run at a different sample rate than the session. PT won't even let you mix files with different sample rates.

It was a good idea to check it for yourself and not just take someone's word for it.

Tom

[3JDamon]

The main advantage of bouncing in session is that it takes a little less CPU than BTD for some reason. When dealing with 48k files I leave them at 48k for mastering, or when not doing 3rd party mastering I wait until all the songs are done and have Batch Converter convert to 44.1, which has just as good anti-aliasing as the "tweak head" setting and I can set it to do all of them while I do something else. I don't know of any sonic benefits to bouncing in session.

Weird, two posts concerning Batch Converter in a row...

[timebandit]

Weird, two posts concerning Batch Converter in a row...

yeah, really! I hope you are getting a sales commission!

[The Weed]

3J,

Do you have a link to Batch Converter? When I Google it I get a lot more than one show up.

Thanks,