in my world...

[bebop]

i am coming to logic from a roland VS multi tracker box

 

in the VS architecture, with a click of a button, you switched between "input" mixer (where you were bringing in audio for recording .. mics, keyboards..etc) and "track" mixer (where you controled all parameters of the recorded track)

 

"aux" was used for routing the audio out to effects or...

 

just normal mix board stuff

 

my question is:

 

in logic, to bring in external audio for recording, I'm finding that making an audio object and calling that channel strip aux or input both do the same thing.

 

which should I use...and why?

[David Nahmani]

You should use neither.

 

In Logic, you record audio on Audio Track Objects. They are the only Channel Strips that have a "Rec" button. All the plug-ins inserted on those tracks process the audio after recording, so you can hear them while you're recording but they're not "printed to tape". They process the recorded audio the same way upon playback, but you can still open the plug-ins and change the settings, or get rid of the plug-ins altogether if you change your mind.

 

You can use Input objects (Or Aux objects with their input field set to an input - same thing) to process the incoming audio before it reaches the track. That means any plug-ins inserted on an Input object is printed to tape: destructive processing. Once recorded, it's too late to change your mind.

 

There's only two reasons why you'd want to do that:

 

1) Artistic choice: some people actually NEED to have limited options, and don't want to be able to come back and change their mind.

2) Saving CPU: once recorded, you can get rid of the plug-ins, and they won't tax the CPU anymore.

 

However, keep in mind you're losing a little tiny bit of resolution when using the plug-ins on an input objects, since then the audio is converted to 24 bits to be recorded in an audio file. When using the plug-ins on a track object, the audio is processed in 32 bit floating point for the whole time it is inside Logic, so you're getting better resolution.

 

Conclusion? Don't use Input or Aux objects unless you need to monitor an input on your audio interface without always having to rec-arm an audio track.

[bebop]

another scholastic response

 

thank you for that david

[Jope]

However, keep in mind you're losing a little tiny bit of resolution when using the plug-ins on an input objects, since then the audio is converted to 24 bits to be recorded in an audio file. When using the plug-ins on a track object, the audio is processed in 32 bit floating point for the whole time it is inside Logic, so you're getting better resolution.

 

This is obviously true for the effects applied before the audio is recorded... But if you want to apply effects to the recorded track then, isn't it calculated in 32-bit-floating-point again? So the "tape print" would be the only bottleneck in the signal flow (and 24 bits fixed are still a lot)?

[David Nahmani]

This is obviously true for the effects applied before the audio is recorded... But if you want to apply effects to the recorded track then, isn't it calculated in 32-bit-floating-point again? So the "tape print" would be the only bottleneck in the signal flow (and 24 bits fixed are still a lot)?

That's exactly what I tried to explain. If you use plug-ins on Input objects, they are applied before the audio is recorded.

 

24 bit fixed is a lot, but when you process, convert to 24 bit, reconvert to 32 bit to process some more.. you lose resolution. How much resolution depends of what kind of levels your signals are, and how much processing you're using on the input object. I wouldn't worry about it too much, but it was just one more argument to not use input objects. Most of the time there's no need to.