Levels change when opening sessions from 1.5 to 2?

I don't know how significant this is, but with nearly all my existing sessions, when I open them in AA2 some of the tracks are showing clipping on the track meters, and very often the Master meter shows clipping as well.

When I look at the waves, they are all well under clipping levels, and I can't hear any distortion, but it is very distracting to have them showing.

I could go through and reduce every fader by the same amount and hope it cures it, but is there an easier answer?

I noticed this too, quite a while back. I'm not sure quite why this is, but it's only related to imported sessions. I did some checking, and the fader levels are all translated correctly. It's certainly the case that the max headroom on the channel faders has been reduced from +24dB to +15, but I don't think that this affects the result either.

There's a change in the way the panning laws work (they are much smoother, and will lose you 3dB in the centre), but this wouldn't affect the overall levels, except when signals are panned hard left and right. I will have to make enquiries...

It's quite noticeable if you import the old AA1.5 theme - the master fader indicator will clip most of the way through the piece, but all the channel fader positions are exactly as they were in 1.5. The quick answer is to reduce the master fader level if you are worried about this - it really doesn't matter about the individual channel levels if you are working in 32-bit. It doesn't really matter about the master level either, come to that - but if you don't like the clipping, that's the obvious one to reduce.

This might be an issue with busses.  As busses are completely different in Audition 2.0, there is no real way to properly interpret and convert an Audition 1.5 bus to the Audition 2.0 bus.  I think the level of difficulty is in how creative the mixer was set in the Audition 1.5 bus.  I think if the bus mixer in the Audition 1.5 session was never messed with, then you can get the same levels in Audition 2.0 by setting the output of each track that used an Audition 1.5 bus to "None".

For example, the Audition 1.5 theme session has every track routed to a bus.  Opening this session in Audition 2.0 will produce clipping.  But if I route every track's output to "None" (with the exception of the Bus track), then the levels are much closer to Audition 1.5.

For example, the Audition 1.5 theme session has every track routed to a bus.  Opening this session in Audition 2.0 will produce clipping.  But if I route every track's output to "None" (with the exception of the Bus track), then the levels are much closer to Audition 1.5.

It might well be a bus issue - but if I unroute all of the bus feeds, and just send the channel outputs to the master directly, I still get clipping with the old theme. So there has to be a bit more than this to it, I think.

It might well be a bus issue - but if I unroute all of the bus feeds, and just send the channel outputs to the master directly, I still get clipping with the old theme. So there has to be a bit more than this to it, I think.

Yep, you're right.  Doing the same in Audition 1.5 produces similar results, but not exactly.  The big difference looks to be the vocal track.  Perhaps one of the effects on it is rendered differently in Audition 2.0?

Yep, you're right.  Doing the same in Audition 1.5 produces similar results, but not exactly.  The big difference looks to be the vocal track.  Perhaps one of the effects on it is rendered differently in Audition 2.0?

I checked that - apart from the precision being reported differently in the dynamics processor, it's all very similar. If you run the vocal track into the bus with everything set at Odb, and no effects, then there's a 0.9dB discrepancy in the outputs that I can't account for - but the bus output is down, not up. The processing levels seem to be very consistent though. I think that this requires even more sleuthing...

I must admit that I'm surprised that I didn't notice this before. And it's something that people will have to be perhaps slightly wary of if they import 1.5 sessions, I think.

I wonder what would happen if one were to import, say, a CEP 2.0 session? AA 1.0 (which I'm still on) has always recognized those fine, with no changes at all.

I guess it was inevitable, given the amount of change that the program has undergone with this upgrade, that everything wouldn't remain perfectly backward-compatible forever.

Still, given the amount of remixing I do, methinks this is one more reason to hang back on the upgrade for a little while.

Chris

If you run the vocal track into the bus with everything set at Odb, and no effects, then there's a 0.9dB discrepancy in the outputs that I can't account for - but the bus output is down, not up.

I want to check this out, but I'm not sure I completely understand what you mean by "discrepancy".  Do you mean between the 2.0 bus output and the output if you don't run it through the bus (still in 2.0)?  Or do you mean between the 2.0 bus output and the 1.5 bus output?

I'm trying to find the level discrepancy you mentioned, but I'm not seeing it.  The mixdown from an empty bus (meaning no effects, eq, etc) in 2.0 vs. the mixdown when not going through the bus in 2.0 are identical in my tests.  When comparing the mixdowns from 1.5 vs. 2.0, I don't see the level difference you are, but the two mixdowns don't cancel each other out perfectly.  Not sure why...

Wait a minute!  I should have thought of this before...   There is still an effect on that vocal track.  Those clips are looped, and therefore, stretched using Audition's stretch effect.  

<homer>doh!</homer>

If you run the vocal track into the bus with everything set at Odb, and no effects, then there's a 0.9dB discrepancy in the outputs that I can't account for - but the bus output is down, not up.

I want to check this out, but I'm not sure I completely understand what you mean by "discrepancy".  Do you mean between the 2.0 bus output and the output if you don't run it through the bus (still in 2.0)?  Or do you mean between the 2.0 bus output and the 1.5 bus output?

I'm trying to find the level discrepancy you mentioned, but I'm not seeing it.  The mixdown from an empty bus (meaning no effects, eq, etc) in 2.0 vs. the mixdown when not going through the bus in 2.0 are identical in my tests.  When comparing the mixdowns from 1.5 vs. 2.0, I don't see the level difference you are, but the two mixdowns don't cancel each other out perfectly.  Not sure why...

I was just looking at the difference in output between a clean send to a clean bus between the channel meter and the bus meter - but...

I figured out where the discrepancy was - it was a matter of what I hadn't turned off, in the end. Now, I can get the feed to the bus to cancel completely by inverting the bus phase - which is as it should be. But I'm still getting too much signal overall on the master channel.

Wait a minute!  I should have thought of this before...   There is still an effect on that vocal track.  Those clips are looped, and therefore, stretched using Audition's stretch effect.  

Er yes, but this is the same in both AA1.5 and 2.0, isn't it?

(I think that it's pretty obvious, but just in case anybody reading this hasn't guessed, Chris knows rather more about this than he's letting on...)

Er yes, but this is the same in both AA1.5 and 2.0, isn't it?

(I think that it's pretty obvious, but just in case anybody reading this hasn't guessed, Chris knows rather more about this than he's letting on...)

 

Yes, I'm on the "inside".

And actually no, the stretch effect has changed in Audition 2.0.  For one thing, it's algorithm has been significantly optimized.  But this does change the resulting waveform, so Audition 1.5 and 2.0 stretch results will be similar, but not perfectly identical.  Although the waveform will be different, it definitely shouldn't be worse.  But don't take my word on it...

-Chris Robinson  (no, not the Black Crowes singer...)

Okay, I think that I've sussed out what's happening.

Basically, I can't make this go wrong in AA2.0 at all - all of the levels are exactly what they should be - the algebraic sums from the track plus the bus come to exactly 6dB more than either one alone - and that's as it should be.

The 'problem' is that the mixer now works correctly - and if you send a signal to the bus, it doesn't mean that this is the only place it goes to - it comes out of the channel as well. This is not what happens with AA1.5 - you only get the channel output from the bus when you send it there, not from the channel itself.

The solution is simple: If you import a 1.5 session into 2.0, and want it to run exactly as it did, then any channel that has a bus send has to be deselected from feeding the master output (under the channel selector, you select 'none'). With the 1.5 theme, where everything is sent to a bus, you select 'none' for all of the channels except Bus A. Now if you play the track, the levels are exactly the same as they would be in 1.5.

The difference is simply in the way that the mixer is configured.

Phew! I was getting worried there for a moment...

Okay, I think that I've sussed out what's happening ...

Good detective work there, Chief.  Makes good sense.

Cool! Nice work there guys - and all sorted out while I caught up on my beauty sleep Thanks very much.