Topic: Cross-Fading MIC/SBD Sources in CEP 2

[mattmiller]

I'm mixing my KM140 and SBD sources in CEP 2.  Everything is aligned, drift-corrected, and sounding great.  The problem is that the SBD source vanishes one verse into the last song.  I'm still working on figuring out what happened there, but, regardless of the cause, I need to blend the transition from MIC/SBD mix to just the MIC source.  I received some good advice to fade out the SBD source rather than having it stop abruptly (and correspondingly fade in more of the MIC source to keep the overall volume consistent).  I imagine there are multiple ways to accomplish this, but I'm hoping somebody can advise me from where I've left off (rather
than starting over with a different method).  See attached screencap showing the multitrack view in CEP 2.  So far, I've:

1.  Normalized each of the sources to -6 for the sake of having equal start points in the mix
2.  This was a pretty good sounding room, but an even better SBD mix.  So I'm going SBD heavy on the mix.  I've gone +5 on the SBD and -5 on the MIC, and I like how it sounds.

So, from this point, what's the proper way to equally fade out the SBD source (top) and at the same time fade in more of the MIC source, and then carry this extra MIC source through the end of the recording?  I realize that this is a cross-fade, but I've never done this and am looking for some step-by-step instructions for CEP 2.

[admkrk]

i would have got my envelopes right before worrying about normlizing, that would be the last thing i'd do.

just fade out the sbd which which will probly mean a corresponding raise in the aud  source. get all that right before normlizing. CEP has to have volume envelopes? use that instead of trying to use fades. 

[Cheesecadet]

I use CEP all the time and I actually like to use the crossfade method rather than using envelopes....  I usually do a 10-15 second crossfade on both sides and it is not noticeable.

Envelopes would work too I'm sure, but I never have really used them so I just stick to the crossfade method...ymmv

[mattmiller]

i would have got my envelopes right before worrying about normlizing, that would be the last thing i'd do.

just fade out the sbd which which will probly mean a corresponding raise in the aud  source. get all that right before normlizing. CEP has to have volume envelopes? use that instead of trying to use fades.

I still plan on normalizing the final mixdown to -0.1, but I wanted to have a 50/50 mix to start with, which is why I normalized each source first to -6.

CEP does have volume envelopes, though I've never used them.  But I've been playing around with them and they seem pretty intuitive.  The approach I'm thinking about is as follows, assuming point A is the point where I want the SBD to start fading out and the MIC to start amplifying to compensate, point B is the point where the SBD disappears and the MIC reaches the maximum volume (equal to the sum of the SBD/MIC sources throughout the rest of the recording), and point C is the end of the recording.

1.  Apply volume envelope to last 10 or so seconds of SBD source, taking it from 100% to 0%
2.  Apply volume envelope to the MIC source, starting at the same point that the SBD source starts to fade out, raising it to max volume over 10 seconds, and holding it at max volume through the end of the recording.

My question, then, would be -- what percentage and I amplifying it by to account for the missing SBD source?  The volume envelopes look to be in percentages, while everything I know about the SBD source is in dB (normalized to -6, and then +5 in the mix).  I assume that's +5 dB?  What is the corresponding percentage that I want to increase the AUD source to so that the overall volume remains constant through the transition?

Or am I way off in my thinking?

[rainingvodka]

I've been using cep 2.1 for years and have done a lot of source mixing with it.  When I mix sources and have to cross fade, I use the envelope adjust drag points to do the fades and for any volume adjustments, I do that by ears on the sound and eyes on the level bars.  Something to remember, imho, is that even if all the "math" is right, that might not give you the best results.  Sometimes the levels should look "off."  Sometimes you want to start the crossfade on one source before the other.  Sometimes you don't want that crossfade to be a consistent amplification.   Etc.  The best thing to do is just spend some time playing around with the program and find the sound that works for *you*

[admkrk]

I've been using cep 2.1 for years and have done a lot of source mixing with it.  When I mix sources and have to cross fade, I use the envelope adjust drag points to do the fades and for any volume adjustments, I do that by ears on the sound and eyes on the level bars.  Something to remember, imho, is that even if all the "math" is right, that might not give you the best results.  Sometimes the levels should look "off."  Sometimes you want to start the crossfade on one source before the other.  Sometimes you don't want that crossfade to be a consistent amplification.   Etc.  The best thing to do is just spend some time playing around with the program and find the sound that works for *you*

i've never used cep but that is pretty much my work flow. i think people get too caught up in the "math" part and don't even listen to the results. i usually have several meters open and listen to each change i make. i prefer to use envelopes also because i can add points to change the amount of fade as needed throughout the duration of it. most times i get better results that way then a straight 0 - 100%.

<edit>

i also use the envelopes to control spikes, etc. in one source or the other throughout the entire recording. saving any overall normalizing or gain adjustments till last if still needed. 

[mattmiller]

I'm finally getting back to working on this matrix and am close to completion, but one thing is holding me back.  I'm using a volume envelope in the multitrack view to fade out the last 5 seconds of the SBD source.  That was easy enough to figure out -- click to put in in the control point at the start of the fade, keep this at 100%, click to put in the control point at the end of the fade, set this to 0%.  However, I can't figure out how to fade UP the MIC source to compensate for it (in multitrack view -- I want to do this non-destructively).  The volume envelope is, of course, a green line across the top of the track, representing 100%.  I can't drag that UP any further to make it fade from 100% to, say, 200%.  Is this not possible?  What am I overlooking?

[Brian Skalinder]

I can't drag that UP any further to make it fade from 100% to, say, 200%.

My vague recollection is that you need to zoom out the vertical axis to provide room above the 100% line into which you may create the envelope points.

[mattmiller]

My vague recollection is that you need to zoom out the vertical axis to provide room above the 100% line into which you may create the envelope points.

That was a promising idea, but it seems that no matter how much I zoom in or out vertically, the volume envelope stays at the very top of the track.

[kirk97132]

I am not sure how much CEP2 is like AA3, but if it is and has automation lanes for the tracks...can you just open an automation lane for that track and then just add whatever gain you need at whatever slope is needed? 

[mattmiller]

I am not sure how much CEP2 is like AA3, but if it is and has automation lanes for the tracks...can you just open an automation lane for that track and then just add whatever gain you need at whatever slope is needed?

A search of the CEP2 "Help" index turned up no results for the term "automation lane".

[admkrk]

can you switch your scale to db instead of %?

you should be able to raise the volume just as much as you can lower it, if that makes sense. in wl the starting point of the envelope is at the mid point of the scale.

[mattmiller]

can you switch your scale to db instead of %?

you should be able to raise the volume just as much as you can lower it, if that makes sense. in wl the starting point of the envelope is at the mid point of the scale.

I haven't been able to find a way to switch the scale to dB.  And having the "baseline" at the midpoint of the scale seems much more intuitive.  But in CEP2, the baseline is 100% and at the very top.  Nowhere to go but down....

[admkrk]

that's strange. it doesn't make sense to only be able to reduce gain with the envelope. you could get around it by setting an initial envelope at the midpoint on all tracks and boosting the overall gain to get it back to were it was, but that seems like more work than it's worth. i've never used cep so i don't know were to look.

[mattmiller]

you could get around it by setting an initial envelope at the midpoint on all tracks and boosting the overall gain to get it back to were it was, but that seems like more work than it's worth.

I thought of that, but didn't want to go that route until I'd exhausted the collective wisdom of TS.com.  In the apparent absence of any other way to do it in CEP2 I've been playing around with this method and it's really not that bad.  The only caveat would be that you need to predict how much headroom you need (i.e., if you know you're only going to want to increase something to 200%, you only need to start with a baseline of 50%, but if you might want to increase something to 300% you need to make sure to start at 33% or below.

[admkrk]

i really hate using percentages, for one thing it's not very descriptive. 50% of 6 is not the same as 50% of 12 for example. it's fine if you're working on the entire file at once, but when you're working on 2 or more files or parts of a file, it just doesn't make sense. i mean if i want to boost or lower a section 3db, i don't want to try to figure out what the % is.

hopefully someone else will speak up that uses cep with an answer. i'm almost tempted to hunt down a copy just to figure this out. 

[Cheesecadet]

In your particular case with the file pic you showed originally...I would just apply a 20 - 30 fade out to the SBD source...then mixdown.

Over that period of time the gradual decrease in clarity from the SBD source would be negligible.  You would obviously hear at the end that there is just an AUD copy...but in this situation you would hear only that anyway.

[mattmiller]

In your particular case with the file pic you showed originally...I would just apply a 20 - 30 fade out to the SBD source...then mixdown.

Over that period of time the gradual decrease in clarity from the SBD source would be negligible.  You would obviously hear at the end that there is just an AUD copy...but in this situation you would hear only that anyway.

That works for gradually fading out the SBD source, but doesn't compensate for the loss of dB's by increasing the MIC source over the same interval.  I accomplished what I needed to accomplish using the suggestion of setting the volume envelopes to 50% for everything to start with.  This allowed me to apply a fade UP to the MIC source to help disguise the vanishing SBD source.  It worked fine, and in the absence of any other way to do it in CEP2 I'll plan on working with 50% baseline envelopes on all my future multitrack work.  That said, if somebody stumbles across this thread and knows a better way to do it in CEP2, I'd love to hear it.