hi I’m a noob at comping. so if I record multiple takes stacked mode into a lane, what’s the best method for comping it? I’ve heard people talk about crossfading parts how would you do that when only one audio piece can be playing at a time? do you have to use multiple tracks?
Here is what to do to begin with - a very basic approach which, after you are comfortable with this, you will develop your own style of sorting, choosing and implementing those choices:
Let’s suppose you’ve got One vocal track, with 6 lanes of recording. Each of the lanes has the same material.
Look at those wave files … you will see gaps stacked more or less vertically. those will be there the singer has paused for punctuation or to breathe.
Turn off ‘snap to grid’, and Select your scissor tool. With the scissor tool, cut vertically through all six lanes at a point of common silence. Go to the next point of common silence and do the same again.
From now on, you will work with your select [arrow] tool, and your mute tool.
Play a small section … between two points of common silence. If you have not yet muted anything, you will hear the bottom - 6th- lane. If you mute the bottom lane, you will hear the 5th lane. Mute that and you will hear the 4th. And so on until you only hear the first.
As you listen repeatedly to that section, mute and unmute in order to make judgement as to what is the best take. Now with your select tool, drag it to the bottom of the pile … as if you are creating a 7th lane for the ‘chosen ones’.
To make comfortable the act of listening to a section repeatedly, set left and right locators at the beginning and end of the section between common silences. Choose Cycle for playing.
Apply that process to the next section between common silences.
And so on to the end of the song.
Here ends the simplest way. There is no cross fading. You are comping discrete phrases.
As your look at you ‘Chosen Ones’ … that ‘Lane 7’, you might want to tidy up those sections by trimming unwanted background noise of breathing sounds, by using the select [arrow] tool to on the part’s lower left and right corners, to drag them shorter.
Note … you might WANT to keep some breath sounds. To delete silence needs to be as mindful a decision as keeping it in.
Beyond that, there are the ‘finesses’
For example “You always been my Slag but toniiite I’m gonna make you my Biaaaaaaatch!!!”
“Biaaaaaaaaaaatch” is a challenging word to sing.
B threatens plosives
tch is at risk either of spitting or of getting lost.
the dipthong transition from iiiiiii to aaaaaaa demands fine control and there are serious questions to be considered as to where the stress should be.
biaaaaaaaaatch, or biiiiiiiiiiiiiiiiatch
A further parameter may come into play if the vocalist, on one or two takes, turns the sung note into a spoken or snarled performance.
Unless the vocalist is present for you to tutor him to do a ‘final’ final take, you might be alone in your studio with the best first half and the best last half of the same word!
At that point, what you were saying about crossfading will be important.
A further finesse: Suppose you have gone through the entire song, choosing the best bits. Now you listen through, and fine that the best bits of the third chorus were infinitely better than any in the first chorus? Then you might want to copy-paste part of whole of the third chorus’s best bits into the first.
There you go … I hope this helps.
But that’s just the beginning:
Cycle record … i.e., creating those 6 lanes … might NOT be about doing the same words on each run. With singing and especially with electric guitar solos, you might be improvising and thus generating entirely different content, with different phrase lengths and gaps.
Here, you will have done different sections of the song, each section betting a rich variety of ‘fitting’ content, which Later, you will use as ‘prime source’ from which to build final ‘lines’ by means of cutting chunks and coordinating and sequencing them. A one bar-long vertical stack of 15 lanes may well contain the material which will form , say 32 bars throughout the song.
Take, for example, the main riff of Cream’s ‘Sunshine of your Love’ …
du du du du du
Du Du Du DUUUUUH Du
du du du du du
Du Du Du deeeeeeeh Du
They might have been cycling a 4/4 with 16 closed hats to the bar, kick on 1 and 3, with snare on 2 and 4. Bass doing 8 to the bar on open E.
Guitarist might have been chugging away on du du du du du for ages before coming up with Du Du Du DUUUUUD Du. Scissoring between them might not have been on his mind. The quantum leap of Du Du Du deeeeeeeeh Du might have been at best an afterthought and at worst, an accidental jump of one octave. Jack Bruce might have given him a dirty look and then they all go home. Into the small hours then might get made the decisions which generate the immortal riff, and that might end up copy-pasted throughout the song. Similarly with Smoke on the Water.
Cycle record and comping method enable you to create your own ‘Construction Kit’
I believe this use of lanes and cycle record is not possible in Cubase 6, which, so I hear, lets you draw selection ‘curtains’ to reveal and audition one-lane ‘choices’ whilst keeping all else muted, as you work your way through the song. You don’t have the freedom to chop and mute at Your Own Whim.
Other posters will need to correct or confirm this.
I don’t know why but I’ve never felt comfortable comping. I would rather play certain sections and, if I don’t like what I played, Stop + Stop (again, to go back to the point where I started) + Undo + Record and do it again. I will, however, split the songs into key sections if needed and record those individually rather than try to do the entire song in one take.
If the take for that section / song is generally good save for a few flubbed sections, I will punch-in / punch-out and apply very tight crossfading to make the transition unnoticeable. But my rule has always been that I have to finish the entire track (not song though) in one sitting so that I’m in the same mood, same EQ, same everything. This way the playing sounds consistent. (Of course, I may erase the track at a later date and redo it…)
All of this is because I try to focus on the overall sound I’m looking for rather than for perfect playing and I feel that, if I comped, I would lose the sight of the forest because I’m focusing on the trees.
Now that I’ve written my manifesto , I will add that I may try it now just because I’m curious about it from the way you are all talking about it.
wow thanks for the posts guys that’s exactly what I needed
Thanks for the informative post Zenda.
I posted this elsewhere but got no replies… thought I might try my luck again within this thread as it seems relevant…
Several people seem to be dissatisfied with the new lane behaviour but I don’t think this question has been addressed directly elsewhere.
I’ve just moved from C5 to C6 and am incredibly confused by the new lane functionality which for me appears to be inferior.
When comping in C5 I would do the following:
- show lanes
- audition takes
- cut segments as desired
- delete the segments that I didn’t want
- resize remaining segments (of my comped “perfect take”) so that the segments overlapped
- apply crossfades
- resize segments / customise the crossfades to give the best sound
In C6 I can’t arrange the segments so that they overlap and hence apply crossfades as I want them.
The new lanes behaviour appears to make comping more straightforward but so far I’m getting inferior results. Am I missing something (i.e. a way to apply crossfades)? Is there another method I should be using for comping? Or is this a loss of functionality?
It’s great that the new functionality is so easy to use and that I don’t need to use crossfades, however sometimes, I do need to use them to get a superior sound.