Thoughts on comping solos from multiple takes

[Billster]

OK, so there was some discussion on a message board for my DAW about editing techniques for creating a comped solo.

 

The guy asking was pointed to some different methods of taking pieces from multiple takes and comping them into a single solo.

 

He replied "thanks for the tips - those look kind of complicated, I have some homework to do"

 

So I cheekily replied "Is learning the editing any more difficult than learning to play a solo from start to finish?"

 

Nobody ran with it. So polite.

 

Anyway, no prejudice - I've done solos both ways. Sometimes I comp, other times I bust my knuckles to get a complete performance in one take.

 

SO, what say you? For taking an extended solo with no fluff, dead spots, or outright clams, is it harder to edit a performance together from multiple takes or to perform from start to finish with no junk?

[Dr. Ellwood]

.

[Billster]

I didn't ask that - I asked which is harder.

 

Though I think I can predict your answer.

[Eric VB]

Well, if you're Lasse Gjertsen (the "Amateur" YouTube guy who splices together single-note recordings), obviously it's easier to edit.

 

I liken this question to the video game "Guitar Hero". You can spend a lot of time and master the game, but will that make you a better guitar player? In the latest issue of Blender, some 11 year old kid completely trounces a pro guitarist in a head-to-head match of "Hero". But who's the better guitarist?

 

As Bill@Welcome Home Studios would ask, "Do you want to be a guitarist or an engineer?" You can hone your studio skills such that a multi-comp'd solo is no sweat, but will that make you a better guitar player?

 

How will your audience react when you can't play the solo as well live as the version they've heard on the studio cut? How will you be received in someone else's studio when you explain that they will have to record you multiple times and spend a good amount of time splicing together the best parts? You wouldn't want to have to fire the guitarist now, would you?

 

However, you might be able to impress at least some studio engineers with your facility for editing.

[Gruupi]

If its a 2 chorus rock solo, then I think one take is the way to go. Its just going to sound fresher and have more balls if its one take. Most of the soloes I have done on rock recordings are "composed" anyway so I learn them then play them. So in this format its easier to do one take.

 

For classical pieces, I have a hard time playing with the red light on. Its not that I can't play the piece through, its more like my playing becomes really stiff and lifeless if I am playing not to make mistakes. I would like to be able to one take these but in reality I try to get a section down that really sounds good, and I almost always end up editing takes together.

 

For the 3 minute improvised jams like our forum jam tracks, its either or. It depends on the day and the track. Some of them are one takes, others are complete edit jobs with many different things spliced together, some are mostly one take with a couple of punch ins. Its whatever works. I want my recordings to sound good, I use whatever tools I have to get the best record of my playing abilities and use the recording to analyze my playing for further improvement.

 

With computer software its really easy to splice things seamlessly, some of my single takes sound more disjointed than pieces I have edited. I understand Elwood's thinking on developing the ability to play one take, it will most likely sound more realistic. But if you are doing so to proove that you can rather than trying to get the best recording possible its not always the best. Its like the old Steve Martin joke "not many people realize that Leanardo pianted the Mona Lisa in one stroke".

 

 

[fantasticsound]

I'm of the belief that both methods are valid, but I try to keep any comp'd solos to at least phrases, not individual notes. I like to improvise and glean the best takes. Frankly, I need more practice to play solos clean. I wish I were a better player, but the fact is I'm not.

 

The beauty of comping is the best ideas from several passes can be used as one. I know from experience that re-recording the complete solo in one take rarely gets all the nuances that really make each phrase great, as originally recorded. I keep practicing to play the final form of the solo as close as possible to the comp'd parts, but it would be foolish to settle for less than spectacular when the recording is being done on a time-table and I can spend more time, later, learning to play it as recorded.

 

I think both are equally hard. Sometimes several tasty licks are wonderful over the rhythm tracks, but can't be connected in a coherent way. You have to decide on one or the other and find a new part to fill in the blank.

 

Playing a great solo straight through is difficult for a number of reasons. Admittedly, I'm not a great improviser. I don't have a large vocabulary of musical phrases, so it's easy to fall into playing a familiar melody I've used before. I'd love to be able to play great solos from start to finish, but I'd also like to be a great lyricist and arranger. Wishing ain't gonna make it so. Practice and study help a lot but I've been at this the better part of 32 years. If I haven't figured it out yet the likelyhood is I'm going to have to settle for working within my abilities, at least for the most part. (I'm not going to become a world class athelete at age 40 either. You can always push your limitations but physics and biology can only be overcome to a degree. I'll also never have the reach of Allan Holdsworth. His fingers are literally twice as long as mine. Should I beat my head against the wall trying to play chords I physically can't reach?)

[miroslav]

Yes...I do both comps and single-take stuff.

 

With comps, I've done 2-3 takes of the exact same lead...and then taken the best portions to create one.

And it's not really a question of "learning the solo"...

 

Often it's the tone and attack/release variations that you will get with 2-3 takes. All three takes can be perfectly executed....but each one has a few outstanding elements that has nothing to do with leaning or playingit's just the way it went down.

 

So...I pull those best pieces and meld them into one.

 

Ohand to really edit well in a DAWyeah, its not easy and it IS very time consuming.

When even you cant notice where the splices arethen youre onto something.

 

 

 

[Gifthorse]

Usually for me, the problem isn't getting an on take it is getting one that I like.

 

Sometimes I will do 2 solos and then if the beginning half is cooler in 1 then we will splice them together. Depends on what kind of song it is.

 

By the way Bill, that one solo is killer.

[Griffinator]

It's definitely easier to comp it together, AFAIC, but I choose the complete solo route because I figure I'm going to have to play it live sooner or later, so may as well get that muscle memory done now...

[Billster]

By the way Bill, that one solo is killer.

 

I know which one you mean from the e-mail you sent, and that's a single take - no comp. Thanx, I am proud of that one.

[Chunk]

Well, if you're Lasse Gjertsen (the "Amateur" YouTube guy who splices together single-note recordings), obviously it's easier to edit.

 

off topic...I watched some of "Lasse Gjertsen" other stuff the "Amateuer" is cool, however some of his other stuff is rather weird.

 

[Billster]

It's definitely easier to comp it together, AFAIC,

 

I think that's right, once you're past the learning curve of the editor.

 

Improvising with sustained intensity for an extended time and developing some sort of emotional arc and structure is definitely a big challenge.

[Mudcat]

I'm of the belief that both methods are valid, but I try to keep any comp'd solos to at least phrases, not individual notes. I like to improvise and glean the best takes. Frankly, I need more practice to play solos clean. I wish I were a better player, but the fact is I'm not....

 

I think both are equally hard. Sometimes several tasty licks are wonderful over the rhythm tracks, but can't be connected in a coherent way. You have to decide on one or the other and find a new part to fill in the blank....

 

Ditto.

 

What's really challenging is comping a solo live to your final mix on analog multi-track. I had to do this with a class project back in 1985 after my sax player showed up after burning a bunch of "hippy lettuce" and was barely able to play at all, let alone assemble a musical sounding take for his solo.

 

My second engineer and I had a lot of fun memorizing and executing all of the punches we had to execute on the three tracks of sax we recorded. We had to do it live to the final mix because there was no open track available for a bounce. It would have been a lot easier on a DAW.

[fantasticsound]

LOL, 'cat! I remember one of my band's demos we recorded at the drummer's (parents') house. Unfinished, concrete basement. We had blankets hanging from the ceiling to separate our amps from the walls and each other, then all of us stood in the "room" with the drums to record the four tracks of rhythm. (Stereo drum mic setup with bass fed to both tracks equally, single tracks for each guitar.

 

We took the tracks back to the singer's home studio to mixdown to stereo, then overdubbed keys, vocals and solos. There was one solo which I really wanted to tap the beginnning to. After several takes with all kinds of hand/string noise between the tapped and picked sections, I finally enlisted the other guitar player to mute any strings I wasn't tapping right up until that section was complete. Worked like a charm and we didn't have to try and fly in a quick run behind the end of the tapping.

[Mudcat]

The Marine Corps' other motto: "Improvise, Adapt, Overcome."

[fantasticsound]

Hey 'cat, I uploaded a snippet of the song, Nothing To Fear to the Neil Bergman, MySpace account. It was from 1990, so excuse the haircut in the accompanying avatar. The digital file was taken from importing a stereo cassette to the computer via the 1/8" soundcard input from a 27 year old cassette player in dire need of a bias adjustment, so the quality is really bad, but I think you'll get the picture. The solo with better clarity is the overdub. The harmonized leads after it were recorded live with the rhythm section, no overdubs. Have a good laugh, guys.

[Gifthorse]

I think some songs call for different approaches. Depends on the kind of band you are, and the kind of songs you write.

 

[miroslav]

Before the DAW...I would just record a guitar track over as may times as needed to get it just right (or any other type of track for that matter).

 

These daysI still track to tape, but thanks to the DAW, I'll only re-record until I know I have a couple of solid takes...never more than three...and then I dump everything into the DAWand massage it.

 

I find that if you have two or three near perfect takes...there's really no reason to beat on it for 12 more takes.

Heck...it's less traumatic tracking that way...

.BUT...you end up spending that saved time/effort on the editing portion.

 

While I can REALLY hate getting into a drawn out editing process...in the end, it really CAN get you that extra 10%-15% of quality that otherwise would be out of reach...or just too time consuming trying to get it without any editing.

 

I don't feel like it's cheating if you comp a couple of takes.

Heck...YOU played both of them...so why sweat it just because you accidentally clipped one of the strings in take 1...and you can easily fix it by using that section from take 2.

 

[Gifthorse]

I agree with you Miroslav, unless you just aren't happy with your 'perfect takes'.

 

I care more about making myself happy when I do solos. We use majority rules on everything so it can be frustrating sometimes. You know cuz everyone can play a little lead. So they want to get in there after 2 takes and tell you what to do. I am open to that so it works out, but I would rather just keep rolling if I am not getting something I like.

 

I usually do get my best takes in the first 1 or 2. I just want solos that make the song better, not worse. Some beer usually loosens me up. Shuts down my brain a little, which is a good thing when I am improvising cuz I am letting my ear guide me instead of my muscle memory and technique. I don't drink and drive though.

[Guitarzan]

i tend to want to do a full take and keep the inspiration complete but sometimes a clam will mess up a good run.

on the other hand since i don't actually do the editing of our stuff there are times that i am less than pleased with the end result.

our other Guitarist tends to chose different stuff than i would.

but then sometimes he makes me sound better.

i am not against piecing things together but i prefer the whole thing to be what my mind is sending to my fingers at the time.

[Gifthorse]

I think it is hard to tell whether a solo is planned/live, or put together by an engineer if it is on time and well executed.

 

I am of the opinion that good technique is part of feel which is part of conveying an emotional phrase dynamically.

 

Zan, nice solo on your son's song.

[Guitarzan]

i had to go back and listen to that. thanks i thought it was kind of simple.

just trying to do something that was un rock and roll. the harmony obviously was a patch on the GT6.

[Squ]

One thing to do is do a bunch of takes, piece together what you like, THEN you learn it, and redo as one take. I've that Satch does this.

 

All methods are fine though.

 

Hendrix comped his solos sometime, like on "Come on"

And of course he played them live with the whole band...

 

As did Van Halen on the first CD, a lot of the leads were live with the band.

[A String]

One thing to do is do a bunch of takes, piece together what you like, THEN you learn it, and redo as one take. I've that Satch does this.

 

That's how Gilmour did it. He claims that he would do 15 (or so) different solos over a song, then pick out the parts he liked, splice them together, play it back a few times so he could learn it, then record the "new" solo in one take. All of his famous solos were done this way.

[Billster]

One thing to do is do a bunch of takes, piece together what you like, THEN you learn it, and redo as one take. I've that Satch does this.

 

That's how Gilmour did it. He claims that he would do 15 (or so) different solos over a song, then pick out the parts he liked, splice them together, play it back a few times so he could learn it, then record the "new" solo in one take. All of his famous solos were done this way.

 

I'm not sure I like that. I'm 50-50 to the concept. I mean, it's grounded in improv, but then becomes compositional with the re-learning.

 

When I'm working out a solo, I'll practice the tune and come up with the concepts I want to use, such as shifting from mode to mode, certain melodic resolutions and turnarounds, fast vs. slow, and so on. As I practice, I'll catch certain things that are definite notes or phrases to use at exact places, but other areas will be more concept driven like "do some Lydian stuff here". I'll practice this stuff for multiple sessions.

 

When it comes time to record, I'll shoot for whole takes that incorporate my definite ideas, and don't contain any train wrecks during the looser areas. If I have to comp something because I didn't get a good combination of the definites and open concepts, I won't re-learn the comp and try to get a "perfect" take that imitates it. To me, the re-learn and imitate takes away the spontaneity of the conceptual areas.

[miroslav]

To me, the re-learn and imitate takes away the spontaneity of the conceptual areas.

 

What's the difference if you play/learn a solo during a whole bunch of dry-run passes....

...or if you actually record those passes and then sit down and pick out the most interesting elements and assemble them into a really well-structured solo...

...and then go back and just record that as the final...?

 

It's really the same thing.

 

If you wanna talk about true spontaneity...then just pick-up and play on the spot...no rehearsals...and keep the recording as it falls.

But who the hell does that...?

 

I think we ALL rehearse what we will play/record....and if you are rehearsing something, then you can't talk bout some spontaneity later on, since the finished solo was cobbled from a bunch of rehearsals...or a bunch of recorded takes.

 

It's the same thing.

 

I think that when you just let yourself go and riff away and record all of that...

...then sit down and pick out the most interesting stuff to assemble out of that...you will end up with MUCH MOR complex and better developed solos that when you try and just play things off-the-cuff.

Often...when you just play cold...your fingers will want to mostly cover well known ground, patterns, runs...etc.

But when you are not playing....and instead assembling...you are then TRULY composing a solo...and often you will come up with neater stuff....IMO.

 

[Billster]

You can have spontaneity within a framework, even if you know what chords and scales you are addressing.

 

Really.

[Squ]

I'm not sure I like that. I'm 50-50 to the concept. I mean, it's grounded in improv, but then becomes compositional with the re-learning.

 

Mmm... at the end of the day, a good guitarist can get a good solo out of a session regardless of the exact method used. I certainly think Gilmour has done some awesome solos, and I'm not going to think less of them based on how he did them.

 

Also, "Spontaneity" doesn't automatically imply "good solo," and the fact that you relearn something that may have come out spontaneously, I would say maintains that spontaneity. I've definitely done takes where something new pops out, and I listen back and think "Well that was cool" but I'm not exactly sure what I did, and of course the rest of the take isn't great. So I learn what I just played, and I have another lick in my bag. Yay!

[Eric Iverson]

I think it's perfectly legit to piece a solo together, but it IS hard to get it to sound natural that way! So Gilmour's method seems well worth trying!

 

I've worked up solos to tunes we'd written, when it was only going to be a chorus long and there was no time to develop anything as an improviser. And sometimes ended playing something different - but the time invested in working up a solo was still time well spent! It helped me get past the lazy lead player's aimless noodling syndrome (ANS for those in the know!)

 

One recording session I played at they had me record a solo literally bar by bar using Pro Tools. At the end the session they said I did a good job, but it sure wasn't at much fun as just going for the gusto! But it wasn't my tune or my session - the singer wanted to bring the demo to radio stations. They had a guy on the session doing some great stuff on a Puerto Rican cuatro - just rattling it off effortlessly. A humbling experience for me, but that's cool. At least I got paid!

 

I must say I admire guys like Django and Wes who could play alternate takes with solos just as good as the ones released, and completely different. Not everybody can do that off the cuff!

 

 

[miroslav]

You can have spontaneity within a framework, even if you know what chords and scales you are addressing.

 

Really.

 

Yeah...and what is a composition that you create in a DAW from guitar solo pieces...a.k.a. track comps?

 

It's a framework.

 

So like I said...it's the same thing.

 

Really.