How much digital editing is there on studio drum tracks?

[Yodelgoat]

I've been laying down drum tracks lately - using real acoustic drums. its a relatievly new thing for me. When I complete what I consider to be a decent take, I find myself tweaking the placement of some of the sounds. I know that on most recordings these days, some editing of drum tracks is considered acceptible. I was just wondering where some of you are on this. I find it interesting that so many drum parts are now totally digitized - either loops or from some kind of sampler. You can usually tell when its a machine. It becomes almost too perfect ( I do think there is such a thing). I used to use a drum machine, but found that the music seems more authentic when I use real drums. How much editing can a person do on acoustic drums, before he can be accused of (the horror!) "using a drum machine"?

I have become a huge believer in real drums, even though its far more time consuming and perhaps less "perfect". Its exactly the opposite of where most music is today. More and more I find myself migrating back to the good old days of Analog Mixers, Outboard processors and cassette tapes (okay, not cassette tapes) but is there some validity in avoiding the heavy use of digitization in music? Admittedly, earlier it was easier to use the machines, and I didnt have a full studio at my disposal, but I sure feel better about using the real things. Real Bass, real Drums, I still do some MIDI for some key parts, but I am even considering losing that as well.

How much computer generation is too much?

[Mikhael]

I think that's up to the user to determine. I may punch in a few things, but I'm one who likes to keep not only the human feel, but the feel of a group of humans playing together. There's an indefinable something that happens when a group of humans are all linked together and firing on all cylinders.

When recording by myself, the first thing I do is lay down the drum tracks. It's hard, playing through the song in my head only, while playing the drums (and usually the cause for a punch-in or two), but then every other part references the timing to the drummer, which is the way it works in a real live band playing.

The next thing for me is the bass, and that's pretty easy. But after that, I'll start sequencing the keys: I use a sequence because I'm not that good of a keyboardist, so I can record a couple of tracks of simple things, then concatenate them together to sound like a decent player. Also, I can fix a few flubs, AND change the patch I'm playing if I decide I don't like it. But it's still essentially a recorded performance, for all that.

Then comes guitar and vocals.

[Rarebird]

Well, once a drum-playing friend stated one of my pieces of music (all done with the computer) sounded like there was a band playing.

[trurl]

Use your ears. If it sounds right, it's right. If it sounds wrong, it's wrong.

[fictionmusic]

Use your ears. If it sounds right, it's right. If it sounds wrong, it's wrong.


That;s my criteria as well. There are tracks I have done on midi which no-one knows isn't a drummer, and there are tracks I have done with a drummer that no-one realizes is edited. It all depends on what works and what the budget and time constraints allow.

[Lebofsky]

I'm all for editing and cleaning up using the tools available. HOWEVER, you have to be tasteful and delicate. It's way too easy at first to make some edits and fix one bad note, like a misplaced snare hit, and think the fixed snare sounds amazing, without realizing by moving the snare around you created a phase problem, or messed with some cymbal decay bleeding into the snare microphone, etc. etc. etc.

In short, once you start editing stuff, you start hearing other people's edits a mile away, and it sucks to hear other people's bad edits, and it sucks to have other people spot your edits. I find most of the time the best bet is to just rerecord the part correctly. Or at least record several takes to a click track (if you don't find that objectionable) and piecing those together takes has far less problems with note placement, feel, cymbal decay...

This all said, there's drum replacement, which is a whole other ball of wax, and something I've been getting into more and more (replacing analog drums with digital drums based on the real performances). This is incredibly useful for fixing bad sounds, but still no solution for bad playing. Nothing will ever beat taking the time to practice the parts, using good microphones, and placing them properly. But editing/cleanup with computers sure lets you get reeeeeally close once you get good at it.

- Matt

[trurl]

I'm all for editing and cleaning up using the tools available. HOWEVER, you have to be tasteful and delicate. It's way too easy at first to make some edits and fix one bad note, like a misplaced snare hit, and think the fixed snare sounds amazing, without realizing by moving the snare around you created a phase problem, or messed with some cymbal decay bleeding into the snare microphone, etc. etc. etc.

In short, once you start editing stuff, you start hearing other people's edits a mile away, and it sucks to hear other people's bad edits, and it sucks to have other people spot your edits. I find most of the time the best bet is to just rerecord the part correctly. Or at least record several takes to a click track (if you don't find that objectionable) and piecing those together takes has far less problems with note placement, feel, cymbal decay...

This all said, there's drum replacement, which is a whole other ball of wax, and something I've been getting into more and more (replacing analog drums with digital drums based on the real performances). This is incredibly useful for fixing bad sounds, but still no solution for bad playing. Nothing will ever beat taking the time to practice the parts, using good microphones, and placing them properly. But editing/cleanup with computers sure lets you get reeeeeally close once you get good at it.

- Matt

I'm personally much more a fan of doing multiple takes on drums and doing edits from them for the reasons listed above. In other words, if you blew a kick drum hit on your main good take 2 in verse 1, grab verse 1 from take 3. That way you can generally get a good clean edit across all the tracks and you're going from one performance to another and cheating less.
On that note, I'm fascinated by listening back to songs from the 70's where you can tell they took sections on drums and spliced them into loops essentially, like Fame by David Bowie- the song goes through once and when it goes back to the intro you can tell it's the identical drum performance all the way through again. I suspect Aerosmith did that a lot as well...

[Polypet]

:up @trurl

back in those days, we were doing that shit with razor blades too, which made it even more exciting :lol

the tools are nice, yes - but as Matt says, nothing beats practicing and recording good performances. digital tools make certain things easier but it can be both a blessing and a curse to have that sort of power (witness pitch correction). they're great if one can restrain oneself from going nuts with them.

needless to say, when you have to make loops or edits and comp takes using the old-fashioned razor blade method, it's sort of self-limiting :)

"digitally record" wisely (to paraphrase the liquor industry),

Kim

[Yodelgoat]

Having recorded with MIDI drums before, and gone back to real drums, I find that you really notice things about machines and samples, like cymbals always sound exactly the same - same velocity, striking the same part of the drum or the cymbal. Funny how when a really good drummer plays he gets the same result. I call that Talent, but when a machine does it - perfectly, somehow I bemoan the lack of human feel. When I hear real drums I enjoy the fact that almost every hit is a unique sound. I understand that some software can actually do this as well.

Now drum replacement? I hear that is just amazing. Each drum gets assigned a better quality drum sound, and its supposed to really work. Whoa, that blows my mind just a little.

[B D]

When I'm recording a drum track myself, if I'm messing something up it's generally always the same spot in the song, for whatever reason. After playing it through a few times and sounding good but always messing up in that one spot, I stop worrying about that spot and do a take that feels good everywhere else, not worried if I stumble in "that spot". Then I'll do some takes not worrying about what happens everywhere else in the song, and concentrate only on the troublesome spot, sooner or later get a good one and use the best bits of both takes. By then I may have understood the problem, so just for the hell of it and for my constant striving to improve my musicianship, I may try it a couple more times with the new knowledge, and try to get one complete take just because I like that idea and it pleases me, though I am not a "one-take" purist by any means.

[Kim Olesen]

I'm all for editing and cleaning up using the tools available. HOWEVER, you have to be tasteful and delicate. It's way too easy at first to make some edits and fix one bad note, like a misplaced snare hit, and think the fixed snare sounds amazing, without realizing by moving the snare around you created a phase problem, or messed with some cymbal decay bleeding into the snare microphone, etc. etc. etc.

In short, once you start editing stuff, you start hearing other people's edits a mile away, and it sucks to hear other people's bad edits, and it sucks to have other people spot your edits. I find most of the time the best bet is to just rerecord the part correctly. Or at least record several takes to a click track (if you don't find that objectionable) and piecing those together takes has far less problems with note placement, feel, cymbal decay...

This all said, there's drum replacement, which is a whole other ball of wax, and something I've been getting into more and more (replacing analog drums with digital drums based on the real performances). This is incredibly useful for fixing bad sounds, but still no solution for bad playing. Nothing will ever beat taking the time to practice the parts, using good microphones, and placing them properly. But editing/cleanup with computers sure lets you get reeeeeally close once you get good at it.

- Matt

+1

[Jacob Holm-Lupo]

To answer your original question: Tons. Quantizing acoustic as well as electronic drums these days is really the norm. It can be done using programs like beat detective or by "manual" editing of drum hits.

The main problem in my ears is that the more precisely the drums are made to fit the grid, the more you loose a sense of groove and propulsion. Most drummers would agree that being able to play precise, to keep the time and to place your hits precisely, is extremely important. But most drummers would also agree that to become a really good drummer, you have to move beyond that, and allow yourself to dance around the grid rather than to be "on" it all the time. Steely Dan said something about that once, that to be a great musician you have to strive for perfection, and then move beyond perfection, to a place where the rules are internalized and you are free to play with them. That's when music comes alive.

THAT's what's lacking in today's quantized music world. IMO.

When I recorded my band's "comeback" album a few years back (White Willow's "Terminal Twilight") I made a decision to leave the drums alone. I didn't to any time-based editing on the drums. You can hear the tempo swinging a bit at times, and you can pick out the odd rushed note. But I think it's all the better for it. The drums on that album really breathe. I'm using the same philosophy for a new album I am recording with another project now. Once you decide to let a performance be a performance, rather than "raw material" for extensive editing, interesting things happen to the music.

The one thing I do occasionally do, however, is to use triggered kick drum samples that I layer underneath the acoustic kick. It's not really audible, but in a busy mix it helps the kick cut through in a way that a purely acoustic kick just can't. People have become so used to hyped, loud drums these days that you have to cater a tiny bit to their expectations.

J.

[fictionmusic]

To answer your original question: Tons. Quantizing acoustic as well as electronic drums these days is really the norm. It can be done using programs like beat detective or by "manual" editing of drum hits.

The main problem in my ears is that the more precisely the drums are made to fit the grid, the more you loose a sense of groove and propulsion. Most drummers would agree that being able to play precise, to keep the time and to place your hits precisely, is extremely important. But most drummers would also agree that to become a really good drummer, you have to move beyond that, and allow yourself to dance around the grid rather than to be "on" it all the time. Steely Dan said something about that once, that to be a great musician you have to strive for perfection, and then move beyond perfection, to a place where the rules are internalized and you are free to play with them. That's when music comes alive.

THAT's what's lacking in today's quantized music world. IMO.

When I recorded my band's "comeback" album a few years back (White Willow's "Terminal Twilight") I made a decision to leave the drums alone. I didn't to any time-based editing on the drums. You can hear the tempo swinging a bit at times, and you can pick out the odd rushed note. But I think it's all the better for it. The drums on that album really breathe. I'm using the same philosophy for a new album I am recording with another project now. Once you decide to let a performance be a performance, rather than "raw material" for extensive editing, interesting things happen to the music.

The one thing I do occasionally do, however, is to use triggered kick drum samples that I layer underneath the acoustic kick. It's not really audible, but in a busy mix it helps the kick cut through in a way that a purely acoustic kick just can't. People have become so used to hyped, loud drums these days that you have to cater a tiny bit to their expectations.

J.


here here!

For the Rebel Wheel we made a concious decision not to use a click track for the last two albums. You can here the tempo move a wee bit (and as I tempo mapped everything I saw just how much and how little that movement was), but I love the feel of musicians interacting with each other and not with a clik track. On the last one, the band recorded live as much as possible, but due to the shifting line-up, I had to track bass with the kit in a few and then replace (or not) with whoever the official bass player turned out to be. I added my guitar, keyboard and vocal parts later. Right now I am in a total live off the floor mode and our next album (to be finished soon) is exactly that.

In other projects I use cliks and I admit I have used the flex function in Logic on a project I just did (where the writer wanted the tempo to increase through-out). We had two songs that did that actually and one was done after the event and one during. I have edited the shit out of projects but far prefer the live-off-the-floor-strive-for-the-unedited-performance a al jazz and classical (or what jazz and classical are perceived as being anyway)

[Yodelgoat]

I have recalled times when using quantization, that it didnt sound right. There are natural pauses, or slight changes in music as it builds in intensity, or slows ever so slightly. Thats why they used things like Tenuto, Staccato, Spicatto, retard - ways to give music more freedom. I havent seen much of that in modern music and I really only know it from having a very thorough music teacher in high school. I dont think most modern MIDI utilizes those lesser known music notations. Certainly Quantization would totally destroy any attempt at using them.

I do have some noticeable clunkers in my drum parts, and I dont mind them within a certain degree, but I just wanted to hear some opinions on just how much editing takes you over the top. Over the past few months of doing alot of drum work, I have reached the point where I thought I would drive myself insane with attempting to clean up beats that are a little too human.

[bp]

I far prefer working on live recordings compared to the perfectionist nature of some studio work. It is even better when it is a live show with video. Usually at that point there is no grid and if you move things around you lose sync to picture.

Regarding drum replacement, I avoid it like the plague. I like to mix drums using the overheads as the first element I bring up and then fill in the close mics. For me it tends to give a more natural sound to the kit. If I must replace a drum due to a bad cable or mic, I listen to the pitch of the drum in the overheads and make sure that I match the replacement's tuning and decay as close as I can. If it is an issue where the drummer didn't quite hit the snare well for one hit (hitting the wrong place on the drum), I'll grab a good hit of the real drum and paste it over the bad hit making sure that I get the timing to match what is in the overheads.

For studio sessions at some point in the tracking day I may have the drummer go through the drums and cymbals one at a time and record a number of articulations for each drum or cymbal with all tracks rolling. That way I can create samples of the real kit used if absolutely necessary with the mixer settings and all the natural bleed intact. I haven't had to use those samples often but if the arrangement changes and I need to create a new drum "performance" for an added bridge or other part I can. This only is done if the drummer or studio is unavailable after the basic tracks are recorded. If you go down that road you get into the business of creating a sonic illusion. That can be perfect if the genre calls for it or can suck the soul and energy right out of it.

All that said, my favorite recordings that I've done have been live to two track with a great, well rehearsed band. Nothing beats that. Those are the recordings I revisit most often because I didn't get burnt out making them over weeks. It becomes a document of a moment in time and all the interaction of great musicians giving their all. If I can capture all of that I am a very happy man.

I have been blessed to work with some great drummers over the years. One of my absolute favorite gigs was mixing George Duke for a TV concert. He had Leon "Ndugu" Chancler drumming that day. During soundcheck I first brought up the overheads. As I brought up the kick mic I could have left it there and it would have been really good. His touch, tone and balance between all the elements of his kit is the very best I have ever experienced. It was an easy day of work that still makes me smile when I think about it or listen to it.

[trurl]

His touch, tone and balance between all the elements of his kit is the very best I have ever experienced. It was an easy day of work that still makes me smile when I think about it or listen to it.
A good drum sound is mostly about a good sounding kit and a good drummer... GIGO! :D

[Gongtopia]

Practice hard. Play the part right. There are no substitutions for a work ethic.

When I use a click (and I use one most of the time), it's usually a funk or hip-hop drum part instead of just a 'click.' That way I have some sort of groove to work with, to play over/under/around. I'm not against making multiple takes and then combining them to form a master take, but I hate 'punching in' something small that should've been played right the first time. Still, if I have a great groove/feel, I'd rather add a punch in than lose the groove on another take. I'm also not against manipulating/moving parts to create what I want. Many of the things I currently am doing are what I call 'collages,' where I record a track, then cut it up and rearrange everything into a new track. Whatever it takes…

[Yodelgoat]

Practice hard. Play the part right. There are no substitutions for a work ethic.

When I use a click (and I use one most of the time), it's usually a funk or hip-hop drum part instead of just a 'click.' Whatever it takes…

Good suggestion. I use a very basic click - 1 click per measure, I generally have a hard time staying on that. A groove track instead of a click may be just what I should be using. With the changes in tempos - is it possible to get a groove track in sync with an existing multi track? Just find a groove and then let the MIDI do its thing?

[Dean Watson]

I think people spend too much time on what they feel is right ( you should have a real drummer ) or wrong ( programmed drums ), rather then accepting or rejecting the music as a whole. I use programmed drums. I'm still not very good at it, but I do spend more time on that then any other aspect of the music - and I enjoy it immensely. I do get comments all the time - pffff "you need a real drummer Man!". My take on this is - if you like what you hear, regardless of how it was 'assembled' then is IS right. If you don't like it, if it doesn't move you, then don't listen anymore.

[sonic]

I have recalled times when using quantization, that it didnt sound right. There are natural pauses, or slight changes in music as it builds in intensity, or slows ever so slightly. Thats why they used things like Tenuto, Staccato, Spicatto, retard - ways to give music more freedom. I havent seen much of that in modern music and I really only know it from having a very thorough music teacher in high school. I dont think most modern MIDI utilizes those lesser known music notations. Certainly Quantization would totally destroy any attempt at using them.

You can do all that with midi if you're willing to put the time in. Of course, it would be faster to have good players play the parts. :)

[Yodelgoat]

You can do all that with midi if you're willing to put the time in. Of course, it would be faster to have good players play the parts. :)

By far the most enjoyable part about writing music for me, is the creative process. Last night I spent about 2 hours trying to get a rather challenging drum part right. It was only about a minute long. I'm sure I could have hired a guy to come in and play it, and saved myself all that grief, but a couple of observations should be made here: First, I'll never reclaim any investment I make in music, therefore it will never be cost effective to pay someone. Secondly, I am a better drummer as a result of all the practice - which actually is the point of the whole exercise. When I started out, I could barely muster the speed to play the part, full of clams. By the time I finally got it "right" (its all subjective) I had accents and sense of rhythm that was totally lacking at the start.

I used programmed drums on a CD I put out in 2005. at that time, I was as satisfied as a person could be with the sounds of the drums. Now, 8 years later, I kind of wish I would have done it with real drums. It isn't that the music is bad, I still listen to the CD and think of it as something I'm happy with. Now after 8 years, I'm probably the only one who still listens to it, and the realization that I am probably my biggest fan (possibly my only fan) I should have done it "right" In several of the reviews I got, the thing they criticized was the programmed drums, and how much better the album would have been if I'd used real drums. I'd probably still be programming away, were it not for those comments, which were not few. Overall I had decent reviews, but the main sticking point was "get a drummer". Since I live in the middle of nowhere in Texas, and I compose music that cowboys and Skynnyrd drummers don't particularly enjoy, Its basically me or the machine. I did play pretty good drums in high school, so it isn't like I was starting from scratch, but I am quite happy with using real, acoustic drums versus programming them. I started this thread asking peoples opinions about how much editing you can do before you need to say "enough", and re-record the part. I understand, and have enjoyed using MIDI drums, and I'm not trying to be a snob about it, I'm just trying to satisfy myself. Its not likely that any of this music will see any distribution. I'll probably just put the finished songs on Soundcloud and be done. I have come to really despise the marketing aspect of music. Its a bit like trying to sell fresh air to a mountain man. I think most of you get that too. There is just so much great music being created, that the only reason to do it myself is because I enjoy it, but why not get it as close to what I want to hear as I can?

[sonic]

Not being able to play drums and nor knowing anyone with the right skills for my music I'm content to program. I'm not too worried about sounding 'real' as my music is largely electronic. However I do program sampled drums in a way that mimics a real drummer and use quite a bit of variation and fills. To fill out the sound and vary it I double and triple track the drums using different samples — from SampleTank and Superior Drummer. That keeps it sonically interesting, though it may not sound 'real'. No worries in my case. It is time consuming though and tiring ... which is why it's going to take me a long time to put together an albums worth of tunes.

[fictionmusic]

I am in a wee apartment now and although it is ideal to work in (above a bakery that is open 8-4 Mon to Sat) as I can make noise after hours, it isn't large enough to have a kit set up (let alone the two I used to have). So it is midi drums for the time being.
As I write a lot of TV stuff and have been using midi drums for years I am totally comfortable with them, and as the samples are so detailed now and often come with totally convincing midi grooves, that isn't an issue. I still prefer real drums for CD releases, but for TV stuff I am happy with midi and quite frankly I doubt many could tell which was which

[trurl]

One of the great benefits of having your own recording setup is the unlimited time to work on things until they're right. I spent 3 years on my solo project and never got it perfect, but at least got it to a stopping point *lol* We did an album once where I programmed all the drums. Then we decided that the cymbals didn't really sound very good, so I went through the whole album and just replaced the cymbals live. Then we decided the hi-hats sounded fake next to the cymbals, so I replaced them (try that some time- it's insanity trying to play just hats in the holes where the hats go). Then we decided the toms had to get replaced. Then the snare and kick. I did that whole dang album playing one drum part at a time. I do not recommend that approach :D :D But we had acoustic drums at the end. If I had any idea I would have just played them from the start.

[Mikhael]

I have been a drummer in a few bands, so the only time I use programmed drums is for songwriting purposes. I record a lot of parts myself nowadays, and have come up with an interesting method for doing it.

The drums get recorded first. After all, the drummer drives the band, right? So anyway, I'll set up a click with the time changes in it, then start playing drums to it. The difference is that I'll play a verse, get it played right, then stop. Then, on 8 separate tracks, I'll record the chorus, then stop. Then record the next verse, stop, and so on. I then merge the tracks together, so that all the bass drums are one track, all the snares are one, etc. I mostly do this because it's a darn sight easier to get one section right, rather than a whole 8-minute track with no other instruments on it. But I end up with a driving set of drum tracks that I then can record the guitars/keys/bass/etc. to, with no click, and the end result sounds a lot like a full band crankin' it out.

I don't have any good examples I'm ready to release yet, as I just figured this out. But so far it sounds REALLY promising. Better than my other attempts, where I played the whole track through on drums, with nothing else but the song in my head...

[Yodelgoat]

I'll set up a click with the time changes in it, then start playing drums to it. The difference is that I'll play a verse, get it played right, then stop. Then, on 8 separate tracks, I'll record the chorus, then stop. Then record the next verse, stop, and so on. I then merge the tracks together, so that all the bass drums are one track, all the snares are one, etc. I mostly do this because it's a darn sight easier to get one section right, rather than a whole 8-minute track with no other instruments on it. But I end up with a driving set of drum tracks that I then can record the guitars/keys/bass/etc. to, with no click, and the end result sounds a lot like a full band crankin' it out....with nothing else but the song in my head...

That sounds really ambitious, but if you can do it, Excellent! I start with an acoustic guitar, and a click, work out all the tempo changes. Add MIDI keyboards, Then Bass - that's my main instrument, Guitars come next, drums and finally Vocals. I've got three fairly epic pieces - in varying states of disarray, and I cant tell you how good it would be to have the drums and bass down nice and tight, and build everything around that. The good news is, bass is about the easiest instrument to edit - at least for me. If it isn't quite tight, its fairly easy to fix. I think with my next attempt I'll take your suggestions to heart and start with drums. Thanks!

[Wilton Said...]

I just posted in another forum about my experience in listening to a bunch of other Indie Bands.

http://www.progressiveears.org/forum/showthread.php/4494-Indie-%28as-in-independant%29-Music-Observation

Many of them had extremely un natural drums sounds both in actual sound and in style. They were so sterile, perfect, compressed and quantized. Add to that, every other instrument was also quantized and pitch corrected. It sucked any life out of the music.

Wilton

[arabicadabra]

>Good suggestion. I use a very basic click - 1 click per measure, I generally have a hard time staying on that. <

Wow I would find that impossible. I need at least the quarter notes (and occasionally the 8th notes) in my phones to give me the tempo I can play along to. Karl makes a sequence up to record to - click track and a MIDI keyboard track, occasionally scratch guitar - just so I know where I am in the song. If I wrote it, I can just use a click; if someone else wrote it, the sequenced keyboard really helps.

[klothos]

The main problem in my ears is that the more precisely the drums are made to fit the grid, the more you loose a sense of groove and propulsion. Most drummers would agree that being able to play precise, to keep the time and to place your hits precisely, is extremely important. But most drummers would also agree that to become a really good drummer, you have to move beyond that, and allow yourself to dance around the grid rather than to be "on" it all the time. Steely Dan said something about that once, that to be a great musician you have to strive for perfection, and then move beyond perfection, to a place where the rules are internalized and you are free to play with them. That's when music comes alive.

THAT's what's lacking in today's quantized music world. IMO.


+1

excellent post.....thread could have ended here

Sometimes I will actually manually shift some of the samples around ( I record my drum tracks with layers of multi-samples), Sometimes I will go further and ever-so-slightly slightly offset the snare track behind to "fatback" a groove......For the record, I am also one of those guys that has fooled experienced drummers with my programming

[Dean Watson]

Kiothos, I'd like to hear some of our tunes then - Can we?

[klothos]

Kiothos, I'd like to hear some of our tunes then - Can we?

Sure, no problem -- I'll do some edits and upload an MP3 of bits-n-pieces (a fade-in/verse/chorus/fade-out) of a few songs either later tonight or tommorrow

[digestif]

I'm late to this thread, but I'll chip in.
I tend to keep a few drum takes and then edit only on a large scale by selecting sections - if I don't like a placement of a hit then rather than move it I'll pick another take for that whole section. My motivation is a mixture of lack of confidence in editing, with all the sustained cymbal noise around, and the desire to keep something like a real performance.
On reference/click tracks while tracking drums I always use create a melodic guide track with midi and have that in the cans - then I can groove around that rather than trying to align with clicks or drum beats. The midi guide also reminds me where I am in the tune, which can be structurally complex.
Another great resource to hear views from some professionals is www.recordingwebsite.com (http://www.recordingwebsite.com/).