Recording Bands (running the sessions to get the best performances)

[karatemanjohnny]

Hi everyone,

I did some recording with my band this weekend, and we are pushing it back a month to do more practicing/preparation. I noticed a few things about how the recordings were run, that our singer mentioned.

The way I usually record bands is to do all of the drums first, then get bass/guitars, then do vocals/solos. Recording all the parts straight through.

But I think it makes very little sense to do it that way, so I'm wondering how other people record bands.

We were doing 4 songs, and on Saturday we did all of the drums. It took 11 hours, minus two hour long breaks. The drummer was exhausted at the end of it. Then today, I did bass, and ended up stopping after 3 songs (only 2 were keepers). Then the guitar didn't go as well as we wanted, and we decided to rework some parts and get more comfortable (we're only a few months together).

The songs are real difficult. I think next time we record, we will do one song at a time. I've heard people say if you don't get it in 5 takes, you won't get it on the last 15.

Here's what I think would be better - what do you think?

track drums, then bass, then guitar, then solos, then vocals.

Repeat for each song.

This will give each player time to rest between playing their parts, and it will have other band members spend less time sitting idle watching other record their parts.

Could some people chime in their opinions on this, and maybe tell their experience recording bands and what order they do things in?

Thank you,

John

[hollywood_steve]

Is it safe to assume that tracking the whole band together is out of the question?

[karatemanjohnny]

Hey Steve,

Tracking the whole band at once is definitely doable, but possibly not all 'keeper' tracks.

We actually did track all at once, but the guitar/bass/vox were all scratch tracks to be redone later. That's how I've been doing it with bands - get the drums first, and the redo the guitar/bass/vox.

- John

[zimv20]

it doesn't work for everyone, but when possible i like tracking everything at once. i regard nothing as scratch except the vocals, and once or twice those were keepers, too.

if the band is used to playing together, imo this technique results in a better overall performance.

[alphajerk]

i have NEVER done it the way you describe. intial tracking is drums, bass, rhythym guitars all live. maybe scratch vocal if they are ques in a song.

if the band cant play a song live then wtf are they going to do live? if they cant pull it off live, i kick them out of my studio [literally, i have sent home bands who werent ready]. i just have no times for putting up with that.

so i can get a whole initial rhythym tracking in as little as 4 hours including setup and i can span it over a couple days. average is about 3 songs in 4-6 hours, and by that time the band is spent, or sometime spent, depending on the number of takes. its rare take 1 isnt the keeper unless a trainwreck brings it to a halt.

as for singer/songwriters, i do it differently due to the nature of them being solo, which i usually track their instrument [usually guitar or piano WITH vox to get the layout of the song. if their pulse just wont work for OD's i map out the song to a click and use bassline quarter notes to play to. then bring in drums, and bass, and any other filler that is needed [solo's, outros, weird shit, etc]


what exactly is your question?

[everybody's x]

I always do drums first (with scratch bass and guitars, but they're set up so if we get a keeper we keep it)
then guitars
bass
vocals
solos
various other weird stuff


I dont worry too much about bleed because I like the vibe of the band playing together


BUT


next thing I'm going to try is cutting guitars,bass,vox to a drum loop and letting the drummer overdub the drums ala Richard Dodd

It will be interesting at least


then I'll probably go back to the "comfortable" way

[karatemanjohnny]

Hey all,

The main thrust of this post was, what do you think about the setup i described, and how do you do it differently?

My reasoning was this, for how I recorded the band - Recording guitar and bass direct, focus on getting a good drum track. Recording the guitar and bass direct, for scratch tracks, gives the drummer a chance to do multiple takes without the guitar/bass players playing along with him.

Same with bass/guitar - after the drums are down, bass player and guitar player can do multiple takes and the drummer doesn't have to play every time. Then, same for vocals.

The problem that happened was, people needed to do too many multiple takes in a row to get it right, because we didn't have the parts down enough.

And, we tried to have each person do four songs in a row - multiple takes on four songs in a row was too much.

So, we decided to push the recording back until we got our parts down better, worked out some more parts for the songs, practiced more.

We're just going to focus on having the parts down much better. What we're trying to decide now is to try and do a demo first, or play some gigs and then do a demo. I think we're going to do some gigs and then revisit recording.

- John

[bloodstone]

Originally posted by everybody's x
I always do drums first (with scratch bass and guitars, but they're set up so if we get a keeper we keep it)
then guitars
bass
vocals
solos
various other weird stuff


I dont worry too much about bleed because I like the vibe of the band playing together


BUT


next thing I'm going to try is cutting guitars,bass,vox to a drum loop and letting the drummer overdub the drums ala Richard Dodd

It will be interesting at least


then I'll probably go back to the "comfortable" way

I've done the "Richard Dodd" method many times with great success. It's a great way to make progress when you're a singer/songwriter like me who can't always get the drummer to come in and do a live take due to scheduling conflicts. It works much better than trying to use a click with everyone playing live.

[brendondp]

Originally posted by karatemanjohnny

The problem that happened was, people needed to do too many multiple takes in a row to get it right, because we didn't have the parts down enough.

So, we decided to push the recording back until we got our parts down better, worked out some more parts for the songs, practiced more.

We're just going to focus on having the parts down much better. What we're trying to decide now is to try and do a demo first, or play some gigs and then do a demo. I think we're going to do some gigs and then revisit recording.

- John

There ya go. The Scouts had it right all along. I'm sure you'll do yourself a massive favour if you can "be prepared"...

Cheers,

bdp

[bobby yarrow]

It's definitely pretty normal to work the way you tried to, getting all drums first, etc. I think most people do keep the bass and guitar tracks, and sometimes they're keepers but for me at least they usually wind up being scratches.

I think this comes out of a couple things: the idea that drums need to be recorded first to work right; the fact that, with a lot of mics on the drums, it takes half a day to set up or whatever; and, the idea that you might not have the personel back in the studio, so you better get what you can now.

That said, everybody works differently, and different bands take different treatment. There are definitely acts who need to record the whole thing at once, vocals and all. When scheduling required, I've even overdubbed drums, worked ok (given the kind of music that it was, and the fact that it was all on a click). I can't think of ever building up one song at a time, the big limit being setting up the drums over and over again. If you can sort of leave the drum setup, or if the setup time doesn't bother you, then there's no reason that couldn't work too.

[Dave Martin]

I have to chime in with the folks who say that the 'one instrument at a time' approach generally sucks. In my opinion, that's not music, it's typing. You may end up with clean tracks that way (or not), but its more likely that you'll have 'clean' individual tracks and no music. Music only happens when people play together.

Instead of trying to get drums, then bass, then guitars, why not get everyone, and fix those tracks that need to be fixed? In fact, most of the sessions I play on (and engineer) follow a fairly well defined work flow. The whole rhthm section records at once; when the producer is happy with the overall feel of the take, then anyone who needs to fix something (a measure, a section, or whatever) will do so. Then those who will do another complete pass (for instance, a second acoustic guitar, an organ string tracks, percussion), will do so all at once. Fix any mistakes on those overdubs, and then move to the next song. A

ny instruments that need to be added later (those that weren't on the tracking session), are scheduled for another time. Another session is scheduled for vocals (and sometimes another for BGV's). If the musicians aren't capable of playing their parts, then they likely shouldn't be in the studio...

(I KNOW this is an ideal situation - and one that usually involves session musicians - but it's certainly doable with a band.)

[Dave Martin]

Originally posted by bobby yarrow

I think this comes out of a couple things: the idea that drums need to be recorded first to work right; the fact that, with a lot of mics on the drums, it takes half a day to set up or whatever; and, the idea that you might not have the personel back in the studio, so you better get what you can now.

Umm...

Haven't you been reading Richard's posts? On some of those records, he had 10 minutes to get a drum sound. And funny, they sound great....

Also, I think that you're missing an important distinction - the drums have to work right, but recording them first doesn't have anything to do with whether or not they're 'right'.

(I know - I said pretty much the same thing in the previous post..)

[Dave Martin]

Originally posted by karatemanjohnny

We were doing 4 songs, and on Saturday we did all of the drums. It took 11 hours, minus two hour long breaks. The drummer was exhausted at the end of it. Then today, I did bass, and ended up stopping after 3 songs (only 2 were keepers). Then the guitar didn't go as well as we wanted, and we decided to rework some parts and get more comfortable (we're only a few months together).

The songs are real difficult. I think next time we record, we will do one song at a time. I've heard people say if you don't get it in 5 takes, you won't get it on the last 15.

Is there any reason that you couldn't 'punch your way to success on the overdubbed parts? You know, play until you make a mistake and then punch in at that point? Or don't people do that anymore?

[wwittman]

First.. there are no rules...
excpet perhaps this one:
Keep it FUN.

You don't want to make it deadly, or a job... which is what seems to have happened.

The trick is to find what "method" works for the band in question.
of COURSE it's always great if you can get the whole band, or even most of it, at once, live... but if that doesn't work then *I* don't believe in flogging them for hours until they "get it"... or give up.
AND I certainly don't believe in "throwing them out"... as I've seen some amazingly good records made by people who were frightfully slow and couldn't play live together in the STUDIO, to save their lives.(didn;t stop them from putting on great live shows either).

If the band gets a better feel by playing together (as many do) then by all means... but if they DON'T then figure out what's happening.
For example, WHY did it take 11 hours to get 4 tracks?
What wasn't happening and what was stopping it from happening?
If the guitar player confuses the drummer, then cut without the guitar player.
I worked with one band where the drummer only wanted to play to a single ref guitar and vocal cues (not actually singing the whole song, just yelling 'here's the chorus')... he knew the songs really well and knew what he wanted to play, he just wanted then to concentrate on HIS performance, not on 'distractions' as he saw it. And in the end they sounded quite spontaneous. It was their ability to play "together" even separately that made it work... and it let us concentrate on each player and ideas more "personally".
If you have to, use the method and record the whole band as guides to a click or drum machine and then overdub the drums... the plus in this method is twofold: one is that you can tailer the foldback mix to whatever he needs to hear to be comfortable and you're not grinding the rest of the band into the ground. Second, by overdubbing, you get to drop in and out as needed.. instead of HAVING to record the whole song all through each time. The click means that the ins and outs are in an easy to find same place each time, instead of a moving target.

I also would agree that you should set up and try to beb ready to go as fast as possible. Again the idea is to make the whole thing feel as non-threatening and fun as possible.
You gain nothing by making everyone feel it's a chore or that they're "no good".
I also think that there's nothing at all unusual about exploring some ideas at the overdub stages, rather than knowing exactly what you want to do before you record. That's part of the fun in this , post-Sgt. Pepper-era.

I'd stop thinking about how you "should" cut tracks and figure out how you NEED to cut tracks with THIS band in order to get what they do best.

[karatemanjohnny]

Nice to wake up to a bunch of great posts. A lot to digest for sure.

We were working to a click and doing punch-ins, but I think we thought it was taking too long and we were editing too much. And what actually caused us to stop was one song that we didn't all like our parts on - plus we were dog tired. We had set a goal to do it over the weeked, but it was too short a time frame for how prepared we were and ended up being way too stressful. So, next time, we're going to take it more slow, and try and do all the instruments in one day.

Setting up drum mics and taking them down isn't a problem, and we can most likely leave them set up. We talked it over and decided to try and do it all in one day.

I agree it should be fun, and should be focused on what it takes to get everyone to feel the most comfortable and play their best.

Thanks,

John

[Sofa King]

Originally posted by karatemanjohnny


Could some people chime in their opinions on this, and maybe tell their experience recording bands and what order they do things in?

I try and get as many keeper performances as I can.
But one thing I ALWAYS do, is I run three takes back to back.

Its much more productive.

If you do one take, its 5 minutes,

then the band meanders into the control room, 3-5 minutes,

they BS a while a few more minutes.

They listen to the take, few more minutes.

then they discuss it,

then they meander back and try take two.



Doing the 3 takes back to back "amortizes" all the BS and meandering time across three takes, not to mention, that the band usually sounds more comfortable on the second or third take.


best
Sean

[brendondp]

Some of the best takes I've ever got personally, and that I've recorded of others, have often been "rehearsal" takes where the band (or the individual) have simply been running through the song with the tape (or hard drive) rolling.

The takes off the floor in those situations have often been quite superior to the ones done subsequently, in part I believe because the pressure of "getting it right" is aleviated due to the fact that everyone has the mindset that it's no big deal if someone screws up or flubs a note or drops a stick.

Everyone relaxes and lo and behold, the groove is there, the feel is right and the track sounds fresh and has a degree of spontaneity that often disappears on take number nineteen - funnily enough. And in my experience, the motivation and attitude of someone facing take number twenty is totally counterproductive to the process of making music.

I have occasionally be known to keep the tape running without the person knowing, sometimes just to see what you get, but often because their self-conciousness disappears to a degree that often results in a superior performance.

Of course, like I said before, the more prepared you can be before you start recording (the more the band has the parts down) the easier it'll be to create an environment conducive to producing great performances.

Just for information's sake, Soundgarden recorded Superunknown on a song by song basis, and all of the Metallica's St. Anger sessions were off-the-cuff takes edited together in post-production - though you can make of that what you will.

I love tracking bands playing together, and bleed is not something I try and eliminate at all costs. But I've also done plenty of layering, and being a dick about one or the other is the last thing a band needs when attempting to make something of importance and longevity that will (hopefully) be fondly remembered.

I hope the sessions go well.

Cheers,

bdp

[bloodstone]

I have had a lot of success punching entire sections of songs with the entire band playing and recording simultaneously. YMMV, but this works well in my humble digital Tascam MX-2424 studio.

[Ted Nightshade]

I really believe that music is something that happens between people. So I always have at least 2 people playing at once for basic tracks. In your case, drums and guitar, drums and bass, drums and guitar and bass, everything including vox, whatever works.

The groove is a mutual understanding between 2 or more people. Two perspectives on the song converge (hopefully) and overdubs work better.

I really think it's quite a discipline to take a song apart and record it in parts. It's not necessarily any easier. For a band that is used to playing together, it can be like learning a whole new way to play. It only makes sense to rehearse for a session where everybody plays their part at a different time, by rehearsing that way too. I find it tremendously disorienting as an artist, so we just do one thing- we perform, same way live as in the studio. That way we can take our experience in the studio right onto the stage, a big advantage.

And recording is a very good way to get a real objective sense of how the band plays together. So I think it's a shame to miss out on that diagnostic opportunity. So many people go into the studio and come out without getting a look at how their act really performs together. Damn shame, I think.

I also really recommend knockin' out at least two or three takes in immediate succession. Not only do you get the advantages the previous advocate of this approach mentioned above, but if you need to splice things together or steal a chord from somewhere, your "parts car" is a performance as close as possible to the one you are stealing parts for. It's the same people in the same room set up the same way just a few minutes earlier or later. I really think that helps if it comes to splice time, and it also helps to knock 'em out 3 takes at a time so no one of those takes is a "big deal" to get anxious about. ****ed up? Oh well, knock it out again, and again, and then take a break and see if anything needs adjusting for the next round.

[alphajerk]

my bare minimum is drums and bass. if i get that guitars can be redone or done later. i just hate not having bass and drums not playing together.... in a band those two should be tight.

[Ted Nightshade]

Originally posted by alphajerk
my bare minimum is drums and bass. if i get that guitars can be redone or done later. i just hate not having bass and drums not playing together.... in a band those two should be tight.

I tend to agree, but I'm fascinated that so many Beatles cuts have the bass overdubbed. Seems the main groove was guitars and drums.

But tracking the bass and drums together really lets you get a good look at how the bass and bass drum get along sonically, and musically, and that is crucial.

[oedipus_complex]

Originally posted by Sofa King
I try and get as many keeper performances as I can.
But one thing I ALWAYS do, is I run three takes back to back.

Its much more productive.

If you do one take, its 5 minutes,

then the band meanders into the control room, 3-5 minutes,

they BS a while a few more minutes.

They listen to the take, few more minutes.

then they discuss it,

then they meander back and try take two.



Doing the 3 takes back to back "amortizes" all the BS and meandering time across three takes, not to mention, that the band usually sounds more comfortable on the second or third take.


best
Sean

Yeah that's a pretty good idea.

I think it all depends on the nature of the band and of the song(s). It's horses for courses. If you want a live band sound, then tracking all at once may work. If you want a more precise sound, one at a time may work. I wouldn't advise doing what some else says, I would advise considering the consequences of each possible method, and choosing the one that suits the band/song(s).

[bloodstone]

Originally posted by oedipus_complex
Yeah that's a pretty good idea.

I think it all depends on the nature of the band and of the song(s). It's horses for courses. If you want a live band sound, then tracking all at once may work. If you want a more precise sound, one at a time may work. I wouldn't advise doing what some else says, I would advise considering the consequences of each possible method, and choosing the one that suits the band/song(s).

I agree with choosing the one that suits the band/song. Anything can work. I've had a lot of success tracking drums, scratch vox & electric guitar live, then overdubbing bass. Sometimes if there's an obvious weak link in the chain it might be best to leave them out of the basic tracking. I did a whole album of material with an artist who tracked drums and DI acoustic live, then overdubbed everything else using players and her own talents on bass & electric guitar.

Maybe the artist in question isn't ready for the studio?

[karatemanjohnny]

Man, this just keeps better and better. Thank you all so much for the replies! Lot of great insight.

Bloodstone - the artist in question, off the original post, was my band and I. We decided to practice more, maybe play out some, and rewrite some parts before doing more recording.

We're going to do more metronome practicing - individually and as a group.

That's interesting to know about Soundgarden, knocking out songs one at a time. We are planning on trying that next, I believe.

Thanks again!

John

[John Hedger]

Originally posted by Sofa King
I try and get as many keeper performances as I can.
But one thing I ALWAYS do, is I run three takes back to back.

Its much more productive.

If you do one take, its 5 minutes,

then the band meanders into the control room, 3-5 minutes,

they BS a while a few more minutes.

They listen to the take, few more minutes.

then they discuss it,

then they meander back and try take two.



Doing the 3 takes back to back "amortizes" all the BS and meandering time across three takes, not to mention, that the band usually sounds more comfortable on the second or third take.


best
Sean

I believe that is a good strategy Sean. I recorded my band this last August at a vintage style Mississippi studio which was set up for "live in the same room" recording. Elvis Costello had recently recorded there just for a day. I was curious and asked how many songs he did and what his recording strategy was. The engineer Will told me Elvis and his band recorded 15 songs in one long day with 5 takes each, bam bam, back to back!!

John Hedger

[418]

Originally posted by wwittman
I'd stop thinking about how you "should" cut tracks and figure out how you NEED to cut tracks with THIS band in order to get what they do best.

Hear, hear!! Why should all bands have to record the same way?