Recording drums - Process issues and questions

Reggae_Mangle

Silver Member
Hey all, hope you had a Very Merry Christmas and a Happy New Year! I know I did!

I'm interested in laying down some drum tracks at my home. I'm not confident enough to pay for a studio just to waste time going over drums over and over.

I will be using a TD30 kit with Superior Drummer 3. I think the sound part of things is pretty much sorted. I am used to the kit, having learnt how to play using it.

The sound side of things sorted out, I have really been struggling with the recording process in itself.

1) For one, I hate having to set a metronome pre-count, rush to my seat while counting and then begin where I need to. This really kills the flow of my performance because I sometimes take more time and sometimes less time to get settled behind the kit.

Is there a better way to do this than rushing after I click record with the pre-count going?

2) I have so far focused my energies on trying to lay down perfect tracks in one go from end-to-end. This is fun, but after take 5 or 6 of a six or seven minute track, I feel like I've burnt up a lot of energy and while some parts are fantastic, others have drifted or there are glitches.

I am really keen to avoid having to look at the midi track and then manipulate notes and velocities and stuff like that.

In that respect, how many of you record in stages? I saw Chris Adler doing it in the studio and I thought that looks promising.

Better for a studio recording than going from end-to-end? This is kind of a moot point since a perfect take from end to end is just very difficult with my skills.

3) When you record bit-by-bit, what is the process like? Do you just start the recording before the start of the next part and then cut and glue so that the transition is seamless? Or do you record the track a few times and comp? Adler seemed to use the former method, but he had a studio guy. I think I'll go crazy if I have to keep going back and forth just to record short sections.

What if I record till a point where things go screwy, then go back a bit and record again for a while till it becomes screwy, etc, till I reach the end of the track? Does that make sense? Or is the sequential approach the best, kind of like groove sections?

4) When you do record bit-by-bit, is there any copying and pasting going on? Or do I record each part sequentially without worry about whether they sound exactly the same? How do you studio cats do it?
 
I'll do my best to answer your questions. I guess my primary question is are you just recording drums or are you recording with a band? Or, are you making a full-band recording where you play all of the instruments?

1) For one, I hate having to set a metronome pre-count, rush to my seat while counting and then begin where I need to. This really kills the flow of my performance because I sometimes take more time and sometimes less time to get settled behind the kit.

Is there a better way to do this than rushing after I click record with the pre-count going?

Yes, re-arrange your room/studio where you can control everything from the seat of your drum throne. Stop that running around mess. The time it takes to do this will be well worth your time IMO. OR, if you have a friend that knows how to run things, have him/her hit "record" for you.

2) I have so far focused my energies on trying to lay down perfect tracks in one go from end-to-end. This is fun, but after take 5 or 6 of a six or seven minute track, I feel like I've burnt up a lot of energy and while some parts are fantastic, others have drifted or there are glitches.

I am really keen to avoid having to look at the midi track and then manipulate notes and velocities and stuff like that.

In that respect, how many of you record in stages? I saw Chris Adler doing it in the studio and I thought that looks promising.

Better for a studio recording than going from end-to-end? This is kind of a moot point since a perfect take from end to end is just very difficult with my skills.

Once again, I'm not sure what you are recording, so finding out will better help me answer your question. All in all, I feel it's best to get everything in one take. When you start piecing things together, it starts to sound "pieced together" after a while. It's like the notes are there, but the "soul" of the song starts to die. This is just my opinion.

3) When you record bit-by-bit, what is the process like? Do you just start the recording before the start of the next part and then cut and glue so that the transition is seamless? Or do you record the track a few times and comp? Adler seemed to use the former method, but he had a studio guy. I think I'll go crazy if I have to keep going back and forth just to record short sections.

What if I record till a point where things go screwy, then go back a bit and record again for a while till it becomes screwy, etc, till I reach the end of the track? Does that make sense? Or is the sequential approach the best, kind of like groove sections?

Yes, if you plan on "punching in" drum parts, you may be better off getting someone in there who knows what he/she is doing when it comes to punching in and out.

4) When you do record bit-by-bit, is there any copying and pasting going on? Or do I record each part sequentially without worry about whether they sound exactly the same? How do you studio cats do it?

Yes, I've done this. Sometimes it works well; other times it comes across as mechanical. How do "studio cats" do it? They get it right in one take most of the time.


Hope this helps!
 
Tip 1: Place a transport on your drums. I use logicremote on an iPad with a gooseneck clamp. Most DAWs have Android/iOS transport apps.

Tip 2: Record the whole song three times through, ignoring f'ups. There will 'usually' be a good take that you can splice together between the three (The verse of take 1, the chorus of take 2, the bridge of take 3, etc).

Tip 3: If there's a complicated fill that's on the edge of your current ability level, play it straight on the main takes, set up a loop on the fill bars, and record the fill using the loop. Splice it in afterwards.
 
Recording alone can be a bit of a bitch. I would try and get a control setup closer to me... Or more likely I'd have someone else do the button pushing. As you mention, by the time you sit down, get comfy, put the headphones on and secure the headphone cable out of the way it can be sort of a pain to get back up. I think there are apps that will let you control modern DAWs, or you could also go the remote desktop route over wifi or network.

As far as recording in parts, if I do that, which is rare, I tend to do it based on how the song is structured. Putting things together works best where there are breaks in the action so you don't have un-natural cymbal cut outs or audible tone differences between cuts on the same track. What I mean is I try and play through at least until a break or big enough gap that stitching can be done elegantly. Not all songs are conducive to this. I'm not a fan of piecing together every little section from different takes, but it's extremely common now. Probably part of why there's so little "life" and too much "perfection" in lots of modern recording processes.
 
1) For one, I hate having to set a metronome pre-count, rush to my seat while counting and then begin where I need to. This really kills the flow of my performance because I sometimes take more time and sometimes less time to get settled behind the kit.

Is there a better way to do this than rushing after I click record with the pre-count going?

I use Presonus Studio One. I set my tune to begin at, say, 30 seconds from the beginning of the timeline. This gives me plenty of time to hit Record, get to my kit and get settled in.

In that respect, how many of you record in stages? . . .

Better for a studio recording than going from end-to-end? This is kind of a moot point since a perfect take from end to end is just very difficult with my skills.

Using the setup above, I set the program to Loop over an expanded area of the timeline. The beginning of the loop is way before the tune starts and the end of the loop is after the tune ends.

When recording begins, the loop function is active. I play through the first take and the playback loops to the beginning of the loop zone and I record another take without having to get out of my seat.

I record as many takes as needed to nail my performance. Sometimes this takes days 'cuz I don't know what I like, or I just plain suck at certain parts and have to practice it.

The actual digital files for each mic/track are very long because they are are entire recording session, but the recording software has internal markers for each separate take.

3) When you record bit-by-bit, what is the process like? . . .
4) When you do record bit-by-bit, is there any copying and pasting going on? Or do I record each part sequentially without worry about whether they sound exactly the same? How do you studio cats do it?

Once I've recorded, say, four takes of a tune and I'm happy with my performance, I listen to each take and figure if I need to swap a fill from take one, for example, onto take 2. Then take the outro from take 4 and place it into take 2.

Studio One (and Pro Tools) make this a very simple process.

I took a workshop at Sweetwater for "How to Record Drums" with Kenny Aronoff. He laid down drums to several pre-recorded tunes and each time he played six or seven takes. The workshop instructor simply selected sections of each drum track and chose "take 2" for some fills, "take 3" for some, etc., until the final result was a performance that never existed.

I asked if this was the norm for non-live recordings (taking selections from various takes) and both the instructor & Kenny confirmed it is. My conceptions about the modern music I like was shattered. Before this, I always thought what I heard was performed in a single take.
 

Attachments

  • SW_Class_187.JPG
    SW_Class_187.JPG
    262.4 KB · Views: 187
I asked if this was the norm for non-live recordings (taking selections from various takes) and both the instructor & Kenny confirmed it is. My conceptions about the modern music I like was shattered. Before this, I always thought what I heard was performed in a single take.

I think it's pretty much customary not only different takes, but quantizing/fixing the drums...not to mention drum replacing or enhancing with samples.
 
I think it's pretty much customary not only different takes, but quantizing/fixing the drums...not to mention drum replacing or enhancing with samples.

"They constantly embrace music only to stifle it. They dissect it, unaware of their preliminary murder of it, and without understanding that the work they have pulled to pieces lacks only one thing, but the most important: life."
 
Building a composite from multiple takes is the accepted norm, mainly because no one can afford the time (and money) to really rehearse any more. Not only that, but in many cases the music will undergo many rearrangements before it is finally considered complete.

Ideally, you want to be able to play the composition from top to bottom without making mistakes, and then you have the luxury of picking favorite sections from each take. If you are going for absolute perfection and even going to the trouble of quantizing, then I really don't see the point. In that case, you should just be programming everything.
 
Building a composite from multiple takes is the accepted norm, mainly because no one can afford the time (and money) to really rehearse any more. Not only that, but in many cases the music will undergo many rearrangements before it is finally considered complete.

Ideally, you want to be able to play the composition from top to bottom without making mistakes, and then you have the luxury of picking favorite sections from each take. If you are going for absolute perfection and even going to the trouble of quantizing, then I really don't see the point. In that case, you should just be programming everything.

One note to add to this...

When working with a producer, they will choose the take(s), and they'll sometimes choose takes that you feel aren't the best one. It's a pretty tough bit of social friction to reconcile the first few times it happens.

The advice I was given for this situation is..... "Its cool, relax. Ride the snake. We're all on the snake together, brothers. If you show fear, the snake will destroy you. But if you trust the snake, He'll take us to the lake. Ride the snake."
 
I'll do my best to answer your questions. I guess my primary question is are you just recording drums or are you recording with a band? Or, are you making a full-band recording where you play all of the instruments?

Making a full band recording by myself. The genre is metal.


Yes, re-arrange your room/studio where you can control everything from the seat of your drum throne. Stop that running around mess. The time it takes to do this will be well worth your time IMO. OR, if you have a friend that knows how to run things, have him/her hit "record" for you.

I'll rearrange things, thanks.

Once again, I'm not sure what you are recording, so finding out will better help me answer your question. All in all, I feel it's best to get everything in one take. When you start piecing things together, it starts to sound "pieced together" after a while. It's like the notes are there, but the "soul" of the song starts to die. This is just my opinion.[/QUOTE]

This is what I feel too. If I can play the track in one take, I'll do it. But this is some really uptempo stuff and I often try things that are beyond my playing level. Sometimes I'll nail it, sometimes I won't.

What I have been doing so far is recording and if there was a glitch, I'd delete the track and record again. After a while, I realised how counterproductive this was from the perspective of a modern studio. So really looking for a way to balance some push-pull in the recording with machine-like precision. Midi programming just doesn't cut it, even if I compare to the recordings I did with glitches.

There is some "feel" to it that programming just can't cop. Or maybe it can be copped with a few non-stop days of tweaking. I just hate manipulating notes with a mouse cursor.


Yes, if you plan on "punching in" drum parts, you may be better off getting someone in there who knows what he/she is doing when it comes to punching in and out.

I could get someone to do it, but someone who knows what she is doing? A little difficult.


Yes, I've done this. Sometimes it works well; other times it comes across as mechanical. How do "studio cats" do it? They get it right in one take most of the time.

Hah, yeah, I figured! Just that LOG video recording some song off Ashes of the Wake I think, and he plays bit by bit and I was like "Woah! Revelation!"

Then I look at guys like George Kollias and I'm screaming my head off by the end of the track.
 
Tip 1: Place a transport on your drums. I use logicremote on an iPad with a gooseneck clamp. Most DAWs have Android/iOS transport apps.

Tip 2: Record the whole song three times through, ignoring f'ups. There will 'usually' be a good take that you can splice together between the three (The verse of take 1, the chorus of take 2, the bridge of take 3, etc).

Tip 3: If there's a complicated fill that's on the edge of your current ability level, play it straight on the main takes, set up a loop on the fill bars, and record the fill using the loop. Splice it in afterwards.

1) I have an iPad, and I use Logic, thanks for Tip 1.

2) This is something I should have been doing. I will retain all recording takes henceforth.

3) Hoho, this looks like fun! Thanks!
 
Recording alone can be a bit of a bitch. I would try and get a control setup closer to me... Or more likely I'd have someone else do the button pushing. As you mention, by the time you sit down, get comfy, put the headphones on and secure the headphone cable out of the way it can be sort of a pain to get back up. I think there are apps that will let you control modern DAWs, or you could also go the remote desktop route over wifi or network.

As far as recording in parts, if I do that, which is rare, I tend to do it based on how the song is structured. Putting things together works best where there are breaks in the action so you don't have un-natural cymbal cut outs or audible tone differences between cuts on the same track. What I mean is I try and play through at least until a break or big enough gap that stitching can be done elegantly. Not all songs are conducive to this. I'm not a fan of piecing together every little section from different takes, but it's extremely common now. Probably part of why there's so little "life" and too much "perfection" in lots of modern recording processes.

I'm really, really scared off making a take "too perfect" by using things like quantisation and nudging notes. That's the main reason I want to play drums instead of programming. The vibe, the dynamics, the human element make the recording sound like a living, breathing thing. I really hate the staccato programming I hear on most (metal) recordings.

Alas, I'm not close to being as consistent on the drums as I'd like, so I think as suggested by everyone, I'll record a few takes from end to end, comp from those and then if there are any errors, I will punch in and punch out.

Good advice on ensuring that there are no unnatural cymbal cuts, etc. That's probably something I would have overlooked.
 
I use Presonus Studio One. I set my tune to begin at, say, 30 seconds from the beginning of the timeline. This gives me plenty of time to hit Record, get to my kit and get settled in.



Using the setup above, I set the program to Loop over an expanded area of the timeline. The beginning of the loop is way before the tune starts and the end of the loop is after the tune ends.

When recording begins, the loop function is active. I play through the first take and the playback loops to the beginning of the loop zone and I record another take without having to get out of my seat.

I record as many takes as needed to nail my performance. Sometimes this takes days 'cuz I don't know what I like, or I just plain suck at certain parts and have to practice it.

The actual digital files for each mic/track are very long because they are are entire recording session, but the recording software has internal markers for each separate take.



Once I've recorded, say, four takes of a tune and I'm happy with my performance, I listen to each take and figure if I need to swap a fill from take one, for example, onto take 2. Then take the outro from take 4 and place it into take 2.

Studio One (and Pro Tools) make this a very simple process.

I took a workshop at Sweetwater for "How to Record Drums" with Kenny Aronoff. He laid down drums to several pre-recorded tunes and each time he played six or seven takes. The workshop instructor simply selected sections of each drum track and chose "take 2" for some fills, "take 3" for some, etc., until the final result was a performance that never existed.

I asked if this was the norm for non-live recordings (taking selections from various takes) and both the instructor & Kenny confirmed it is. My conceptions about the modern music I like was shattered. Before this, I always thought what I heard was performed in a single take.


Thank you, this looks like a good way to get the ball rolling.

How exactly were the takes put together though? Just a cut and paste? Or was there a more fluid approach to ensuring that everything came together as a contiguous piece of music? This is something I really dread when it comes to studio work, I dislike doing it even for guitar. Everything seems to come across as mechanical when I do it, so hoping there's a better way.
 
I think it's pretty much customary not only different takes, but quantizing/fixing the drums...not to mention drum replacing or enhancing with samples.

I'll be using Superior Drummer 3, perhaps with a bit of layering using some of the kit pieces I have in BFD3, or maybe just the cymbals.

I may do some manual fixing of timing, but I'd like to avoid any hard quantisation if I can help it.

Something I shall definitely experiment with, thanks!
 
Building a composite from multiple takes is the accepted norm, mainly because no one can afford the time (and money) to really rehearse any more. Not only that, but in many cases the music will undergo many rearrangements before it is finally considered complete.

Ideally, you want to be able to play the composition from top to bottom without making mistakes, and then you have the luxury of picking favorite sections from each take. If you are going for absolute perfection and even going to the trouble of quantizing, then I really don't see the point. In that case, you should just be programming everything.

Thanks, I think I'll do a bit of both, i.e. record from end to end several times, comp, identify errors and then punch in an punch out/record in a loop to correct those bits.

I don't see the point of heavy quantisation either. I dislike the way it becomes very inhuman. Perhaps that's because of the way I do it...
 
One note to add to this...

When working with a producer, they will choose the take(s), and they'll sometimes choose takes that you feel aren't the best one. It's a pretty tough bit of social friction to reconcile the first few times it happens.

The advice I was given for this situation is..... "Its cool, relax. Ride the snake. We're all on the snake together, brothers. If you show fear, the snake will destroy you. But if you trust the snake, He'll take us to the lake. Ride the snake."

He's old, and his skin is cold.

I wish I had a producer, but this is strictly a solo project on a tight budget. The luxury I have is time. At the same time, the problem I have is a lack of deadlines.

Thanks, though, I think I'll do some reading on why such decisions are made.

Much obliged, gents!
 
Last edited:
Gracie! I was working on a song yesterday and pretty kicked with the idea of playing end to end several times to come up with a perfect take. May still require some punching in and out, but the feel is good.
 
Back
Top