Drums after other tracking?

  Hi there

I am a guitarist and not a drummer. I am more a songwriter and less of a producer, however this modern age has me squeezed into engineering and producer and I'm getting much better thanks to a lot of painful years of learning.

My question today is about laying down drums. I like to use Session Drummer 3. I was taught, or rather the opinion given to me was to do your drums first before other tracking. I understand the logic of this. Yet, I'm not inspired and my drum tracks are very boring (doing them in step sequencer).

So I like the phrase "There is more than one way to do something right!"

What I'm thinking to do is just do a click track, which I know they do in pro studios. I guess that would be just be a 4/4 closed hi-hat in tempo with the project. Then I'd like to my other tracking - guitars, bass, pianos, whatever...

Coming back to drums I'd like to play them live, not live as in real drums, but live as in my fingers on my midi keyboard. Now, here is the part that I could use help with. Of course, if I play them live chances are they will be a bit off, even with a click track. I've heard of this thing called 'quantize' though I've never used it and don't really know how. But basically, my notion of it is that I could select say my midi track and click quantize and it would analyze my 'hits' and then nudge them to say a certain snap point.

Does this sound reasonable? And any help with learning how to use quantize would be appreciated.

Looking forward to your reply and expertise! Thank you,
Ryan

If'n it were me, I'd lay an instrument track to a click track and then play the drums to that.
Then redo that instrument to the drums.
Quantizing will take whatever you play and move the notes to the nearest grid setting. ie: 1/16, 1/32, 1/8 etc.
Quantizing en-mass will also remove all realism and feel to your track. 
Quantize is not a magic bullet. It is just another tool that you probably should learn the ins and outs of.

We all have a lot of "painful years of learning". trust me. you're just beginning.

Wow you all are great and I love your approaches. I'm glad to hear the simple approach of just setting up a click track and then doing my other tracking to it and then PRV and quantize some keyboarded drums. I always felt the step sequencer was unnatural and I could never think incredibly mathematically to write a very good sequenced 'fill' or snare 'roll'. But I can do all that playing my midi keyboard. I like what you said about not quantizing 100%.... that would unnatural too. I will play around with it and see what I like. I will have to look into the groove quantize for the future. I also thought there was a lot of stock to the suggestion to record everything and then get the drums how I'd want them and then retrack everything. Yes, it's a lot of work, but it's a good suggestion on an important song. And really, how many times do I end up retracking quite a bit anyways so that my full mix works together! So even though it sounds like more work, I'm doing it already! I will experiment tomorrow on a small project and see how I fair! Thank you!!!!

This is close to my normal method.  Also good advice on quantizing. I have had pretty good luck at just zooming in on clips with timing problems.  With drums I'm usually working with 9 or 10 tracks.  Quantizing can also cause other problems such as phasing issues.  In a lot of situations I just quantize the measures with noticeable problems. 

timidi


If'n it were me, I'd lay an instrument track to a click track and then play the drums to that.
Then redo that instrument to the drums.
Quantizing will take whatever you play and move the notes to the nearest grid setting. ie: 1/16, 1/32, 1/8 etc.
Quantizing en-mass will also remove all realism and feel to your track. 
Quantize is not a magic bullet. It is just another tool that you probably should learn the ins and outs of.

We all have a lot of "painful years of learning". trust me. you're just beginning.

Just my opinion: quantization is a great evil, a scourge that needs to be stamped out.

I'm not talking about fixing mistakes, that's a fundamental part of the process of production. The difference is in the definition of what constitutes a "mistake" versus normal human variation. The former sticks out and detracts from the song, the latter is essential for establishing groove and feel.

The biggest drawback to the one-man-band approach of building up a song track-by-track is that you lose the interaction between the virtual "players" that would normally occur with a real band playing together. 

In the real world, each musician would be taking cues from one another, altering their timing and dynamics based on what the others are doing. A crash to emphasize a word in the vocal, a pause in the bass line to open up space for a drum fill, coordination between drums and rhythm guitar to reinforce the overall rhythm, synchronizing kick and bass, a lead guitar fill to answer a vocal phrase, and coordinated dynamics between all instruments. These things tend to get lost when you construct a song in layers. 

The defense is to lay down simple parts initially and then revisit each part later. I will often re-do all of (or parts of) my basic rhythm tracks after all instruments have been recorded, attempting to make them more complementary to one another in the context of the full ensemble. 

Drums in particular are an ongoing process, rather than a single step. The song may start with a basic drum pattern that's little more than a click track. That'll get continuously modified as the song is constructed, right up till the end. The very last thing I do before calling a song "done" will usually be some minor drum edit.

This is just a roundabout way of saying that I think your proposed method is a good one. Start with a click, then come back to the drums later on, and continue to refine them throughout the process. Don't be afraid to scrap and re-do entire sections that aren't fitting well.

But whatever you do, don't quantize! (OK, if you must, quantize just the kick, but stop there.)

Political correctness dictates that I have to make an exception to my no-quantize rule for EDM and hip-hop. Some genres expect machine-like quantization, so if you're into those genres, disregard everything I've said so far. And go sit in the corner until you come to your senses.

bitflipper




The biggest drawback to the one-man-band approach of building up a song track-by-track is that you lose the interaction between the virtual "players" that would normally occur with a real band playing together. 


Amen

But whatever you do, don't quantize! (OK, if you must, quantize just the kick, but stop there.) 

Amen

Political correctness dictates that I have to make an exception to my no-quantize rule for EDM and hip-hop. Some genres expect machine-like quantization, so if you're into those genres, disregard everything I've said so far. And go sit in the corner until you come to your senses.

and AMEN!

gotta love Bit's insightful commentary. 

gotta love Bit's insightful commentary. 

AMEN!




Mike

i don't know if sonar's audiosnap has gotten easier to use since S6PE but if you are a guitarist i would recommend using it in reverse - lay down some rhythm guitar (acoustic is best for catching the spikes, even if you don't end up using it) and then use that to line up your sequenced drums.

didn't see this mentioned skimming through, but the corollary to "don't quantize" is find a groove. no matter what genre, even happy hardcore trip hop glowstick, there is a groove. sometimes (as in the happy hardcore etc.) it's a lot more dynamics (and very fast) and less about timing. sometimes (slow blues) it's a lot more timing and far less dynamics. but there always is one, or its boring. 

i remember back in high school, when i was still sequencing MIDI, playing a song (which was perfectly quantized of course) for a couple of my friends (girls) and being taken aback when they started dancing to it. (After all, this was serious work.) years later, i realized a very important fact: people want to connect to your art. even my perfectly quantized sequenced-over-many-weeks really-not-that-good intended-to-be-a-serious-yet-upbeat song had enough emotion and life and dynamics to go straight past the "brain" and right to the soul, somehow. 

guitarhacker said something very important: write the song before you start recording it. if it's not alive in you, if it doesn't come out of you naturally, if it doesn't make you smile and take you somewhere amazing, it won't do that for anybody else and no amount of recording goodies will change that. the only counterpoint to that is, record your inspiration. doesn't have to be sonar, you can do it with a handheld recorder or even your smart phone.

when you've got the song written, when it's part of you, then record it with your key instrument, as a scratch track. go from there. click track, lay drums on top of the scratch track and audio snap, whatever. just get the song recorded and the rest will come pretty easy, because it will either fit, or it won't.