Take folders, comps, edits, printing (For an idiot)

[Whizzinby]

I avoided using takes when I started recording stuff because I couldn’t quite wrap my head around on how to use them. I think that made me woefully inefficient because I’ve been manually repositioning the play head and hitting record…every take… until I can half ass live with an entire part. But I forced myself to use it the other day by setting a loop on the part. After destroying a few take folders I managed to fumble around how to edit in and out of certain takes in the folder when I fat finger a chord etc. So really cool. Progress. But questions:

1. Is that how you guys initiate takes, setting a loop?
2. Is there a way to have the recording count-in repeated with each pass? The way I’m doing it, it just immediately starts the new take once the loop ends. (Which is fine for some riffy stuff, but tricky on others)
3. Is there a way to break a specific take out of the take folder? Ex, if I’m trying to get two solid takes, that will eventually get hard panned L/R and by random luck nail two takes, is there a way to pull one of those takes entirely out and assign it to its own track? (Leaving the other good take in its existing track folder?)

Also I was watching a video with CLA doing a mix and some of the tracks were “printed” (I think that’s the terminology) to audio tracks already. Do you guys ever do that for anything while mixing/mastering? Generally I’m running all my tracks in real-time being processed by whatever I have in the chain. I was trying to think of how that would be advantageous for a simp like me, and it seems like it could be beneficial on tracks that I’m running Neural plugs and not the AxeFX, and to just print it down, to eliminate the need to run the plug-in. (Which could be good for system resources, on some of my mixes I could fry an egg on my Mini trying to run a bunch of Neural plugs) And, how do you actually do this, is it in the bounce menus?

And lastly, on the same video he seemed to have a single track that contained two individual hard panned tracks. Is that right/possible? So typically for distorted parts I will have two tracks each with their own individual mono takes, hard panned L/R respectively. Is there a way to collapse those into one track while still maintaining their individual panned settings? That would eliminate a ton of track clutter if so. Maybe I was misreading what I saw, but it looked like two entirely different waveforms within the same track. My eyes lit up with potential glee.

Edit #Logic

 

[LeftyLoungeLizard]

Hmm. Could be semantics, but the way I do things, a "take" is a track in the DAW. I usually shoot for a good/perfect take for every guitar track I do, and that's all the way through the entire timeline. But some tunes/pieces are so complicated, so layered or so difficult (meaning a tough passage that I don't have down cold) that I will break things up into chunks timeline-wise.

When it comes to doing multiple takes (if needed), I will create a new set of tracks for each (new) take. For example, I'll do a doubled rhythm part (L and R) using 2 tracks, then create 2 more tracks in the DAW and do the doubled part again. So "takes" in my case mean "more tracks"; then pick the best. But more often than not, I'm doing this because I'm trying a different tone/gear and want to hear the difference. If I'm fucking up the part, I'll just delete the take and do it again until I get it right, and this, for me, happens by far the most when I'm having to start the tune/piece with just the guitar (before drums etc come in) and have to be in the right groove timing-wise - 99.5% of the time no click track or similar. Also, to save CPU usage if there's too much going on, I'll disable FX on tracks I'm not using.

On a solo that's difficult (whether one-pass, or broken into chunks), I will have up to 3 extra attempts (3 more tracks), and pick the best. If I don't get what I want in 3 extra attempts, I will delete the worst one and keep going until I have what I want.

Sometimes you have to do multiple takes to get into the groove of things or get it right.

I only "print" (freeze) tracks if I need lighten up the CPU load in a multitrack session with tons of plugins running real-time.

I should also add that I do a "hybrid forward-backwards" mix style; meaning, I will get levels and eq/comp all the tracks/busses first, then apply the "2" FX (typically Mastering EQ, Bus Comp, tape), then mix "backwards". Meaning, I will listen to the entire thing with all the mustard on it and if necessary, go to say the rhythm guitar bus and tweak the Pultec EQ some, or add a little more low end to the kick drum, etc. I usually don't need to "print" (freeze) tracks because my PC has plenty of horsepower, nor do I want to "print" (freeze) tracks because of my mix style. Everything is "liquid" and remains so, so that I can tweak. I will then export a clip (.mp3) from this. The reason to do "backwards mixing" is because no track is an island unto itself; how does everything sound together? That's what matters. Could have a killer doubled rhythm guitar part perfectly EQ'd etc sounding great by itself, then throw it into the full mix and it sounds like shit. Gotta adjust things with everything playing and every FX running.

 

[PLX]

The "Toggle Repeat" function in ReaperDAW is what I use on certain parts to figure out what intervals/patterns are going to work.

I guess this is the same thing you're talking about.

 

[MirrorProfiles]

Do you guys ever do that for anything while mixing/mastering? Generally I’m running all my tracks in real-time being processed by whatever I have in the chain. I was trying to think of how that would be advantageous for a simp like me, and it seems like it could be beneficial on tracks that I’m running Neural plugs and not the AxeFX, and to just print it down, to eliminate the need to run the plug-in. (Which could be good for system resources, on some of my mixes I could fry an egg on my Mini trying to run a bunch of Neural plugs) And, how do you actually do this, is it in the bounce menus?

What DAW are you using?

I would say it’s absolutely advantageous to commit plugins and processing down - not just for CPU resources, but because when it comes time to mix, you shouldn’t be thinking about vocal tuning or edits or IR’s or tweaking amp settings.

The more simple you can make the session, the more seamless you can navigate it and control the mix. There’s no need for loads of tracks and tons of control when it comes time to mix, lots of those decisions can be committed to a track when it’s in a good place.

And lastly, on the same video he seemed to have a single track that contained two individual hard panned tracks. Is that right/possible? So typically for distorted parts I will have two tracks each with their own individual mono takes, hard panned L/R respectively. Is there a way to collapse those into one track while still maintaining their individual panned settings? That would eliminate a ton of track clutter if so. Maybe I was misreading what I saw, but it looked like two entirely different waveforms within the same track. My eyes lit up with potential glee.

Yeah you can do this, just pan each mono part L and R in the stereo track. You can actually be even more smart arse and use 5.1 (or higher) to hold more tracks in the same file. Some people do this with multiple mics and DI’s and just mute/balance the individual channels as needed.

Just seen you’re using logic. All the stuff you mentioned is possible, honestly I’d just search some youtube tutorials and read the manual a bit as all that stuff is possible and quite adaptable to your needs. Logic does require you to bounce tracks to get a stereo track, in most other DAW’s you can just drag and drop 2 mono regions to a stereo track.

BTW Logic is a fucker for having TONS of stuff running at once. the more you commit (Bounce in Place in Logic speak) the better. It can do all kinds of goofy shit with ARA and flex editing running, I don’t ever trust it at all until it’s just playing back audio files with nothing going on behind the scenes.

 

[mbenigni]

Great topic. The first time I saw the comping features in Logic it blew my mind.

A couple of versions ago, Ableton introduced the same for Live, and I thought I'd make constant use of it. But I never actually managed to get comfortable with it. Part of this is a constant awareness that my PC is underpowered and my SSD space is tiny. I get distracted watching all of that audio pile up while I hack away at a part.

I prefer to keep bloat in check by managing things manually. If I'm working on a tricky part, I'll typically create a duplicate and then toggle between recording to one track or the other. I make rough edits as I go, and when that part's done, I can copy paste to back to the original track, or automate levels, or bounce... depending on what the material demands.

As for repositioning the head and hitting record... I haven't found anything consistently better. But I have been able to streamline the process by adding a key binding (e.g. "r") for record, and adding markers or moving the loop block so the header automatically returns to where I need it when I "Ctrl-z, r". Leave that marker/ block a little earlier for a count in comprised of whatever audio happens to precede.

 

[MirrorProfiles]

Just as someone who used Logic for years and then switched to Pro Tools, PT Playlists are BY FAR the best way of handling multiple takes and comping them into a master take. Quick Swipe comping is kind of fun but its a bit of a nightmare for proper organisation, and as ever with Logic you have kind of crappy control over crossfades and edits while in quickswipe mode.

Logic added playlists more recently but they're pretty horrible to use, and when you combine Logic's playlists with quickswipe comping you can get into a really messy situation. The best advice I can give with quick swipe comping is to try and choose your comp early on, and then use flatten, rather than flatten and merge. You can also export or move the comp to a new audio track and then mute+hide the old one. Get all that clutter out of your session as soon as it makes sense to.

 

[FuzzyAce]

1. Is that how you guys initiate takes, setting a loop?
2. Is there a way to have the recording count-in repeated with each pass? The way I’m doing it, it just immediately starts the new take once the loop ends. (Which is fine for some riffy stuff, but tricky on others)
3. Is there a way to break a specific take out of the take folder? Ex, if I’m trying to get two solid takes, that will eventually get hard panned L/R and by random luck nail two takes, is there a way to pull one of those takes entirely out and assign it to its own track? (Leaving the other good take in its existing track folder?)

1. I never use the loop take function personally.
2. some DAWs may have a click in before each loop repeat? not sure which though?
3. I'm sure you can but I just haven't explored it enough

Glad I could help!



As far as bouncing, this is something I do a lot. Song dependent, of course, and especially when running virtual instruments or resource hungry plugins. I prefer committing to something but I always keep a copy of the original track but disable/hide it if I need it later on.

 

[Whizzinby]

What DAW are you using?

I would say it’s absolutely advantageous to commit plugins and processing down - not just for CPU resources, but because when it comes time to mix, you shouldn’t be thinking about vocal tuning or edits or IR’s or tweaking amp settings.

The more simple you can make the session, the more seamless you can navigate it and control the mix. There’s no need for loads of tracks and tons of control when it comes time to mix, lots of those decisions can be committed to a track when it’s in a good place.

You do this “dry” (for the lack of a better word) just the amp sim, or do you also bake in some of your EQ moves? I assume you just want the core amp tone, to give you freedom to make any eq/compression moves later.

Logic does require you to bounce tracks to get a stereo track, in most other DAW’s you can just drag and drop 2 mono regions to a stereo track.

Ok, I’m going to deep dive this, because if I can collapse two mono panned tracks to one stereo track (that’s retain the individual pans) it’s going to be the best thing ever. My brain doesn’t work well with the clutter of tracks and this will greatly simplify the mix process.

Side note I just figured out the X-fade between segments, which is another moment of…

 

[Whizzinby]

1. I never use the loop take function personally.

I was just messing with it and on that particular project the project would crash after the 3rd-4th loop. Might have to fall back to the ole manual trigger per take if that’s not an aberration for that specific project. Was kinda hoping to be able to do 7-8 passes then just edit/comp any imperfections and potentially get 2-3 tracks out of one go.

 

[MirrorProfiles]

You do this “dry” (for the lack of a better word) just the amp sim, or do you also bake in some of your EQ moves? I assume you just want the core amp tone, to give you freedom to make any eq/compression moves later.

What I tend to do is record each guitar to its own mono track, comp together a master DI track. Once I've got all the guitars recorded and tones dialled in, Ill usually commit the plugins and quite often the FX too.

Once the sounds are committed, then Ill organise them so that both rhythm tracks are on one stereo track, lead guitars might go on another stereo track, clean guitars on their own one. Sometimes Ill put cleans and leads on the same track if they dont come up very often or need much work doing. Its sort of playing tetris just to give myself less distraction.

Generally, I think of it like one track should be a single "part". That part may be made up of several performances but really they're combined together where I only need one fader for it. Same is true of backing vocals, there might be stacks of layers that go into them but at a certain point I can just have a single stereo fader to balance them in.

A good tip I got from Tom Elmhirst is to try and get your session so that when you look at the mixer page you can see all of the faders on your screen at once. Thats usually 30-40 tracks which should be PLENTY. Very few analog consoles had more tracks than that, and in the tape days you'd need to sync a few machines to take full advantage of it. If its more, it just means that you are giving yourself control over things that probably don't need it.

 

[FuzzyAce]

Was kinda hoping to be able to do 7-8 passes then just edit/comp any imperfections and potentially get 2-3 tracks out of one go.

I can definitely see the appeal of that approach. I may dive into the function at some point, but we get stuck in our habits.

 

[FuzzyAce]

What I tend to do is record each guitar to its own mono track, comp together a master DI track. Once I've got all the guitars recorded and tones dialled in, Ill usually commit the plugins and quite often the FX too.

Once the sounds are committed, then Ill organise them so that both rhythm tracks are on one stereo track, lead guitars might go on another stereo track, clean guitars on their own one. Sometimes Ill put cleans and leads on the same track if they dont come up very often or need much work doing. Its sort of playing tetris just to give myself less distraction.

Generally, I think of it like one track should be a single "part". That part may be made up of several performances but really they're combined together where I only need one fader for it. Same is true of backing vocals, there might be stacks of layers that go into them but at a certain point I can just have a single stereo fader to balance them in.

A good tip I got from Tom Elmhirst is to try and get your session so that when you look at the mixer page you can see all of the faders on your screen at once. Thats usually 30-40 tracks which should be PLENTY. Very few analog consoles had more tracks than that, and in the tape days you'd need to sync a few machines to take full advantage of it. If its more, it just means that you are giving yourself control over things that probably don't need it.

I work in a similar way. I do like grouping tracks and sending to a separate stereo bus so may end up with 4 or 5 buses give or take. Often I like to volume (or effects) automate separate buses when it makes sense.

 

[MirrorProfiles]

I work in a similar way. I do like grouping tracks and sending to a separate stereo bus so may end up with 4 or 5 buses give or take. Often I like to volume (or effects) automate separate buses when it makes sense.

I used to do it with lots of busses too, and even adopted a naming scheme to make stuff as visually obvious as possible (on top of using track colours and icons).

Mix session layout and suggested naming conventions | New Zealand Association of Music Producers

Prior to digital audio workstations taking center stage in studios, engineers would make meticulous session notes on paper and package these notes with the master tape. The reason for this was to allow a transparent transfer between different personnel working on a project. The theory goes that...

www.nzamp.org

^This is amazing to adopt for your sessions.

but over time I just found having loads of busses meant more room to have conflicting automation, more chance to miss some routing, and leaving options open didn’t really benefit me at all. It’s either correct or it isn’t, I think it’s generally good to avoid the mindset of “it’s kind of there but i’ll come back to it”. Sessions can get huge and unwieldy so easily and at some point, someone has to make sense of it.

 

[DrewJD82]

Man, there’s a lot of stuff I don’t even know exists that I could probably utilize but I think I’m pretty suck in my ways at this point. I never do comps and my ‘take folder’ is my recycle bin.

I just do take after take until I hear what I want then move on. I learned a while back that leaving myself with options equals giving myself more excuses to spend forever getting something done.

 

[FuzzyAce]

I used to do it with lots of busses too, and even adopted a naming scheme to make stuff as visually obvious as possible (on top of using track colours and icons).

Mix session layout and suggested naming conventions | New Zealand Association of Music Producers

Prior to digital audio workstations taking center stage in studios, engineers would make meticulous session notes on paper and package these notes with the master tape. The reason for this was to allow a transparent transfer between different personnel working on a project. The theory goes that...

www.nzamp.org

^This is amazing to adopt for your sessions.

but over time I just found having loads of busses meant more room to have conflicting automation, more chance to miss some routing, and leaving options open didn’t really benefit me at all. It’s either correct or it isn’t, I think it’s generally good to avoid the mindset of “it’s kind of there but i’ll come back to it”. Sessions can get huge and unwieldy so easily and at some point, someone has to make sense of it.

I would definitely change a few things if I were sharing session files with others but that rarely happens these days. I do try to keep my track count down, nothing worse than having to scroll through a ton of tracks even if I know exactly what is their purpose. Plus I always seem to enjoy the songs where minimal tracks are used -- they always sound better to my ears.

 

[Whizzinby]

What I tend to do is record each guitar to its own mono track, comp together a master DI track. Once I've got all the guitars recorded and tones dialled in, Ill usually commit the plugins and quite often the FX too.

Yeah I think poor verbiage on my part. By “dry” I mean the core guitar tone (amp/cab/effects) but no EQ/Comp or mix related moves. (The latter of which you wouldn’t be able to undo when mixing)

A good tip I got from Tom Elmhirst is to try and get your session so that when you look at the mixer page you can see all of the faders on your screen at once. Thats usually 30-40 tracks which should be PLENTY. Very few analog consoles had more tracks than that, and in the tape days you'd need to sync a few machines to take full advantage of it. If its more, it just means that you are giving yourself control over things that probably don't need it.

That makes sense. Generally I haven’t had that problem until recently now that I’m breaking the drums down slightly. (I was, for longer than I’d care to admit, just having all the drums route to one fader from GGD, and trusting the balancing they defaulted from the plug) This was good for simplicity (one fader) but not for being able to eq/comp each part of the kit individually. I’m now routing them to four tracks: kick, snare, Tom’s, cymbals, which is probably still too few, but it is giving me more flexibility. But the knock on effect is it’s a lot more tracks to micromanage and the projects get more complex. Drums are a pain in the ass.

 

[MirrorProfiles]

Yeah I think poor verbiage on my part. By “dry” I mean the core guitar tone (amp/cab/effects) but no EQ/Comp or mix related moves. (The latter of which you wouldn’t be able to undo when mixing)

Yeah, it kind of takes some common sense - there might be some eq or processing that you know you are going to do so you may as well just commit. and there might be other things that you know you’ll need to refine a bit or may be subject to changing once a mix comes together. Often if it’s broad strokes eq like a neve or helios or api i’ll commit as it’s “part of the sound” rather than balancing it against other stuff.

That makes sense. Generally I haven’t had that problem until recently now that I’m breaking the drums down slightly. (I was, for longer than I’d care to admit, just having all the drums route to one fader from GGD, and trusting the balancing they defaulted from the plug) This was good for simplicity (one fader) but not for being able to eq/comp each part of the kit individually. I’m now routing them to four tracks: kick, snare, Tom’s, cymbals, which is probably still too few, but it is giving me more flexibility. But the knock on effect is it’s a lot more tracks to micromanage and the projects get more complex. Drums are a pain in the ass

with drums, I approach it similarly. Early on I might have lots of channels and as I progress, I can usually settle on a blend of mics that should be pretty close to what I need.

That usually consists of summing all kick mics to their own track, all snare mics to their own track, toms ideally down to 2 or 3 tracks, Overheads on a stereo track, hats/ride/cymbal spot mics I usually keep mono but it depends on their role in the song, and room mics I’ll sum to a stereo fader. I find the sooner in the process I make those blends, the more I can build on those decisions with other sounds and the production sounds more finished sooner. There’s a time and place for having control, but getting things most of the way there and then simplifying things is MUCH better for giving your brain space to focus on important stuff.

 

[Whizzinby]

I just do take after take until I hear what I want then move on. I learned a while back that leaving myself with options equals giving myself more excuses to spend forever getting something done.

I have a great habit of playing a part passably right until I hit record. The record button guarantees I will fat finger timing the start, or my personal favorite, just completely forget the part entirely. (which I’ve otherwise been playing without issue for the last five minutes. )

I think takes might help because once I’m in a loop, I’m just playing freely and thinking less.

 

[DrewJD82]

I have a great habit of playing a part passably right until I hit record. The record button guarantees I will fat finger timing the start, or my personal favorite, just completely forget the part entirely. (which I’ve otherwise been playing without issue for the last five minutes. )

I think takes might help because once I’m in a loop, I’m just playing freely and thinking less.

Hahahahahahha that’s like me sitting down to start a new song and I’ll noodle away thinking I’m recording everything only to realize I never started recording and I’ve already forgotten all the cool riffs that fell out.

I’ve tried the looping/comping before and just let it fill up with takes, but I found I wasn’t giving myself enough of a break between takes and my fretting hand would cramp up faster.

 

[Sleezy E]

1. Is that how you guys initiate takes, setting a loop?
2. Is there a way to have the recording count-in repeated with each pass? The way I’m doing it, it just immediately starts the new take once the loop ends. (Which is fine for some riffy stuff, but tricky on others)
3. Is there a way to break a specific take out of the take folder? Ex, if I’m trying to get two solid takes, that will eventually get hard panned L/R and by random luck nail two takes, is there a way to pull one of those takes entirely out and assign it to its own track? (Leaving the other good take in its existing track folder?)

1. Yes, I use loops for takes. Typically so I don't have to keep starting a section over again. I typically stop when I nail it and use the last take. On solos I might pick and choose bits from different takes. Reaper makes this easy.

2. Yes, you can program a count in before the loop... in Reaper anyway.

3. There are many ways to break a track out. I typically just duplicate the track, pick the take I want from each track and hard pan them.