Curious About How You Guys Mix, What Order/Workflow??

This forum is for discussing Propellerhead's music software. Questions, answers, ideas, and opinions... all apply.
BeckettColt
Posts: 1
Joined: 19 Mar 2020

Post 19 Mar 2020

Just trying to get a feel for what you guys have been taught or believe is the right "order" of your mixing, especially with effects and automation.

Do you automate your BOUNCED tracks? Do you Bounce every track to go to pre-master? Do you have some MIDI tracks and AUDIO tracks then bounce those?
Do you automate your bounced AUDIO tracks only? Or do you Automate BEFORE bouncing?
Do you keep your tracks pretty dry until you bounce? When do you use Inserts vs. Sends?

I just want to see what people believe is the right order or what works best for them, what their habits are.

User avatar
Raveshaper
Posts: 1085
Joined: 16 Jan 2015

Post 21 Mar 2020

This is something everyone does "to taste" the way they like it, so there isn't exactly a single "right" way.

The way I mix is at low volume in my headset. If it sounds balanced and has the energy I'm trying to accomplish without having to crank the volume, it'll only sound even better if I do turn it up.

I do a lot of parallel processing and all channels that make up a given instrument or sound are bundled into a bus channel.

My way of mixing and mastering is a little weird. I lean pretty heavy on surgical EQ and send the meat of each sound into a specific frequency range in a sort of DIY comb filter made out of 10 channels. These channels boost the "sweet spots" for each type of instrument or sound.

The result of the 10 channels is then merged with the dry master bus before being sent to the mastering patch.

My approach favors dynamic range and headroom rather than loudness.
:reason: :ignition: :re: :refillpacker: Enhanced by DataBridge v5

User avatar
chimp_spanner
Posts: 2412
Joined: 06 Mar 2015

Post 21 Mar 2020

BeckettColt wrote:
19 Mar 2020
Just trying to get a feel for what you guys have been taught or believe is the right "order" of your mixing, especially with effects and automation.

Do you automate your BOUNCED tracks? Do you Bounce every track to go to pre-master? Do you have some MIDI tracks and AUDIO tracks then bounce those?
Do you automate your bounced AUDIO tracks only? Or do you Automate BEFORE bouncing?
Do you keep your tracks pretty dry until you bounce? When do you use Inserts vs. Sends?

I just want to see what people believe is the right order or what works best for them, what their habits are.
Yeah as Rave said, no right answer. Me, I basically never bounce. I just run everything 'live'. Mostly because I'm indecisive and always want that option to change notes or whatever. Although there's something to be said for committing to audio; less resources, for one. Easier to manage visually/from an organisation perspective, and maybe it helps separate the writing and mixing phases. Hmm. Maybe I should try it haha.

Anyway yeah, all of which is to say, just whatever works for you. Most everyone will have their own way.

User avatar
Rising Night Wave
Posts: 115
Joined: 03 Sep 2019
Location: Vransko, Slovenia

Post 22 Mar 2020

1. I do a loop
1b. when I add another mix channel I level the fader
2. when I am finished with loop I do stereo image for each channel an panning and level master fader
3. I do a song (construct the song)
4. I do some FXs, fading automation, etc
5. when I am finished with song I go to:
5a. FX return section (solo each track and do the thing: cha cha cha :D )
5b. FX send (EQs) (solo each track and do the thing again: cha cha cha :D )
5c. then go to Master Inserts
5d. and finaly go to Master Compressor

6. then I upalod the song to SoundCloud
7. and IF no plays THEN GO TO frustration
8. frustration :evil: buuuu!

9. IF frustration TOO MANY THEN I GO TO hell
10 HELL: I want to break my computer, want to sell my copy of Reason 10, I start to kick my butt till it hurts and so on... till it is too much THEN I GO TO 1. and start doing new loop

I still love Reason a lot :thumbs_up: :puf_wink: :essentials: :essentials: :essentials: !
Rising Night Wave @ SoundCloud
HW: Asus ROG Strix GL753VD | M-Audio M3-8 | M-Audio Uber Mic | Shure SRH1840 | Shure SE215 | LG 49UK6400
SW: Win10Home64 | Propellerhead Reason 10

rorystorm
Posts: 27
Joined: 06 Jul 2019

Post 22 Mar 2020

wasn't there another thread about this just recently?

I've also got a weird method... so:

First off, I do a rough mix progressively during the recording. This is largely because I tend to write as I go rather than having a pre-arranged idea in my head of how the piece will sound. I do some surgical eq-ing on each instrument - and if I can be bothered on each drum sound but I'm quite lazy I generally just do one eq for the whole drum machine which does me no favours, I know. Plus I also do a rough balance of what the volume for each instrument should be. I find it quite distracting as I'm recording if an instrument I previously worked out doesn't sit right. I'll also go back and rework some of the eq on previous instruments if they don't sound good with what I've recorded later.

Once I'm happy I've finished recording I'll go back at a later stage and do a more exact mix.

For this I do three parallel channels in the mixer for each instrument (and for each drum sound if I'm feeling OCD). Each one goes to its own like, master bus. The first channel is the full mix of the instrument. The second one has a low pass filter somewhere right down the bottom end. The third has a high pass filter right at the top end. Where the filters sit is kind of dependent on the frequency range of the sounds the instrument is producing, and also in comparison to the other instruments. I'll also fiddle about with gain staging each channel. Plus I'll adjust the gain so that, for example the low passed channel for the bass isn't too loud and pushes up the bottom end.

Then I'll eq the master bus for those three channels, maybe add some compression if need be but often not. This is where I do things like put a 30 hz bass cut on the drums so they don't compete with the bass or cut out the low end of a guitar if I'm using one or what ever. Plus the bus is where I do sends and panning.

I find this makes the overall sound richer - bigger low end - and clearer - more emphasised top end.

Normally I tend to work in projects so I'll wait until I've finished what's either an ep or an album, and then do this in one big batch. It's a balance between not exhausting my ears and making sure I have consistent mixes across all the tracks. Dependent on the number of instruments, it normally takes me about half an hour or so for each track. I also have a few things I know I do wrong so I adjust for them even though I can't 'hear' them at the time, like for some reason I mix the drums twice as loud as everything else, so even though it sounds right to me at the time I'm always turning them down.

After that I'll do a rough mix down, and master it in T Racks.

Then I'll have a listen to everything I've done, in one go. Again this is to make sure the levels etc are consistent and also to start getting a feel for what order the tracks will go in on the album. Then I'll go back and listen again this time making detailed notes about what changes to make. Sometimes these can be quite drastic, like cutting chunks of the track out or ending it sooner, but most often it's like, turn down the kick drum at 2.53 or eq the bass more. Normally I leave this part for a couple of weeks at least so I lose familiarity with the tracks and get a bit of objectivity around how they sound. For this I use some mildly shit blue tooth speakers, rather than my monitors, for that does it still sound good on a car radio kind of exercise.

After that I'll do what should be the last mix, give it one more listen and make any final adjustments if needed. Then I'll publish it on my bandcamp page and....no.......one......will.........ever.............listen..............to..................it.

oh! the other thing I do is that, if there's a specific 'sound' I want for a project I'll make a template for it that has an effect or two going right into the master mixer - mostly an eq but I quite often put Scream unit on the tape setting to give it some warmth. Or both.

Hope that's useful.

StrokerX
Posts: 29
Joined: 30 May 2017

Post 23 Mar 2020

Rising Night Wave wrote:
22 Mar 2020
1. I do a loop
1b. when I add another mix channel I level the fader
2. when I am finished with loop I do stereo image for each channel an panning and level master fader
3. I do a song (construct the song)
4. I do some FXs, fading automation, etc
5. when I am finished with song I go to:
5a. FX return section (solo each track and do the thing: cha cha cha :D )
5b. FX send (EQs) (solo each track and do the thing again: cha cha cha :D )
5c. then go to Master Inserts
5d. and finaly go to Master Compressor

6. then I upalod the song to SoundCloud
7. and IF no plays THEN GO TO frustration
8. frustration :evil: buuuu!

9. IF frustration TOO MANY THEN I GO TO hell
10 HELL: I want to break my computer, want to sell my copy of Reason 10, I start to kick my butt till it hurts and so on... till it is too much THEN I GO TO 1. and start doing new loop

I still love Reason a lot :thumbs_up: :puf_wink: :essentials: :essentials: :essentials: !
HAHAHAHA...that was awesome.

For me...I'm still not even sure what "send" in send fx means or how it's different from an insert fx, or if I'm even saying it right. LOL. and I started with Reason 4. I'm sure I do a few things in 10 steps that could be done in less.

User avatar
mcatalao
Posts: 1167
Joined: 17 Jan 2015

Post 24 Mar 2020

Hi.

I prefer to make a hard wall separation between the production processes. This helps me to keep focused on the different aspects of the song, and go from working more musically to a more technical and finalised song.

So here's what workflow for a song (completely made by me including vocals with lyrics) would look like:

1 - Create hook, melody, harmonies and lyrics.

I like to do this with the most simple way. Depends on the mood, what inspiration brings me, so i can be in my living room at the piano or in the studio, with reason with only a piano loaded. If there are already lyrics, i try to make a melody for them and then find a good harmonization, if there aren't lyrics, I just improvise riffs at the piano and sing stuff alongside. When something interesting comes along, i record it or write everything to paper. Pretty much 70's style writing, chord progressions, lyrics, alternative lyrics etc. If i'm in the studio everything is recorded in a draft file in reason, put in a folder with a word doc with the lyrics. At this point, the song is pretty much laid down, musically speaking (melody, chord progression, lyrics, and even some lead instruments ideas are already here).

2 - Sequencing, record vocals and other instruments.
Here i start by rearranging the song based on the paper or the draft project. I'll go from the piano to drums and bass, then guitars, strings, and so on. At this point the idea takes real form, from a simple piano and vocals song into a complete song with the parts defined. This is the part where Reason's Block's are really handy. So, the form of the song takes place also here, from verses to bridges to breaks and outros and endings and so on. I like to think about panning at this point always to keep the stereo field cohesive, so most times, each instrument in a given region will have a pair, sort of a friend that sings along or talks in sort of a counter point between them. I do this to the instruments i know i want in stereo, i won't ever pair lead vocals, bass, snare and kick drum, and probably piano if it is a strong instrument. Lead instruments in bridges or big hooks won't have pairs, but can have voicings. Finally i record vocals, starting by the main vocal, second vocals, then the choir if i want bigger stacks of vocals (all the vocals are doubled, even the main vocal). I always aim for the best take possible so i punch in a lot (not automatically because reason unfortunately does not have auto punch in...).
I try my most not to do mixing tasks at this process, i don't add inserts, i don't parallel compress, nothing… i just do panning and basic levelling if necessary (and if i'm recording, the signal has to go about -18 dbfs with everything at nominal values - so a little bit of gain staging here is necessary when you work with mics, preamps, di's or any other kind of gear that requires level matching).

3 - Edit, project cleaning and mix preparation
So at this point, i edit all recordings, take care of comp'ed tracks selecting the best of best, take care of out of tune notes on vocals and recorded instruments. I also take care of crosses between multiple clips, and take care of noises and stuff that gets in the way.
I end this stage with the mix preparation, I then reorder all tracks and mixer channels if they are out of order, colour code all my tracks and channels, and create busses in advance for the mixing stage (i usually will have a bus for drums and percussion, another for guitars, another for piano and keys, another for synths, a group for main vocals as usually i might have multiple vocal takes, doubling and so on, and finally a group for second vocals and choirs, a group for effects. If it is pop i usually have an additional group for pre master effects, like filtering and stuff like that.

Somewhere in this process i also make a big decision: If the project is getting high on CPU, i'll bounce all tracks to a new project to start the mix fresh, probably between tuning stuff, and color code it to make the busses out of the bouced tracks.

4 - Mixing
Here's the real mixing part, everything till now was song development and preparation. Stuff that if you are only mixing for a client you don't have to do, thought the 3rt part is important no matter what. The third part of my process, and the beginning of the mixing session are of paramount importance, because these are the parts when you get to know your song. If the song is yours, you have an opportunity to take a look at it more objectively, if it isn't, all the navigation through the project, colour coding, ordering, editing and so on, will make you know the track well, and start to imagine how the track will sound.

And this moment there's another very important task to take care. I separate volume in two distinct types, absolute volume and apparent volume. And i created a little rule for myself that has worked for me quite well. I take care of absolute volume in the gain knobs on the mixer and insert paths (either with the output volume of the last insert in the chain or in the absence of it, i use a selig gain as the last insert device), and i take care of apparent volume in the faders. So at this point my faders still stay quiet at 0. And i start taking care of every channels, put the song playing at the highest level of each track and with the gain knob i make sure every thing stays at -18 to -15 dbfs (you have to check this in solo, and look at the mixer's leds in dbfs mode or look at the big knob monitor in the rack).

So, i took care of absolute level. Every decision i make from now on has both technical and aesthetically base.
It's important to say that most times i work with the SSL, with Eqs before inserts. So, now i start taking care of the space of the instruments in the mix, and their relation to one another. Making use of the busses, i will commute a lot between soloed instruments, full instruments, soloed busses, and so on. At this point i go on a loop of tweaks, but i start by eq'ing the channel. I start by filtering the top and bottom frequencies, so almost every single channel will have the HPF and LPF on, even a bass and a drum kick will have HPF on. I start by using the roll the HPF until i hear it kicking in, affecting the bass region of the instrument, then take it back until the frequencies i "feel" i need for the character of the instrument are there. I try to work out problematic relations with filter rather than over compress or side chain at this stage. So kick and bass, i can make space for the kick only with a HPF on the bass, a bit before the lowest note on the song. As i might do all this soloed, but i always go back to the bus, and to the full song. Its very important to test all filtering and eq against the song, so that you check the space of each instrument and it's relation. The device i'll spend more time in the mix is the eq as it has the potential to do so much good or so much evil to a mix in a simple tweak of a knob.

I like to add compression after the eq is more or less defined, because compression will be affected by the tweaks you've done on the EQ and imho you need the dynamics to affect the full signal.

Most times i don't work with patches, i prefer to work the sound down up. But i have bypassed chains for vocals, bass, piano, etc, that then i evolve with the need of each channel. So for vocals, i have a chain with Selig Leveler, Selig deEsser, and selig gain. I'll put this one on the insert after the eq, and rework the vocal level and take care of sibilance. I do the same for other instruments untill the sound of each one and the relation of it with other instruments, is good.

Positioning your sounds in the mix, vertically and horizontally is very important. Positioning horizontally is done by panning the instrument right or left. I seldom pan in LCR (left-center-right), to achieve maximium stereo separation. LCR is great for various reasons. It works great either if you are using monitors or headphones, because you always work for maximum wideness, leaves a lot of space for the stuff you don't want to mix stereo (kick, bass guitar, snare, main vocal all get in a vertical position that they don't "clash" between them. And this is why i like to think instrumentation in pairs, because all the remaining instruments now get on this side positions except when working out lead instruments. Third thing LCR is great for, is that you don't have to worry too much with phasing as you won't have much similar information on the cross talk region, so the mix will be less prone to phasing issues.
But if EQ gives you a sense of space and verticality on the frequency range, all this must sound on the same space, and for this use reverb and delay to mimic reflections of a virtual room. For that i usually start by having 2 reverbs as sends on my template, usually a long reverb with some predelay, with hpf at 600 to 1200Hz. Second a short hall or room delay. The idea of 2 reverbs is to mimic the room behaviour, if i have a lot of reflections from the first reverb, i "feel" it is more near to me, if i feel more long reverb it is more away from me. But most of the instruments will get some amount of each of these one. Other stuff that are at my send section are a time sync'ed strong delay, that i like to use on some synths and guitars, a different delay for vocals that i use for effects, an additional reverb for vocals, with a longer pre-delay and more density, and a chorus or unison to thicken some instruments, choirs or other stuff. Some of these could be added on groups or as parallels, but these are enough. If i feel in need for example, a specific drum reverb, i put it as a parallel of the drum group.

At any given point you always have to take care of apparent level, so you have to re position your fader after these tweaks. But every time you move your fader, don't forget that you need to do this in relation to the song. So try at maximum to take care of this apparent re-Levelling with the whole song playing, not with a soloed instrument or group.

You have to listen and re-listen everything, and at some point you might start to get tired, loose objectivity, or have some trouble to achieve a specific goal. It's time for a cup of coffee or tea, and clean your ears, go for a spin, take some time, hear other types of songs, or nothing, move a bit and so on. Anything that helps you achieve serenity and objectivity.When you come back to it, start by taking a listen to all the song, feel it developing. Try to understand what parts of the song need more punch or pull up your audience and vice versa. Maybe you need the chorus to be stronger, or the vocal to be more intimate at some point. You can make your faders work for you by automating them from the apparent level up down and return. Or add some parallel compression for spice the drums a bit, or add a delay to a vocal at the end of a chorus, or for effect. At this point anything counts to push up the song to a better musical product. If you're working for a client, everything you do here has to be with his/her consent, so ask him/her if this or that is a good idea. But at this point as i was saying anything goes as long as it is technically valid and it brings novelty and adds a good experience to the listener.

You can see once you take care of cleaning the project and the absolute levelling is done the process gets a little less ordered. That's because every decision you make here will affect the whole mix, and you shouldn't be afraid to go back and redo this or that. It's normal to question a previous decision, and you have the opportunity to redo some stuff if needed.

The next step of the process would be the mastering process. My most important advice is don't do it. Give that job to another person for the sake of objectivity. Specially if the song is yours, you should step down and let another person do it.

Hope this helps.

User avatar
motuscott
Posts: 1900
Joined: 16 Jan 2015
Location: the New York

Post 24 Mar 2020

1. Check stash
2. Call my guy
3. Wait what was the question?
Vlad the thread stopper

User avatar
motuscott
Posts: 1900
Joined: 16 Jan 2015
Location: the New York

Post 24 Mar 2020

Old school, sure
But there's something to be said for that
Vlad the thread stopper

User avatar
BradfordMoeller
Posts: 129
Joined: 06 Oct 2016

Post 25 Mar 2020

I have 3 methods:

-If I'm being spontaneous/ experimental, my composition stage is blended with my mixing stage, all decisions are intertwined, then I master as normal
-If I'm being calculated, I will compose fully, then bounce-in-place, then save-as a new file, delete the non-bounced media, then mix, then master. (a method which this community helped me perfect)
-Sometimes, especially if I am mixing/mastering for a friend, I will solo each track, mixing each track individually to taste, then listen to the whole, and make small adjustments to the mix, and then master.
:reason:)))

User avatar
selig
Moderator
Posts: 8449
Joined: 15 Jan 2015

Post 25 Mar 2020

I basically have two current approaches: mixing from scratch, or mixing as I go. I mix from scratch when working on other artists, and mix as I go when mixing my own music.

From there I have two approaches for setting basic balances, additive (traditional, starting with all faders down) or subtractive (more based on a "mix as I go" approach). Additive is how I typically mix from scratch, but not always - currently mixing a song from scratch doing it with a more subtractive approach, which is to say I set peak levels to hit -12dBFS on all tracks, then start pulling down instrument that are too loud, and lifting instrument that are too soft (often using Selig Leveler for this part). With that approach you're always listening to the whole mix, with exceptions for setting important details like a threshold on a gate or similar where you need to be certain you're getting the expected results.
In all cases, I first do only the obvious fixes/repairs/prep, such as cleaning tracks in the time line (removing any talking/etc during the track, fixing obvious mistakes), applying any obvious hight pass filtering, leveling at the clip level, using Selig Leveler for more consistent levels on vocals or bass etc.

As for levels, many already have heard me harp on about no waiting for the Input Gain knob on the SSL mixer to set levels - levels should be consistent before they ever hit the SSL. So audio track not recorded at the reference level of peaks around -12 dBSF get adjusted there. The quickest way to set audio level on clips is to simply normalize all tracks then reduce clip gain by 12 dB. You don't have to listen through or search to find the highest peak, it's already done for you - at this point you know all audio comes into the SSL at no higher than -12 dBFS so no need to do anything further (unless you add EQ or Dynamics, in which case I use Selig Gain to correct levels since the EQ and Dynamics don't allow it). For ALL instruments in the Rack the same rule applies - the output of EVERY instrument should hit the same peak level of -12dBFS, and any effect added should be compensated for so the output of each devices is at -12 dBFS. This not only leaves headroom for the mix, but also allows doing A/B comparisons or bypassing/deleting any device without affecting overall track levels.

The general approach I use is to do as much as possible with as little as possible, before moving on to EQ/Dynamics and delays/reverbs/chorus type effects. So I spend more time at the beginning just adjusting levels and then adding very basic/gentle EQ etc. when necessary. When things get even closer, I'll consider automation and add the Master Compressor. Then finally, when all but the final polish touches are done I'll add some brick wall limiting (Ozone 9 Elements, these days) and maybe some widening if it works on the mix in question.
Selig Audio, LLC

User avatar
hurricane
Posts: 1222
Joined: 14 Oct 2017

Post 25 Mar 2020

I like to split the writing and mixing process. Back when I first started I used to do everything at once. That wasn't working for me because I would never finish anything. So now I get all the parts down and the arrangement as close to finished as possible - and I do it dry, using no effects and I disable any onboard fx on synths. And then when the song is done, I make myself stop - no more tinkering. I then bounce everything to mono and import the tracks into a NEW project, where I have my send fx template already loaded, then I gain stage, and start mixing. I always put the Waves SSL channel on all the tracks first, and then usually the Waves SSL Bus Comp on the master. Automation happens here. That's it.
https://soundcloud.com/hurricane-music | https://soundcloud.com/hurricaneaudiolab
Mac mini (Late 2012) | Logic Pro X | Reason 10.4 | DX7 | DX7s | TX7 | CZ-1 | JP-8080 | Virus Classic | Roland XP-80/XP-30/RD-700/TR-09 | UR22mkII | Emagic AMT8 | SL990 PRO

User avatar
moggadeet
Posts: 55
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 25 Mar 2020

I approach mixing as a Tetris game in three dimensions.

- Volume / dynamics
- Frequency
- Panorama

The task is to place elements in these three dimensions.

Elements playing at the same time? They should not overlap in frequency. If they do, put them separated in the stereofield. If one is mono, put a Haas on the other. Or pan them to the sides.

Elements with same frequency content in the middle? Make them play at different moments / duck one with the other.

And so it goes.

Regarding levels I swear by the constant peak reference method of Gilles. With it I never touch the trim in SSL and mostly just two or three faders at most for any mix.

User avatar
adfielding
Posts: 661
Joined: 19 May 2015

Post 26 Mar 2020

I tend to mix as I go, though I tend to try and get things sounding as "correct" as possible before it hits the mixer - I like to think of it as mixing in the rack. Once I think I'm more-or-less finished with a tune I'll generally give it a final pass with the SSL mixer, as there'll invariably be something that stands out as sounding a bit "off" after I give it a bit of distance.

But yeah, I don't think there's a right or wrong way of doing it. I've tried a few different approaches, and this just happens to be the one that works for me.

User avatar
guitfnky
Posts: 1921
Joined: 19 Jan 2015

Post 26 Mar 2020

top-down mixing, 100%. I guess that's what some people are calling mix-as-you-go. basically, pick the best sounds for the job, setting panning as I'm writing something out, and only reaching for faders or processing when it needs it. most of the synth and plugin sounds out there sound great right out of the gate, so it's a bit of a waste of time to start mixing from scratch, usually.

as I listen, I make mental notes of things that need to be addressed (usually fairly obvious stuff, like, 'the bass timing on that one note in the second verse needs to be cleaned up', etc.), or add clip labels for stuff that's more subtle (correcting a slight pitch issue that's gone unnoticed for awhile, for example). this stuff usually takes a backseat while I'm building the song.

once I have the arrangement laid out, I'll rerecord any audio tracks, since typically as I'm arranging, I'm either using Blocks, or flying a lot of clips. typically at this stage, I'll make sure I'm happy with any drum fills/transitions, and that none of them are identical.

then once everything is arranged and recorded, I start going through and tweaking the stuff I've made note of earlier, if I haven't already. now I'm firmly in listen mode, and really just trying to see if anything that needs attention jumps out at me. if I hear something, I make more notes to fix them, and keep listening, then go back and address them. I don't solo tracks and listen for every detail, because no one cares, as long as everything's working together in the full mix. usually the last step is tidying up the audio clips to ensure they're all faded in/out, and then applying a master fade out at the end of the track.

  • Information
  • Who is online

    Users browsing this forum: CommonCrawl [Bot], stefan, teddymcw, Trendiction [Bot] and 9 guests