PlanetZ forums - scopeusers.com

Hi gents,

This is a simple question but the answers are probably trivial. I find it hard to find a good setup or decent method to mix down, find myself doing it different each time.

So what do we got? There's VSTi and recorded sources which can be mixed on the sequencer's mixer, except for the ones that need more SFP processing, they get their own busses. Both the sequencer mixdown, SFP processed sequencer stuff and SFP synths can be mixed on DSP. After that, I have 10 outputs to an analog mixer where the final stereo mix is recorded from.

I usually start with a very basic setup. The sequencer is empty, no tracks or busses pre-made except for 4 ramp (cfr. Flexor) outputs. Also the MIDI remote is set up, and a couple of devices are programmed to be controlled from within SX.
The SFP projects contains just 24 ASIO sources, 10 ASIO destinations, and the MIDI routing is also layed out as far as possible. Next I usually throw in the STM2448, but that depends on the size of the project I have in mind.

One could set up tracks in the sequencer that are preset to contain drums, MIDI as audio, to contain synths etc, and route them out to different ASIO busses. I prefer to do this as the project grows, I feel a little restricted and pushed in one preset direction when all stuff is already layed out.

But then... the mixing. Stuff is runnin' and pumpin' all over, the song starts to work out. Kick, snare, hihat go on their own buss to the analog mixer, so I have direct control over these principal drums. Next, I think some grouping needs to take place, outside of the frame of routing that goes on for send effects, intermodulation etc.

In a classic rockband, one could first balance the drumkit, then the drums with the bass. Then add and baland with drums and bass a guitar and a voice. In the eletronica soup that I tend to make however, there is no such thing as leads or well defined groups of instruments with their almost predefined place in the song. How do I balance all this stuff easily? How do you manage all the blips, toms and sweeps? How do you group them? Do you make arrangements as in 'these sounds are the hook', 'these other sounds are the foundation', 'here's a bunch of accents' and group these together and mix them each on their bus?

Now that I'm starting to make more complex projects, this is getting quite confusing.

Looking forward to read your thoughts on how to nail these mixes down!

at0m.

Hi at0m|c.

I'm the wrong person to answer your question, since I really prefer rock music with only 3-4 voices.

Nevertheless from listening to lots of orchestral music and "big" rock mixes over the years, I would say that you should set a few voices as the focal points, and mix everything else relative to them.

If you want a fantastic rock example, pick up California by Mr Bungle. It is so dense that it gives me "mix nightmares". But it still sounds dynamic, coherent, and very musical.

OK I'll be quiet so the electronica folks can speak!

Cheers,

Johann

I like the musical content of california . but the mastering is too compressed to my taste.
but it's far better than the last tomahawk ! which is horrible concerning the mastering !
the level war !!!!

it's a tough issue.. Recently, I'm quite comfortable with a template project in SX that routes Sampletank's chan 1-8 to STM2448. On SFP side I have around 32 ASIO in and 4 ASIO out. I don't bother with mixing anything in SX, it just makes things too complicated. Starting from this, I just add as I go along.

About grouping things together, I tend to group according to EQ and comp type... which is a rare match so I usually end up with lots of audio channels anyway.

It really depends on how hi quality the project needs to be. With quick projects, I can get away with putting all basic drums in one channel.

But I guess as you know from my music, I don't use too many tracks for my tunes so you may be facing issues that I'm not familiar with. I usually stay under 16 tracks most of the time, unless it's an orchestral tune.

The opinions expressed by nprime are not necessarily those of the network, it's producers, or frankly anyone else on this planet:

I'm a firm beliver in simplicity.

The single most important button in mixing is the "mute" button.

You have to constantly ask the question "what is this doing for the song".

When it seems like you just can't get a acceptable balance the reason is often because there is just too much going on. The human ear gets confused and doesn't know what to focus on, so the listener gets fatigued very quickly. If someone is singing you really can't have much more than drums, bass and one chordal instrument happening. My own rule is no more than 3 or 4 things at once, it just creates a nightmare (which might be what you are after, who am I to say?)

It's partly MIDI's fault, because it is so easy to keep layering up sounds.

If you've got things in the mix that have be -9 Db relative to everthing else...then it probably didn't need to be there in the first place.

The second most important button is "bypass". As musician/producers we have a tough job. We listen to our song hundreds of times. We think it will sound better if we change something. We should all have a big poster over the computer that says "DIFFERENT IS NOT ALWAYS BETTER". And nowhere is this truer that when it comes to effects.

Most especially reverb. If there is one thing I have learned over the years it is that reveb seems to do exactly the opposite of what you expect it to do. It only makes an instrument sound "bigger" when it is soloed. Excess reverb on individual tracks will result in mud, and the rest of the mix will be mired in it. it won't sound bigger it wil sound flatter.

Especially if your next step is to brutally limit the stereo signal to get more "volume". Punch requires dynamics. You cannot have dark without light. You cannot hear "loud" if you don't have "soft" for contrast. Loud all the time just isn't perceived as loud for long, the ear fatigues and starts tune the monotony out.

"Damn it!" you say, "I paid for these plug-ins and I'm going to use as many of them as I possibly can in everything I do! How dare you suggest that people can't tell when I have that subtle little chorus sound on the string layer in that giant multi layer chord? Of course people can hear that I boosted the arpeggiateing bass sound by 2 dB at 600 Hz just in the choruses. These little things make all the difference in the world!"

I know, I know, but's it's all going to be all right...it's O.K., we're from the VST clinic, You just calm down now and step away from the computer, good, good...O.K. now you just let us put this "special" little white jacket on you, that's right, now you're just going to feel a little prick in your arm...that's better now isn't it?

Now we can tell you the truth, only other musicians can tell, and even then I often can't hear what somenone has done from one mix to the other. It's only apparent when you there and part of the process. Layman do not hear these things. They feel a beat, which they either shake their butts to or not, and they hear the singer, if there is one, or the lead line if there is not. They are not conscious at all of the mix subtleties. It's heartbreaking I know, I cried for days when I finally admitted it to myself, but it's the truth. It's why your pelvic affiliate cannot understand why you spend untold hours working on what to them sounds like the same song.

Sorry to be the one to break this news to you, but someone has to, it had to be here, it had to be now. Production is seldom what makes a great song. It's all in the writing, I'm afraid. The beat the melody, or the lick.

Everything I say here is a set of rules that were meant to be broken, and do not apply to many forms of music. nprime is in no way legally responsible for any damages arising from people failing to use all of their plug-ins, all of the time. Do not ingest this material, if the material is accidentally ingested I suggest a thorough cleansing with beer, until everything you read here has been erased.

Yahoooo! It's great to be a musician here and now!

R


<font size=-1>[ This Message was edited by: nprime on 2004-09-04 16:46 ]</font>

<font size=-1>[ This Message was edited by: nprime on 2004-09-04 16:48 ]</font>

Sounds like you had a tough day at the studio nprime..

When it seems like you just can't get a acceptable balance the reason is often because there is just too much going on. The human ear gets confused and doesn't know what to focus on, so the listener gets fatigued very quickly. If someone is singing you really can't have much more than drums, bass and one chordal instrument happening. My own rule is no more than 3 or 4 things at once, it just creates a nightmare (which might be what you are after, who am I to say?)

Totally agreed! 3-4, maybe 4-5 is the max, if you don't count the individual percussive elements. If you need to layer something, do it in complete unison. I hate it when there are too many ineffective parts just filling up obsolete space. 1 part playing a well balanced part is much better than 5 parts trying to fill each other in. (because you already have 5 other parts) It's one important lesson that'll help the writing, and the mixing.

<font size=-1>[ This Message was edited by: kensuguro on 2004-09-04 22:16 ]</font>

oops

<font size=-1>[ This Message was edited by: kensuguro on 2004-09-04 22:14 ]</font>

how do you edit "oops", did you spell it wrong?

R

Less is better!

less is better ? I think more is bettter, the bigger the universe for your audience to play/dance in, the easier they will get lost.

i've settled on the stm2448, and the asio2 32 64 drivers, 16 source and 2 dest.


everything stays SFP if possible, which is fairly easy as the only vsti i use is microtonic, which goes straight to the 2448 anyway.

i have u-control uc-33 for inputting automation

I sometimes bounce synths to audio if extra reverbs etc. are needed for the mix.

Optimaster, PsyQ and sometimes some grapheqs, back in to cubase for a stereo track.

pretty stock standard method, but there ya go

A tune I've been working on over the last couple of weeks has helped me to agree very much with what nprime had to say.

It's a 'chilled D&B' number - I got the tracks laid down in about two days. I then spent the next week 'trying to add to it':

"Damn. There's only drums, bass and a pianoey sound in that section - surely it needs something else!"

Nothing I tried worked. There was a constant voice in my head saying "Actually, that sounds OK, maybe it doesn't need anything else..."

A super-subtle soft pad maybe? Nope.

In the end, I just concentrated on getting my rather minimalist piece sounding as nice as I could and left it at that.

This is the frirst time I've tried using the STM2448...I've used the STM1623 up until now.

I'm really not sure about the built-in (channel) compressor(s). I tried using it on the bass & bass drum but however I set it up, it actually reduced punch and presence. I gave up and used the standard SFP effect as an insert in the end. I think I might go back to using the 1632 mixer as I just have a 'better feeling' about it.

Royston

I'm really not sure about the built-in (channel) compressor(s). I tried using it on the bass & bass drum but however I set it up, it actually reduced punch and presence. I gave up and used the standard SFP effect as an insert in the end. I think I might go back to using the 1632 mixer as I just have a 'better feeling' about it.

I talked to creamware about this. They say, the algorithm are exactly the same. And so same DSP use etc.
But the ranges of the knops are different.
The different sound comes from the slightly different settings you make. Ajusting the attack time is a little bit more difficult in the channel comp.

Nice description of my situation with almost every new song, Atomic, except for the outboard mixer

My solution is to record every track (again) as soon as I feel sure about it's function and sound.

Mostly I keep the reverb part 'life', because it's too delicat to decide about it too early.
This way I can remove the VST's and VSTi's along the process, as well as the DSP consuming synths in SFP.

This gives room for mixing tools as PsyQ, SPL Transient designer, HQ reverbs and Optimaster.

General principle: less is more (indeed), and if things grow out of controll, record in audio, but keep things recallable, to overdo a recording if needed.

Tip: keep track of bouncing processes by saving them exactly after you did the recording, save SFP and Sequencer with identic name, like 'Songtitle/instrument(group)/bounce'.
This makes it easy to go back to that stage

And everytime again I resolve to keep my next song really simple, with just headlines and a good grooving beat...
So far I failed always
cheers.

Like nprime said =P.

Except, I must admit, I can easily hear stuff that sits at -9dB/-12dB. Sometimes the Devil is in the details =P. Subtle is ultra-mega powerful. Especially if it's something that gets played on fairly large soundsystems.

But yeah, less definitely is more. I've spent the last years trying to put as little as possible into tracks, and it's been really rewarding.

Go nprime and symbiote!!!!

This is all great information! I love reading how other musicians setup their studio. Before I got SFP (last week) I had a Live template with Sonic Synth 2 hosted on one midi track, eight midi tracks for recording parts with SS2 instruments, and eight audio tracks for recording the audio coming from the SS2 instruments. If I needed real drums, I'd add a Jamstix track. That was it. Now that I have SFP, I'm trying to figure out the best way to integrate it into my setup. I created a new Live template that has my trusted SS2 track, one midi track for each SS2 part, and one audio track for the SFP mixer. This way I can send everything to the 1632 for adding CW effects. I just haven't figured out how to get everything back into the Live master track to record. If SFP had a MIDI sequencer and audio rendering, I wouldn't even need Live. It would be a complete music making package.

Peace /
Chisel

really an excellent thread...

haaa, splendid atom, have you tried the 48/96 big mixer? anyone? The current project we're remixing is quite big and running out of tracks with the 24/48.. we kinda like working out 50 tracks clusterfucks

The 48|96 works very well. (As does the 48S surround mixer.)

ppl - in what kind of cases you using more then 2 asio destinations?