After I have mixed down my track at 24bit 96khz.. i usually bring it into this software called Har-Bal to push down any frequencies that are sticking out and pull up any frequencies that are buried, I also roll of super low frequncies, usually at somewhere between 35hz or 45hz and mold the low end to have a decent 6db/octave curve. with a little bit of a dip to allow the kick to do its magic. The high end is usually rolled off sharply at somewhere between 14-18khz to end at 20khz (do this only if you are going to be dithering down to 44.1khz later. Then I render that and bring it into soundforge. My mastering chain is usually some SSL compressor clone followed by izotope ozone followed by wave L2 or L3.  The SSL is usually just to glue things together a little bit. usually only a db or two of gain reduction.. fast attack.. usually auto release. Sometimes I will parallel compress it so its not so heavy. I will then bring it back up to about -3db and start focusing on ozone. I dont usually use the presets they have. On a decent mix I can usually get away with two bands of compression, separated at somewhere between 200-400hz. The high band may or may not be engaged at all. The low band is generally really low ratio.. 1:2 or 1:3 with a medium attack of 100ms or so and a medium release of 200-500ms. 2db of reduction or so. This gives the low end some spring. If Necessary I will push up the makup gain on the low end a little. All of this should be done while the limiter is pushed up to reducing 2db or so. Smooth setting (3ms or so) is usually good and the meter is usually in RMS mode and should be peaking at around -12db. after giving the low end the spring, then I can focus on molding the mids and highs with eq. After all that. Usually I can get another extra few db out of L2 without hurting the tone much. If I need to, I will tweak the eq a little bit. I set the dithering to 16bit type2 moderate in L2 or L3 and render it. If its gonna be mp3 I usually set the ceiling to -0.3db. The encoder is gonna smash it a little. After Rendering, It should be good to respample at 16bit 44.1khz.. but i usually just mp3 at 16bit 48khz. Soundforge seems to introduce some weird noise at 44.1 regardless of your dithering. So if im gonna go to 44.1khz i will use an external tool. So that's pretty much it..;)

Ummmm.  I press record then render.  Imma go feel inadequate now.

Ditto

ah yeah. totally on the subtractive EQing bit.. always better to cut rather than boost.. and usually not more than 3db when mastering unless you are using some fine point FIR EQ or something to be surgical. as far as compressors sounding good.. well. just gotta find the right settings and try to understand the character of whatever compressor. no matter what. if you compress you are going to be introducing some distortion.

Some really interesting stuff here; the main reason I came here was to improve to at least adequate level and knowing the workflow can help.

For me, I start in LMMS and finish in Audacity with nothing in between, but that was just my one song.

I'm a bit tired, but here's a rant :-)

I work from a pretty modest studio setup where the roof is quite high and it's a very "live" room.  I've floated a "soundcloud" above the op, but sound still bounces around a bit.  I've popped dampening on the backs of the LCD and made some panels from Acoustisorb 3 (covered) on sections of the walls - but not the whole lot.  So it's still got a bit of air to the space.  I think with the emphasis on spending big on speakers and sound modules that people forget that a few hundred bucks on room treatment is way more beneficial.  It'll outlast your speakers and certainly your computer...

I find a good listening spot is vital too - which stupidly can come down to a good solid computer chair which invites you to sit in it.  Mine was an 18th birthday present years ago and I've never looked back.  Also getting your speakers away from the wall (a bit) but a meter or so in front of you in important.  I edit & mix dialogue for films / TV / conferences during the day so I've got an older set of Tannoy actives - before they went cheap - and have them sitting on a pair of Mission's (equiv of the 796se).  So I combine Hi-Fi speakers and Monitors when I mix.  Yes.  Recommended technique?  Uhhh... no, it's almost in the "NEVER DO THIS" category!  But I don't care, it works in my space and sounds great.  And that's the thing, you gotta find what works for you.  On top of that I like everything in front of me at once like in a "real" studio, so I have a few screens to compensate for my... uhh... pet hate of hidden windows.  It's not ideal for mixing at all - but at the end of the day, if I want to mix/master something huge, I'll take it somewhere pro. 

You can't always stick by the rules though... often because a lot of documented techniques rely on having semi-decent gear and software.  Back in the 90's when I was still in primary school and computers still had "turbo" buttons, I had a Tascam Porta 03.  It's mix capability was volume / pan. 

Everything had to light up the 0db LED and touch +6 on record else the hiss would take over.  It would distort a bit at +6 but I saved my pocket money for good tapes.  I was literally mowing the lawn for TDK SA90's.  I didn't have aux sends or even EQ so all decisions were made during recording.  I'd do a series of 10 second "test" recs with a beat, bass line, guitar / whatever and check it all lined up then tweak various settings on amps, etc, do a test again, then write down all the settings to use.    I used to turn Dolby NR on for selected tracks when I recorded (mainly drums / clean guitars) and turn it off again on replay.  Then I'd throw the biasing out on the record deck when I was doing the mixdown.  This would saturate the master - preferably a TDK MA60 - and get more balls.  It was all black magic and experimentation back then - weekends of fucking around with settings as an 11 year old. 

To this day I still commit to "a sound" on recording.  Partly because I do most composing with waveforms that I've played live and recorded.  I play a sequence, tighten up the notes if I was out, group the phrase, position the phrase on the timeline.  If I want to change note pitches in a sequence I just isolate the waveform and nudge it up or down by x %.  Octave down is div 2... everything else is div 1.x where x is the number of semitones div 12.  I figure my notes out by calculator and I sometimes throw individual note tunings out a bit to thicken up the sound and add imperfection.  This is just how I happen to write music (I won't expand on this technique because I know I'm the only person who works like this!!!)

I usually work at 48khz with 24 bit mixing.  And master at 48khz, dropping back to 44.1 before final limiter is applied.  My master clock is set to 48k for broadcast work and there is less jitter working at your native sample rate.

Once everything is mixed,  my mastering is done in Audition 1.5.  I do not use 2.0 or later!  The primary compressor in the original Cool Edit Pro / Audition is still one of the best IMHO - very mathematical and non-transparent.  I usually use EQ with an british EQ plugin or run / sum a mix though my mates Hill Audio Concept 8400 (which is down the corridor).  This is usually to set the master brightness, get the lower mids in shape and set the overall tone.  Sometimes external summing, esp on more rock or bluesy stuff works a treat, but I haven't had time to do this with WB.  Occasionally I add a bit of transient management and then a healthy dose of multi-comp limitor to slice off the fur. 

I usually avoid slamming my tracks with the multi-comp limitor.  (Simply using L3 is a good example)  It should already sound slamming before the point of final limitor - don't be fooled by the volume going up - you might be losing a lot more power than you realize.  Pull the brick wall down so it's approx the same level as the input.  What are you losing?  Is the emotion and space still there?  I then go and tweak sections of waveform AFTER I'm "mastered"... especially on build ups.  I sometimes recompress and double limit heavy sections in tracks too. 

One thing I do which is archaic is I always master in a destructive wave editor.  I never master with an effects chain.  I process and commit; and if I revisit something earlier on the chain, then I undo, undo, undo.  It's just how I work.   I like to see exactly what's happening to the waveform after I apply something and I have a second 27" monitor dedicated to spectral analysis and phase metering.  I'll often drop in a LPF @ 100hz and check nothing too crazy is out of phase down there.  With modern subs passing at up to 120hz and then summing, you gotta make sure it's in phase in the arse end too.   Most of my mastering workings are about visualizing the speaker cones in my head and working the mix from a "moving air" point of view. 

All the saturation and warm fuzzy stuff happens at the point of recording as far as I'm concerned.  The Big thing to remember is that the master compressor at mastering stage will fatten things up and the limitor will pick up the prickly peaks, so don't get caught up on crunching the sound too hard at the mix / record stage.  Parallel comp on drums or a vocal harmony grouping can often be a great technique to get things bigger from the bottom up, rather than slamming from the top down.  But that's more of a mixing thing.....

I think a lot of people expect "mastering" to fix things that should be addressed in the iso's.  I use Reaper as my primary DAW... for everything mix related.  Within each of the isolated tracks during mix I'm running pass filtering on anything which lacks a low freq element, and compressors (if they need it) but most importantly just make sure the instruments sit well together.  A lot of a good master comes from concentrating on the tone at the source of the sound rather than butchering it with EQ or a comp after.  Sometimes that means committing to a sound before it hits the soundcard.  I've got an Orban 424, which is actually a TV broadcast box designed for compressing news desk close mics / lecterns, but it sounds great on a vocal chain or bass guitar alike. 

I got it for $20 at an auction of old TV station equipment.  It's one of those weird super expensive & over engineered boxes made for the broadcast sector in the 90's which sounds amazing in a studio.  And mofo's from TV stations throw things like this out these days.  I've also got DBX mic pre's, and this lil Behinger one with it's tube switched for an Electro Harmonix one.  I usually just run guitars / bass /synths via the DBX in a lot of situations.  In the case of my Jem, I just run it DI.  Lots of times it's just a matter of plugging in and listening... maybe recording a section quickly and previewing it.  Sometimes DI works great.  Sometimes micing up an amp or running though a pre works better - really depends on the track.  Sometimes a little bit of comp just adds the power and confidence the sound needs.  Certainly with vocals, outboard can be great for a vocalist as their voice monitors as being big and powerful and they are less likely to flinch when they sing loud and it sometimes means you'll get more out of them!  (or yourself if you are singing)   That can actually go for a few instruments and sources, but be careful not to overcompress because you can't undo this once it's recorded.  Some people like to actually monitor back though a compressor, but record the audio direct... that way you get the emotional kick back from a comp, but you can choose your own options in post.  Yes.  I get emotional over compressors.

Like George Bowles points out, bass managment never goes to far astray.  I find MaxxBass is worth the dosh (mine came in a multi-pack 9 years ago). 

Basically it reprocesses everything below about 70hz - depending on the settings - and recreates the bass harmonics to sound fatter.  But it also compresses the s**t out of the bass freqs so you still have the character of the sound in the mids, but with a more confident wall-of-sound bottom end.  Sometimes this works, but other times it can be muddy and you need to ride the maxxbass fader.  Sometimes it's better just to leave a sound raw.  Different bass managers work in different ways too... so some will work in situations where others don't.

One of the most profound mixes I have ever seen (on video) was a Led Zeppelin recording where the guy puts on a 1" 16 track multi and pushes all the channels up to around 0db flat.  Everything sounded pretty good without any EQ or additional effects and I try to exercise that philosophy too.  I might record a bit bright or a bit bass heavy to get a vibe right but I'll be conscious of that.  When I'm playing I want to get lost in the mix and I want things to sound studio-esque while I'm tracking.  The less I need to play with a sound once it's recorded, the better.

All this said - most of my Weekly Beats are tracked are mixed in a hurry.  It's teaching me things, mainly what I can get away with, but also frustrating me that the mixes are a bit sloppy and hap-hazard.  But this is the idea isn't it?  Constrained time?  "Just do iiieeet"?!

Finally, a book I found helpful - esp if you like using mics:  Mixing with your Mind by Mike Stavrou.  It's full of whacky first principle techniques that are really fun to read about.  Chasing the flame, flavors of mic, whole section on parallel compression and lots of studio stories.  A good read for sho.

Last edited by cTrix (November 9, 2012 10:32 am)

My setup is as follows: Two screens, FL Studio 10, a few downloaded plugins, and a cheap pair of Sennheiser headphones. All I've learned about music has been in the past three/four years messing around on FL Studio, and whatever I've managed to learn from band class (french horn!). So, yeah, I have NO FREAKING CLUE what the heck you guys are even talking about. I wish I had some way to learn about mixing and engineering a good song, but being in high school still, and not knowing any experts in real life, it's hard to learn stuff. Any suggestions on where to go to understand all of this?