Mastering compression advice needed..

[pachi2007]

Ruari I agree with you, those are Bob opinions and workflows. I also agree about the way they are presented.But I still think it´s the best book out there, expecially for somebody who wants to know what mastering is about. I think there a lot of concepts that are excepcionally exposed (like dynamics discussed here).Maybe it´s not useful for you who are a profesional but I still think it´s unvaluable
I also agree that you´ll learn more attending a few sessions but what´s the chance for somebody who´s starting? Would you let somebody who´s not a client  sit next to you? If so I´m in... and a few friends too 

Sorry I double posted still trying to figure out who attachments and stuff works  ;D

Cheers

Pachi,

I think the big issue I have with how mastering is presented by Bob and on the internet is that people want to complicate the issue a lot.  Bob talks about techniques like Multiband expansion which are not used by the majority of working mastering engineers.  I would say to everyone beginning in mastering - keep it simple.  One EQ, one good compressor, de-esser and a limiter.  Start there and that is really all you'll need for 95% of jobs.  You won't need multiband very often, or stereo widening or so many of the other "darkarts" techniques that are supposedly the preserve of the mastering giants.

If you ever come to L.A. you are welcome to sit in on a session with me!  I'm sure that if you are serious about learning you can find someone to sit in with where you live.

I understand and I totally agree. I know there are many pros that talk at the forums as if God touched them to speak the truth and they get an unbelieveable respect from the crowd, then use that respect for selling themselves or their stuff. I´m not saying everybody does but a few names come to mind. Gearslutz is one of those forums where you see guys asking things like : Please Mr.X is my C12 ok for recording shakers or should I get a better one?- Yes little grasshopper, C12 is not bad but I prefer this new mike that I endorse that I think it´s the holy grail for shaker recording, but that´s just me of course...
Speaking about mastering I remember one ad of a plugin endorsed by a pro (don´t remember who) that read something like "since I have this plugin my mixes don´t go to the mastering house anymore"
So I undestand what you mean and in fact, looking back at my post I see it sounds a little bit gearslutzish  ;D but what I meant is that I had to read it and go back to it several times to get all the info inside, not "this is The Truth from the man."

Ruairi thanks for your offering, that´s very kind. I work at a recording studio and we always send our mixes to mastering though I admit I´m forced to bring levels up myself with low budget works but I´d never call that mastering in a million years. I also hate to do it for many reasons though I love the process and I´m always trying to improve.
There´s just one mastering studio in the small city I live and for what I´ve listened my grandma can do it better,next closer one would be a 5 hour drive.So I take your word, my wife spent a week at LA last year for work so who knows... it would be awesome.

Pachi

 

[skal1]

ruairioflaherty  i think you've come up with the best idea yet , that is , to find  a ME willing to let me sit on a session or 2.

So if anyone  in the forum knows of  a ME willing to let me sit on a few session that would be great , so i as my  finance dictates i will not able to jet across the world for this ;D.


So i am from uk  and i live Birminghams.


cheers lads good thread..


skal1

 

[ruairioflaherty]

I'm sure there are lots of good folks in your area.  Another good idea is to go to mastering sessions with any bands you know.  I traveled from Ireland to London to sit in on sessions many years back and also dropped in on sessions in L.A. when I was traveling.  Both by offering to babysit projects for bands.

Good luck.

 

[evilcat]

Hi guys!

Very interested thread, here. I'm not a mastering engineer (I record and mix only) but sometimes I put a maximiser on a mix so my clients have a "loud" mix to listening to once they're at home ! But I've never liked Waves maximiser as they tend to make very agressive mids. I recently discover the ISL dynamic processor by Nugen Audio and I have to say it's a fantastic plug-in. I record and mix all in analog, and rarely use plug-ins, but this one is really great, it can be very, very transparent even with 6-9 dB of GR. It's a broadcast oriented plug we bought for TV commercials mixes, but of course it works on everything. Every one should give it a try, you can have free-trial for a month.
When you talked about digital limiting and clipping, which processors exactly ? Are they hardware ? Plug-ins ?

Best,

Ben

 

[barclaycon]

The best advice I can give is to locate the best sounding records you can and then find out who mastered them.
Then you can look into the philosophy, working methods and equipment the guy has used (or not - as the case may be).
I agree upto a point that Bob Katz seems to have a lot to say for himself, and he seems to say it as though it is fact rather than just his opinion, but he does have some good ideas and his general philosophy seems sound. However,  I would urge you to check out his discography to see if you like the way he does things before closely following his 'instructions'.
Another guy who seems to mouth-off a lot about mastering and clearly has a lot of time on his hands is Steve Hoffman (judging by the amount of pontificating he does on forums etc.) Some of the things he says are just plain barmy e.g. 'tape hiss is our friend' (certainly no friend of mine!) and 'Sonic Solutions is the enemy'. (Clearly he doesn't know how to operate it).
Again, a bit of listening research will pay dividends on whether you think his wisdom is valid or not.

Just so you don't think I'm knocking all mastering engineers, I would say that the best guy in London is Ian Cooper at Metropolis. He is someone who genuinely improves the final product and makes it sound loud without crushing it to death.
Not always an easy thing to do!
Checking out the website and equipment lists for clues.
(Masalec mastering compressor for example).

 

[labak]

Hi!

Doing mastering for long and what I could tell!

You are not mastering with your eyes. You can read books to learn basic mastering technics and tools, find great vu-meters to control the job done but do not work with your eyes stuck on, never forget Learning by doing!

You are mastering with your ears. Listen to tracks and tracks, have your references in each styles of music and instruments where you could join a personal feeling to sounds / frequencies. Try to reproduce / find again them with low amount of EQ. Like said before, EQ is 97% of the job here. Learn principes of psycho acoustics effects on ears done by frequencies.

Once again compression is used to give a global shape for a track, adding amount of controlled harmonics, control peaks…

Trying to increase RMS with a compressor tends to give non transparent frequency response. Lows are way to much strong than highs.
Once again learning by doing. Take notes, I have a book full of them.

Hope this helps.
There are no Magic black box or Tools coming from Mars! The only one which is the most personal and build on your own is your ears listenning music on speakers you know very well and in a room you can stay for hours and you know each positions.

Sorry for my "frenglish".

Best.

 

[labak]

When you talked about digital limiting and clipping, which processors exactly ? Are they hardware ? Plug-ins ?

It depends of your tools!

It's like in a kitchen you have different Knives, anyway.

In fact I'm not a big Clipping fan, but you need to know how to do it properly if the client ask for a review much louder.

I'm taking 1db of gain reduction here and there (Clipping my AD, using my L2 hardware, using plugings). I nerver go more than 2db of GR on a gear.

Doing it in the AD says that you need a converter that support this! The most issue with not "Mastering Grade" AD is the analog stage, the input before conversion, that sometimes distorts before converting. You need an analog stage that leaves the transients and peaks unchanged.

In Digital domain, it's way more different because you are working with machine that have "look head" and are doing it on the samples with High bit depth (48 to 64 bits) that is really a lot of possibilities to code your level differences when limiting. They are made for

I never do this using analog limiter, way to slow and non transparent when hitting them really hardly. Sometimes to have a more accurate sidechain on my compressor, kind of look ahead process, I'm sending a delayed clone of my track to master in it.

Hope this helps.

Best.

 

[abbey road d enfer]

For someone starting out I'd strongly advise forgetting about RMS and particularly the idea of target RMS.  I've been making records for just over 18 years and mastering for 10 years and I have never, not once measured the RMS value of a track I've mastered. 

I couldn't agree more!
RMS is loudness. You can't get one without the other. One of the goals in mastering is "making it louder", which results in increased RMS.
OK, so, what is "making it louder" ?
A compressor? A limiter? A gizmo? A louderizer? No.
There is only one thing that "makes it louder": GAIN.
Turning up the gain makes the peaks clip, so one has to find something that prevents clipping, guess what? Some kind of compressor-limiter.
The whole game in the loudness war is there: how to increase the level, which involves increasing gain, whilst keeping peaks under control.
This may also involve EQ and particularly HPF: VLF energy takes a lot of space very inefficiently, since it is not properly reproduced by most equipment (and very often does not carry useful musical information) and can play havoc with any piece of the electroacoustic chain.

 

[Gold]

It's also good to make a distinction between level and loudness. Level is pretty easy to define. Loudness is not. When I'm working the general idea is to get loudness without increasing level. That's the trick.

 

[joaquins]

For someone starting out I'd strongly advise forgetting about RMS and particularly the idea of target RMS.  I've been making records for just over 18 years and mastering for 10 years and I have never, not once measured the RMS value of a track I've mastered. 

I couldn't agree more!
RMS is loudness. You can't get one without the other. One of the goals in mastering is "making it louder", which results in increased RMS.
OK, so, what is "making it louder" ?
A compressor? A limiter? A gizmo? A louderizer? No.
There is only one thing that "makes it louder": GAIN.
Turning up the gain makes the peaks clip, so one has to find something that prevents clipping, guess what? Some kind of compressor-limiter.
The whole game in the loudness war is there: how to increase the level, which involves increasing gain, whilst keeping peaks under control.
This may also involve EQ and particularly HPF: VLF energy takes a lot of space very inefficiently, since it is not properly reproduced by most equipment (and very often does not carry useful musical information) and can play havoc with any piece of the electroacoustic chain.

Sorry, but I understand you can get higher loudness WITHOUT increasing RMS... As Fletcher-Munson said once, if you boost bass and treble level percived (loudness) will be higher without increasing the level it self... Loudness control in many home stereo make something like this (some ones just boost low freq) so when you hear it at low levels you get the sensation you are listening at higher level... Music mixed and mastered to be heared at high levels have less bass and treble than music meant to be heared at lower levels, sometime you listen a song of a heavy metal band and mix is not complete unless you turn up the level (or turn on loudness option on the stereo)

JS

 

[abbey road d enfer]

Sorry, but I understand you can get higher loudness WITHOUT increasing RMS... As Fletcher-Munson said once, if you boost bass and treble level percived (loudness) will be higher without increasing the level it self...

I'm sure they didn't say that...when you increase bass and treble, the overall level increases, so both peak and RMS increase. Statistically, 50% of the energy is concentrated below 350 Hz (it varies with the type of music, but is valid for most western music), so when you increase bass and treble you would have to decrease the midrange content enormously in order to keep the overall energy constant.

Loudness control in many home stereo make something like this (some ones just boost low freq) so when you hear it at low levels you get the sensation you are listening at higher level...

We are all here familiar with the role of the "physiologic" switch in HiFi equipment, and your description is right: it compensates for the lack of sensitivity of the ear at extreme frequencies /low level. It doesn't get louder (at least not more than the actual increase in RMS value).
Now we could discuss for ages the works of Zwicker & Fastl and show that altering the spectrum in the way you suggest can show some increase in loudness with no increase in rms value, but this change of balance would be a major one, making it actually unusable for music.
About 15 years ago, a company came with a product that was intended to increase the loudness of spoken messages in aircrafts. Their unit was based on Swicker & Falst research; it analysed the content of speech and background noise to boost the bands that were dominant in terms of intelligibility and cut those that were unnecessary. It was a total flop; in subjective evaluation tests, most of the listeners commented that the balance was so altered it made the voice unpleasant and anxiogenous!

 

[Gold]

One of the reasons to have a human operating the controls is that loudness is subjective. There are so many variables that it unfortunately comes down to taste.

I think the ITU standard is a very good loudness standard. It was very carefully thought out. It can still only tell you what's there. It can't tell you how to process audio for increased loudness without increased level.

 

[etheory]

The secret to a good mix is mental-disorder-like focus and insane amounts of time practising.
I have spend literally from 10pm-2am for the last 2 years practising (I work a day job and have a girlfriend, so at some point I have to tend to those things also), and before that, probably even more hours a day, almost 7 days a week just practising production/mixing/mastering.

No matter what you read or what people try and sell you, there is a reason why some people are REALLY good and everyone else isn't.
The same reason an olympic runner is better at running than you are. It's a combination in innate ability, determination, patience and a never-ending dedication to the art.

You have to think like the guys that still forge samurai swords by hand, or the monks who still practise kung-fu. You can't get truly good at something by reading or going to classes - you have to DO it. And a LOT.

Whether you like crazy electronica genre's or not, one of the best electronic groups out there production wise is Noisia - and these guys spend hour/days/years/whole lives doing nothing BUT practising.
If you can get your RMS level's to half of theirs and retain the clarity/definition/punch/in-your-face-ness that they do, than you might get an appreciation of what it takes.

My mixes these days are WAAAAAY better than they used to be, purely though Doing first, Reading second, and acknowledging that I still have a LONG way to go third.

It all starts with the production and the mix. Depending on genre your frequency selection and dynamics per sound are WAY more important than a bus compressor or maximizer. You will never get "that" sound by concentrating on the end - you have to concentrate more on the process of getting there. There is no easy win - you can only get good at this by doing more of it.
Genre-dependent things like gating/side-chain compression on individual elements to create space, carving out holes with EQ to fit in other elements, and subtle controlled distortion/harmonic generation will make your mixes sound a LOT louder than any mastering compressor or limiter.

After all that, you then send it to a mixing and/or mastering engineer. I regularly send my mixes to a professional to get a stem mix done (costs about $100, which is VERY cheap), and listen VERY carefully to their feedback, and adjust and repeat, over and over and over again. And I'm only just starting to be proud of the results I'm getting.

I am quite young, but I've been writing music and producing on a computer since I was 9 years old (I have the files to prove it), which was 22 years ago. I've practised that entire time. It's taken almost that long to realize how much I still have to learn and how little a mastering plugin can help me get there.

my2c.

P.S. for an example of the current mixing/mastering work I've done, this is my latest work - http://leesafar.bandcamp.com/album/laundry-after-midnight - I co-wrote, recorded, produced, built the equipment, mixed and mastered almost all that material over the last two years. You might not like it, which is fine, but if you do, then I'd be happy to answer any questions you have about any facet of the mixes or the mastering techniques/gear used on that album.

 

[abbey road d enfer]

"mental-disorder-like focus"

Though you have a questionable way of presenting it  , I tend to agree with what you say. In particular "It all starts with the production and the mix". We're all familiar with the famous saying "We'll fix it in the mix", today it's more like "We'll fix it at mastering". Most of the good records are the result of a vision from the start by the producer of what will be the end result. This way, the whole production process from start to finish should go relatively smoothly - I'm not saying it doesn't take time and effort.
This is 99% of the time. I also had some productions turning out completely different (for the better) than my original vision.

"and listen VERY carefully to their feedback"

This is paramount in improving one's skills.
As much as I can, I submit my mastered files to the producer/artist/musicians. They have all a specific idea of what the product should sound like. Most of the times they recognize an improvement they expect, but sometimes they notice the product has taken a different color/flavour that they may not be entirely satisfied with. In such a case, I'm happy to oblige.

I'm a little reluctant sending stems to the ME, fearing he ends up with a balance quite different that what I envision. Do you send rough mixes with the stems for reference?

"an example of the current mixing/mastering work I've done"

I've listend to the Lee Safar title; although not really my style, I liked it, and I recognize the process you describe: continuity. From writing to mastering via arranging and mixing, there is a common thread.
I have only one criticism: vocals seem to be too much autotuned to my taste. Maybe you'll tell me there is 0% autotune on vocals and I'll haul myself out  ???

 

[tv]

...
I have only one criticism: vocals seem to be too much autotuned to my taste. Maybe you'll tell me there is 0% autotune on vocals and I'll haul myself out  ???

Maybe it was heliconed?
I find it amusing, though that entire mixing proces is starting to resemble what guys that did "multimedia" audio (like a decade plus something ago) had to do to the "mix" to get it across the tiny speakers. I remember bouncing loops from cubase and getting them into samplitude because it had quite supersonic dynamic fx, which I liked better than waves then. "we" were often looked down by the studio pros, but it is quite funny because almost everybody in popular genres is essentially "mastering the subgroups" these days, like we had to do. And yes, Noisia, I read that they are pre-distorting the hf-band to get it across the maximizer so it doesn't end up all mushy. Been there (parallel filtered distortion buss), but with slightly lower freq band. You can get similar crispy effect (but different in vibe) if you overdrive a hi-mid band in a multiband. I also used some custom designed and built "recording preamps" for things like gtrs and bass, similar to what I used in a live setting to squeeze more impact "on dial".

But hey, ... blasphemy
It worked

Of course, it annoyed the hell out of proper studio guys..

 

[Gold]

You have to DO it. And a LOT.

Yes. In the end there are no shortcuts. If you want an incredibly loud and good sounding record that is enjoyable to listen to, it starts with the songs and arrangements. Something like Green Day doesn't happen by accident. If you have to do a lot at the end (mastering) then you have missed the opportunity for the best results possible.

 

[etheory]

Though you have a questionable way of presenting it 

I have a habit of forgetting about things like political correctness. In my workplace it basically doesn't exist, however this is the Internet, and there are people who can get offended by that stuff, sorry

I'm a little reluctant sending stems to the ME, fearing he ends up with a balance quite different that what I envision. Do you send rough mixes with the stems for reference?

Getting a different balance to what I envision can be as useful as them getting it exactly the same.
The issue is that when you listen to something long enough you lose all perspective. So many sensory actions are dependent on others, including even mood.
I read a lot of forums, and constantly read about how "some DAWs sound better than others". Now unless DAW's are somewhat of a religion to you (which is perfectly fine), you'll understand that this can be easily refuted scientifically, and has been. That's right, Ableton Live sounds EXACTLY the same as Pro Tools. You can even null the resulting files to prove it, and plenty of people have. But how things look, and how you feel, change your perspective enough that I often find a professional second opinion is a great thing.

I started the first song on the album (my remix of her track "I'm Here", the only track off of the album that existing before we started - and the one we used to gauge how we'd work together) and had it stem mixed 3 times before I liked what the ME did with it. I didn't know anywhere near as much about mixing as I wanted to (at least, that's what I thought....), but I knew I didn't like the first 2, and, to some extend, I ran out of money before it got to the point I wanted it to, but I had to let it go - that's what happens in projects with deadlines. The next one, "We Jump", was done in 2 goes, and I took everything I learnt the first time, and the ME said it took far less effort to get it sounding well-balanced that time. The final one I had mixed professionally was "Shackles", and again, it was even better. From there, out of money, I HAD to mix and master everything else on the album. "Touch" and "Crave The Night" were produced by other producers, but I mixed and mastered them using all the techniques I learnt by talking for hours and hours to the ME.

Now it's quite possibly that "I'm Here", "We Jump" and "Shackles" are the best mixed and sounding tracks on the album, but, all I know is, that I went LEAPS and BOUNDS after the experience. Abbey I'd have to assume you are far more experienced than I am, but, I would still highly recommend it if you are on the fence. In fact, I wouldn't even necessarily give a track to an ME to finish it. You can always use their ears as a second opinion and then still finish it yourself. I see no hard or fast rules for that.

I've listend to the Lee Safar title; although not really my style, I liked it, and I recognize the process you describe: continuity. From writing to mastering via arranging and mixing, there is a common thread.
I have only one criticism: vocals seem to be too much autotuned to my taste. Maybe you'll tell me there is 0% autotune on vocals and I'll haul myself out  ???

I am absolutely over the moon that you've said you like it (even if only a little bit  ). It's always wonderful to know someone else has gotten something out of it. Sincerely, thank you.

As for the vocals, well, you do what you can. There are strengths and weaknesses to every performance, and, the ones I worked with were both recorded in challenging scenarios, and often completely out of my control (a lot of the time she was in LA when I was here in Sydney, so I couldn't always yell "could you please do that bit again, you were a bit flat or sharp etc.). So you make do with what you have. There is around 8hrs per track just of take-editing, volume automation and manual sibilance removal. I'm not sure why they call it "auto-tune", as the process was anything BUT auto  . I used the manual mode and drew in every note and amount of tuning manually. In some cases I had to push it quite far to ensure it was in the right key and at the right tone. So any artifacting you hear is likely due to the distance the note was pushed. I strived a LOT for natural phrasing, trying as much as possible to avoid quantized sliding and other things that I HATE. But, again, you do what you have to do sometimes. I certainly hope doesn't sound like something by a current RnB artist or something. But I concede I still have a lot to learn, so I'll take that on board.

 

[etheory]

But hey, ... blasphemy
It worked

Of course, it annoyed the hell out of proper studio guys..

My only two comments would be:

1.) I don't believe in audio blasphemy. If it sounds good, it is good, obviously (I know you know that). However, as weird as it might sound coming from someone who listens to hardcore bass-ripping DnB music - I don't care for loudness wars of the RMS-type. Deadmau5 has been one of the most interesting artists to come about of late on this topic, since his mixing and mastering are PHENOMENAL, but NOT loud in an RMS sense. Just well-balanced. This is a revelation for EDM. In fact, even some of Noisia's stuff actually has quite a bit of dynamics going on, even though it sounds like it's absolutely slammed, when you listen for the subtlety, it's there. It's amazing how you can make something sound loud, without changing it's RMS value at all. Perception is a funny thing to play with, and generally doesn't need a compressor or a limiter to happen.

2.) What is a "proper" studio guy?

 

[abbey road d enfer]

I have a habit of forgetting about things like political correctness. In my workplace it basically doesn't exist, however this is the Internet, and there are people who can get offended by that stuff, sorry

You can be sure I haven't been offended the least; just a pique.

Getting a different balance to what I envision can be as useful as them getting it exactly the same.
The issue is that when you listen to something long enough you lose all perspective. So many sensory actions are dependent on others, including even mood.

yes, I hear you. But I have yet to find an ME who wouldn't start by crushing everything. I'm too stingy for experimenting on my own budget. I like it when a producer/artist pays for it. 

As for the vocals, well, you do what you can.

I understand that, the frustration at having to spend hours mending what others have butchered. When it happens, I try as much as I can to let these people know that I had to work extra hours because they didn't, not in an acrimonious way so as not to irritate them, just to let an imprint in their subconscious.

I'm not sure why they call it "auto-tune", as the process was anything BUT auto  .

Maybe just like an automobile still requires you to turn the wheel and push the pedals...

I certainly hope doesn't sound like something by a current RnB artist or something.

Certainly not. I must admit I recently became more and more sensitive to autotune on vocals, but I must also confess that I'm sometimes guilty of it, when the only choice is between a sore note or and artefacted note.

 

[ruairioflaherty]

The secret to a good mix is mental-disorder-like focus and insane amounts of time practising.

OK, I'm going to be a pain in the ass and disagree with the first part. I've watched some of the very best at work and have done this myself for a good long while.  In my experience you do not get to a great result by concentrating ever harder, that's just another part of the Gearslutz myth.

Truly great music is written, arranged and performed, not mixed.  If you have great music find a good balance and get out of it's way, perhaps add a little salt and pepper.  Ff you do not have good music no amount of "mental-disorder-like focus" will make your mix worth listening to.

I do realize I sound like an old man to save any of you the trouble pointing it out.