PlanetZ forums - scopeusers.com

I'd like to open this thread about mastering. How do you perform this final step?
1. Do you record your final mix in the DAW (eport or merge) or externally with Wavelab or even VDAT?
2. How do process your mix (MASTER) and in which order (EQ/normalize/compress)?
3. Do you use OptiMaster or PsyQ or any other comparable tools (e.g. L1)?
4. How do you monitor? Speakers or headphones or both?
5. How do you check your Master: analyzer, different speakers, car stereo? Do you a/b often?

Share your knowledge and experience!

I'm a bloody amateur in mastering, but perhaps you continue with reading:
I do the final mix in my DAW (Pulsar1 + Luna2) with a mastering compressor from SonicTimeworks as an insert in the master channel. So the mix grows very crisp and strong and lively ...
Sometimes i record this mix once again with an analog tape simulator, then it sounds warmer.
Greetz, Thomas

I forgot to say: To check the result, I try to hear my music at the DAW, the HiFi, the ghetto-blaster of my wife, in the car and in many locations as possible.
Mastering is good with both, monitor AND earphones.
If you have mixed a long time, stop working, go bed and continue next day - with tired ears you will get some strange results.

I'm interested in this subject too.

I doo know about this link though filled with articles on every subject regarding making music on your computer:

http://www.studiocovers.com/articles.htm

Perhaps you will find it usefull.

Thomas

I don't consider myself to be at all qualified for this matter (there are people who ONLY do mastering.. so) but I'll try to offer what I know. This is one BIG topic, and it's going to be very long. But bear with me. It's mastering we're talking about. (crowd goes "Oooo" )

People expect different things out of mastering. It could range from applying minimal EQ just to keep the sound consistent througout the album, to EQ plus mega-boost in RMS power so that the tune rocks, all the way to using it as a part of the sound creation process. So the setup and required technique varies as much as the different styles of tunes being written all over the world. I'll try to be very general.

Here's my setup. Some may agree, some may not. But it just works for me, for now:
CW Parametric 4ch EQ
Celmo Subbass (or self resonating highpass filter tuned to the "woofing" frequency -> 50hz or so)
CW Masterverb
DADEV Multicomp (or Finaliza)
If you use Finliza, maybe you can add a limiter in the end.

The relationship between the devices work like this:

• EQ+Multicomp work together
• Subbass+Reverb work together.
• Multicomp+limiter work together like mad.
• EVERYTHING before Multicomp affects multicomp.

So, as you can see, the whole thing is rather messy and chaotic. The relationship goes in circles.

Let's start with Subbass+Reverb. It's very optional. I don't use it that often unless the materiall has to sound very boomy. (but we're all addicted to bass so.. hehe)

Only use subbass if the source material truely has no 40-100hz zone. If it does, it's better to bring it up during multicomp. If it doesn't, give it some. Put it through subbass, lower the original signal a bit. Be very conservative when you add subbass, because once the frequencies are there, you can then fiddle with it in multicomp phase. Being conservative means imagine yourself lost in the jungle and you only have 1 chocolate bar to survive a month. Take too big of a bite, and you're going to have trouble further ahead.

The reverb unit after the subbass is to compensate for when the bass signal is too short. (shorter than 1 wavelength, which is like 100-150ms, depending on its pitch) If you feel that the bass is too "not there", too short and.. not punchy, but just plain snappy, it's time to use the reverb. But you want only the 40-100hz zone to be reverberated almost unnoticably, so make sure you set the lowpass filter on masterverb all the way down. Fiddle with the right rev time and ER settings. (pre-delay should be 0) Again, be very conservative.

Again, subbass+reverb shouldn't be used too often. It's just too easy to mess things up. The best would be for your source material to already have the bass frequencies in it. If it does, though, by pass subbass+reverb and switch to EQ+multicomp+limiter combo.

EQ.. I use it to mainly compensate for lost mid treble to high treble. You can mess with the bass in multicomp.

Multicomp is the hardest part of them all. Things like optimaster, T-Racks do this part automatially, or have pre-set parameters. Doing this by hand takes an enormous amount of practice and experience. But don't let automation take this part over because it's the most enjoyable part!

I usually do multicomp channel by channel, checking the total mix as I go from channel to channel. Depending on how you want the material to sound, the priorities are different: (in order of priority)

• LOUD= mid, treble, bass
• boomy= bass, mid, treble
• transparent= treble, mid, bass
• thick= treble, bass, mid

These are very general though... so just use these as starting points, or if you have priorties of your own, use them by all means. The priority here, refers to the amount of mashing you do in the comp phase.

However, aside from the priorities, each band has its own characteristics that you should consider.

Bass: You want the bass to breath.
Attack/Release: Give it lots of attack time if you want to highlight it.
Threshold: Typically can be set pretty low, because you can carve alot of bass, and still have it dominate other things.
Ratio: Depends how much you want to mash it.

Mid: Most varying part, depending on material.
Attack/Release: Keep release at minimum value if you want to be very loud and coarse, but if you want to keep things under order, set release very long (150-200ms, or even longer) This prevents obvious pumping.
Threshold: Don't have it react to everything. Look for the loudest part of the tune, and adjust so that the comp reacts alot. Play the softest part of the tune and make sure the comp dosn't react. But of course, you can mash everything as well if you want to be loud.
Ratio: This is the most noticable part, so don't set ratio too high. Try these combinations if you're troubled...

• high ratio+high threshold+little gain (for minimal intervention)
• low ratio, low threshold, lots of gain (for natural intervention)
• high ratio, low threshold, lots of gain (for maximum intervention)

Treble: Can make your tune very uncomfortable if you mess up.
Attack/Release: Don't mash it too quickly. Give it around 30ms or more. Release an be fairly quick without being too obvious. But if you've got shimmering pads, give it longer release time.
Threshold: Set it low to mash it, and make things irritatingly loud. Since treble is usually pretty low compared to other bands, it's ok to set threshold pretty low.
Ratio: Try to keep ratio low. An obviously mashed treble will mess up the definition of your tune. Usually, just using gain will give you ok results.

The last process in the link would be a limiter , because the multicomp process may cause some strange peaks. And just to keep things under well order. Here are two extreme examples:
Smooth Overall will smooth things down.
Attack/Release: 0ms attack, long release (200 or so)
Threshold: Mid-low. Reacts to lots of the louder parts
Ratio: Very low

Clinical Fix will remove specific peaks.
Attack/Release: 0ms attack, short (90 or so)
Threshold: High. Reacts only to the peak you especially hate.
Ratio: Infinite

So that's it I think. Mastering is really hard to explain. I learned what I learned so far, spending about 2 years experimenting... and believe me, I learn something new with every tune. It's ALWAYS different. No one trick ever works. My earlier experiments really sounded like crap. I hated myself for it. But recently, I'm finally able to produce some OK results. It's still hard though, to say I can beat the better "whiz it right through me baby!" finalizing solutions. So I must practice some more.

One thing you learn by experimenting with the mastering process is to make a good mix to begin with. Some mixes are naturally easier to master than others. Even though they are all good mixes. The difference is hard to tell. So a "good mix" here, doesn't point to a good mix in terms of artistic, creative quality. It's more along the lines of non-exaggerated signal distribution, non-extreme compression.. Things like that. Especially extreme compression. Something that's compressed out of proportion during mix phase will be a big headache during multicomp, because essentially you're compressing twice, and that means you'll get even more extreme results. The problem arises when there's something in the background that you want to deal with, but there's a ultra-compressed clap in your face. You'll hate yourself for make the clap so snappy. By mastering your own mix, you creat lots of dilemmas between the mastering "you", and the mixing "you". But when they both start to work with each other, that's when you'll get good masters.

Remember, the first 50 tunes are going to sound like crap. Some may even sound so bad, you'd swear the crap was left out to rot for... sheesh! Just waaay too long! But there's so much mastering can teach you about sound in general, you'd want to do it even if meant turning your awesomely executed tune into nutrition for plants. If you think the result sucks? Just pump it through the "easy" solution!

<font size=-1>[ This Message was edited by: kensuguro on 2002-11-04 11:00 ]</font>

While I've got some time on my hands (which I don't.. at the moment), I'll also write what I know about monitoring. Again, I'm not qualified in drawing the lines here too. So please read this casually. (as with all my other "suggestions")

Of course, you'll need a good set of monitor speaks that you've learned to its bones.. and hopefully down to every line of its DNA code. I'll say before anything else, that a multi-billion dollar monitor is no good to you if you haven't learned it.

Me, I work with Event 20/20 bas. It's all I could afford. Before that I worked with 10M. Before that, I worked with bulky old second hand Yamaha 300M (or something like that, it's out of production now). I also use AKG K240 headphones. The 300M died because I decided I had to monitor louder than a club, just for the heck of it.

The single most important part of monitoring, is volume. You must keep the monitoring volume 100% consistent. Ok.. maybe this was a bit obvious. But it may not be for some people. Why the volume matters. It's because volume will mess up your perception of frequency content, it will mess up how the speaker reacts to these frequencies... Man, these two combined can really screw you up. The back of your monitors.. Change it once and for all. Spend as much time as you need to get it done right, and don't even go near it after you've caliberated them. The fader controlling the signal going to your monitors. Lock it with some chains, and have green beret rangers guard it. Also, get yourself a gun so you can shoot yourself if you're ever tempted to touch it.

Now.. it's easy to get caught up with issues about acoustics. Frankly, there's only so much you can do at home. The situation will be quite different if you've got the pleasure of owning a studio. But as long as you're a bed-room tracker, forget about acoustics. Just make sure you cover anything that's got hard surfaces, and make sure you don't have too many beer bottles between you, and the monitors. Yes, that means even your most beloved cannot come between you and your monitors. Make sure the center of the monitors are at your ear level (or close). Then, work like mad, so you can get yourself an acoustically proper studio.

Headphones- are totally useless for any kind of tweaking in the frequency domain I humbly suggest. Well, if your neighbors really hate you, then I guess you can try EQing with headphones... but I doubt that'll get you an album contract. I only use headphones to check for distortions that the monitors may have missed. (clicks can sometimes get by.. and also bass distortion) And also to check for stereo imaging. Personally, I've given up on trying to make tunes sound good on headphones. They're just too limited. I mean, even pretty nice headphones like my AKG K240 are limited.. think about listening to the mix on teeny-tiny Sony earphones that come with a brand new Discman... If you find yourself trying to tweak your mixes for these earphones, use the gun on yourself and make life easier. If you've touched the monitor faders in the process? Empty the clip on yourself.

Checking your mixes can be hard. Every single type of speaker has its own character. They all have different response times. You'll have to forget about crappy speakers. There's nothing you can do to save them. But even those aside, there are a multitude of speakers you have to take into account. For me, the solution isn't to play my tunes on every single speaker I can lay my hands on. Because the solution should be within your mastering process. It's the multicomp phase that makes the difference.

This is the bottom line. As long as you have a very dynamic, very vibrant mix that sounds no less than magnificent on your professional monitors, there's no way a consumer hifi can reproduce the signal. In essence, you have created source material that can never be reproduced. The difference that causes this, as I've said, is the response time of the speaker. How many millimeter can the speaker move in a given amount of time. But there's one way to make pro speakers and consumer speakers act similar. That is: Don't make the speakers work too hard. That means, to make the material less dynamic.

Let me explain. (maybe I don't need to, but as I said, just let me ) Say that an instant drop of voltage (or volume) causes a speaker to move 3mm. (just imaginary numbers) And a monitor speaker can do this in 5ms, where as a consumer hifi will do it in 15ms. Now, if you tweak the voltage so the speaker only has to move 1mm, and the monitors do it in 1ms, and the consumer hifi does it in 2ms, you've effectively decreased the performance gap between the monitor and the hifi. (the non-linear drop of reponse time is a result of physics) So basically, it can be said that the lower the dynamic range, the less the speaker works, and therfore the better the performance on consumer speakers.

Fortunately, multicomp does this fairly well while keeping things accurate by using many bands, and the limiter after it will mash it down even more. Scientifically speaking, there's an incredible drop in signal accuracy by doing so, but if you can do it with no obviously percievable artifacts, cool! The more you mash, the better it will sound on El-Cheezo brand speakers. (but it will NEVER sound good on headphones )

Even by using multicomp, though, doesn't mean you shouldn't check on almost any type of speakers you can get your hands on. But just remember that it becomes much easier (scientifically) to get consistent results this way.

Ok, this is about all I can say about monitoring and mix checking. Again, this is only one way to view this. Just one difference in the formula will mean you'll have to think of a different way of working. Like mastering for a specific type of setup... ie the big club boxes. That requires a whole different set of ways to monitor and check your mixes. So please only think of these as guidelines.

But one rule is always the same. Don't change the monitor volume!! Remember, winners don't change monitor volumes. (The changing of monitor volumes is considered a serious federal offense)

<font size=-1>[ This Message was edited by: kensuguro on 2002-11-04 12:22 ]</font>

Hey Ken, great 'article' - seems to become an interesting thread. I'm going to print and read it tonight.
Thanx!

"You must keep the monitoring volume 100% consistent"

I'm sorry but I disagree. This is my philosophy:

Check your mixes at different volume levels. Check your mixes on as many different speakers in as many different environments as possible. Check your mixes in mono on cheap crappy speakers. Check your mixes in mono on expensive speakers. Check your mixes on your headphones. Don't be afraid to mix using headphones. But always check the results on speakers as a final test. Use a commercially recorded CD as an A/B comparison (the idea is to pick something similar in style to the music you are mixing/mastering just to calibrate your ears). If you can afford to, put the project aside for a few days and then come back to it with a fresh perspective. You'd be amazed how different things can sound a day or two later.

Here's my bottom line suggestions: read as much as you can on the subject. There are quite a few excellent websites on the subject. Spend a few days browsing for them and I think you'll find plenty.

Take note of ideas that seem to repeat again and again as they will start to carry more weight. Keep those ideas that ring true to you and discard those that don't.

Learn to use the tools you have at hand. Don't be afraid to bend the rules if it helps you to understand the difference between a mediocre sound and a great sound. Your ears will "learn" much faster that way.

Listen to as much commercially recorded music as you can and spend time training your ears to determine where and how certain instruments fit in the mix. Don't be afraid to experiment. Trust your ears but be mindful of things like phase cancellation which may be hard to detect.

Check with some of the better mastering studios in your area and ask if you can sit in on a session or two. You can learn an awful lot just watching and listening.

More importantly: have fun learning!

You're right. There are so many ways to tackle this subject. And the strange thing is.. they all seem to work. It's all a matter of taste, and situation. It's good to be very open to new ideas, like krizrox says.

And by the way, I say don't change monitoring volumes when you tweak parameters. After the final mix is done, yes, check with all sorts of combinations.

<font size=-1>[ This Message was edited by: kensuguro on 2002-11-05 04:05 ]</font>

Yes - it's a very subjective process. Ask ten people how they do it and you will get ten different answers. Mastering is an art. And art is in the eye and ear of the beholder.

In my own experience, it took me about a year before I felt comfortable (or confident) in my own abilities. And I'm still just nothing more than a pimple on Bob Ludwigs' ass

The more you do it, the easier it will get - like anything else in life.

Thanks Ken and Krizrox, great readings!!

I'm sorry to cross-quote. Can we get this topic into the Study Forum?

On 2002-11-04 17:48, Steve-o wrote:
http://getimo.de/linkpage2/html/tuts_common_katz.php
One of the best tutorials about mastering available imho.

Here's another link to mastering tips, at Future Music this time:
http://www.futuremusic.co.uk/fm_mmusic.asp?ID=5884

Plenty to read now

I was fiddling with Finaliza today.. this device may have quite a lot of potential... The gui sort of gets in the way, but it shoud be possible to do something serious with this.

>> Here's another link to mastering tips, at Future Music this time:
>> http://www.futuremusic.co.uk/fm_mmusic.asp?ID=5884

Great link, great articles!
Thanx AtOmic!

All the descriptions on how to build a good, optimal mixing are valid and we learn from each other this way, it’s great. The curious thing is that every one, even if apparently in contradiction to each other, makes sense to my understanding. All the ideas conform sort of a river that comes to converge, as a whole, into the See of Mastering, where we can find everything said about mixing, being mixed.

I think, as said before, that every idea is valid by itself but of course, applicable to a particular context. All the ideas are great, but within it’s own context and sometimes almost useless out of it.

I can see nobody has referred to different music stiles, instruments. The procedures are definitely different from one musical world to another.

If we talk about classical music using direct instruments to microphones, we are in a completely different world and the whole process is different from the beginning to the end which is mastering. In classical music for instance, engineers use as little FXs and processing as possible, sometimes they use absolutely nothing but an enhancer, sort of Waves Ultramaximaser, and that’s it! Nothing else. In Techno music we are going to use a whole bunch of plug-ins FXs to biff up the sound.

I can’t separate in fact, the beginning from the end. I mean, I can’t separate the way I produce the “timbers” and “recorded” sounds, the way I apply “FXs” to the MIDI instruments, from the mixing and mastering stages. They all are together, a single process cos they influence each other much.

There is the kind of music you are processing to take into account when mixing and there is the “machine” that will reproduce the music too, and this is also important. If you compose for TV, which is the world of multimedia sound, you have to work in a completely different way that you would if you produce a Trance track, to be played in dancing clubs.

I do not have a fixed way of doing anything cos of this. I don’t like much to stick too hard to a specific method, cos methodology is good for a learning ship process that makes you free, but can become a prison if you follow it constantly doing everyday the same thing. I would do process repetitions if I was the owner of a commercial studio or something like that, for jingles and this sort of things, but I would not do it for my own music or anything being rather creative, innovative and so forth. I prefer to use my ears above all.

My ears are WAY BEYOND anything, the most important plug-ins I use to do my master tracks. I trained myself to not trusting a plug-in by it’s name, or because it is supposed to do something great; I rather try something and don’t care about how much I have paid for it, or which is the market name, or what people think about it. This is perhaps why I love to peel off the labels of commercial jars… Of course, I think about all this things when I buy something, but not when I’m working.

When I’m working, I put labels out of my face and use my EARS. I become extremely passionate listening with an incredible concentration and I would be tremendously bothered if somebody would interrupt me during this process… I feel like a surgeon doing sort of a dangerous operation, I become quite tense and even nervous when mixing the final mix, but I’m enjoying it… it’s not unpleasant at all. It’s similar to the tension you feel when going to play life facing a big audience, something like that… Then I’m ready.

I think your attitude when mixing IS also important, your ATTENTION, your consciousness directed to what you are doing without thinking about what you are going to eat, or which TV program I’m going to see tonight, or the bloody bill I have to pay tomorrow if I still want electricity at home!

I once talked about a balance concept, I do believe in it strangely, but instinctively not intellectually. In fact, I have observed that the little I know, sometimes can become an insidious artefact between my free creativity and spontaneous feelings, and the final mix. If I start thinking what people will think about my mixing, or think about the market, or think about money, etc., I’m certainly killing my creativity, so first of all, I need to be free of any preconceptions and anti-artistic ideas of the like.

Mixing is marvellous cos it’s difficult! Every little thing you do, will be important. Anything at all! If you pan an instrument just a little bit to the right, everything will sound different and the VERY sense of the song can change. This is why when you are mixing, you are still composing; as a mater of fact: YOU HAVE FINISHED YOUR COMPOSITION WHEN YOU HAVE YOUR CD PLAYING YOUR SONGS AFTER THE FINAL BURN, AND ONLY THERE!

I have been working and trying hard to learn how to master for about five years and a half now and I can say I’m just going into it… It’s real difficult to do an excellent job in mastering, it requires such a deep attention, such a silent mind that it’s difficult to be there doing a transparent job. Transparent in the sense that nothing interferes your ideas and what you feel and hear, disturbing the final result of your work.

I didn’t know mastering was such a wonderful thing. It really is enthralling! Completely fascinating!

Krizrox or anyone .

what do you mean by writing :

phase cancellation.

what is it?

btw very helpfull thread .

cheers
omb

It's where signals within the mix (say, the bass track and the kick drum track) or perhaps the left and right channels of the final mix have different phase. This will cause anything from lower than optimal output, to complete cancellation of the signal, where you don't hear anything at all. (this is rare)

Basically, phase goes from 0degrees, to 180 degrees. 180 degrees is where the signal is flipped upside down. So imagine mixing one sin wave, with a 180 degree phase shifted (or called phase inverted) sin wave. The 180 degree shift means that wherever the original sin wave has hills, the shifted one has valleys. (complete opposite) So, if you mix them together, the hills of the original sin wave merely fill in the valleys of the inverted signal, causing complete silence. That's complete cancellation.

I’m experimenting with recording vocals directly including it’s already melt EQ and reverb onto them avoiding adding reverb on a later stage. I get a better incredible natural sound. We tend to work too much vocals and they get unnatural, spoiled. I think I’m going to keep working this way for a while. Of course, there are FXs vocals, but this is something else. Fore real vocal, the only think I would recommend, if you want it completely natural, is a mastering tool sort of an enhancer like the Waves Ultramaximaser or something similar.

I’m experimenting with recording vocals directly including it’s already melt EQ and reverb onto them avoiding adding reverb on a later stage. I get a better incredible natural sound. We tend to work too much vocals and they get unnatural, spoiled. I think I’m going to keep working this way for a while. Of course, there are FXs vocals, but this is something else. Fore real vocal, the only think I would recommend, if you want it completely natural, is a mastering tool sort of an enhancer like the Waves Ultramaximaser or something similar.

I’m experimenting with recording vocals directly including it’s already melt EQ and reverb onto them avoiding adding reverb on a later stage. I get a better incredible natural sound. We tend to work too much vocals and they get unnatural, spoiled. I think I’m going to keep working this way for a while. Of course, there are FXs vocals, but this is something else. Fore real vocal, the only think I would recommend, if you want it completely natural, is a mastering tool sort of an enhancer like the Waves Ultramaximaser or something similar.