Let's all sit in a circle and talk about mastering

[Cyxeris]

Would anyone be interested in discussing mastering chains? Since I'm almost at that point, I thought it would be timely to open a discussion about it. Here's the basic rundown of my preliminary efforts at piecing one together to suit my needs.

Each track:

Waves LinEQ Lowband, low cut/rolloff circa 35-45 hz
Waves LinMultiband compression

Mains:

Waves LinEQ Lowband, low cut/rolloff circa 35-45 hz for DC elim.
Waves LinEQ Broadband, for overall balancing and shaping
Waves L2 for finalization

I'm still playing with other ideas like mastering reverb and so forth, but I am trying to approach it as minimalisticly as can be afforded. I'm also playing with issues like should I eq pre or post-multiband compression, should I bother low-cutting twice, and so on. Really, these are issues of taste, but this is the most holistic part in the process, the bottleneck, the funnel, so getting this right is far more important than, say, getting the EQ on one of the synth parts in one of the songs right.

Since I can hardly come close to imagining the idea of thinking about beginning to consider referring to myself as anything remotely like a mastering engineer, I figured "hey, let's touch base with the gang, and see how long it takes for the thread to mutate into a discussion about globalism and the New World Order!"

Cyx

[mosc]

I don't know how one can make rules on how to do mastering. I think it depends on the nature of the material. My bias would be to do nothing but normalize the digital information. Then, and only then, take actions to improve areas where you perceive problems. No problems, no actions.

Compression, reverb, and eq all either remove something or add something. They can be good, but just because we have them isn't necessarilly justification enough to use them.

As for DC removal. If you don't have a DC problem, why worry about it? Some people have expensive playback gear that can really handle low frequencies in the 20 Hz range. Why roll off?

Since the 1950s, pop music recording engineers have been compressing the hell out of vocals. I'm not saying this is bad, but we've ended up with people singing very badly because they know the compressor will take care of their inability to control their voices.

Guitar players are like singers in this regard. Often the overdrive their amps. This is a form of compression. Many guitar players can't control dynamics by playing their instruments.

As you can tell, I think all this is a sad state of affairs. So put away your reverb, eq, compression and all those other gimmicks.

If you are mastering for music to be played in a car, please disregard this post. Compress the hell out of it, roll off the bass, and even if the music doesn't have it, add a beat.

[Kruge]

I like kittens.


scnr

[Cyxeris]

[mosc]

Congrats, Kruge...
Well done...

[Cyxeris]

I have a minimalist mentality about things. Use what you need, and that's all. I haven't decided if I need the glossy sheen of a 0.3 second mastering verb or not. I do, however, mix with the intent of mastering, so certain decisions are made during mixdown that take into consideration what will happen at the mastering stage, so that the two processes become more like a single process, which I prefer. Compression in the mix is sparse. in a 40 track mix, I may use compression on 3 tracks, for instance. Eq is used just enough to balance things out and emphasize specific tonal elements. Mixing is like sculpture to me, I love it. I love the change of pace from composing and polishing to molding and finalizing. Both can get old and tedious, but the see-sawing of processes keeps things fresh.

Cyx

[elektro80]

mastering chains? I guess some mastering engineers have basic chains they use over and over,, but you should tune this for your music.. not use other people´s multipurpose setups.

You can sort out some "probs" before final mix and mastering. check for excessive energy in voices.. in the multitracks.. which are outside the essential signal band you think you really need in there. Like.. are there any subbass flops or growls.. fliterflapping.. whatever.. which peaks a signal.. or adds to its "energy" outside of what you want there.. such happens all the time.. controlling those can be smart.

You should decide for how loud this is going to sound. Too loud means too little true dynamics.. and this means too compressed.. but.. what you actually do with the mix is not always what is percived. if you lift sections which one musically thinks of as soft.. or expressive.. these sections van often work better... or not.. it all depends.

Multiband compression is smart.. but it is a lot harder than the product blurbs tell you.

You might want to use a compressor like that Waves l thingie for both lifting the overall loudness.. but also for the soft and subtle tasks like rounding off excessive peaks.

What usually makes the most mess are all those peaks all over the place which pushes the main volume of the track down. I guess you should plan the track in loudness and see what happens if you do this or that in various sections. It can often be smart to work with a copy of the project at this stage and try to compress and process parts which should match each other.

if you decide to use single band compressors for certain types of processing you should test this using an eq at the input stage.. and try even a 5 or 10 band fiter.. changing the bandwidth composition of the signal before you compress it can often help a lot.

Subbass is cool. You can plan several types of bass in the songs.. like some are compressed like in modern music.. with a hard rolloff.. and others start below the usual rolloff.. The big problem is the typical resonance frequencies of ordinary budget stereo speakers. This is typically just below the ordinarry pressure peak of pop music which is at 85-95Hz. What you can do is test this with various home speakers and try to develop your own technique to fade the bass in hard.. below where ordinary speakers give up.. and at the same time control the overall loudness in the song.. like if you add a lot of shit down there which ordinary speakers do not handle at all.. the energy from this added bass will steal from the overall dynamics bandwidth. Well.. chances are you can manage to make a mix and master which has that extra sub oooompf when played on good stereo speakers.. but hopefully doing this in a way that does not hurt budget gear.. without making the budget gear explode. many radiostations over here have their own "mastering chains" just for this type of use.. they run all music thru filters and limiters and multibandcompressors in order to make the shit sound even louder.. but also removing all that very very deep bass. Subbass is not radiofriendly.

What you can do is get some CDs and LPs made by artists you like.. and throw these into your DAW.. and check out on meters and virtual scopes and whatever.. how they really work.. look for clues and tricks. there are only clear rules when it comes to certain genres.. like various strains of pop. Well.. classical music is a bit special.. It is a lot harder to record and master classical music than what one might think.

When it comes to reverb.. you can use this in order to create effects.. which have noithing to do with creating more room in there.. rather to create tails here and there to make details more "detailed" etc.
You could try to sidechain the reverb from certain voices and run the pure reverb output thru some limiters.. compressors.. whatever.. and play around with this.
An old trick with modules/ modulars is using seperate signal chains with envelopes in order to shape stuff like reverb .. like.. you run a heavily compressed/brickwall limited reverb signal thru an envelope which behaves slightly different than what the main synth signal does. ( the heavy reverb can roll off before the main signal does.. ? like just snapping past the main bass boom when the main signal envelope enters the "sustain" step..? You figure out something.. ) And you can even sidechain this and run a feed into a filter set up to change filter settings for every 16th or whatever. ..Mmm what I mean to say is .. you can use reverb as an integrated part of the instrument.. rather than throw it in later. You can use reverb in order to create fatter signals without really add what people usually percieves as reverb.

Budget speakers and resonance.. read up a bit on speaker design and like.. bass reflex theory .. You will be amazed.

[Cyxeris]

[elektro80]

I was not thinking about "modern" stuff.. rather the music you love and cherish whatever its age. LPs have pretty bad dynamic range.. so you probably learn how they did what they did by listening some of those old LPs you like.

Personally i use a lot of compression and limiting and such, but I started with this for live use in order to make a better mix. It is quite possible to use compression as a part of the patch and you will figure this out if you listen to some old LPs. My favorite setup is having at least one compression chain for each and every synth. i use compression as part of the patch.. and then for controlling all those filter howls and flipflops...
There are a lot of classic synth sounds you cannot create without using some extra limiters and compressors. Also, in certain kinds of patches you MUSt have limiting/compression in order to control the signal before you route it into the next device/module/efx processor.

Anyway, this has nothing to do with mastering as such...

[Cyxeris]

[elektro80]

[themoors]

I t like the presets in t-racks, thats about as close as i get to mastering, I also admit to being the owner of an aural exciter.

[Cyxeris]

[Cyxeris]

[paul e.]

rock music is loud and sounds best when driven by volume..not much dynamics in rock music to begin with

the best classic rock is often the result of distortion [even distorting the mixing boards inputs] and tapesatuartion and 'maxing the vu meters' beyond 0db to get that fat signal spreading out beyond the track lanes on the tape

but today, with 0db brickwall in digital, loud rock music often ends up clipped and somewhat harsh and maxing the 'vu meteres' means 'forceful' sounding mixes- as the naturally louder signals like drums and guitar cannot 'saturate' beyond the 0db limit

but the problem now is, naturally loud and sonic sounds like heavy guitar/drums will sound limp and wimpy if you record them at a an RMS of -12db..they will lack a lot of the spunk the should have

so, what happens is most of the sounds in rock/pop music really only has a range of about -10db to 0db to operate in...making for very limited dynamics in a mix


this is the fault of the digital recording realm and not rock music per se

[paul e.]

a further point about vocals and compressing

ofetn compressors are used to bring up the volume and presence of a vocal track in relativity to the other sounds

if you have a 24 track rock mix, compression on the vocals will be absolutely neccessary in order to hear them

because, with all that energy generated by amplified distorted instruments, a lone voice or two will be like screaming in to the eye of a storm...and you might have to scream just to hear your self

so often compression is to give the vocals enough gain to sit up above a huge amount of loud sounds..and to increase intelligibitly of sibilants that can gets lots in dense mixes

and mosc, i know many people who can sing very very well and do not use compresors to compensate for a lack of talent, but need to use them for the above reason [especially in particularily dense mixes]

as well as guitarists who use overdrive sounds but also know how to control dynamics [john scofield comes to mind]

in fact jimi hendrix showed us all how to use distortion dymaincs . [ i.e. riding a feedback wave] so the two are not mutually exclusive

[Cyxeris]

[paul e.]

what i am saying is that current digital standards are not allowed to go passed 0db, but certain music *needs* to..like rock music..like it did in the analog days

which is why we are hearing so many clipped waveforms in dense rock mixes. not really as a result of spl wars..[the radio station's limiter keeps all songs the same volume mainly anyway]

dvd a will make the situation better as you will have increased headroom, perhaps allowing enough sonic space for these naturally very loud sounds to exist as loud sounds without being square waved

as for electronic music, this is not a problem generally

[Cyxeris]

Also, anyone, do feel free to correct any of my technical points if I am in error. Thanks.

Cyx

[elektro80]

DVD_A ... arrrgh.... I will try to find you guys some info and opinions on this.. it is not as great as it is supposed to be.

The O dB limit.. well..

Sure.. in DAW speak this is where there are no numbers left.. and pushing the level above 0dB creates digital clipping.

Old analog gear added softclipping/saturation/artifacts when pushed close to selfcombustion. It is mainly in our own minds that we have created the idea that these artifacts means "loud".. and this can be done in the DAW too. - and without overcompression..

Compression is in itself just.. compression.. but compressors are used for adding "analog warmth" or whatever the ads say... You can use compressors in other ways too

[paul e.]

[paul e.]

adding this caveat. 'without using plug ins emulating analog gear'

as these in my opinion are silly and do not work

[Cyxeris]

[paul e.]

cyxeris

[paul e.]