self mastering...loudness

[FlowersForHuman]

I know that the answer is probably "send it to a professional" but...

...my mixes (i write and record all types of music) are always noticably quieter than my favorite "professional records".......

...I always seem to be around 6-9 db too quiet for my comfort zone after mastering. Nice and loud, but quiet enough for everyone to notice a difference. Can anyone help reassure me on what to do to remedy this?

I basically track into a motu 828mkII into cubase, mix in the box, digitally export a stereo mix, bounce it back in, normalize to 0.0, eq/compress/limit it just above the point of noticability, export again, bounce it back in, normalize again, and then export the "final master."
Everything is in-the-box.

I do this because 1) I assume that above 0.0 "volume" the track will digitally distort and that, as long as I normalize to 0.0, the track (at least its peaks) will be at the maximum volume, and 2) that as long as my mix is solid I will have the loudest possible result.

Of course I do so and my result is obviously not the best possible because somehow the pros alwas 1-up me (or more like 6db-up me).

Is my thinking flawed? Should I just keep pushing the master mix fader on the final export? The normalizer only goes to 0.0 and I get a peak indicator on anything above 0.0db so I assume that somewhere distortion is happening or could potentially happen.

Do the pros just push it and listen to every split second of everything over and over until they get the loudest possible result without audible distortion? Seems counter-productive and counter-intuitive in the digital realm to do so...

Maybe the fact that I have huge peaks in my mix waveforms should tell me something?

Help! What do they do? What do you do?! What do I do?!?!

[ThePitz]

Disclaimer: I am an extreme amateur with little real-world experience. However - i'm a compulsive information scourer with a reasonable grasp on theoretical info.

First of all - it seems like you're doing an unnecessary amount of exporting and importing and whatnot. If i'm mastering something myself - I think the cleanest way to do it is by putting all your mastering on the master outs. There's no need to export and import over and over (unless you need to bounce stuff off 'cause you're computer can't handle the multiple plugs, I suppose.)

Second - if you absolutely must compete with super super super loud masters (which is inadvisable in terms of sound quality) I would:

1. Make sure everything that can be high-passed, is. There's a lot of low, inaudible rumbly information on tracks that don't need it. While inaudible when tracks are soloed - this stuff builds up and then when you send it to your compressor on the master fades - the low, mushy and inaudible stuff activates your compressor - which causes it to do a bunch of undesirable things.

2. Once everything is high-passed. Compress the individual tracks more. If you have a snare hit that peaks a lot higher than everything else - when you send it to your master compressor it has to squish that down a serious amount before it starts to work on anything else. Better yet - instead of compressing that snare hit - duck it with volume automation, if possible.


You are absolutely correct, though. Nothing should go above 0.0. Actually - I recall hearing some arguments that you should try to keep your stuff .1-.3 db below that.... Not sure why... Or if that's correct. Just seem to vaguely recall it...

Again. I'm no expert... But if a gun was held to my head and some dude was saying "Make this track LOUDER!" ...

I'd probably try to judo chop or roundhouse kick the gun away... Maybe use nun chucks if I had some available.... Or a ninja star... Or some jedi shit. Nobody puts baby in a corner.

But if my hands were cuffed to the desk... I would high-pass, compress, limit and duck the individual tracks. Squish the meat, cheese and bread before squishing the whole sandwich.

-pitz

[solo-bration]

If you see drastic peaks in your mixes, you could probably work on your individual tracks in general (as the previosu poster advised). A "good" mix tends to 'look' even, especially in the loud sections, and NOT like the spine of a fish

You don't want to exceed ZERO or you will start getting clips/distortion. not sure if this format works for cubase, but this is a plugin from Massey that is good for "pseudo-mastering". Its a brickwall limiter and it's quite forgiving. It's also free. This could ruin your mixes, but if used tastefully, help 'glue' things together and make them seem louder.

http://www.masseyplugins.com/L2007.html

disclaimer: everything is subjective. compression is the devil. there's no right or wrong.

[MoreSpaceEcho]

keep in mind that most "pro" records these days are seriously no exaggeration a good 6db too loud. they sound like comical shit and are painful and tiring to listen to.

no one's gonna think you have a small cock if you put out a record that's noticeably quieter than whatever major label shitstorm they were listening to previously. 'kind of blue', '1000 hurts', 'spiderland'...these are all "quiet" records and i don't think too many people complain about them.

that said, pitz' advice is good. also, you don't need to do all that (or any) normalizing. get your gain via the compressor and the limiter. also look into clipping...you can stick a clipper in between the compressor and the limiter and eek out another db or two. with some (percussive, distorted) types of music, clipping sounds a lot more transparent than leaning on the limiter too much. the clipper in the voxengo elephant limiter is good.

[decocco]

Maybe the fact that I have huge peaks in my mix waveforms should tell me something?

Bingo!

The only way to get an ass-kickingly loud master that is not totally ruined is to have an ass-kickingly loud mix. This requires heaping amounts of delicious compression and fader rides during mixing. Parallel compression and bus compression (multiple busses, not just the mix bus) are your friends.

If you have a mix that is very dynamic and then you slap a limiter across the mix bus to try and make it way louder, your nicely balanced mix will be instantly destroyed. Bye bye snare drum. Bye bye ALL transient sounds. You'll end up with mush.

Like ThePitz said, high passing is your friend. You won't want to high pass everything, but if you've got a lot of 60Hz info in your electric guitar tracks, this is totally eating up headroom in your mix.

It's possible to make mixes that are still plenty dynamic enough for rock and roll, but you must mix with loudness in mind. There is no need to do all this normalizing and exporting and so forth. This is not helping you. Normalizing (in most cases) should be banned.

[ott0bot]

It's funny how most of us understand the having a good dynamic mix with lower peaks and less compression sounds better, but 90% of musicians will say "it's too quiet" or "it doesn't sound professional" as a result.

I've been using a minimal amount of limiting on the master bus when doing my standard mixing, and monitoring the incoming signal so it peaks around -6 to -4 and average around -18 to -12. It hasn't seemed to turn my mix into mush. And it is pleasing the bands. Then i suggest having it mastered, or I'll quasi master it with a limiter, compressor and sometimes an eq on the master bus and give them both files. I'm not a pro and don't have any great outboard stereo compressors/limiters so I'm forced to do it all in the box as of now.

When I analized the signal it appeared to peak around -1 or -2. Is this bad? I would assume not since most of the "professional" songs I ran through my analizer seems to be hitting right at zero and have a much less limited dynamic range than my mixes. And my mixes are only slightly quieter on a standard stereo than "real" records.

I know for mixes it's generally recomended to stay in the -6 to -18 range, but...
For you mastring engineers out there, what do your finished masters end up peaking at?

[MoreSpaceEcho]

-.3

nothing "wrong" with peaking at -1 or -2, but there's kinda also no real reason not to have it peak right under zero.

[Nick Sevilla]

It's funny how most of us understand the having a good dynamic mix with lower peaks and less compression sounds better, but 90% of musicians will say "it's too quiet" or "it doesn't sound professional" as a result.
records.

Next time a musician complains about this, have them mix the song.

I did once, the muso NEVER complained again. He could not even get close in balance and levels to what he considered my "shitty" mix.

Also, most musicians could not express in common everyday language what "professional sounding" actually is. Next time one complains, ask them to describe in detail what they think is "professional standard"

Cheers

[MoreSpaceEcho]

nobody ever complains when they just listen to their mixes. it's only when they do a direct a/b with the new Whatever record and then they get insecure and think no one will listen to their record if they have to nudge the volume knob a 64th of an inch. retarded.

not an attack on the OP or anyone else. area man just venting. i'll show myself out now.

[Ryan Silva]

-.3

nothing "wrong" with peaking at -1 or -2, but there's kinda also no real reason not to have it peak right under zero.

Don't some consumer CD players distort at -.1?

Just a rumor I heard.

[Ryan Silva]

keep in mind that most "pro" records these days are seriously no exaggeration a good 6db too loud. they sound like comical shit and are painful and tiring to listen to.

no one's gonna think you have a small cock if you put out a record that's noticeably quieter than whatever major label shitstorm they were listening to previously. 'kind of blue', '1000 hurts', 'spiderland'...these are all "quiet" records and i don't think too many people complain about them.

that said, pitz' advice is good. also, you don't need to do all that (or any) normalizing. get your gain via the compressor and the limiter. also look into clipping...you can stick a clipper in between the compressor and the limiter and eek out another db or two. with some (percussive, distorted) types of music, clipping sounds a lot more transparent than leaning on the limiter too much. the clipper in the voxengo elephant limiter is good.

It's all about Tom Waits 'Swordfishtrombone' I've never heard a more dynamic mix. Ahhh the sound of "Space"

[Waltz Mastering]

For you mastering engineers out there, what do your finished masters end up peaking at?

When using a bw limiter I'll usually peak at -.2

If you plan on making mp3's form the master it's better to set the ceiling for the source files at -.4 so they don't over shoot.

[FlowersForHuman]

nobody ever complains when they just listen to their mixes. it's only when they do a direct a/b with the new Whatever record and then they get insecure and think no one will listen to their record if they have to nudge the volume knob a 64th of an inch. retarded.

EXACTLY!!!!

Okay, thanks guys...great advice/reassuring information here. Of course after I sobered up and A/B'd again today I didn't feel like my "volume penis" is lacking in any great way...after all, I spent a little money on some prosumer gear (that I get to keep!) for my records and the platinum-selling records I'm comparing them to probably cost more than my house just for the sounds, so no biggie...but I wanna keep the thread going here because some interesting points came to light and I am about to record a bunch of bands...

..also for the record...I think my mixes sound great, just not very "bold" volume wise, and I certainly understand the volume wars thing (I actually wrote a term paper on it in undergrad) and thats why its so frustrating to me that I can't get my tunes to translate to LOUD and still sound NICE!! I disagree that MOST "pro" recordings are over-compressed because I own dozens of them that all seem to sit in this realm of "beautiful sounding" AND "loud." I understand hi-pass filtering and compression and all that so here goes with questions...


1) better mixes...to get it strait...if my mixes sound awesome to my ear but have lots of high peaks and aren't as loud as I want it, I basically have to remix them if volume is my only concern, right?

2) same topic...one main difference I see when A/Bing to "pro" records is that no matter what the style ("loud/soft"), there tends to be more "energy" accross the range of hearing (in bass, mids, highs) on all the recordings I lust after, as opposed to mine which sound nice but not ballsy...ie both Pantera's "Reinventing the Steel" and Beck's "Sea Change" have vast amounts of energy across the board, and both come across as super loud. Probably just my mixes "suck" then, eh?

3) I learned my way of "mastering" through trial and error because it theoretically made sense to me...no one every taught me how....What is a downside to normalizing?

4) As far as the master bus is concerned...what is a typical signal chain for you guys out there and how do you approach the act of mastering?




ALSO>>> If you want to know what got this whole ball rolling...here is the song that I wanted to be LOUD!!!

http://www.exemplaryrecords.com/media/a ... 20City.mp3

[uncle bastard]

My two penn'orth: thanks to reading TapeOp,I have taken to mixing at -12db. Once I've got everything sitting nicely together the way I want at that level (with Final Mix tweaked and strapped across the master output ), I'll export to Audacity and use that to bring the highest level to 0db. I haven't been doing it long, but so far it has given me full sounding results with a good dynamic range and no clipping. I never normalise.

[MoreSpaceEcho]

1) better mixes...to get it strait...if my mixes sound awesome to my ear but have lots of high peaks and aren't as loud as I want it, I basically have to remix them if volume is my only concern, right?

you don't have to remix per se, but it wouldn't hurt to go through the tracks and automate the volume envelopes of any really big spikey hits, and this will perhaps gain you a couple db more level.

2) same topic...one main difference I see when A/Bing to "pro" records is that no matter what the style ("loud/soft"), there tends to be more "energy" accross the range of hearing (in bass, mids, highs) on all the recordings I lust after, as opposed to mine which sound nice but not ballsy...ie both Pantera's "Reinventing the Steel" and Beck's "Sea Change" have vast amounts of energy across the board, and both come across as super loud. Probably just my mixes "suck" then, eh?

don't be too hard on yourself. the bill for the mastering of 'sea change' was probably more than you've spent on your entire studio.

3) I learned my way of "mastering" through trial and error because it theoretically made sense to me...no one every taught me how....What is a downside to normalizing?

the downside to normalizing is there's really no upside. if you have a mix where nothing peaks above -6 for the entire song except one snare hit right at the end that hits -.1...normalizing that will gain you exactly 1/10th of a db in level. not really much help.

4) As far as the master bus is concerned...what is a typical signal chain for you guys out there and how do you approach the act of mastering?

as far as the master bus in mixing i don't have anything on it, i just mix. in mastering, a typical basic chain would be eq, compressor, limiter in that order. you listen and adjust and try to do as little as possible. a/b any processing against the dry source AT THE SAME LEVEL. the same level is crucial because if one file is louder than the other, you are going to think the louder one is better. with the levels matched you can more objectively evaluate what you're doing with any of the processing.