Using compression on final mix..

006

Member
Jan 10, 2005
8,952
0
36
Ok, so I'm starting to rip hair out. I'm trying to figure out how to make my mixes go to 0db...intead of having a bunch of peaks and valleys in the waveform. I use to be able to do it...and now I can't figure it out. Somebody told me to use some compression. Anybody have some good starting points? Or screenshots maybe, I have the Waves Rennaissance RCOMP, as well as the normal DigiDesign one ( I don't think you guys would recommend that one though, hah. ) and a couple of other plugs. I would kill for an actual hardware unit, but all in due time. Latah.

~006
 
Hmm, well I've found this to be one of the most irritating processes of mastering... getting the compression just right. I've only ever been able to nail it well once and that was because the mix itself was good.

What I normally do is find the point where most of the 'useless' peaks occur, so if that's -10db then you set the compressor threshold around there. Experiment with ratios until you find something that doesn't sound 'boomy' as this controls how much the peaks are attenuated. With attack and release times, I have trouble with these too... I've usually found for metal that a shorter attack time is better, but it really depends on the mix.

It seems to be quite common practice to compress the final mix multiple times until all dynamics have been pressed out of it and it's just a wall of sound. If you're going for this kind of sound then you'd probably want to lightly compress 2 or 3 times until the final effect is what you achieve.

To get that final bit of volume out of your recording make sure that you use a limiter of sorts to chop the remaining peaks off and pump your recording as high as it will go. I recommend setting the threshold at -0.3db
 
Yeah thats actually not too far from what we send radio stuff to the ad companies at the studio. Normally it's at like -0.6db. Thanks for the input. It's something I haven't been able to figure out for home recording stuff. I can do it at the studio, but it's a lot different there. I figured I would have to take the multiple comps approach, as you said. Thanks for confirming my convictions on that one, now that means more tedious work, :p . The limiter idea sounds great. Speaking of volume, thats mainly why I posted. I always notice that my home recordings are never as loud as stuff that I have ripped and put on my iPod. Actual label signed band's music I mean. It's always like...I want to say 5dB below or more. It's just annoying. But anyway, thanks for starting me off. Latah.

~006
 
i'll tell you what you want to know.. for a price.... you must stop using the word "Latah" forever. mwuhahahahaha :devil:
 
unless you have something like a L3 (or l2 or l1) maximizer brick wall limiter or a tc electronics finalizer (i am only talking BUDGET mastering gear here) then moonlapse suggestion pretty much hit the nail on the head. if you want to preserve more dynamics you might want to try using just one compressor across the stereo buss hit about 2-3db of gain reduction, eq to taste on the next insert if you dare and then try something like the maxim limiter or another compressor set to limiting. set the limiter threshold to between -.3 and -.1 db and use the make up gain on the compressor (or the eq and make sure you arent clipping any plugins) in the first insert to crank it up until you have just a bit of gain reduction on some of the peaks and you should be pretty damn loud by then man. see if you can track down an L3 though that will make your life ALOT easier.
 
LI, L2,L3 all good. Many guys I know that stay in the box use this-Vintage Warmer. Once you learn how to use it. ALL your mixes will be as loud/and with dynamics as the big boys. Oh yeah thier 'mix pack' bundle is awesome too for 'mastering'/'finishing' your mixes.
 
Hey Kenny--
Any tips on using the VW for that application? I have it but haven't used it much yet, and the presets seem kind of extreme. I need to sit down with the manual and learn the controls more thoroughly, because they're pretty unorthodox.
I do love the "Snare" preset on toms, oddly enough...
 
normalizing mixes is a no no.

heres why. Normalization works by reading the peaks in a track. When you normalize a track there may be inherent peaks (ones the occur from DC spikes or any number of things) that are much higher than the audio in dB and quite possibly are inaudible or have been covered by a crossfade etc... normalizing analyzes the peaks of the audio and brings those to the set level and keeps the entire audio track in relative loudness of those peaks (if that makes sense). as far as i know there is no normalize feature that can judge the dynamics your audio and adjust itself accordingly yet haha. would be nice though. normalizing can make your mixes a great deal louder, but not in the right way. also any signal processing like that will cause some signal degredation anway.
 
Frank'nfurter said:
Btw, what does that mean, Latah?
Ur....slang for Later :loco:
Hey Matt first thing I did when I got V.W. was go to the back panel and set the 0vu reference level to -14dbfs (oh so very important!!!!)
Then tweaked the saturation and release knobs. Set overs to 1. I got it awhile back and other than Tritones Hydratone, the Vintage Warmer was the best price to quality plug-in I've ever bought!!!! :headbang:
 
Frank'nfurter said:
Btw, what does that mean, Latah?
yeah.. slang for "Later".. kind of urban, hip-hop, rap language. i hate it badly. :yuk:

auf Deutsch, "Später". so, it's like saying "bis später"
 
James Murphy said:
yeah.. slang for "Later".. kind of urban, hip-hop, rap language. i hate it badly. :yuk:

auf Deutsch, "Später". so, it's like saying "bis später"
:Spin:

Just for curiosity lata (h) means can (like can of worms, can of beer, etc)
in portuguese.

please stop with that!!!
 
Holy crap. It's just a habitual thing I've done for years on forums. Lol. Yeah.

~006
 
axeman720 said:
normalizing mixes is a no no.

heres why. Normalization works by reading the peaks in a track. When you normalize a track there may be inherent peaks (ones the occur from DC spikes or any number of things) that are much higher than the audio in dB and quite possibly are inaudible or have been covered by a crossfade etc... normalizing analyzes the peaks of the audio and brings those to the set level and keeps the entire audio track in relative loudness of those peaks (if that makes sense). as far as i know there is no normalize feature that can judge the dynamics your audio and adjust itself accordingly yet haha. would be nice though. normalizing can make your mixes a great deal louder, but not in the right way. also any signal processing like that will cause some signal degredation anway.
Thank you for the infor sir :)

God I feel like an idiot.