Looking for some advice, please. I’ve received some wave files for a mix that were recorded with a lot of dynamics and many peaks sometimes reaching -.02dB. Sure, it’s better than flat and overly compressed signals, but I’m still left with the challenge of taming and preparing the tracks for mixing. Ideally, I’d like a signal with no more than around -10db or so for the peaks. Here’s the question: what are the one or two best options for dealing with this? Some compression (even serial processing) that targets these transients and peaks? Manual gain adjustment (a lot of targeted work)? Lower the entire track volume (which will also reduce RMS)?
If this was Midi, I could handle this in a number of ways; for example, using the Project Logical Editor or tweaking the higher midi velocity, etc. But, when it comes to audio, I’m being somewhat cautious and want to avoid processing the signal too much. While I can whack the peaks, I need to preserve the RMS (body of the sound) as much as possible. Any best practices to approach this? Any special techniques or processing tips? Thanks!
Dealing with dynamic, high peak wave files
-
- KVRist
- 483 posts since 17 Dec, 2013 from The Netherlands
Try using a compressor with seperate controls for peak and RMS release times. TDR Kotelnikov has such controls and also a controlknob to prevent the peak-compression affecting the RMS
-
- KVRist
- Topic Starter
- 41 posts since 15 Nov, 2005
Thanks! Do you think it best to approach this with a compressor plugin, or use a manual editing technique? As I think through it, I'm beginning to fret about the work required if I have to manually correct 20+ wave files.
-
- KVRist
- 483 posts since 17 Dec, 2013 from The Netherlands
Can't say which method is best, using a compressor, or manually working peaks. But I do know that you have only so much time and that a compressor got invented to take over such a manual task.
So it's really up to you, I guess.
Sometimes I do work manually but only on selected tracks like de-essing a lead vocal and riding the volume fader instead of compressing.
So it's really up to you, I guess.
Sometimes I do work manually but only on selected tracks like de-essing a lead vocal and riding the volume fader instead of compressing.
-
- KVRist
- 251 posts since 7 Feb, 2017
Maybe some envelope shaping. Compressor & limiter also works.
-
- KVRist
- Topic Starter
- 41 posts since 15 Nov, 2005
Thanks for the comments! After reading various things, I'm currently inclined to do the following: buy Fabfilter Pro C2 (a "clean" compressor with great visual aid) and apply several instances in series, each one targeting and taming the peaks just a bit. Cheers.
-
- KVRAF
- 4711 posts since 26 Nov, 2015 from Way Downunder
-0.02 is so close to digital clipping. The first thing you should do is just turn the volume of the file down.
-
Hermetech Mastering Hermetech Mastering https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=7418 - KVRAF
- 1619 posts since 30 May, 2003 from Milan, Italy
How does it sound? Instead of looking to fix a problem that doesn't exist, using clipping, compressing, or limiting (all of which are non-linear processes, and will add nasty intermodulation distortion to your mix), why not just mix everything at a lower level and leave the peaks in there, if there's not an audible problem? The mastering engineer can get the loudness later, no need to worry about that, just get the mix sounding great and stop worrying about how the waveform looks. If it sounds too quiet, turn up your monitor gain.
-
- KVRist
- Topic Starter
- 41 posts since 15 Nov, 2005
Thanks good people. I appreciate your advice. Perhaps Hermetech is correct: I'm being too concerned about something technical that doesn't really need to be addressed, particularly if the musical aspect is not calling out for help. Still....because I'm curious and because it's fun, I wonder what tools and approach might be in order to tackle the issue. Cheers to all.
-
Hermetech Mastering Hermetech Mastering https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=7418
- KVRAF
- 1619 posts since 30 May, 2003 from Milan, Italy
In mastering, when peak reduction for loudness is required, for me it's usually a combination of digital limiting and clipping, and analogue compression and clipping. EQ is great for perceived loudness increase, but tends to add as many peaks as it reduces, even when cutting, due to phase.
-
- KVRist
- 243 posts since 17 Sep, 2006
Sorry, but "the mastering engineer can make it loud later" is a very wrong attitude.
The mastering engineer can only increase the loudness of a given mix so much, depending on the crest factor of the mix. So if you dont care about loudness in the composing and mixing stages of a project, the loudness of the result will also be quite limited. The mastering engineers can add some loudness at the mastering stage, but a mix that is at lets say -3dBFS and has a loudness of -30LU will never come out of mastering as loud as a mix that enters mastering stage at -18LU for example. And I dont mean by aleady being squashed and overcompressed, but because the composition and arrangement was already done with loudness in mind and the peaks where already tamed per channel in the mixing stage.
Cheers,
Codex
The mastering engineer can only increase the loudness of a given mix so much, depending on the crest factor of the mix. So if you dont care about loudness in the composing and mixing stages of a project, the loudness of the result will also be quite limited. The mastering engineers can add some loudness at the mastering stage, but a mix that is at lets say -3dBFS and has a loudness of -30LU will never come out of mastering as loud as a mix that enters mastering stage at -18LU for example. And I dont mean by aleady being squashed and overcompressed, but because the composition and arrangement was already done with loudness in mind and the peaks where already tamed per channel in the mixing stage.
Cheers,
Codex
-
Hermetech Mastering Hermetech Mastering https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=7418
- KVRAF
- 1619 posts since 30 May, 2003 from Milan, Italy
Sorry? How come?
I have do disagree, as someone who has mastered thousands of tracks over the last decade. The level of a track (peak, RMS or whatever other way you want to measure it) before mastering has absolutely nothing to do with how loud the mastering engineer can make it. It's about the quality and denseness (number of elements) of a mix. Every mix has a "loudness potential", but its level before mastering has nothing to do with that.
You can focus your mix on having a low crest factor if you want to, and some styles such as Psy Trance really call for this, but the OP didn't state his genre or style etc. so I think my point holds. That was that how a mix looks is not as important as how it sounds, and if it sounds good, don't worry about how peaks are looking on the waveform. If you look back you'll see the OP agreed with me.
Loudness, or how waveforms look should be the last thing on a mix engineers mind, just focus on sound quality/an amazing sounding mix. Any mastering engineer worth his salt is going to be able to add transparent pearlesque gobs of that shit later on, using the various methods I mentioned above.
I have do disagree, as someone who has mastered thousands of tracks over the last decade. The level of a track (peak, RMS or whatever other way you want to measure it) before mastering has absolutely nothing to do with how loud the mastering engineer can make it. It's about the quality and denseness (number of elements) of a mix. Every mix has a "loudness potential", but its level before mastering has nothing to do with that.
You can focus your mix on having a low crest factor if you want to, and some styles such as Psy Trance really call for this, but the OP didn't state his genre or style etc. so I think my point holds. That was that how a mix looks is not as important as how it sounds, and if it sounds good, don't worry about how peaks are looking on the waveform. If you look back you'll see the OP agreed with me.
Loudness, or how waveforms look should be the last thing on a mix engineers mind, just focus on sound quality/an amazing sounding mix. Any mastering engineer worth his salt is going to be able to add transparent pearlesque gobs of that shit later on, using the various methods I mentioned above.
-
do_androids_dream do_androids_dream https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=164034 - KVRAF
- 2908 posts since 26 Oct, 2007 from Kent, UK
+1Hermetech Mastering wrote:How does it sound? Instead of looking to fix a problem that doesn't exist, using clipping, compressing, or limiting (all of which are non-linear processes, and will add nasty intermodulation distortion to your mix), why not just mix everything at a lower level and leave the peaks in there, if there's not an audible problem? The mastering engineer can get the loudness later, no need to worry about that, just get the mix sounding great and stop worrying about how the waveform looks. If it sounds too quiet, turn up your monitor gain.
Basically exactly what I was going to post
Yes, Just do your mix with plenty of headroom and stop worrying about what the waveform looks like!
-
do_androids_dream do_androids_dream https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=164034
- KVRAF
- 2908 posts since 26 Oct, 2007 from Kent, UK
That's the only part I might disagree with in Herm's post - but I think he was using it as a broad generalism. Yes, If you really want your track to be 'loud' then it most definitely does start much earlier than the mastering stage - even before mixing - sound choice is very important.Alphacodex wrote:Sorry, but "the mastering engineer can make it loud later" is a very wrong attitude.
