Meta Normalize

since WL6 the functionality of Meta Normalizing has changed. since then WL8 (i skipped WL7) and now WL9, i can NOT re-create the behaviour or results that i want to achieve.
any given montage containing various clips of different RMS or Integrated Loudness:
i want the Meta Normalize function (MN) to analyse the clips and bring up the quietest clip to its max peak without clipping and then adjust all other clips (with a higher RMS) to be attenuated to match the RMS of that quietest file.
the result is a montage where the apparent loudness of all clips is equal to the quietest clip, which has been boosted to its max peak level.
somehow in WL6 that was a possible setting that i can not achieve with the MN dialogue in WL8/9.
instead I get WL8/9 matching the loudest clip in RMS, then lowering the master output to avoid clipping.
it achieves a similar result, but displays a distorted visual because most of the clips are just showing as “sausages” without any resolution or detail due to the boosting far above the individual 0db line.

any help with that?
it would work i think if the WL8/9 dialogue would give the option under Clips: Match quietest clip, instead only Match loudest clip.


Kamalski, I know we’ve both been in on discussions about this in the past.

I tried your suggestion to match quietest clip, because it makes sense. (In Wavelab 8, sort of simulating such a setting by visually deciding which is the quietest clip, selecting that clip, and then selecting the setting “match focused clip” in the meta-normalizer). It works to a degree, but what if the quietest clip is too quiet? The whole album will be brought down to that quiet level, with no good way to get everything up relatively to peak near zero, like in Wavelab 7 and Wavelab 6. (besides trial and error with LUFS values).

I hadn’t looked at the meta-normalizer in Wavelab 9 until today, but it looks like PG has put in a new feature that does what you and others have been asking for, as similar to Wavelab 6 and 7 as possible given the difference in the default RMS analysis settings in WL 6 and 7, and default LUFS analysis settings in Wavelab 8 and 9.

There’s a new checkbox called “force equal loudness”, and a new factory preset called “All Clips Same Loudness and Below 0db Peak”. Load that factory preset and I think it will do what you want.

You could download Wavelab 7 and the Wavelab 9 trial and compare (the results are not exactly the same, but that’s to be expected I think with RMS default settings vs LUFS default settings, but Wavelab 9 is probably more “right” now with LUFS rather than the Wavelab 7 RMS.

Anybody who’s asked for this, I’d like to know what other people think.

thanks bob99
that did indeed work as I wanted it.
but only available in WL9, which i have already but didn’t switch over from WL8.5
as WL9 has such a different look and workflow i did not have the time to learn it.
this is actually one of my main criticisms for software evolution in general. as a professional user I do NOT need a new look and fancy little “new ways” of doing things. i need the software to follow ingrained learnings and patterns. this is how mastery in any field is achieved: learning and using your tools over and over. to deviate from this can only be justified if something drastically different and new is introduced that brings brand new capabilities to a software. everything else is just catering to a clientele of geeks and hobbyists. (ProTools, as much as it lacks in many ways, is keeping the “old” workflows consistent over its evolution as a software!)

thanks for your help

If you really want to do this in Wavelab 8, there’s a way, but it takes 3 steps. I’ve made some preset screenshots below.

Make presets as in the pictures. You’ll need to have a clip selected in order to make the Meta 2 preset, because it requires the “only selected clips” checkbox.

  1. Load and run the Wavelab 7 Meta 1 preset.

Don’t play the montage yet (!), as it’ll be incredibly loud. I started with Preset 1 at 0 LUFS (super incredibly loud), but decided -3 LUFS was plenty as I’ve never seen anything that loud (I think I’ve seen something that was -4LUFS once, but that’s crazy). But -3 LUFS is still very loud so don’t play the montage until you’ve run Meta 2 and Meta 3.

1a. Look on the clips tab to see which clip was raised the most by the Meta 1 preset. Select that clip. (that was the quietest clip, going from your suggestion).

  1. Load and run the Wavelab 7 Meta 2 preset.

(leave the selected clip selected for step 3).

  1. Load and run the Wavelab 7 Meta 3 preset.

That’s it. It’s fairly easy once you get used to it.
Wavelab 7 Meta 1.PNG
Wavelab 7 Meta 2.PNG
Wavelab 7 Meta 3.PNG

thanks bob99. very helpful!
however it need an additional step before you 3 steps:
i need to determine which is actually the quietest clip.
for that i have mad an additional MN preset that needs to be run first
the attached images show the original mon
step 1 shows the settings of the MN preset that results in showing that the quietest clip is the one clip that was boosted most! now remember that clip and undo the MN. step 2.
now continue with the 3 MN presets as introduced previously.
so it is actually a 5 step process, but it gets the results i need.

original mon.png
step 2 undo and select.png

Great Kamalski, if that works better for you. But my Step 1 and Step 1a determine the quietest clip too, as they’re all going to a common LUFS value. The FINAL result should be the same with just the 3 steps, unless I’m missing something.

if you did not choose the quietest clip in the first place, your 3 steps only will not get the desired result. checked it out. if you choose one of the louder files the results are different

Thanks so much for taking a look at this Kamalski. I’ll check it out and get back to you. I’m still thinking an artificially high LUFS target like -3 LUFS or 0 LUFS without peak constraints in Preset 1 works just as well as “match loudest clip” in determining which is the quietest clip, and that the quietest clip doesn’t need to be selected until after Step 1:

  1. Run Preset 1

1a. Look at the clips tab to see which clip was raised the most by Preset 1 (raised the most, because it was the quietest clip). Select that clip.

  1. Run Preset 2

2a. (Leave the selected clip selected for Step 3)

  1. Run Preset 3

If our presets are exactly the same, that method works for me, but like I said I’ll check it out again.

The only reason I can think why it wouldn’t work as expected is if one of your original clips is louder than -3 LUFS, which is hard to believe, but I guess maybe could be the case depending on extreme level or processing. Try changing the setting in Preset 1 to 0 LUFS, follow the 3 steps (including 1a), and see if that doesn’t work as expected. But again, don’t play it while the clips are raised like that because it’ll be extremely loud.

(that LUFS value in Preset 1 is just an arbitrary artificially high value I chose to cover nearly all super loud possibilities of the original clips. It could be set to 0 LUFS or +3 LUFS or +6 LUFS, doesn’t matter. Technically I guess it should be set to the highest value the program allows, but I didn’t want to do that if someone could accidentally play super loud gains like that in the middle of a multistep process. A mutistep process that eventually reduces those gains for normal playback.)

EDIT: Something gets weird if all clips are targeted higher than about +10 LUFS in Wavelab 8, then attempted to match a normal level clip. The loudness matches are not correct to the normal level clip. Maybe that’s why the LUFS levels are limited to a maximum of 0 LUFS in Wavelab 9, so it’s probably a good idea to use a maximum of 0 LUFS in Wavelab 8 when doing this. That should cover anything anyway I think.

thanks bob99.
i tried again with your suggested 3 step program and it works as desired. i made a mistake previously.
so thanks, 1 step less!