Old 01-16-2019, 04:51 PM   #41
Coachz
Human being with feelings
 
Coachz's Avatar
 
Join Date: Oct 2010
Location: Charleston, SC
Posts: 12,770
Default

Quote:
Originally Posted by karbomusic View Post
Sure, if the peaks are near zero but only if they are near zero. But there is some value to knowing how loud those values actually sound IMHO.
Yep. 😎😎
Coachz is online now   Reply With Quote
Old 01-16-2019, 06:13 PM   #42
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 7,272
Default

Hard limiting to RMS can be pretty pretty weird, and is probably not what you actually want. The longer your window, the longer it takes for the detector to register that you've crossed the threshold (in either direction). Some of that can be dealt with using proper lookahead (precomp in ReaComp), but it can still be strange. I've used it for certain things, but would never consider doing so on a full mix except as a special effect. I do a lot of long RMS (and precomp) compression at mostly very low ratios, and that is a good transparent way of reducing the dynamic range, but at ratios big enough to be called limiting it can be pretty unnatural.

What you actually need (as mentioned above) is to render your thing and then analyze and/or normalize to reach your target, and if you can't get a DR (peak minus integrated average) with that, use more conventional dynamic processing.

Quote:
Originally Posted by Coachz View Post
Peaks not near zero I'm always down about 13 lufs peaking at like -10. My monitors volumes are right where I like them. I had an spl meter before Armageddon here. Now I wing it.
Then your monitors are about 10db too loud. It's that simple. You should be able to hit 0dbFS without blowing things (or yourself) up. If you want it that loud for other reasons (I can think of a couple), maybe try turning down the level of the Hardware Send on your Master.

Last edited by ashcat_lt; 01-16-2019 at 06:28 PM.
ashcat_lt is offline   Reply With Quote
Old 01-16-2019, 06:33 PM   #43
Coachz
Human being with feelings
 
Coachz's Avatar
 
Join Date: Oct 2010
Location: Charleston, SC
Posts: 12,770
Default

Quote:
Originally Posted by ashcat_lt View Post
Hard limiting to RMS can be pretty pretty weird, and is probably not what you actually want. The longer your window, the longer it takes for the detector to register that you've crossed the threshold (in either direction). Some of that can be dealt with using proper lookahead (precomp in ReaComp), but it can still be strange. I've used it for certain things, but would never consider doing so on a full mix except as a special effect. I do a lot of long RMS (and precomp) compression at mostly very low ratios, and that is a good transparent way of reducing the dynamic range, but at ratios big enough to be called limiting it can be pretty unnatural.

What you actually need (as mentioned above) is to render your thing and then analyze and/or normalize to reach your target, and if you can't get a DR (peak minus integrated average) with that, use more conventional dynamic processing.


Then your monitors are about 10db too loud. It's that simple. You should be able to hit 0dbFS without blowing things (or yourself) up. If you want it that loud for other reasons (I can think of a couple), maybe try turning down the level of the Hardware Send on your Master.
Great suggestion. Hardware send is a great place to do this. Thanks
Coachz is online now   Reply With Quote
Old 01-16-2019, 08:23 PM   #44
woogish
Human being with feelings
 
Join Date: Feb 2011
Posts: 693
Default

try multicompression
woogish is offline   Reply With Quote
Old 01-17-2019, 04:03 AM   #45
Coachz
Human being with feelings
 
Coachz's Avatar
 
Join Date: Oct 2010
Location: Charleston, SC
Posts: 12,770
Default

Quote:
Originally Posted by woogish View Post
try multicompression
Thanks but I"m trying to move into the LUFS world since it appears directly designed to measure loudness it would seem a loudness limiter using LUFS would be very logical.
Coachz is online now   Reply With Quote
Old 01-17-2019, 11:00 AM   #46
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 7,272
Default

Quote:
Originally Posted by Coachz View Post
...it would seem a loudness limiter using LUFS would be very logical.
But it's not. I dig where you're coming from, but I'm telling you it doesn't actually work in practice. Have you tried it yet? I mean if you actually like what it does, then fuck me, but I'm pretty sure you won't like it nearly as much as other ways.
ashcat_lt is offline   Reply With Quote
Old 01-17-2019, 01:11 PM   #47
Coachz
Human being with feelings
 
Coachz's Avatar
 
Join Date: Oct 2010
Location: Charleston, SC
Posts: 12,770
Default

Quote:
Originally Posted by ashcat_lt View Post
But it's not. I dig where you're coming from, but I'm telling you it doesn't actually work in practice. Have you tried it yet? I mean if you actually like what it does, then fuck me, but I'm pretty sure you won't like it nearly as much as other ways.
I've started to use LUFS meters on the master to see what my current loudness is but that's it. Is there something not to like with setting a mix to -14 lufs or do you mean trying to limit channels by using lufs?

What downsides do you experience ashcat_lt ? Coming from an engineering background I tend to migrate towards the shiny thing......oh look 30 band eq....then I end up using a 4 band because it works better for me . :-)
Coachz is online now   Reply With Quote
Old 03-23-2024, 12:52 PM   #48
FelixMagi
Human being with feelings
 
FelixMagi's Avatar
 
Join Date: Mar 2016
Posts: 82
Default

Old thread new plugin: LufsLimiter

Freeware, certainly the commercial version is the complete one.

https://vst.saschart.com/lufslimiter...ive_normalizer

FelixMagi is offline   Reply With Quote
Old 03-23-2024, 02:32 PM   #49
Sound asleep
Human being with feelings
 
Sound asleep's Avatar
 
Join Date: Nov 2009
Location: Montreal, Canada
Posts: 9,052
Default

To me LUFs limiter seems less desirable, because due to the nature of loudness, even the momentary setting is an average.

So, regular rms, is much faster, and sample peaks are instantaneous.

Therefore making something louder, and disallowing peaks to exceed a certain level, is good.

For LUFS, lets say you get to a section of -13 LUFS, what should your limiter do? it's a little different, because you can't just lop off the tops. it's not peaks we're looking t, it's over a time. So, the "limiter" would really essentially need to turn down the input. and you could make a normalization that scans the file, and normalizes so no momentary lufs above -14 occur, but no limiting will have occurred.

But let's say you want to try and hit -14 consistently, so, riding the input, let's say. This might be more what you'd want, but it's still weird in my mind, because let's say you have 200ms of real loud, 200ms of real quiet, then 200ms of real quiet again.

The first 400ms will need to be limited to stay at -14 lets say, so what about the 2 200ms sections? they will need to be counted with the loud 200ms and the quiet 200ms, so, what do you do?

I think due to its nature, its difficult to limit that way.

However, what you could do, which may be what you'd want, is set a lufs type limiter that will set the level of limiting to achieve a desired loudness level. So, you set it to -14, hit play, and it will move the threshold wherever it needs to be, to sit at -14 integrated. you could loop a short item as well. Then your threshold will be set to that setting, and you leave it there.

But, i don't believe this is generally desirable. I think most of the time, it's preferable to set the amount of limiting based on the sound once level matched.

It might be cool to be able to set reaper meters to lufs and then push into a limiter. you could do that with meter plugs.

But if you never want it to exceed a certain lufs, you'll need to find the loudest part, set it there, or print the track and run normalization. but of course that won't affect dynamics.

a script could probably allow you to set a limiter, and a meter plugin which is your script plugin, and this would control on a parent track, and then it would ride the child fader to prevent going over -14.

But ultimately, i think you'd prefer just going with your ears.

This is my understanding of how it works, anyway.
__________________
Slava Ukraini
Sound asleep is offline   Reply With Quote
Old 03-23-2024, 02:43 PM   #50
Coachz
Human being with feelings
 
Coachz's Avatar
 
Join Date: Oct 2010
Location: Charleston, SC
Posts: 12,770
Default

Same here, I really got hooked on ReaLimit at -1 truepeak as last in the chain with my LUFS-I meter after it. I just compress ahead and get dynamics where I want and lufs at -13.5 and all is good.
__________________
Track Freezing Scripts

Coachz Repo
Coachz is online now   Reply With Quote
Old 03-24-2024, 01:42 AM   #51
FelixMagi
Human being with feelings
 
FelixMagi's Avatar
 
Join Date: Mar 2016
Posts: 82
Default

Here's why I disagree:

Even the peak limiter averages over a certain window and has an envelope, but the window for peaks is narrower than for LUFS. RMS uses even larger windows close to Lufs. But let's remember what LUFS means: it's an indicator of human perception. The human ear doesn't perceive peaks but also a certain window of samples, so it has latency. The increase is always gradual, plus there's a Look Ahead there (in the commercial one).

Another aspect you haven't taken into account is that standard window for integrated LUFS, have you seen how narrow the one used in the program is? Below standard.

Quote:
Originally Posted by Sound asleep View Post
So, regular rms, is much faster, and sample peaks are instantaneous.
Your statement is false, you can set the program to respond as in the case of RMS window, and the peak limiter is not instantaneously responsive. Only clipper is instantaneously.

The limiter doesn't cut the peaks, have an envelope, only clipper does, pay attention to this confusion.

It seems like you haven't tested the program and haven't read what the manufacturer says, you're theorizing too much without trying.

From what I've observed, the big difference is that this type of limiter works well with large attenuations per track and less on the master. This aspect seems to be quite relative, maybe I'll come back later after testing it on other music genres. Another difference would be in the envelope, the attack and release are larger.

Personally, I’ve used it for trance music and the results are very good. For live performances, it's already irreplaceable. Pay attention to the news, to benefit we must jump over old approaches and patterns. "I know" is not a good aproach.
FelixMagi is offline   Reply With Quote
Old 03-24-2024, 01:51 AM   #52
TonyGlover
Human being with feelings
 
TonyGlover's Avatar
 
Join Date: Mar 2016
Posts: 70
Default

Quote:
Originally Posted by FelixMagi View Post
Pay attention to the news, to benefit we must jump over old approaches and patterns. "I know" is not a good aproach.
Interesting what you say.

I have been looking for this type of program for some time.
TonyGlover is offline   Reply With Quote
Old 03-24-2024, 02:29 AM   #53
TonyGlover
Human being with feelings
 
TonyGlover's Avatar
 
Join Date: Mar 2016
Posts: 70
Default

Sound asleep all peaks limiters have an attack, it cannot be instantaneous.

The Lufs measurement window is of course larger, but that is the reason why this kind of limitation complements well the peak limitation.
TonyGlover is offline   Reply With Quote
Old 03-24-2024, 11:03 AM   #54
Sound asleep
Human being with feelings
 
Sound asleep's Avatar
 
Join Date: Nov 2009
Location: Montreal, Canada
Posts: 9,052
Default

Quote:
Originally Posted by TonyGlover View Post
Sound asleep all peaks limiters have an attack, it cannot be instantaneous.

The Lufs measurement window is of course larger, but that is the reason why this kind of limitation complements well the peak limitation.
Yes they can, with look ahead. Even without look ahead, they can be pretty instantaneous. L1 and L2 are both like that.
__________________
Slava Ukraini
Sound asleep is offline   Reply With Quote
Old 03-24-2024, 11:08 AM   #55
Sound asleep
Human being with feelings
 
Sound asleep's Avatar
 
Join Date: Nov 2009
Location: Montreal, Canada
Posts: 9,052
Default

Quote:
Originally Posted by FelixMagi View Post
Here's why I disagree:

Even the peak limiter averages over a certain window and has an envelope, but the window for peaks is narrower than for LUFS. RMS uses even larger windows close to Lufs. But let's remember what LUFS means: it's an indicator of human perception. The human ear doesn't perceive peaks but also a certain window of samples, so it has latency. The increase is always gradual, plus there's a Look Ahead there (in the commercial one).

Another aspect you haven't taken into account is that standard window for integrated LUFS, have you seen how narrow the one used in the program is? Below standard.

Your statement is false, you can set the program to respond as in the case of RMS window, and the peak limiter is not instantaneously responsive. Only clipper is instantaneously.

The limiter doesn't cut the peaks, have an envelope, only clipper does, pay attention to this confusion.

It seems like you haven't tested the program and haven't read what the manufacturer says, you're theorizing too much without trying.

From what I've observed, the big difference is that this type of limiter works well with large attenuations per track and less on the master. This aspect seems to be quite relative, maybe I'll come back later after testing it on other music genres. Another difference would be in the envelope, the attack and release are larger.

Personally, I’ve used it for trance music and the results are very good. For live performances, it's already irreplaceable. Pay attention to the news, to benefit we must jump over old approaches and patterns. "I know" is not a good aproach.
You're right, i am not referring to that limiter. Idk how it works, i never tested it, and i have no plans to ever test it.

What i meant by peaks being instantaneous, is you can have a limiter that will just not allow samples above a threshold, and when the samples go above that, it prevents them from exceeding the ceiling. There may be some intersample peaks, but it doesn't require an amount of time to elapse, before determining "this is too loud" if a sample is above the threshold, it is above the threshold. That's instantaneous. I'm not sure exactly how they are programmed to work, under the hood.

For lufs, you need time to elapse, in order to define a number, because lufs are inherently like that. Samples are at a given value immediately, and that's that.

If the plugin uses a below standard window size, then it's not LUFS. the window size is part of the definition of LUFS, momentary, or integrated. You can make the window whatever you choose for it to be, but it is a standard, you cannot decide. So, if you change it for your plugin, then your plugin is limiting loudness, based on LUFS style of loudness measurement, but it does not follow the LUFS standard.
__________________
Slava Ukraini
Sound asleep is offline   Reply With Quote
Old 03-25-2024, 02:38 AM   #56
FelixMagi
Human being with feelings
 
FelixMagi's Avatar
 
Join Date: Mar 2016
Posts: 82
Default

Lufs limiter can have look ahead, why not? As I believe, many plugins like this will appear.

Window size it can be changed even in the standards. Read again what Lufs mean, it's not limited to a single window size.

If you haven't tried the program to see that even integrated it's about the same for different window sizes, these are empty and false words. The goal is to limit in Lufs not measurement.
FelixMagi is offline   Reply With Quote
Old 03-25-2024, 04:28 AM   #57
TonyGlover
Human being with feelings
 
TonyGlover's Avatar
 
Join Date: Mar 2016
Posts: 70
Default

Without a test, the discussion is useless and unproductive.
TonyGlover is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 04:38 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.