Loudness wars cause a free meter to be created

flatfinger said:
http://www.dynamicrange.de/en/download


This is a nice peak meter/RMS with the twist of also having a averaging "crest" bar in the middle ...............the price is right !!
Click to expand...

IMO the loudness wars is for the birds. I've heard so many songs destroyed by dynamic multi band compression. It really has become sad. Get a good clear mix. Add a few DB to it and call it the day. If the listener wants volume, well, there is a lil knob that's specifically called "VOLUME". Use it.

Gabriel
 
Gabriel_0 said:
IMO the loudness wars is for the birds. I've heard so many songs destroyed by dynamic multi band compression. It really has become sad. Get a good clear mix. Add a few DB to it and call it the day. If the listener wants volume, well, there is a lil knob that's specifically called "VOLUME". Use it.
Click to expand...

Did you actually click the link before you jumped to conclusions? The website is one of the many advocates against the loudness war.

I like the look of the meter :) It has a 'dynamic range indicator' to encourage you to keep the dynamics in your music... the meter fades to red as the dynamic range becomes squashed up.

Thanks for the link :)
 
mattr said:
Did you actually click the link before you jumped to conclusions? The website is one of the many advocates against the loudness war.

I like the look of the meter :) It has a 'dynamic range indicator' to encourage you to keep the dynamics in your music... the meter fades to red as the dynamic range becomes squashed up.

Thanks for the link :)
Click to expand...

Gotta admit, no I did not, my bad. I was pretty wrapped up in answering PMs so I just gave my thoughts on the loudness wars (one of my main pet peeves) is all.

All the best
Gabriel
 
Haven't tried the real-time one yet, but the offline was is pretty sweet.

I'm thinking of putting up one of those labels on my album when it comes out.
 
I love the loudness. I am working on a 10 in-line compression technique that will squash the mix into a DC signal. Shit, pure DC is my thing. I named my band "DC"

I don't like AC/DC because they have an AC component to them that rubs me the wrong way.
Long Live DC!!!!!!!!!;)
 
Great little tool. I got the off-line one happening, but I can't get the real-time one to show up in my list of plug-ind in REAPER. I've copied the folder in all my plug-in folders but it's still not coming up in REAPER.
 
My guess is that there will be a couple of sad ironies here: first, that this meter will probably be used most often by those trying to push the DR number as low as possible ("Awsome! I got my mix up to a DR of 4, and it rawks!"); those that believe in the wonders of dynamics usually don't need a meter to tell them they have it right; second, that those who use it the most will often be those that could use the phase correlation meter at the top the most, but who have no idea how to read that part of the meter.

G.
 
Glen ,
you're the eternal optimist !!! Most of the stuff I listen to ( nothing much past '95!) is 12 at the lowest . Kind of sad ... entire generations thinking music should be a sonic buzz saw that never lets up , without spaces and tons of unatural symmetric odd order distortion ... tisk , tisk !! I guess I'm not real optimistic either !!:(
 
Interestingly, all of my classic rock CDs hover around the 10 range. Deep Purple and The Beatles are consistently 9, some Tull albums are aruond 15, others are around 10-12. Death Magnetic is holy shit...
 
flatfinger said:
Glen ,
you're the eternal optimist !!! Most of the stuff I listen to ( nothing much past '95!) is 12 at the lowest .
Click to expand...

Seafroggys said:
Interestingly, all of my classic rock CDs hover around the 10 range. Deep Purple and The Beatles are consistently 9, some Tull albums are aruond 15, others are around 10-12. Death Magnetic is holy shit...
Click to expand...
There is nothing necessarily "classic" about a "classic rock CD". First off, anything recorded before around 1981 or so was recorded and mastered only for vinyl at best. Anything that old that winds up on CD has had to be remastered with that ME's/labels idea for what constitutes a good-sounding remaster, which often bears only a passing resemblance to the original dynamics-wise.

What makes it worse is that this remastering is often performed several times over the years and for different geographical release or different labels. This is true for music from all eras; I have three different versions of XTC's "The Ballad of Peter Punkinhead" that all sound quite different, and a couple of different versions of some old Ray Charles stuff, some of which sounds absolutely horrible - far worse than the original vinyl ever did.

A lot also has to do with the source from which the CD has been mastered. Did it come from the original masters, and if so form the first, second or even third generation of the master? There are some crapola CDs that have been recorded and remastered from vinyl press and not even from master tapes. And so on.

As far as the optimist in me, think about it. The only people who "need" a crest meter are those who are interested in squashing their mixes. Those of us who prefer to let the dynamics breathe don't need a meter to tell us where the mix wants to go - in fact the meter can't tell us a damn thing about the proper dynamic range for any given mix.

I'm all for dynamic range and hate the loudness wars, you guys all know that. But assigning numbers to a disc is pretty meaningless, because crest factor and dynamic range depend upon content, not upon a measurement.

G.
 
Oh yes I'm fully aware that all of my shit has been remastered in recent history. Though The Beatles strikes me as odd, since those masters are from 1987, when the war first started, but still having a dynamic range of 9? Even Metallica's Black Album, which came out 4 years later, has more dynamic range than that.
 
I don't mean to pick any nits here, just to make sure we're all using the same vocabulary...

It kind of bothers me a little that they put that big DR right under the crest factor meter, because that is actually a measure of crest factor, not the actual dynamic range of the song. What's the difference? Crest factor is the difference between the RMS "average" level and the peak level. Dynamic range is the difference between the quietest part of the song - or the noise floor - and the loudest peak. So to say that a song has a "dynamic range" of 9dB when that is actually it's crest factor is technically incorrect. If the song peaks at 0dBFS, has an RMS of -9dBFS and has a noise floor of -60dBFS, it actually has a full dynamic range of 60dB.

I know that's kinda splitting technical hairs, but it's probably good for everybody to understand that difference off the bat in order to avoid possible confusion later in this or other threads.

It also may help explain low "DR" numbers on older recordings. There's plenty of old vinyl that is simply quieter stuff but does not necessarily have more dynamic range or crest factor than new stuff. For example, it's not unusual to find some old vinyl and even the phono preamps through which they are played to have a full dynamic range of only 55dB or so. They may RMS at, say (just for example), -17dBFS at unity gain structure through the converter, but peak at -8dBFS at unity converter gain. Their crest factor is only 9dB, but because their total dynamic range is only 47dB (55dB - 8dB), the crest factor as related to the total range is relatively large at something around 20%.

A digital recording, with a canvas (at 24bit) of 140dB and an quality analog path with a potential (with proper gain structure) of probably some 65dB with an identical crest factor of 9dB means that the crest factor is only 13% of the total analog range and only about 6.5% of the total digital range.

G.
 
Another factor here is just how little meaning can be put into an RMS measurement of an entire song. An average volume as averaged across three minutes or so tells us almost nothing about the actual dynamics within any given part of that three minutes. While this is an extreme example, it makes the point: three minutes of sine wave or pink noise averaged to an RMS of -9dBFS will sound entirely different from a three minute track which has a running RMS average of -18dBFS 50% of the time and an RMS of 0dBFS the other 50% of the time, yet the later will also have the same average RMS for the entire track.

For a bit more real-life of an example, let's say that we have two songs where, if we were to exclude the peaks, have the same average volume. One song is at 120BPS, the other at 180BPS, and at every beat there's a heavy peak shooting up to 0dBFS. Basically they have the same kind of dynamic content, but because the second song has 50% more peak energy than the first, that will drag the average RMS reading for the second song up. Same kind of dynamics, same point-to-point crest factor, yet for the overall track the RMS will read higher and the crest factor will therefore read lower.

This also means that the average non-peak volume of the second song could actually be lowered, with the real-time crest factors increased, making the song more "dynamic", but the average RMS and crest factor readings for the entire track would be no different than the louder, lower crest factor first track. Higher dynamics, more perceived "range" but no change in the numbers.

G.
 
Last edited:
SouthSIDE Glen said:
Another factor here is just how little meaning can be put into an RMS measurement of an entire song. An average volume as averaged across three minutes or so tells us almost nothing about the actual dynamics within any given part of that three minutes. While this is an extreme example, it makes the point: three minutes of sine wave or pink noise averaged to an RMS of -9dBFS will sound entirely different from a three minute track which has a running RMS average of -18dBFS 50% of the time and an RMS of 0dBFS the other 50% of the time, yet the later will also have the same average RMS for the entire track.

For a bit more real-life of an example, let's say that we have two songs where, if we were to exclude the peaks, have the same average volume. One song is at 120BPS, the other at 180BPS, and at every beat there's a heavy peak shooting up to 0dBFS. Basically they have the same kind of dynamic content, but because the second song has 50% more peak energy than the first, that will drag the average RMS reading for the second song up. Same kind of dynamics, same point-to-point crest factor, yet for the overall track the RMS will read higher and the crest factor will therefore read lower.

This also means that the average non-peak volume of the second song could actually be lowered, with the real-time crest factors increased, making the song more "dynamic", but the average RMS and crest factor readings for the entire track would be no different than the louder, lower crest factor first track. Higher dynamics, more perceived "range" but no change in the numbers.

G.
Click to expand...

The meter and the stand alone app are doing seperate things . They say that the stand alone app is where to get the "official " DR rating and the stand alone app is described as working like this

From the manual pg4

In order to determine the official DR value, a song or entire album is scanned. A histogram ( loudness distribution diagram ) is created with a resolution of 0.01 db. The RMS - "an established loudness measurement standard-"(emphasis mine ) is determined by gathering approximatley 10,000 pieces of loudness information within the span of three seconds ( added by me ...; thats about how long the stand alone app crunches the data before arriving at a dr number ) From this result, only the loudest 20% is used to determine the average loudness of the loudest passage .
Click to expand...


There is allot about how there approaching it in the manual , and I beleive they are aware of the problems involved in some songs that are real quiet at the start and then build into a badass cresendo ... the first 2/3 would make the last 1/3 of super thick average give a total and erroneous dr number if you just did a rms measurement on the whole file without regard to the structure.
 
OK, I missed the link to the manual the first time I went there. I just gave the relevant parts a quick read. Hanoing only just read it, some of it I got to think about a bit, but I did find much of what is on pages 7 and 8 to be especially honest and relevant:
The DR value therefore represents the amount of density/thickness (loudness) more than the dynamic range of the song.
Click to expand...
and
Since the measurement does not take psycho-acoustic elements into consideration, very dynamic music (such as classical music) is less suitable for DR scaling...This equivalence would only be possible if psycho-acoustic parameters were incorporated which would necessitate long and costly research and development.
Click to expand...
I'd argue that their example of classical music vs. pop music is only an extreme example to make the point, but the problem with such numerical measurements also exists between much closer related genres and even within genres.

Perhaps the problem may not be as acute between Buddy Guy's bluesy "It Feels Like Rain" and Eric Clapton's bluesy "I'm Tore Down" as the difference between Vivaldi's classical "Autumn" and Alice Cooper's pop rock "School's Out", but it's still going to be there to a significant degree because the density and dynamics are still very significantly different.

In fact, I'd put a Telarc recording of Vivaldi's "Spring" up against many pop/rock songs for density and dynamics as not all that different, so even their example of classical vs. pop is not a solid one.

It is up to the song and the content, and assigning a qualitative meaning to a numerical value is jumping a step too far, IMHO.

G.
 
Seafroggys said:
Interestingly, all of my classic rock CDs hover around the 10 range. Deep Purple and The Beatles are consistently 9, some Tull albums are aruond 15, others are around 10-12. Death Magnetic is holy shit...
Click to expand...
I have old Deep Purple CDs that I bought in the late 80's and they are no where near as loud as the lastest versions of the same CDs. They have been remastered and ruined.

There is a greatest hits that has a version of Hush that sits around -9dbfs rms! There are so many things wrong with that, I don't know where to start.
 
I am very fortunate in that I got most of my music collection before the loudness wars . I have most of the stuff I want from my younger day , and in true old codger fashion , have little intrest in what the whipper-snappers are listening to.

Not only am I a dinasaour to them yutes because of my taste in music ; but I'm also a weirdo because I PAID FOR MY MUSIC:o:o.
 
Farview said:
I have old Deep Purple CDs that I bought in the late 80's and they are no where near as loud as the lastest versions of the same CDs. They have been remastered and ruined.

There is a greatest hits that has a version of Hush that sits around -9dbfs rms! There are so many things wrong with that, I don't know where to start.
Click to expand...

I almost wish I got a lot of the 80's releases for most of my albums, but they are either hard to find or were just mastered poorly, regardless of how loud it is. It took awhile before the industry knew what the hell they were doing when making CD's. Also, the new releases have cool bonus tracks, which is always a plus.

I do think my In Rock and Fireball are older CD releases, mainly because there's no bonus tracks, they're quieter, and sound thinner, almost harsher. My Machine Head, Who do We Think We Are, and Burn releases are warm and full sounding, but hot.

I'd almost want to say the new releases sound better, even with a high RMS.
 
You must log in or register to reply here.
Back
Top