The Loudness War: Give It Up!

It's the original mix. Only the box set has a stereo remix for Layla.

The Platinum SHM-CD and old Polydor CD are light years better than the terrible compressed HD download.

 

Are you certain about that? I once owned a single CD release with the remix.

 

I think it comes down to what many of us on this forum have said before. The reason this continues to happen is twofold

1) Money
2) Nobody outside a small percentage of people cares

 

1. The "wars" have thankfully saved me a ton of #1
2. I am happy to be in the small percentile on this one - since it's my wallet and my ears. I continue to find maximum joy in finding those treasures from yesteryear as issued with excellent dynamics and no Advil requirements

VP

 

Not positive but I have the following masterings and they all have the original mix:

Dennis Drake Polydor mastered CD
Polydor 90's remaster
MFSL
SACD
Deluxe Edition
Platinum-SHM CD
HD Download

The only one that I have that has the stereo remix is from the 20th anniversary box set.

 

I suppose both time wasting and enjoyment are in the eye of the beholder.

I think you just described rap/hip-hop.

 

Or maybe most people just listen to the music and don't care about any of that, they just want something in the background to fill the silence. The bottom line is the people that care about this stuff just don't make up a large enough slice of the pie for the industry to care what we think. They leave that up to the Mofis and AFs of the world.

 

And the limited dynamic range was causing that so he turned down the treble. I believe he was confusing boosted hi frequencies with dynamic range compression/limiting.

 

Actually, the first really-compressed CD I came across was Pet Shop Boys Very in 1993, with an average DR of 9 - extreme, for that era. I think that's right at the margin where things start to go from loud to impossibly distorted.

Their latest - Electric - has a DR value of just 5... I find it to be exceedingly unpleasant to listen to. Fortunately being electronic dance music, most of the energy is concentrated in the bass, which makes it a little less ear-shattering...

The Nightfly is an early digital recording and still one of the best-sounding recordings I've ever heard. The carryover of '70s production techniques and styles means it lacks most of the shrillness that characterizes so much of what was recorded in the '80s (both analog and digital).

 

Cymbals can sound absolutely awful and unpleasant when the recording gets crushed to a brickwalled DR. It's not necessarily because the treble in the recording has been boosted. It's because the compression and limiting and other processing can make the treble (mostly the cymbals, but also other parts of the percussion, and can affect other parts of the music too) very unpleasant to the ears. I can understand the need to turn down the volume to lessen that unpleasant treble. This isn't true for all compressed DR recordings. Is definitely true for some.

 

I wonder if squashed DR is simply one of the things we will reflect back on and say, "well that's just how they liked their music during that era." Much the way I think of music recorded in the 1980s... shrill, brittle, with way too much high frequency extension. Digital audio was a brand new toy. I think '80s era recordings kind of reflect that. Even though I wish that music were recorded differently, I still enjoy Richard Thompson's "Daring Adventures" and "Amnesia" immensely, to a certain extent in spite of how it sounds.

I guess what prompted this musing on my part, and apologies in advance to those who think I am wasting my time, is that I analyzed my Fountains Of Wayne albums using DR today. All their records have a DR8 except the most recent, "Sky Full Of Holes," which is a DR6. I absolutely adore SFOH. But now I cannot help wondering how much more I might enjoy it in a more dynamic form. The mastering does seem a bit "in your face" compared with their other offerings. Maybe I will reflect back and think, "Well, that's just how folks wanted their music to sound in the 2010's."

Your mileage may vary.

 

Completely agree. At one time I owned a recent Japanese import of Procol Harum "Live with Edmonton..." I don't know what the DR was, but it was very squashed and to my ears completely unlistenable, mainly because the cymbals sounded horrid and were always competing with everything else in the mix. The MFSL version only has a DR8, but it is at least listenable. I found the difference to be dramatic and the DR difference is probably no more than +1 or +2.

 

The Nightfly album TT DR value is actually DR16...

 

You are correct, sir.

I think I just found a bug in DR. The TT DR value it displays in the GUI is not the same value in the file. Wow.

It appears to display the reading for the last song analyzed, not the value for the entire folder. How freakin' braindead is that? Jeebus. Save me from developers curating the user experience.

Here's the file listing:

----------------------------------------------------------------------------------------------
Analyzed folder: /Users/Mike/Music/iTunes/iTunes Music/Music/Fagen, Donald/The Nightfly
----------------------------------------------------------------------------------------------
DR Peak RMS Filename
----------------------------------------------------------------------------------------------

DR15 -2.21 dB -19.33 dB 01 I.G.Y..m4a
DR16 -1.69 dB -20.55 dB 02 Green Flower Street.m4a
DR14 -2.98 dB -20.99 dB 03 Ruby Baby.m4a
DR15 -3.70 dB -21.22 dB 04 Maxine.m4a
DR15 -3.47 dB -21.33 dB 05 New Frontier.m4a
DR17 -1.20 dB -19.99 dB 06 The Nightfly.m4a
DR16 -2.71 dB -21.35 dB 07 The Goodbye Look.m4a
DR18 -0.05 dB -20.58 dB 08 Walk Between The Raindrops.m4a
----------------------------------------------------------------------------------------------

Number of files: 8
Official DR value: DR16

==============================================================================================

 

From the database, it looks like DR18 is for the most dynamic track, not for the entire album.

 

But you guys all realize I hope, that the DR number, is not really all that indicative of anything for sure.

Its whole implementation ( even thought it has a noble idea ) is somewhat flawed.
Some types of music that are the most dynamic, are rated far lower.

It does NOT measure dynamic range, but merely how dynamic a "portion" of the signal is, at only a certain range of loudness.

In other words, it is ENTIRELY possible to have a very wide dynamic range, and still have a mediocre number.

It actively ignores a big portion of the actual sound level range, and is only interested in a chunk of the loudest parts.

A song that starts at a whisper, and gradually gets louder to a very dramatic full volume ending, will rate mediocre, even though it has the widest range there could be.

I do not put all that much stock in DR ratings. There is something better.

 

No, I don't think I realized any of that. I assumed it was a simple arithmetic sum of the loudest value minus the smallest value in each song file. It certainly takes long enough to process each file. Still, it is (marginally) better than nothing I suppose. What do you feel is a better tool?

UI gaffes like the one noted above just p*ss me off in the worst way. In this day of applications with no documentation, the UI absolutely must make sense and be well-behaved. I flunk TT DR on both counts. Sorry, developers being in charge of the user experience are the bane of my professional existence. People who code classes, properties and methods, usually have no freakin' clue how an actual human being would use a product. So you end up with this opaque plastic UI wrapping the properties in the order that the developer wants to process them. That is almost never (!) the way an actual human being needs to interact with a product.

OK, just step away from the keyboard...

 

Well the "extended" treble in the 80's were loudness of that time.

Some 70's and 80's stuff are also a bit bass-shy because they were recorded with vinyl in mind. If the album was longer say more than 50min they had to reduce the bass to make it fit.

 

Grant,

It has been my experience that recordings with lower DR tend to be less pleasurable to listen to than those with higher ones. That's why I tend to stay away from them, especially if the are "re mastered" release of older recordings. My goal is to avoid "crushed" records.

I agree that you can have a recording with a DR=12 that is really poorly mixed, EQ'd and mastered, Hopefully that factor will be reported in reviews

 

Until a better solution is implemented, the DR database is a great tool to detect highly compressed releases and avoid disappointing purchases. That's what it was created for, not for finding the most dynamic CDs.

 

Yes....and you can use the DR ratings and view a wavform with Audacity to determine this as well. The wavform will show that delicately quiet whisper...then the gradual building of the sound. They're both tools to show you what you're hearing and to identify variances in mastering of volume levels.

It's very rare when I've heard something that, TO ME, sounds better with a low DR rating if there's a mastering with wider dynamics available. Of course there are exceptions as I HAVE heard...but generally DR can be a valuable tool. Just like wavforms that also get dissed by those who contend that it's ONLY about the music...not the technicalities. To me they BOTH go hand in hand. I have to enjoy the music FIRST..but the listening experience can be ruined by a lousy mastering. To others, they may not care.

 

Vlad the Impaler. There's a guy who's taken some artists I admire down a sonic road that I don't.

 

Agreed. To say (as Grant did) that "the DR number tells you nothing" is a gross oversimplification. It certainly conveys something about the relative dynamic range of two or more song files. It may be less accurate than we might like, but it is a straight forward apples-to-apples comparison, and as such, does tell me something. It just doesn't tell me everything.

 

There is no bug in TT DR plug-in for foobar2000. You need to make sure, however, that the collection of tracks you are processing uses the same encoding/bit-depth/sampling-rate & has the exact same Album tag too...

 

I really think the problem is with the terminology, loads of people think "compression" is the overall term for what we are talking about it isn't, I'd differentiate between "compression" and what I'd call "unwanted audio distortion" or "audio destruction", compression when used on individual instruments can be great and give them extra punch (eg bass and drums), when applied to the extreme on the overall recording it destroys it. So now when I talk about compression I mean the effect on an individual element in a recording (or a few), I just call the "Loudness" thing audio destruction.