I don't understand the obsession with DR meters

[mitchco]

DR is but one aspect of several for what makes a good recording, mix and master. It certainly is not the only aspect.

Maybe some folks are not sensitive to hearing compressors "pumping and breathing" along with the music. Or sensitive to peak limiting where the transient attack is so severely limited it doesn't sound remotely like a drum kit. Maybe some folks are perfectly fine with that. I know I am not, especially when one is used to listening to recordings with reasonable dynamic range. Then along come bands like Oasis in 1994 with the Producer admitting to using dynamic range compression "more than would normally be considered professional" says it all to me. Maybe folks like that crushed wall of sound or are simply used to it as we have been mostly listening to that ever since.

I have an acoustic drum kit in my listening room along with my stereo. Aside electric bass and guitars, I like to play along with tracks from the stereo. Dynamics comes from how you play. Playing in bands, one learns where to play quieter and louder depending on the part song and what the band members agree on where to bring up or down the dynamics. My point is, all of those nuances that bands work on for hours can be lost due to excessive dynamic range compression. For me, that just ruins the tune. Maybe I am hypersensitive to it as I twisted compressor and limiter knobs in the studio control room for over a decade and know intimately the signature sounds of these devices, especially when abused. I would be the first to admit guilt of overusing these as well.

While DR is not the end all beat all, it does have a meaning for me and the enjoyment of the music. I feel there is a happy medium as discussed by mastering engineers like Bob Katz who can influence the industry. But it does not help when every DAW has mastering presets set for "crush" as default.

 

[Pio2001]

I've tried a dynamic range meter (foo_dynamic_range for Foobar2000). It seems to me that it works best with tracks that have a constant playback level from end to end. It gives a good indication of the peak vs average level.
But when the level varies a lot along the track, it becomes useless. The most ridiculous number was given on a classical recording. Probably the most dynamic track among the 7500 items of my playlist. Edward Grieg - Peer Gynt, I Dovregubbens Hall, by Neeme Järvi, 1987.
It is the famous "in the hall of the mountain king". It begins with a clarinet playing at -60 dBFS RMS, and ends completely clipped after a furious crescendo.
Comparing the RMS level of the clarinet with the RMS level of the fortissimi at the end, the difference is 50 dB. Thus, it should be a DR50 recording.
The DR meter says it is DR11 !
That's less dynamic than many Depeche Mode or Kraftwerk tracks, that are measured between DR14 and DR17.

For the rest, I prefer a lot dynamic recordings. But among the less dynamic ones, some are much better compressed than others. For example Stellamara - Resulina (DR5) is a very well compressed track. It sounds very good. I've even got a DR2 track that sound acceptable : The Enigma TNG - Welcome to Planet Gliese. It sounds grossly compressed, but it remains enjoyable.
On the other hand, Trobar de Morte - Excalibur acoustic version (from Beyond the Woods, the acoustic songs) is DR7, but it sounds horrible. Here the compression destroyed everything.

I think that it depends on the original dynamics. If the original song has no dynamics to begin with, it can fall into the DR6 category while the recording is high fidelity, while a DR15 original song can be butchered if the dynamics is decreased to DR6.

 

[Sal1950]

I've heard a number of people make comparisons on a apples / oranges basis.
It's near impossible for the layman to compare the sound of two recordings where only the DR has been heavily compressed.
Either folks are saying "I have recording X with a DR5 and it sounds great, but have never heard it left in a more natural range of DR11 and up. Or they have compared a remastered DR crushed recording against the original and prefered the crushed but have no idea how many other changes were made during mastering to FR, etc.
If you take 2 examples of recording X and only crush the DR on one with no other modifications, the higher DR one will sound more alive and real. There''s no reason to accept the destruction of DR on CD's or HD downloads/streaming.

 

[Fluffy]

think that it depends on the original dynamics. If the original song has no dynamics to begin with, it can fall into the DR6 category while the recording is high fidelity, while a DR15 original song can be butchered if the dynamics is decreased to DR6.

That's a good point. Some things never had any realistic dynamics to begin with. this is true for some electronic music, and also with the very noisy metal genres where it's pretty much a wall of sound. In those genres, the measure for sound quality is very different from other more acoustic stuff like jazz or classical. They should not be judged by the same parameters.

 

[Sal1950]

That's a good point. Some things never had any realistic dynamics to begin with. this is true for some electronic music, and also with the very noisy metal genres where it's pretty much a wall of sound.

Then there wouldn't be anything to debate, no compression would be needed. Just record it and keep the level under clipping

 

[Pio2001]

It is also true with acoustic recordings.
In the album Parfums Ottomans by the Al Kindi ensemble, the first track, Taksim Tambur Neva, is a solo instrumental played on tanbur, with DR15, while the track Billazi Askara, that is a short vocal solo with drums, is DR6.

It just happens that the singer maintained a constant loudness during the song, while the drums are mixed in the background, while the tanbur player played with more variations in loudness.

Or maybe the vocal tracks were more compressed than the strings ! That's quite possible too

 

[Dogen]

Compression is a tool and recorded pop music, for the most part, wouldn’t sound good without it. But I much prefer it to be apples to individual instruments, not the final mix.

 

[Dimitri]

Some of the tools that are employed to achieve these ends are compression and limiting. On the far end of the chain, after the musician has played their tune and the engineer mixed it and the producer was happy and went home, there sits some audiophile and analyzes their hard work with a very basic algorithm that spits out a single number. And for him that single number is the absolute arbiter to whether the people that made the music did a good job or not. This is utterly ridiculous.

Would it be different if the algorithm was more "sophisticated" and spit out a few numbers rather than one?
DR, "loudness war", brikcwall limiting" were not part of my vocabulary many years ago.
I didn't start by looking at DR numbers going "hmmm this is bad".
I bought a CD after stumbling on a youtube video and heard a couple of songs.
Put the CD into the CD player and what came out was puzzling.
Looking around about "bad sounding CDs" I realized there was a "war" raging I was unaware of.
Looked up old albums and sure enough original releases have higher number than newer ones.
That is what is utterly ridicilous.
Along with the people that attempt to "justify".
Eventualy all the current music will need to be re-released so it can be re-sold, once more.
There is a good chance "restored to it's original dynamic range" will be the "new" selling feature.

 

[Blumlein 88]

Create an allpass IIR filter with e.g. f=60 Hz and Q=0.7 and apply it to a heavily compressed music track, e.g. Prodigy - Diesel Power with DR=5.
You get a resulting track with increased DR=12. So the DR has improved pretty much.
But the modification is just phase shifts. The compression is still there and I would wonder if you hear dynamic differences.
So IMO the DR value is an indicator only. It is not a well-defined criteria for quality.

I've already said DR is one metric and not a one number to guide on quality. That some DR12 or better tracks can sound bad. And that the DR software can be gamed. OTOH, never heard anything rated DR6 that wasn't crushed to death. Even if you figure out a way to game that and get a wider range recording to pop out a DR6 score all you have done is game the software.

What I typically do is what Uli says, dump it in a sound editor and what is what on compression/limiting is apparent visually. People like metrics and a single number to hang their hat on however. The DR software works in the sense that once you get into 6 or 7 or less you have a squashed to death recording.

I don't get this industry pervasive idea to max out loudness on any music you touch. I don't think even raunchy heavy metal benefits from excessive compression. If however you think it does, I still cannot see how someone would do this with pop vocals or acoustic/voice music, but they do.

Maybe in 40 years when all who think different have died we can have streaming music with somewhere between 2 and 4 bits with some noise shaping in the format. It is all you'll need, and it will save lots of bandwidth.

 

[Blumlein 88]

Then there wouldn't be anything to debate, no compression would be needed. Just record it and keep the level under clipping

Obviously you aren't a successful modern mastering genius. If you get a recording that naturally has no dynamics, you jump for joy and apply heavy limiting to get it SUPER LOUD!

 

[LTig]

Any musicians in this thread? How about drummers? Or if not a drummer, ever stand by a drum kit being played?
Which drum sound do you prefer in this demonstration? The ones that sound lively or the the ones that sound neutered?

Yes, Yes, yes, the original of course.

 

[2020]

This is a strange conversation. I don’t believe musicians set out to write songs with any level of DR in mind.

Actually, they are doing pre-mastering techniques that take into account it will need to be loud later by designing certain song structures that respond better to loudness down the line. I rememeber readinag about this going on 9 years ago.

TSo what you have left is music that is recorded loud and with compression or music that is recorded with the full range of the instruments/effects still in tact.

They use a lot of pre-mastered sounds and fuckery to get commercial pop to these blistering levels while still sounding smooth in a way.

If you want your music louder, that’s why there is a volume knob. I’d be very surprised if anyone ever enjoyed a compressed version of a track compared to full range when volume matched. So it’s not that the goal is high DR, it’s that all else being equal, generally higher DR is better.

This really does play into consumer electronics though. You start making things dynamic and you lose a lot of volume control on cheap devices.

 

[2020]

The point I would like to make is are we attempting to create a High Fidelity recording or not?
I've never been to a modern heavy metal concert but I imagine even they have higher dynamics live than they do on disc.
Live music whether electronic or acoustic has a very wide dynamic range, without it, it loses it's life and illusion of reality.
Music for listening in a loud environment like a car can benefit from a reduced DR, but that's not what you want at home on an expensive rig you assembled to create a high fidelity reproduction of live music. There has to be better answer than crushing everything you can purchase to a DR6!
We've seen it happen everywhere, witness what was done to the Jackson Browne - Running on Empty album when it was remastered for "High Definition".
Original CD

HDTracks 24/192 "High Definition" Download

Are you freakin kidding me, This is supposed to be a "premium release" that you pay extra for ??? I don't think so.
What a shame the labels are ruining some of the great recordings from the past and then have the balls to charge us more for them. All in the name of better playback on inferior equipment in a poor environment.
What happened to the music industry of the 50-80s that constantly worked to make better, more realistic recordings ie "The Absolute Sound"
I'm actually shocked by the people here saying "so what, what deference does it make?". Shame on you for accepting this practice.

Amen. High resolution audio is a myth anyways. There is NO PLAYBACK BENEFIT to it.

 

[LTig]

Any musicians in this thread? How about drummers? Or if not a drummer, ever stand by a drum kit being played?
Which drum sound do you prefer in this demonstration? The ones that sound lively or the the ones that sound neutered?

Being a (bad) drummer I prefer high dynamic recordings. Just had a listen to the 2012 recording of Jon Lords Concerto for Group and Orchestra. When listening at an SPL where you can still here the quietest instruments at the begin of the first movement then the final is breathtaking. This is how a rock drum snare sounds.

 

[Sal1950]

Thankfully there are a few in the industry that have worked to reverse the trend.
On Kscope's site while speaking of the new 3 disc box remaster by Steven Wilson of Porcupine Trees - In Absentia this is written,
"The first CD features Steven Wilson's 2017 remaster, which has considerably less compression and limiting for a more dynamic listening experience."
https://kscopemusic.com/artists/porcupinetree/
There are some in the industry devoted to delivering a quality High Fidelity product, even some younger ones.

 

[Fluffy]

Thankfully there are a few in the industry that have worked to reverse the trend.
On Kscope's site while speaking of the new 3 disc box remaster by Steven Wilson of Porcupine Trees - In Absentia this is written,
"The first CD features Steven Wilson's 2017 remaster, which has considerably less compression and limiting for a more dynamic listening experience."
https://kscopemusic.com/artists/porcupinetree/
There are some in the industry devoted to delivering a quality High Fidelity product, even some younger ones.

Steven Wilson is one of the main proponents for this whole dynamics trend. His albums used to sound great, but in the last years everything he touches is getting this ultra-dynamic treatment that makes everything very fatiguing to listen to. I think one of the reasons he does that is because this anti-loudness war trend, and making stuff more dynamic is a way to get more customers. Exaggerated drum hits doesn't make things more realistic or higher fidelity, it's just following a trend.

 

[Blumlein 88]

Steven Wilson is one of the main proponents for this whole dynamics trend. His albums used to sound great, but in the last years everything he touches is getting this ultra-dynamic treatment that makes everything very fatiguing to listen to. I think one of the reasons he does that is because this anti-loudness war trend, and making stuff more dynamic is a way to get more customers. Exaggerated drum hits doesn't make things more realistic or higher fidelity, it's just following a trend.

Maybe some loud pink noise modulated no more than 3 db would smooth your overwrought ears.

Otherwise you have some bizarre idea of how fatigued hearing works. Heck even OSHA understands this.

 

[LTig]

Exaggerated drum hits doesn't make things more realistic or higher fidelity, it's just following a trend.

Those "exaggerated drum hits" are a slightly compressed version of the real sound. What you think a drum should sound has nothing to do with reality at all.

 

[Fluffy]

Those "exaggerated drum hits" are a slightly compressed version of the real sound. What you think a drum should sound has nothing to do with reality at all.

I know exactly how drums sound in reality thank you very much, and it's unbearable.

Actually, why do we even need to mix things? Let's just play everything at their original SPLs. I'm sure an acoustic guitar will sound fine playing alongside a full drumset. And only opera singers would be allowed to sing, because no one else would be able to cut through the SPLs of the rest of the band. We should record rock albums in the audiophile way – two spaced microphones in the room and nothing more. This will give us the most realistic and authentic sound experience. I wonder why no one actually record albums this way.

Oh, right, I remember why. Because it will sound like shit. Modern music depends on modern recording techniques, and one main goal for these techniques is to not actually play the instruments at their real relative loudness, because that's insanity. No one really wants to hear the full volume of a drumset in their living room. Have you tried to actually put real drums in your house? Well I did. And the whole building heard it. In fact they still heard it when I put them in the basement. If you would reach actual playing levels of acoustic drums every time you turn on your hi-fi, you would lose your hearing very quickly. And obviously they would realistically be waaaay louder than any unamplified instrument.

You can go around and ask any professional sound engineer you want why they put compression on drums, and they would all tell you that they need it to tame them down and mix well with the rest of the band. And on the way it makes them fuller and more musical – it calms down the initial attack and brings up the tonal resonances. Leaving the drums untouched and uncompressed at all is unheard of in most studios – expect the bizarre niche ones that cater to the exotic audiophile listeners that need confirmation that their hi-fi system can reach ear-bleeding levels of SPLs at every snare hit.

 

[LTig]

I know exactly how drums sound in reality thank you very much, and it's unbearable.

Actually, why do we even need to mix things? Let's just play everything at their original SPLs. I'm sure an acoustic guitar will sound fine playing alongside a full drumset. And only opera singers would be allowed to sing, because no one else would be able to cut through the SPLs of the rest of the band. We should record rock albums in the audiophile way – two spaced microphones in the room and nothing more. This will give us the most realistic and authentic sound experience. I wonder why no one actually record albums this way.

Oh, right, I remember why. Because it will sound like shit. Modern music depends on modern recording techniques, and one main goal for these techniques is to not actually play the instruments at their real relative loudness, because that's insanity. No one really wants to hear the full volume of a drumset in their living room. Have you tried to actually put real drums in your house? Well I did. And the whole building heard it. In fact they still heard it when I put them in the basement. If you would reach actual playing levels of acoustic drums every time you turn on your hi-fi, you would lose your hearing very quickly. And obviously they would realistically be waaaay louder than any unamplified instrument.

You can go around and ask any professional sound engineer you want why they put compression on drums, and they would all tell you that they need it to tame them down and mix well with the rest of the band. And on the way it makes them fuller and more musical – it calms down the initial attack and brings up the tonal resonances. Leaving the drums untouched and uncompressed at all is unheard of in most studios – expect the bizarre niche ones that cater to the exotic audiophile listeners that need confirmation that their hi-fi system can reach ear-bleeding levels of SPLs at every snare hit.

I think you're overreacting here. You may have noted that I wrote "slightly compressed". I know very well that it is impossible to capture the full dynamic range of a drum or a symphonic orchestra for playback in a home system. Therefore compression is required, and a skillful mixing/mastering engineer can work wonders to make a recording sound fantastic on a home system.

However, as I interpret your postings, you prefer a totally crushed down sound. Please don't take this as a critizism of your personal preference - you can prefer whatever you prefer. But also please don't tell me that this crushed sound is an accurate representation of a real drum - it's far from that. There are quite a handful of very good mastered recordings where the drums sound much more real than in the crushed down example in posting #20.