DR Dynamic Range

[Doug G.]

REM's "Monster" was the first CD that made me aware of the "loudness wars". It is terrible.

Doug

 

[ar surround]

Is this what brickwalling looks like as a waveform? Visually, to me it looks like it is. Although it doesn't look too bad when I blow it up. Or is it?

The DR of this track is only 7 according to the Dynamic Range DB site. That's an interesting figure given that the first 40 seconds of the song definitely has dynamic range.

 

[HomerJAU]

That‘s it. You can see the DR7 right there in the waveform.

 

[ssully]

Brickwalling looks much worse than that. Only parts of that waveform look 'bricked'

Compare to, say, Oasis's What's the Story from 1995. *THAT* is brickwalling.

 

[ar surround]

But what I don't understand is why some of these "highly" compressed remastered CDs sound so much better than the original CDs.

Take for example, The Moody Blues song Blue World, which is on the album The Present. That song on the original 1983 CD has a DR of 12 vs. 7 on the 2008 remaster. Yet the 2008 remaster has remarkably more clarity and "perceived" dynamics than the 1983 disc. Also, upmixers such as Logic7 seem to be able to extract more out of the 2008 remaster. And I can crank it without any listening fatigue. The 2008 remaster sounds so much better that I gave away the original CD.

I find this a very confusing phenomenon.

 

[HomerJAU]

Compression makes it sound louder, with more low end. Listen to both for long periods and the 2008 won't be so enjoyable.

Turn up the 1983 and you'll here a bit more detail and hear the bass.

 

[ssully]

Compression makes it sound louder, with more low end. Listen to both for long periods and the 2008 won't be so enjoyable.

Or it might, because there may be improved EQ (apart from compression) or better tape source, compared to the earlier release.

 

[par4ken]

I have found that brickwalled CDs never sound as good as those that aren't. Mild amounts of compression might go unnoticed and might fool some into thinking that they sound better due to the higher volume level. To me those early releases still sound better (even with their source limitations) than the latter remastered ones, which are fatiguing to listen to.

I didn't realise the extent of the problem until I actually investigated by looking at the waveforms. Without exception all the really bad sounding CDs were brickwalled!

I am very leery of purchasing any newly remastered CD, and those Blu-spec discs from Japan are among the worst! It seems that releases from the UK are among the best. Labels like Prog Temple, Cherry Red etc., have been consistently good.

And don't get me started on HDTracks, you can no longer trust releases from them!

 

[4-earredwonder]

What can one 'reasonably' expect from an industry whose artists and sound engineers work in a VERY HIGH DECIBEL industry. Hearing loss can even occur at a young age and is hereditary as well.

Really wonder how many could pass a hearing test? But there are almost invisible state of the art hearing aids available.......

I do try to purchase RBCD reissues from companies like Rubellan, Cleopatra, etc. which seem to dispense with the loudness wars.....Agree, those Japanese Blu Spec CDs are suspect. Just purchased a James Taylor RHINO boxset pressed in the UK [6 discs] and they really do sound SUPERB.

 

[Sal1950]

Or it might, because there may be improved EQ (apart from compression) or better tape source, compared to the earlier release.

I fully agree.
I've often mentioned there's much more to SQ than a simple DR number can reveal. We all hate what's been done to many modern remasters and new releases alike. But beyond that there are tons of really excellent sounding recordings with really disturbing DR's.

I didn't realise the extent of the problem until I actually investigated by looking at the waveforms. Without exception all the really bad sounding CDs were brickwalled!

Possibly another really clear example of sighted bias?
You see the waveform distortion so hear even more SQ issues with later listens. ?

 

[ar surround]

Compression makes it sound louder, with more low end. Listen to both for long periods and the 2008 won't be so enjoyable.

Turn up the 1983 and you'll here a bit more detail and hear the bass.

This is where I think that much of the credence put into this DR index collapses. Please tell me if I'm full of it. Some of the site gurus @sjcorne et al please chime in.

As illustrated in post #85 above, the song Blue World on the 2008 UK remaster has what we consider a relatively poor DR of 7. The waveform looks close to being brickwalled. Yet when I demux the song, none of the stems look anything close to being brickwalled:

And as I noted earlier, the 2008 remaster of Blue World with a DR of 7 sounds miles above the 1983 CD with a DR of 12. And I've never experienced any listening fatigue with the 2008 remaster. I suspect that:

1) This DR meter in Foobar, or wherever it comes from, does not perceive what the human ear hears and should be taken with a grain of salt.

2) Demuxed waveforms tell a better story of what we are hearing than either the stereo waveform or the DR meter.

One thing that surprised me since I've been putting together my own 5.1's from stems it the perceived increase in dynamic range when converting from stereo to multichannel with stems. (Also, @Sonik Wiz and I had some discussions on how dynamic range increases when applying pre-synth.)

This is certainly a more interesting topic than I had previously thought.

 

[par4ken]

Possibly another really clear example of sighted bias?
You see the waveform distortion so hear even more SQ issues with later listens. ?

What bias? The CDs in question sounded terrible, period. I checked the waveforms afterward (not before) and without exception they were brickwalled! This was all done before even knowing about DR value, nothing but the bad sound affected my judgement!

CD's that sound good are either not brickwalled or very lightly brickwalled. The DR value seems to correspond closely to what I see with the waveforms and so IMHO the DR rating is a useful tool.

 

[Sal1950]

CD's that sound good are either not brickwalled or very lightly brickwalled. The DR value seems to correspond closely to what I see with the waveforms and so IMHO the DR rating is a useful tool.

Maybe your more sensitive to DR issues?
In any case your so locked into this position there's no room for debate.
But I will stand my ground when I say there's more to SQ than a simple DR number reveals which shows only the margin between soft and loud. It tells you nothing about inner detail resolution, FR, imaging ablity, etc etc etc.

 

[HomerJAU]

I side with @par4ken

Maybe I’m more sensitive about DR. I agree there‘s more to good sounding music than DR. But DR 7 sounds pretty crappy to me compared to DR15’s in those early CDs!

I think just about everyone with a half decent system would choose 15 over 7 in a blind volume matched test.

 

[Sal1950]

But DR 7 sounds pretty crappy to me compared to DR15’s in those early CDs!

Sure.
But there are just as many DR15's that sound like crap.

 

[HomerJAU]

Sure.
But there are just as many DR15's that sound like crap.

But low DR will ruin a great album. Example: Flaming Lips - Yoshimi Battles the Robots (DR7)

 

[sjcorne]

But low DR will ruin a great album. Example: Flaming Lips - Yoshimi Battles the Robots (DR7)

It's definitely loud, but Yoshimi in both stereo & 5.1 isn't so awful-sounding to my ears. I find the The Soft Bulletin nearly unlistenable though.

 

[Sal1950]

But low DR will ruin a great album.

Ruin?
So nothing else matters, not even the performance?
Kind of extreme, no?
I'm not familiar with Flaming Lips but I had no problem listening to DR5 Tears For Fears - Tipping Point when it came out, I would in no way call it "unlilstenable". Was I happier when SW released his remasters, sure.
Just like over at ASR, a low or high measured SINAD isn't the last word in measured or listened component performance.
JMHO.

 

[AYanguas]

That‘s it. You can see the DR7 right there in the waveform.

The waveform of compressed/brickwalled audio looks like when a hairdresser trims the ends of the bangs

 

[ar surround]

What bias? The CDs in question sounded terrible, period. I checked the waveforms afterward (not before) and without exception they were brickwalled! This was all done before even knowing about DR value, nothing but the bad sound affected my judgement!

CD's that sound good are either not brickwalled or very lightly brickwalled. The DR value seems to correspond closely to what I see with the waveforms and so IMHO the DR rating is a useful tool.

But low DR will ruin a great album. Example: Flaming Lips - Yoshimi Battles the Robots (DR7)

It's definitely loud, but Yoshimi in both stereo & 5.1 isn't so awful-sounding to my ears. I find the The Soft Bulletin nearly unlistenable though.

So I'm confused. How can a track have a DR of 6 or 7 when the demuxed stems are all DR> 10? Take Blue World again as an example: