DR Dynamic Range

[par4ken]

I'm inspired to start this thread due to the influence of many other posts involving sound quality. We can debate the merits of tubes vs solid state, DSD vs PCM, analogue vs digital etc. What a lot of it comes down to in the end is Dynamic Range. Many, many people have been turning to vinyl over the years. Many of them are too young to be doing so for nostalgic reasons. They do it because most of todays music (CD and downloads) are mastered like CRAP!!!

When CD's were first introduced they promised perfect sound forever. They never quite lived up to that ideal though, the first players sounded electronic all the bass was there full and solid, but the highs were unnatural. CD players evolved with better DAC's and analogue circuitry oversampling etc. As CD players improved CD mastering deteriorated.

I first noticed that something was amiss when I tried to make a CD from CD tracks from a "best of" combined with vinyl album tracks (added to recreate the original album). No matter what I did the CD tracks were always louder. I thought that tick and pops in the vinyl were making the normalized vinyl level lower than the CD and tried squashing the peaks with the vinyl tick and pop filter software, I was using Wave Corrector at the time. The vinyl was always lower. Eventually I dropped that approach and just ripped the entire album from vinyl.

Latter I started noticing that the vinyl in many cases trounced the sound quality of the CD. Every time I opened a bad sounding CD in Audition or Audacity the result was the same, brick walled and even clipped audio in some cases. Instead of a nice rounded sound envelope all you got were flat maxed out waveforms from start to finish!

I noticed that even some HD tracks are affected as well. Every one of my vinyl rips tested so far show a DR value of 13 to 14. I was horrified to see that the DR rating of the Rolling Stones "Sticky Fingers" from HD tracks was only 6!!! On another site someone posted about it, he complained to HD tracks. HD Tracks reluctantly refunded his money but stated that it was a one time thing as there was nothing wrong with the download!
Most HD Tracks that I've tested so far range from 10 to 12 DR value. IMHO high rez downloads should at least equal the DR of vinyl but really should exceed that figure. CD's came in with the promise of perfect sound and greater dynamic range than was possible with vinyl but with the exception of a handful of audiophile labels CD's have never met that promise. From posters here I understand that classical music has largely escaped this loudness madness and that's a great thing.

I never gave much thought to the DR rating until recently. I went to add the DR meter to Foobar I found a notation about it "Warning: This component is known to be problematic, due to repeated crash reports, and should be installed with caution." in addition the link didn't work. After some searching I found a working link. Well the meter works great on all PCM files, that I've tested. When I try it on .dsf or SACD.iso it crashes.
I tried to convert an iso to flac the way I used to do with Foobar and the program also crashes. I'm not sure why the DR meter would effect SACD conversion. Hopefully the bugs can be worked out. I intend to check all my CD's and downloads so that I can find replacements for those with low DR values.

 

[J. PUPSTER]

I'm inspired to start this thread due to the influence of many other posts involving sound quality. We can debate the merits of tubes vs solid state, DSD vs PCM, analogue vs digital etc. What a lot of it comes down to in the end is Dynamic Range. Many, many people have been turning to vinyl over the years. Many of them are too young to be doing so for nostalgic reasons. They do it because most of todays music (CD and downloads) are mastered like CRAP!!!

When CD's were first introduced they promised perfect sound forever. They never quite lived up to that ideal though, the first players sounded electronic all the bass was there full and solid, but the highs were unnatural. CD players evolved with better DAC's and analogue circuitry oversampling etc. As CD players improved CD mastering deteriorated.

I first noticed that something was amiss when I tried to make a CD from CD tracks from a "best of" combined with vinyl album tracks (added to recreate the original album). No matter what I did the CD tracks were always louder. I thought that tick and pops in the vinyl were making the normalized vinyl level lower than the CD and tried squashing the peaks with the vinyl tick and pop filter software, I was using Wave Corrector at the time. The vinyl was always lower. Eventually I dropped that approach and just ripped the entire album from vinyl.

Latter I started noticing that the vinyl in many cases trounced the sound quality of the CD. Every time I opened a bad sounding CD in Audition or Audacity the result was the same, brick walled and even clipped audio in some cases. Instead of a nice rounded sound envelope all you got were flat maxed out waveforms from start to finish!

I noticed that even some HD tracks are affected as well. Every one of my vinyl rips tested so far show a DR value of 13 to 14. I was horrified to see that the DR rating of the Rolling Stones "Sticky Fingers" from HD tracks was only 6!!! On another site someone posted about it, he complained to HD tracks. HD Tracks reluctantly refunded his money but stated that it was a one time thing as there was nothing wrong with the download!
Most HD Tracks that I've tested so far range from 10 to 12 DR value. IMHO high rez downloads should at least equal the DR of vinyl but really should exceed that figure. CD's came in with the promise of perfect sound and greater dynamic range than was possible with vinyl but with the exception of a handful of audiophile labels CD's have never met that promise. From posters here I understand that classical music has largely escaped this loudness madness and that's a great thing.

I never gave much thought to the DR rating until recently. I went to add the DR meter to Foobar I found a notation about it "Warning: This component is known to be problematic, due to repeated crash reports, and should be installed with caution." in addition the link didn't work. After some searching I found a working link. Well the meter works great on all PCM files, that I've tested. When I try it on .dsf or SACD.iso it crashes.
I tried to convert an iso to flac the way I used to do with Foobar and the program also crashes. I'm not sure why the DR meter would effect SACD conversion. Hopefully the bugs can be worked out. I intend to check all my CD's and downloads so that I can find replacements for those with low DR values.

For sure a lot of later bricked CD out there, which is why I've been actively collecting the earlier CDs from the 80's or MFSL, as most of those are much better.
For those that don't already know, here's a good resource for DR numbers:
http://dr.loudness-war.info/
I'm also getting back into my vinyl collection after decades of on "a hiatus" status.

And since I'm also getting into converting Quad albums now into flacs/wav etc. it's good to remember that if you add silent Center and LFE channels for some players to accept, that it alters the DR numbers to a lower number. I'm doing one now that measured with just the four channels it's a DR 14 on some songs; but with the added silent channels drops it down to DR 9 which does not reflect the true number.

I also don't believe I've had any DR meter crashes on my Foobar2000 dsf files (but it may be converting to PCM first I'll need to check my settings )

 

[sjcorne]

And since I'm also getting into converting Quad albums now into flacs/wav etc. it's good to remember that if you add silent Center and LFE channels for some players to accept, that it alters the DR numbers to a lower number. I'm doing one now that measured with just the four channels it's a DR 14 on some songs; but with the added silent channels drops it down to DR 9 which does not reflect the true number.

I find that the DR meter is generally misleading for multichannel audio--all it does it take the DR value of each individual channel, add them together, then divide by the # of channels to generate an average. As you say, this is obviously an issue for quad albums with an empty center/sub as the average reading is being skewed by those two extra channels with a DR value of 0.

It's also problematic for surround mixes where the front channels are less dynamic than the rears, or vice versa. Take for example, "Born Under Punches" from the Talking Heads' Remain In Light--it shows as DR14 overall, but the DR reading of each channel breaks down as follows:

FL: 10.9
FR: 10.3
C: 18.4
LFE: 14.5
RL: 14.7
RR: 14.9

It's also pretty clear from a visual perspective that the fronts have less breathing room than the rears. I don't think that's necessarily a bad thing in this case, as the front channels carry the rhythm section and that compression gives it some more 'punch'. My overall point is that when it comes to multichannel, I wouldn't get too caught up in what the DR meter says.

 

[Soundfield]

Isn't the measurement of the dynamic range of a recording entirely meaningless unless you know the DR of the original performance?

 

[Plan9]

Isn't the measurement of the dynamic range of a recording entirely meaningless unless you know the DR of the original performance?

It's good for comparing different masterings of the same mix. It isn't a measure of sound quality in absolute terms.

Also, vinyl playback adds anywhere from 2 to 5 DR points depending on the rip -so it's not very useful to compare it to digital masterings.

 

[HomerJAU]

And since I'm also getting into converting Quad albums now into flacs/wav etc. it's good to remember that if you add silent Center and LFE channels for some players to accept, that it alters the DR numbers to a lower number. I'm doing one now that measured with just the four channels it's a DR 14 on some songs; but with the added silent channels drops it down to DR 9 which does not reflect the true number.

This is only albums with DSF I believe as the conversion does not make an empty channel.

You can get the correct SACD rip DR by by removing the silent channel then add a new one and the DR value is correct. (Music Media Helper has a tool to fix the silent channels).

For FLAC the silent channel(s) is empty and is ignored in the DR calculation.

A few of us fixed most the DR values for SACDs listed in the QQ Hires DR report recently

https://www.quadraphonicquad.com/files/QQ Hires Poll with DR.pdf

 

[Plan9]

I posted more of my thoughts about it here: Black Sabbath - Vol. 4 Super Deluxe Edition (with some Steven Wilson mixes) Feb 12th, 2021

 

[J. PUPSTER]

This is only albums with DSF I believe as the conversion does not make an empty channel.

You can get the correct SACD rip DR by by removing the silent channel then add a new one and the DR value is correct. (Music Media Helper has a tool to fix the silent channels).

For FLAC the silent channel(s) is empty and is ignored in the DR calculation.

A few of us fixed the DR values for SACDs listed in the QQ Hires DR report recently

I was addressing two different subjects there Garry; and the DR meter must handle wav that way then, from what I'm experiencing if not flac?

 

[J. PUPSTER]

I posted more of my thoughts about it here: Black Sabbath - Vol. 4 Super Deluxe Edition (with some Steven Wilson mixes) Feb 12th, 2021

Yup, that's what I was thinking about the Bass issue also from your post.

 

[Soundfield]

It's good for comparing different masterings of the same mix. It isn't a measure of sound quality in absolute terms.

So if a particular 'mix' has twice the DR it's twice as good as another 'mix', even if neither remotely reflects the DR of the original performance?

 

[HomerJAU]

More info on SACD DR issues in this thread and how to solve:

https://www.quadraphonicquad.com/forums/threads/dynamic-range-dr-for-hires-releases.27682/

 

[Plan9]

So if a particular 'mix' has twice the DR it's twice as good as another 'mix', even if neither remotely reflects the DR of the original performance?

Recording -> mix -> mastering

The players performing the music are recorded on separate tracks. These recordings are then mixed together. The result is the mix (in mono, stereo, quad, 5.1, Atmos, etc...).
The "original performance" doesn't exist until it has been mixed, otherwise it is just a bunch of disparate tracks, or instruments on their own.
The mixes are then mastered, and it's what you usually hear on vinyl, CD, streaming Blu-Ray, etc... and what is analyzed by the DR meter.

Something with twice the DR points as something else doesn't mean it's twice as good, it just means that it has twice the dynamic range... as measured by the DR meter. There are also other ways of measuring dynamic range and loudness.

 

[Soundfield]

The "original performance" doesn't exist until it has been mixed

Oh, you're talking about the artifice of electronic pop music, sorry. Any live acoustical performance would have a measurable, DR at the time. A Hi Fi recording of such would, by definition, attempt to accurately replicate that DR, not enhance it for dramatic effect.

 

[par4ken]

It's good for comparing different masterings of the same mix. It isn't a measure of sound quality in absolute terms.

Also, vinyl playback adds anywhere from 2 to 5 DR points depending on the rip -so it's not very useful to compare it to digital masterings.

I think that it is a useful tool in helping to predetermine sound quality as long as the vinyl is a clean copy and the ticks and pops have been removed. DR ratings are not absolute but do seem to give a very good indication of what mixes have been overly compressed. By my testing I don't think that most vinyl is overly compressed but recent releases from digital masters could be called into question.

 

[par4ken]

So if a particular 'mix' has twice the DR it's twice as good as another 'mix', even if neither remotely reflects the DR of the original performance?

Most likely it will sound much better but of course DR alone is not the only factor.

 

[par4ken]

I find that the DR meter is generally misleading for multichannel audio--all it does it take the DR value of each individual channel, add them together, then divide by the # of channels to generate an average. As you say, this is obviously an issue for quad albums with an empty center/sub as the average reading is being skewed by those two extra channels with a DR value of 0.

It's also problematic for surround mixes where the front channels are less dynamic than the rears, or vice versa. Take for example, "Born Under Punches" from the Talking Heads' Remain In Light--it shows as DR14 overall, but the DR reading of each channel breaks down as follows:

FL: 10.9
FR: 10.3
C: 18.4
LFE: 14.5
RL: 14.7
RR: 14.9

View attachment 60508

It's also pretty clear from a visual perspective that the fronts have less breathing room than the rears. I don't think that's necessarily a bad thing in this case, as the front channels carry the rhythm section and that compression gives it some more 'punch'. My overall point is that when it comes to multichannel, I wouldn't get too caught up in what the DR meter says.

I think the best use of the DR meter is to compare different versions of the same stereo recording. Great for weeding out those overly compressed CD's. Again it's not an exact measure of Sound Quality, just another tool.

 

[HomerJAU]

By my testing I don't think that most vinyl is overly compressed but recent releases from digital masters could be called into question.

Vinyl has typically always had good DR. Modern digital recordings, especially in the later 90s and 2000s are just about always compressed badly. This is one of the reasons vinyl has had its resurgence.

 

[par4ken]

Oh, you're talking about the artifice of electronic pop music, sorry. Any live acoustical performance would have a measurable, DR at the time. A Hi Fi recording of such would, by definition, attempt to accurately replicate that DR, not enhance it for dramatic effect.

Nobody said anything about enhancing DR just preserving it.

 

[par4ken]

And since I'm also getting into converting Quad albums now into flacs/wav etc. it's good to remember that if you add silent Center and LFE channels for some players to accept, that it alters the DR numbers to a lower number. I'm doing one now that measured with just the four channels it's a DR 14 on some songs; but with the added silent channels drops it down to DR 9 which does not reflect the true number.

You would think that adding silent channels would not effect the DR value, or would increase it (as average RMS level would be lower). Odd.

 

[J. PUPSTER]

You would think that adding silent channels would not effect the DR value, or would increase it (as average RMS level would be lower). Odd.

Read @steelydave's thread, that's where I learned about it...

https://www.quadraphonicquad.com/fo...nge-readings-and-how-to-interpret-them.23281/