Jump to content
Sign in to follow this  
mitchco

Article: Dynamic Range: No Quiet = No Loud

Recommended Posts

Thanks for an eye-opening (and 'ear-opening') article. I am a listener, not an engineer, so much of this flew over my head, BUT, when it got to the 'Green Numbers are Good' and I saw a few recordings listed from my collection. I gave 'Back in Black' a listen, the 'Brothers in Arms'. And although you lost me in technicalities I heard what you are talking about.  I will look up the Green numbers and know they are Dynamic, not Loud...

Share this post


Link to post
Share on other sites

After reading your article on DR analysis I wanted to see what my collection looked like.  I set up a field in JRiver to show DR results and went about testing about 4200 CDs.  What I did different was to break down the difference between Redbook, Hi-Res PCM and Hi-Res DSD.  I have also analyzed just the releases I have from Blue Coast Records.  Here are my results:

 

RedBook  1982 CDs

 
1-7  227   (11.45%)
8-10  780   (39.35%)
 
Thus 0-10 (BAD)  1007  (50.76%)
 
11-13   782   (39.46%)
14-18   194   (9.79%)
 
Thus 11-18 (Good)  976  (49.24%)
 
 
 
 
Hi-Res  2200 CDs
 
1-7  86   (3.91%)
8-10  592  (26.91%)
 
Thus 0-10 (BAD)  678  (30.82%)
 
11-13   1142   (51.915%)
14-18   382   (17.36%)
 
Thus 11-18 (Good)  1524  (69.27%)
 
__________________________________________________________________________________________
 
 
Hi-Res  PCM  1313 CDs
 
1-7  79   (6.02%)
8-10  407  (31%)
 
Thus 0-10 (BAD)  486 (37.01%)
 
11-13   668   (50.88%)
14-18   160   (12.19%)
 
Thus 11-18 (Good)  828  (63.06%)
 
 
 
Hi-Res  DSD  887 CDs
 
1-7  7   (0.79%)
8-10  185  (20.86%)
 
Thus 0-10 (BAD)  192  (21.65%)
 
11-13  474   (53.44%)
14-18   222   (25.03%)
 
Thus 11-18 (Good)  696  (78.47%)
 
 
__________________________________________________________________________________________
 
Looking at Just DSD releases from Blue Coast Records:
 
Hi-Res  Blue Coast DSD  68 CDs
 
1-7  0   (0%)
8-10  0  (0%)
 
Thus 0-10 (BAD)  0  (0%)
 
11-13  26   (38.24%)
14-18   42   (61.76%)
 
Thus 11-18 (Good)  68  (100%)

Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites
36 minutes ago, wwaldmanfan said:

Your effort is admirable.

What is a "hi-res CD"?

Redbook is 16/44.  Hi-Res is anything that is 24 Bit, i.e. 24/96 or 24/192.


Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites

@freddie40 Very interesting. How many of those are non-classical? I did something similar in JRiver. Made a playlist with everything I have that is classified as rock, blues, alt, etc., but contains no classical or jazz type recordings. Sorted by DR with a total of 16,564 titles. DR11 or less = 13,106. About 20% of my collection is DR12 or greater.

Share this post


Link to post
Share on other sites
1 hour ago, mitchco said:

@freddie40 Very interesting. How many of those are non-classical? I did something similar in JRiver. Made a playlist with everything I have that is classified as rock, blues, alt, etc., but contains no classical or jazz type recordings. Sorted by DR with a total of 16,564 titles. DR11 or less = 13,106. About 20% of my collection is DR12 or greater.

 

I have no classical or Country.  They are all Rock, Jazz or Blues.

 

BTW:  The DR values of the Albums in my collection are an average of the DR of the songs of each album.  Not necessarily easy to do in JRiver.

 

Dave


Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites
2 hours ago, mitchco said:

@freddie40 Very interesting. How many of those are non-classical? I did something similar in JRiver. Made a playlist with everything I have that is classified as rock, blues, alt, etc., but contains no classical or jazz type recordings. Sorted by DR with a total of 16,564 titles. DR11 or less = 13,106. About 20% of my collection is DR12 or greater.

 

If you looked at 1-10 and 11-20 I bet you results would be much closer to mine.

 

Dave


Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites

I've been on a re-buying tear this past year replacing my bad/poor DR music with good DR. I check out what I own w/the DR meter and if its bad go to the DR site and find a good issue, comb Discogs, eBay, SH forum and buy it. I think I've bought about 100 CD's this year not to mention SACD's. The difference in sound is significant and I'm quite happy I've reviewed my older music. All the new stuff I immediately check DR.


Furutech GTX-D>J River 26>Curious USB>Chord Hugo TT2>Meze Emyrean Copper Cable>Meze Empyrean Headphones>my ears>audiophile brain

Share this post


Link to post
Share on other sites
1 hour ago, mitchco said:

Hi Dave, I sorted on DR10 or less = 11,390 titles, which is a about 70% of my collection. Cheers, Mitch

 

Part of the difference is the type of music that we own.  Some music is recorded much better than others.  It just depends.


Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites
34 minutes ago, KDinsmore said:

I've been on a re-buying tear this past year replacing my bad/poor DR music with good DR. I check out what I own w/the DR meter and if its bad go to the DR site and find a good issue, comb Discogs, eBay, SH forum and buy it. I think I've bought about 100 CD's this year not to mention SACD's. The difference in sound is significant and I'm quite happy I've reviewed my older music. All the new stuff I immediately check DR.

 

 

:)

 


Crystal Clear Music Tweaked Mac Mini / Yosemite -> JRiver 22 -> Ayre QB9DSD -> Bryston BP26DA -> Bryston 4BSST2 -> B&W 802Di | Transparent Reference XLRs, Transparent Super Speaker Cable, Maple Shade USB cable

Share this post


Link to post
Share on other sites
1 hour ago, KDinsmore said:

I've been on a re-buying tear this past year replacing my bad/poor DR music with good DR. I check out what I own w/the DR meter and if its bad go to the DR site and find a good issue, comb Discogs, eBay, SH forum and buy it. I think I've bought about 100 CD's this year not to mention SACD's. The difference in sound is significant and I'm quite happy I've reviewed my older music. All the new stuff I immediately check DR.

That's great, nice to have the stuff from before the label Godzilla's were eating all the DR.  :)

Since I became aware of the dynamic range site some years back I always check it before buying anything to get the best I can. Don't forget to contribute back and put up as much info as you can on stuff in you collection that hasn't been already listed.. We all benifit.

Cheers


"The gullibility of audiophiles is what astonishes me the most, even after all these years. How is it possible, how did it ever happen, that they trust fairy-tale purveyors and mystic gurus more than reliable sources of scientific information?"

Peter Aczel - The Audio Critic

no-mqa-sm.jpg

Share this post


Link to post
Share on other sites

Very curious after reading the counts above, since I'm a chronic arithmatic... Just looked at counts in my library. There's rounding up, and oddly very little rounding down... 

 

DR 1-7          13%

DR 8-10        29%

DR 11-13     41%

DR 14-18     17%

DR 19+         .5%

 

Or - DR 13-Up    30%

 

FWIW, I used the DR rating assigned by JRiver, since when I've checked specific albums, it has matched what I've seen in the DR Database.

 

I was surprised to see what was in the rarified DR of 19+, and found I had a number of cuts that were higher than 20, some nearing 30.  Almost all of them were percussion ensemble stuff, although I've got some studio masters (from friends' recording sessions) that are up there, because they'd not yet been finally engineered (and DR compressed) to send off to be burned to CD.  One album was all 20 and over, and it was a percussion ensemble performing just with hand claps and wood blocks and sticks. I can tell you that when I listen to some of those albums, I find I'm continually fiddling the volume, because practically there is no volume setting that lets you hear the whole thing - you'll be happily listening and then suddenly someone slugs every percussion instrument they can get their hands on, or suddenly it's so soft you're not sure if the cut is over.  I guess there is such a thing as too much DR for "normal listening."

 

A few gross generalizations.  Sleater-Kinney, Alabama Shakes, and a number of other very different types of music, like Pentatonix and Ben Folds 5, are all in the bottom DR sector.  Almost all the jazz is between 8 and 15.  Almost all the classical is 10 and up.  In categories like bluegrass, world music, or alternative, it's all over the place.

 

I'm going to submit a few of the crazy outliers to the database.  They didn't have any of the percussion ensemble stuff.

Share this post


Link to post
Share on other sites

@ednaz As mentioned in the article, mastering engineer, Bob Katz has an excellent article on an integrated approach to metering, monitoring and leveling practices. This helps answer both at what listening level one wants to be listening to, plus the addressing dynamic range issue. Highly recommended for you and your recording friend.

Share this post


Link to post
Share on other sites

First and foremost, thanks for another great article and insight Mitch!

 

I'm wondering if you could do a follow up article on the differences between crest DR which you (and DR database uses) and R128 DR. I found different opinions everywhere, at least on JRiver's Wiki, they basically state they bother measure differently and need to be looked at in different scenarios with neither being the final say on DR. But on the ROON forums its a very different story where everything about crest DR is not relevant and R128 (some say combined with waveform analysis) is the only way to look at it. Where there are 1000 examples of where I think this is wrong I tried to look in my collection of different versions of the same CD where this is best illustrated and I finally found one, Def Leppard Hysteria. I have both the 1987 and 2006 remastered versions and the nice thing with this example is that R128 shows negligible difference but DR is almost double on the original compared to the remastered version and I don't think I have to tell you which sounds better. And because the peak levels are close to the same it's easy to test without adjusting the volume, granted I did want to turn down the volume on the 2006 version. I won't explain the differences from the technical side, I'll leave that to you.  

 

Andy

P.S. and I just noticed in the pic that I didn't label the albums properly, the top is the 1987 version and bottom being the remastered (with bonus songs cutoff). 

Hysteria.jpg

Share this post


Link to post
Share on other sites

Hi Andy, thanks for your comment. For the article, I arbitrarily chose crest factor DR to get the point across about overly compressed sound, with a handy database to compare. Interesting example you have there :) Let me give it some thought for a potential follow-up article. Cheers!  Mitch

Share this post


Link to post
Share on other sites

Mitch: Nice Article! Like yourself I'm an old school engineer with over 32 years in the Biz.

Also like yourself I have been fortunate to be brought up on all the same kinds of gear as yourself and have experienced the technology arc and the early adoption of Digital Audio in my studio career.

 

Although, there is nothing quite like the sound of  a Studer going through a classic Neve Console, most of the engineers of today will really never get to experience what it really sounds like. With some modern plugins and dsp I figure they can get at least 85% of the way to what we could experience back in the day.

 

The so called loudness war has actually been around for quite a long time in the pop industry dating back to the sixties where Radio Programmers, Producers and Engineers tried to get their singles to sound as hot as possible on AM and FM radio but as you know and pointed out so eloquently when digital came along it became a new ball game as far as dynamic range and what could be captured.

 

I am in complete agreement and can do nothing but echo your sentiment. Unfortunately  a lot engineers and enthusiasts will never have the chance to be brought up in the studio system as you and I were, but what they do have is this wealth of information and powerful tools at their fingertips.

 

I sometimes wish that modern DAW's didn't have so much visual information that the user can access.

all we had back when we started was the Analog VU Meter and the occasional Peak Meter and our ears.

I encourage people working in production today to go back and listen to older CD releases and listen to older records and really study the Dynamics and study the EQ and Sonical qualities of the music.

and to stop paying so much attention to the visual information and just listen.

Share this post


Link to post
Share on other sites

Maybe I'm missing something.

 

So, if I check the DR database or use the TTD Range Meter on Creedence Clearwater Revival - The Best of Creedence Clearwater Revival (the one with the green cover), I'm getting an average DR of 14. However, something like Billie Joe + Norah - Foreverly scores a DR of 8 with a lot of clipping...or Robert Plant / Alison Krauss - Raising Sand also only scores a DR of 8, again with a bit of clipping. Now, my music is driven by Audirvana into Dynaudio XEO 5 speakers...at best, introductory audiophile grade and perhaps even that's being a little generous. But, to my ears the CCR sounds kinda flat and lifeless whereas Foreverly or Raising Sand has some life, depth and stage presence. I'm not trying to discount this article...far from it and I have many examples where reworked pieces should have probably been left alone if it's not going to be done properly. My point being, I'm not sure I'm convinced that I can look at this database as a be all to end all, see green numbers and be assured that the files are going to sound great. Conversely, perhaps lower scoring numbers won't necessarily sound like crap. As a disclaimer, the XEOs aren't the type of speaker that will get your neighbour across the street come marching over with clenched fists...these were meant for low SPL levels so perhaps I'm not getting the chance to suffer from listening fatigue. Or, perhaps years of rock concerts have taken their toll on my ears...or perhaps I'm simply missing something.

Share this post


Link to post
Share on other sites

Hello @cdnguy, thanks for your comment. You may have missed this sentence in my article, "I am not saying that dynamic range is the end all beat all attribute that determines the overall enjoyment of a musical piece, but there is a minimum bar that should be met on any musical piece."

 

Of course there are other factors. In the case of the CCR double disc, which I own and have played to death, the mix and master is indeed "flat" or in record production terms, "dry".  I had asked a remastering engineer from HDTracks, who got the 1/2" CCR analog tapes, what he was planning on doing to give the mix more depth and there was not much that can be done as the recording is generally bright sounding, with little depth (i.e. delay and reverb in production terms), and already printed to tape. So frequency response shaping and effects also play a role in how a mix/master sounds.

 

Dynamic range compression is "candy" to our ears. A little bit generally tastes good, too much and ... is my point.

Share this post


Link to post
Share on other sites
Quote
8 hours ago, mitchco said:

"I am not saying that dynamic range is the end all beat all attribute that determines the overall enjoyment of a musical piece, but there is a minimum bar that should be met on any musical piece."

 

Ah yes, totally missed that...sincerest apologies. I suppose it's not just my ears failing me.

Quote
8 hours ago, mitchco said:

there was not much that can be done as the recording is generally bright sounding, with little depth

 

Too bad...I really like their music.

Share this post


Link to post
Share on other sites

Coming to this read very late. I have been accused being a walking music lexicon, having started with The Beat Club on B&W German TV in the 60ties. From all the countless music recordings I have listened to in every imaginable format since then, there are still only a handful good recordings that stand out. The articles mentions a few. But compression is just one of many problems that make music less enjoyable, or in some cases unlistenable despite brilliant music artistry. I hate to say it, and it is really not my preferred music style at all, Jazz in the Pawnshop is a prime example where the sound is "right", DR included (I have no idea what the actual rating is). The reason for the good sound is the simplicity of the recording method, a few Neuman U47 in perfect "stereo" alignment, plus KM56's for the piano, Studer mixer, NR unit and two Nagra tape recorders. 

 

Multi-track recording of rock music is obviously a different beast, but I still believe that simplicity should be the first order to generate a truely acurate and engaging image of a recording session. It starts at the source, where the music is recorded, live on stage or in the studio. Everything downstream just adds a level of alteration of the orginal sound. The MQA folks have a point wanting to retain control over the entire downstream until it hits our ears. 

 

Interestingly, I find examples of superior DR in some of today's EDM productions, completely electronically generated with sophisticated software that allows the artist to maintain complete control over the music recording and mastering process. Just because of this, I like to listen to pounding, high DR musical EDM content on streaming Internet radio stations, in 24bit/96 FLAC format. Straight from the horse's mouth, unaltered and enjoyable. 

 

Cheers, 

 

 


Digital: Gustard X20u, Burson Conductor V2+, Chord Mojo, Dragonfly Blk/Red, TDA1387, Allo Piano 2.1/KALI, DigiOne

Share this post


Link to post
Share on other sites

I am guessing what I have to say has some connection to the article as I am not too in-tuned with mastering and mixing techniques but no doubt that I have lost the full ability to ever listen to studio recording albums ever again. The overall quality you really can't budge, the added reverb tail is also not too bad but something is missing and I am guessing its the dynamics.

 

I agree my room is more like a studio control room which doesn't help with recreating a the perfect musical space but adding some lateral reflections and increasing the overall reverbrance isn't going to magically make the studio track sound spectacular. Contrary to this, I do listen to a lot of late 90' electronic music which I feel actually sounds good in my room and although I can still ascertain differences in the quality of the mixing especially with added reverb it seems once I switch back to say Jazz, Rockn'Roll or any other music that can be played live with instruments and a vocalist just doesn't cut it with a studio recording.

 

Once I switch to live recordings, the envelopment and spaciousness is something different, of course you can still hear a difference in the recording, mixing and mastering but mostly you get a dynamic feel with natural reverberance. Now, I am guessing this phenomenon is the lack of dynamics which Mitcho is talking about and it is really disappointing cause I am also guessing that studio recordings have the potential to be mixed with the studio recording feel but with the right dynamics not to make it sound to dull.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  



×
×
  • Create New...