Dynamic range of 'old' CDs

[MeloManiac]

Tonight I put in a CD I've had for many years, going back to 1992, a recording of Beethoven's 6th and 8th (see: https://www.discogs.com/Beethoven-HarnoncourtChamber-Orchestra-Of-Europe-Symphony-No-6-Pastoral-Symphony-No-8/release/10515973)

This CD starts rather quietly, so I always have the inclination to turn up the volume, but then soon the full power of the (humble) HK430 with my Heresy 1972 forces me to turn it down again. Even then, my brain seems out of balance because of the large differences in volume (what I call dynamic range), I guess that's because I listen a lot to Spotify, which doesn't have this 'problem'.

Of course, it is actually the other way round: the dynamic range of the old days is correct, while Spotify (and I think their competitors also), have raised the volume of the quieter sequences. 

CDs can have extended dynamic range, while recordings on vinyl are limited physically (too loud can make the needle jump out of  groove).

 

Any thoughts on this, or corrections, are welcomed!

[tigerwoodKhorns]

First CD that I ever bought was Crime of the Century.  It has a note stating that the new CD technology will show the limit of the analog master tapes, so don't complain about the quality. 

 

One of the best sounding CDs that I have and it actually has dynamic range. 

[codewritinfool]

What you are describing is audio compression.

Google “loudness wars” to learn how this ruined many releases over the years, especially in the early 90’s.

I have some CD’s of the same album and same artist from different years and they are all dynamically different.

[babadono]

No corrections are necessary, I don't think. Dynamics are routinely squashed in recordings, see "loudness wars". The CD medium is capable at 16 bit quantization of 96dB dynamic range but this is rarely if ever utilized. Music is supposed to be dynamic, the more dynamic it is the more live it seems. Going to live performances, although hard nowadays with COVID, is a good way to keep "in tune" with what music is supposed to sound like.

[Chris A]

A thread I started a few years ago where I did some statistics on dynamic range vs. year of release...

 

One of the graphs I presented in that thread showing the dynamic range (crest factor) for phonograph records ("DR Ave. Vinyl") vs. CDs ("DR-Digital Ave") by five-year increments:

 

 

Note that phonograph records can't support compressed dynamics like CDs (i.e., the format can't be abused as much as digital format) because the needle will literally jump out of the groove because of the limitations of the format itself.

 

Another graph showing how there is bimodal statistical characteristic for dynamic range into two apparent frequency populations--"loud" and "soft".  This implies that there are some genres that show excessive dynamic range compression, while other genres are not showing that compression nearly as much (the data as of 2013).  I bet you can guess which genres they are...

 

 

The above investigation started me down the path of demastering tracks...first to recover deep bass attenuation in records and CDs, and then I found the other issues almost simultaneously, notably boosted highs from 1-5 kHz:

 

 

Chris

[MeloManiac]

21 minutes ago, Chris A said:

Note that phonograph records can't support compressed dynamics like CDs (i.e., the format can't be abused as much as digital format) because the needle will literally jump out of the groove because of the limitations of the format itself.

 

Your graphs seem to suggest that phonographs, vinyl records, have a much wider dynamic than digital carriers. 

I now remember this older thread o yours, BTW. Is that correct? 

 

[Chris A]

What you've missed is that the reason for the compression of dynamics is to push the average loudness to be louder.  Phonograph records can't get any louder than they already are because of the needle tracking thing prevents further compression/pushing the tracks louder.  In fact, phonograph records almost can't handle the full audible frequency range, and require 40 dB of equalization/de-equalization (i.e., the RIAA curve) to de-emphasize the bass and to boost the treble on the vinyl itself to increase the average loudness level in the grooves, then the reverse EQ curve is applied in the phono preamp during playback:

 

 

Phonograph record mastering is usually controlled by separate guy (other than the mastering guys) whose job it is to alter the downmix tracks they receive from either the mixing guys or the mastering guy--before the mastering guy cranks down on the dynamic range even further to make the tracks as loud as possible (even though no consumer that I know asked them to make the tracks louder--since the AGCs in their portable disc players and MP3 players bring the average loudness down by default nowadays and no one typically listens to the tracks as loud as the mastering guys are making them). 

 

The phonograph mastering guys alter the intermediate tracks they receive so that they will play on real records (...I kid you not...).   So they largely have been left alone to do what they need to do to get the tracks on vinyl, and not to make everything as loud as possible before they receive the tracks.  That's why the records have a higher dynamic range. 

 

The CDs, by virtue of their ability to be abused much further, win the "race to the bottom".  It's not the fault of the Red Book CD format (44.1 kHz/16 bit) that the people using the format have made consistent the institutionalized abuse. 

 

CD format is actually quite good if you don't do that "make it as loud as possible" maneuver that popular music CDs seem to always get nowadays.Go listen to CDs having very high dynamic range--like the albums of Flim & the BB's, the album called James Newton Howard & Friends, or Michael Hedges Aerial Boundaries.

 

Chris