The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
CD mastering in the 1980s, why such low levels?
Old 12th June 2019
  #1
Gear Addict
 
basehead617's Avatar
CD mastering in the 1980s, why such low levels?

I’m continually amazed by the low overall level of CD rips of the original mid-late 80s releases.

I’m not talking about classical recordings or things that genuinely used a lot of dynamic range, but just overall incredibly low average levels and well below 0 peaks.

What was the process of mastering back then that was responsible for this? Was there just no reliable way yet to brick wall limit and an aversion to mastering stage compression that they just had to turn things down until a whole track didn’t clip at a given input?
Old 12th June 2019
  #2
As far as I'm aware there wasn't an easily available brick wall limiter until Waves released the L1 in 1994 (there may have been Weiss or Junger products before then, I don't claim to be an expert). Without a brickwall limiter people treated digital clipping as the potentially catastrophic phenomena it was. Also, most CD players back then were for home stereos which easily had enough power to cope with these "low" levels. Consumer speakers nowadays expect to be fed a brickwalled signal, and so don't tend to be as powerful.
Old 12th June 2019
  #3
Lives for gear
 
avare's Avatar
 

Why do you want them slammed against the (brick) wall?
Old 12th June 2019
  #4
Lives for gear
 
Adam Dempsey's Avatar
 

Verified Member
It was partly due to the sources themselves of most initial CDs (often 2nd generation EQ'd-for-vinyl master tapes, of mixes with healthy dynamic range where compression reigned for its sound, not "loudness"); partly that few wanted to push or clip the harsher sounding 16 bit A-D converters of the time (although optimizing level for 16 bit A-D was beneficial – it later became a thing to lightly clip then drop the master -0.1dB or so); it was partly that an over or even 0dBFS on a CD Master (1630 U-matic tape) would be deemed "illegal" and rejected at the plant, risking costly delays; and partly due to the absence of look-ahead digital peak limiting.

Remember, they weren't "low level" – only relative to what came later.
Old 12th June 2019
  #5
Lives for gear
 
Justin P.'s Avatar
 

Quote:
Originally Posted by Adam Dempsey View Post
Remember, they weren't "low level" – only relative to what came later.
Right. I don't remember any complaints at the time. People were just (generally speaking) happy to have the music in a more convenient format, with less noise, and that wouldn't wear out as easy as previous formats.
Old 12th June 2019
  #6
Gear Guru
 
Thomas W. Bethe's Avatar
 

Verified Member
On my Avocet controller there is a "0" db setting on the volume knob. If I listen to music before the "loudness wars" the setting of "0" db works well. If the music is done after the loudness wars started I have to use the minus -10 db or lower setting for comfortable listening level. I guess that answers the OPs question.
Old 12th June 2019
  #7
Deleted e999d8e
Guest
Quote:
Originally Posted by Justin P. View Post
Right. I don't remember any complaints at the time. People were just (generally speaking) happy to have the music in a more convenient format, with less noise, and that wouldn't wear out as easy as previous formats.
Yep.
Old 12th June 2019
  #8
Deleted e999d8e
Guest
Quote:
Originally Posted by basehead617 View Post
I’m continually amazed by the low overall level of CD rips of the original mid-late 80s releases.

I’m not talking about classical recordings or things that genuinely used a lot of dynamic range, but just overall incredibly low average levels and well below 0 peaks.

What was the process of mastering back then that was responsible for this? Was there just no reliable way yet to brick wall limit and an aversion to mastering stage compression that they just had to turn things down until a whole track didn’t clip at a given input?
I thought CD's were plenty loud when I first started collecting them, especially compared to LP's. I had to adjust the volume on my amp consistently downward to accommodate my CD player.

Now people expect everything to be audible above the noise level of a moving car. There's no accounting for taste, I guess.
Old 12th June 2019
  #9
Lives for gear
 
Greg Reierson's Avatar
 

Verified Member
This thread reminds me that most young engineers have no idea what a VU meter is.

Transferring analog masters to digital where 0VU ~ -20dBFS insured that the source noise floor was above the digital noise floor and that the source peaks were below the max digital level. It was a good match between the two formats with minimal in-band compromises.

Why we now chop off the top 10dB and cram it all up at the top of the digital scale is the real puzzler.
Old 12th June 2019
  #10
Lives for gear
 
Justin P.'s Avatar
 

A new Bob Ludwig interview just posted today that is relevant to this thread:

https://bobbyowsinskiblog.com/2019/0...wig-mastering/
Old 12th June 2019
  #11
Lives for gear
From that article: "We’re all afraid of the over levels...."

Hahahaha, Bob was definitely not afraid of the overs, I have loads of '90s records he mastered and there's clipping on almost all of 'em. That's not a criticism, they all sound great, I just thought that quote was amusing.
Old 12th June 2019
  #12
Lives for gear
 
matucha's Avatar
Quote:
Originally Posted by Thomas W. Bethe View Post
On my Avocet controller there is a "0" db setting on the volume knob. If I listen to music before the "loudness wars" the setting of "0" db works well. If the music is done after the loudness wars started I have to use the minus -10 db or lower setting for comfortable listening level. I guess that answers the OPs question.
Wow, talk about different calibration. 0dB on my avocet is usable only with very low volume material (production/tracking time). With mastered stuff it's like -30dB for me.
Old 13th June 2019
  #13
Lives for gear
 
Adam Dempsey's Avatar
 

Verified Member
Quote:
Originally Posted by scraggs View Post
From that article: "We’re all afraid of the over levels...."
And one reason it was feared by mastering engineers is that the CD master (U-matic 3/4" tape) literally would fail the error check report by the Sony analyzer, resulting in another (real-time) pass, fingers and toes crossed for no glitches/un-correctable errors, then delivery and potentially very costly delays.
Old 13th June 2019
  #14
Lives for gear
And yet there's overs on virtually everything!
Old 13th June 2019
  #15
Lives for gear
 
Adam Dempsey's Avatar
 

Verified Member
Quote:
Originally Posted by scraggs View Post
And yet there's overs on virtually everything!
Not when they dropped the output -0.1dB!
Old 13th June 2019
  #16
Gear Addict
 
basehead617's Avatar
Quote:
Originally Posted by Adam Dempsey View Post
And one reason it was feared by mastering engineers is that the CD master (U-matic 3/4" tape) literally would fail the error check report by the Sony analyzer, resulting in another (real-time) pass, fingers and toes crossed for no glitches/un-correctable errors, then delivery and potentially very costly delays.
Among these replies that seems like the likely 'real' reason.
Old 13th June 2019
  #17
Lives for gear
Quote:
Originally Posted by Adam Dempsey View Post
Not when they dropped the output -0.1dB!
I believe you and I'm not trying to be argumentative! I just have a zillion cds on a drive on the studio computer, Wavelab meters set to display true peaks, and almost any record I put up, from say 1990 onward, there's peaks over 0.

I'm guessing the Sony was analyzing sample values and not ISPs? Or....something? I'm outta my depth here!
Old 13th June 2019
  #18
Gear Addict
 
basehead617's Avatar
Quote:
Originally Posted by Greg Reierson View Post
This thread reminds me that most young engineers have no idea what a VU meter is.

Transferring analog masters to digital where 0VU ~ -20dBFS insured that the source noise floor was above the digital noise floor and that the source peaks were below the max digital level. It was a good match between the two formats with minimal in-band compromises.

Why we now chop off the top 10dB and cram it all up at the top of the digital scale is the real puzzler.
Someone can correct me if I'm wrong, but I'm pretty sure the D->A converters of the time were particularly poor with low-level signals.

So being 'quiet' in the PCM data actually sounded pretty bad. I have a 1984 Technics CD player and some of these CDs in question you can hear the awful quantization noise.
Old 13th June 2019
  #19
Gear Head
 
S_mask's Avatar
 

There's a special window of high-quality CD mastering before the 'look-ahead' brick wall limiters took over the Loudness Wars and yet after converters got better sounding than they were initially when Sax decried the 'digital madness'. 1990's Question and Answer by Pat Metheney (Bob Ludwig), 1994's Talking Timbuktu by Ry Cooder and Ali Farka Touré (Bernie Grundman), and Joao by Joao Gilberto (Antonio Barroso) are three examples of high crest factor CDs that 'breathe' and that invite repeat listens. They present CD as a hi-fi medium, which it technically is. Bob Katz tried to save it with his K-System of monitor-gain-calibrated level metering, but the loudest common _dominators_ trump wisdom.
Old 13th June 2019
  #20
Lives for gear
 
SmoothTone's Avatar
 

Quote:
Originally Posted by scraggs View Post
I'm guessing the Sony was analyzing sample values and not ISPs?
This'd be it. I don't think oversampled metering became a thing until around the turn of the century.
Old 13th June 2019
  #21
Lives for gear
 

Verified Member
inter sample analysis, not oversampling...

jünger digital limiters had their max. output set to -0.29dB fs if i remember right - the best feature was not the limiter though (it sounded horrible) but their upward compressor (or multi-loop compressor): i'm still using their gear today for live mixinging, broadcasting and mastering!
Old 13th June 2019
  #22
Lives for gear
Quote:
Originally Posted by Justin P. View Post
A new Bob Ludwig interview just posted today that is relevant to this thread:

https://bobbyowsinskiblog.com/2019/0...wig-mastering/
"The Daft Punk record [Random Access Memories] is not heavily squished either, compared to other electronica records. We raised the level, but it’s not insane."

this is why Daft Punk has made do a second mastering and final mastering by Antoine Chabert "Chab".

The Bob's mastering has been judged too dynamics.


Sacred Bob
Old 13th June 2019
  #23
Lives for gear
 
SmoothTone's Avatar
 

Quote:
Originally Posted by deedeeyeah View Post
inter sample analysis, not oversampling...
ISPs are detected via 4x oversampling.
Old 13th June 2019
  #24
Lives for gear
 
lowland's Avatar
 

Verified Member
A couple of early 90s CDs I still regularly use as references at about +10 over my usual listening level:

Ry Cooder/Vishwa Mohan Bhatt - 'A Meeting By The River' (1993)
Recorded to 1" stereo analogue on location
https://en.wikipedia.org/wiki/A_Meeting_by_the_River

Jan Garbarek - 'I Took Up The Runes' (1990)
Recorded by the great Jan Erik Kongshaug at Rainbow Studios, Oslo
https://en.wikipedia.org/wiki/I_Took_Up_the_Runes

Last edited by lowland; 13th June 2019 at 12:56 PM..
Old 13th June 2019
  #25
Quote:
Originally Posted by dinococcus View Post
"The Daft Punk record [Random Access Memories] is not heavily squished either, compared to other electronica records. We raised the level, but it’s not insane."

this is why Daft Punk has made do a second mastering and final mastering by Antoine Chabert "Chab".

The Bob's mastering has been judged too dynamics.


Sacred Bob
It's still a couple of db lower than most other modern records in the genre, though.
Old 13th June 2019
  #26
Lives for gear
 

Verified Member
Quote:
Originally Posted by SmoothTone View Post
ISPs are detected via 4x oversampling.
thx - i didn't know what technique was/is used for detection; i thought you were mixing up things...
Old 13th June 2019
  #27
Lives for gear
 
Greg Reierson's Avatar
 

Verified Member
Quote:
Originally Posted by basehead617 View Post
Someone can correct me if I'm wrong, but I'm pretty sure the D->A converters of the time were particularly poor with low-level signals.

So being 'quiet' in the PCM data actually sounded pretty bad. I have a 1984 Technics CD player and some of these CDs in question you can hear the awful quantization noise.
Oh, for sure there were terrible converters available in that time and dither wasn't well implemented yet. I had a Fostex DAT as late as 1988 that had the worst sounding grungy converters I have ever heard. The zero cross distortion in the ADC was so bad they intentionally added a DC offset to stay away from it. Really, really nasty sounding converters. But they were not the norm. Most were fine and the bad ones were only objectionable on very quiet stuff like classical. Typically not bad for pop / rock. But I'm sure there were plenty of other exceptionally bad ones out there. Still not a good reason to compromise an otherwise good sounding master for a broken playback system.

The first two CDs I bough were Talking Heads "Little Creatures" and Don Henley "Building The Perfect Beast". '80s levels and still sound fantastic today.
Old 14th June 2019
  #28
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
yeah those first Sony R-DAT Recorder ADCs were harsh and clanky to my ears, then if the mix engineer hit the “emphasis” button, oh boy... sheeesh.

i suppose when you transferred the analog tape to digital back in those days, the level of 0VU on the tape vs the input level of the ADC was everything.

still one of my very favorite CDs from 1985 is Michael Hedges “Arial Boundaries” super quiet and very dynamic by todays standards.

the other one was Level 42 “World Machine”.

at first we found the “perfect” digital sound at 16-bit quite interesting, much like a dog turning it’s head up on one side.

we’ve been working our way back to analog ever since, hence all the modeling plug-ins at high bit rates.

jt
Old 1 week ago
  #29
Moderator
 
jayfrigo's Avatar
 

Verified Member
Quote:
Originally Posted by Hippocratic Mastering View Post
As far as I'm aware there wasn't an easily available brick wall limiter until Waves released the L1 in 1994 (there may have been Weiss or Junger products before then, I don't claim to be an expert).
The Sony DAL-1000 (Digital Audio Limiter) was the "go-to" limiter in the early days.
Old 1 week ago
  #30
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by avare View Post
Why do you want them slammed against the (brick) wall?
Because the six people out there who still actually buy music (and the six thousand music supervisors who license it) audition it over the internet. Loud wins. This wasn't a thing back then.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump