Skip to main content

Topic: Current CD mastering practices (Read 40519 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
Current CD mastering practices
Reply #25
CNN  has an article  about the industry and how decisions are made. I realize it's not about the engineering in particular, but the mindset helps explain the engineering, IMHO.

Mark

  • Erukian
  • [*][*]
Current CD mastering practices
Reply #26
I sorted foobar's playlist by album gain, and i found a lot of my techno records at top. Early Underworld has a postive gain, up to +4, although that doesn't mean the samples used havent clipped prior to being used, but the dynamics are definately left in there. I just listened to their cd dubnobasswithmyheadman, lots of faint background harmonies start to whisper at 70dB, at 80dB it's like i'm hearing a different recording, tons more detail that i never thought existed.

I wouldn't be surprised if a lot of techno pre-2000 was recorded and mastered by the small indie-techno labels without the use of any compressor/limiter. The DJ's probably had it cranked so loud anways they heard all the details.

-Joe

  • moozooh
  • [*][*][*][*]
Current CD mastering practices
Reply #27
‘The Wall’ mastered by Fidelity Labs goes up about 10—11 dB after replaygaining. I still don't get it: how do they manage to make the disc so quiet, yet preserve all the dynamics?.. :\
Infrasonic Quartet + Sennheiser HD650 + Microlab Solo 2 mk3. 

  • esa372
  • [*][*][*][*]
  • Members (Donating)
Current CD mastering practices
Reply #28
Quote
‘The Wall’ mastered by Fidelity Labs goes up about 10—11 dB after replaygaining.
 
My MoFi copy of The Wall shows an album gain of +1.35db...

Is "Fidelity Labs" the same as "Mobile Fidelity Sound Labs"?
Clowns love haircuts; so should Lee Marvin's valet.

Current CD mastering practices
Reply #29
Quote
OK, this is going to sound very stupid, but remember that I am new here.  I understand that clipping is when the volume tops 0 dB, but what does that sound like?  I really have no clue what you guys are talking about.

-Phil
[a href="index.php?act=findpost&pid=319096"][{POST_SNAPBACK}][/a]
If you want to know, there's one easy way. Take an MP3, load it into MP3Gain, and adjust the gain to around 106 or 112dB. Then listen to the MP3.
This won't permanently damage the MP3; remember, MP3gain is completely lossless and reversible.
Quote
Norman people don't give a fuck about how a CD sounds.  Infact, I done a few silly test wich seemed to show they prefered fucked up music.
[a href="index.php?act=findpost&pid=319173"][{POST_SNAPBACK}][/a]
I know what you mean! I've had times, especially back in high school, when I'd pass my CD player to a friend so they can sample my music collection, and they'd turn the volume up as high as it can go, and turn on the old "bass boost" feature. The result is horrible. To paraphrase Rip Rowan, it's more white noise than music. All the bass drum pumps are reduced to dull thuds enveloped in static. And yet this is how they like it.
  • Last Edit: 14 August, 2005, 01:41:07 PM by NeoRenegade

  • GeSomeone
  • [*][*][*][*][*]
Current CD mastering practices
Reply #30
Quote
I do have a Personal Theory: the music is being engineered to be played in cars. Or iPods on noisy subways.[a href="index.php?act=findpost&pid=319065"][{POST_SNAPBACK}][/a]

This is not so far fetched. My (similar) theory is that CD's are made to sound like people would hear them on the radio/MTV. To me the worst compressed sound is heard on many radio stations.

I'm reluctant too to buy a CD nowadays. I was shocked how clipped the latest Coldplay (X&Y) album was. There are some quieter parts that give some relief, but 80% is limitted flat with severe clippings.
In theory, there is no difference between theory and practice. In practice there is.

  • HotshotGG
  • [*][*][*][*][*]
Current CD mastering practices
Reply #31
Quote
I forgot to include one of the most important casualties of compression;

Ambience, timbre, acoustics .... space.

Whether produced electronically with reverb/echo, or by genuine venue or studio acoustics, they are often vital for the 'atmosphere' created by a group/ensemble, or even simply to hear the timbre of a paticular voice or instrument.

They vanish almost into the noise-floor where left intact - very difficult to 'listen for', almost subliminal.

Heavy compression crams them up into into the rest of the signal, and as good as destroys them.

Of course, to be aware of these 'cues' you have to be at ease, and in an environment that allows you to hear them, using replay equipment (player, amp, speakers) of sufficient resolution.

The industry seems hell bent on denying us these pleasures. Is it stupidity? Malice? I don't really care anymore - they make me ANGRY either way.


I absolutely agree with you. My trapt self-titled sound exactly like this, well produced yes, mastered horribly distorted guitar is clearly masterd louder than anything else (their was virtually no noise-floor.)  I was amazed to find out how well The Back To The Future soundtrack was mastered and produced though. It's a shame, because mastering engineer's can control these things to I mean they are the ones with the "multiband compressors". A lot of 80's recordings used to use a fair amount of reverberation both natural and electronic and the resulting timbre sounded so much better, because of it.

Quote
This is not so far fetched. My (similar) theory is that CD's are made to sound like people would hear them on the radio/MTV. To me the worst compressed sound is heard on many radio stations.


I think it's good combination of both the industry and other ridiculous factors like that.
  • Last Edit: 14 August, 2005, 07:34:14 PM by HotshotGG
budding I.T professional

  • moozooh
  • [*][*][*][*]
Current CD mastering practices
Reply #32
Quote
Quote
‘The Wall’ mastered by Fidelity Labs goes up about 10—11 dB after replaygaining.
 
My MoFi copy of The Wall shows an album gain of +1.35db...

Is "Fidelity Labs" the same as "Mobile Fidelity Sound Labs"?
[a href="index.php?act=findpost&pid=320048"][{POST_SNAPBACK}][/a]

Ugh, yeah, right… Mobile Fidelity Sound Labs, that is. 
Still, my copy is very quiet indeed: replaygain_album_peak=0.354797.
Infrasonic Quartet + Sennheiser HD650 + Microlab Solo 2 mk3. 

  • Cyaneyes
  • [*][*][*][*]
Current CD mastering practices
Reply #33
Quote
Quote
Quote
‘The Wall’ mastered by Fidelity Labs goes up about 10—11 dB after replaygaining.
 
My MoFi copy of The Wall shows an album gain of +1.35db...

Is "Fidelity Labs" the same as "Mobile Fidelity Sound Labs"?
[a href="index.php?act=findpost&pid=320048"][{POST_SNAPBACK}][/a]

Ugh, yeah, right… Mobile Fidelity Sound Labs, that is. 
Still, my copy is very quiet indeed: replaygain_album_peak=0.354797.
[a href="index.php?act=findpost&pid=320378"][{POST_SNAPBACK}][/a]


My MFSL The Wall (gold disc, catalog number UDCD 2-537) is +1.34, album peak 0.999969... do you have a different MFSL version? 

Current CD mastering practices
Reply #34
Hi, I'm a lurker of this site from time to time, specifically I'm a big user of LamedropXPd (extreme setting) for supreme quality mp3's. I love watching the kbps readout on foobar2000 fluctuate  Anyway check out these wavforms from a recently re-released metal album originally from 1991. The record co just HAD TO go and ruin a classic and "update" it so it sounds blaring, harsh, and actually becomes irritating to listen to. GOOD JOB! MISSION ACCOMPLISHED! I'll be hunting down an original copy. Here's the "waveform" and the "db waveform"
  • Last Edit: 16 August, 2005, 01:28:52 AM by Hermit-ically Sealed

  • Lyx
  • [*][*][*][*][*]
Current CD mastering practices
Reply #35
oh
my
god!

Maybe even that animation on loudnessrace.com was an underestimation of what would follow... if i read that db-waveform correctly, then this track has 10db dynamic range? wtf?

- Lyx
  • Last Edit: 16 August, 2005, 02:00:29 AM by Lyx
I am arrogant and I can afford it because I deliver.

Current CD mastering practices
Reply #36
Quote
oh
my
god!

Maybe even that animation on loudnessrace.com was an underestimation of what would follow... if i read that db-waveform correctly, then this track has 10db dynamic range? wtf?

- Lyx
[a href="index.php?act=findpost&pid=320405"][{POST_SNAPBACK}][/a]
It's sad. The current CD's and re-issues of originals that have been compressed to the point of destroying the very thing they're supposedly trying to sell (good music), seem to be the path of the future. Whatever, If these record co's want to self destruct and continue on their path of destroying music so be it. Good thing I have almost no desire for modern music especially the most affected (pop crap). I'll gladly pay much more for a rare used OOP cd that the record co wont make a penny from.

  • BradPDX
  • [*][*][*]
Current CD mastering practices
Reply #37
I have produced several CDs and been a participating musician in dozens. The practice of using large amounts of "global" compression (compression applied to the entire mix and not just particular tracks) is a phenomenon with roots back into the 70's.

Let me add something really quick and important here: totally uncompressed recorded music is relatively rare in the pop/jazz/folk world and has been for a long time.

The essential psychoacoustic problems being addressed are 1) scale and 2) noise floor. Issue 1 is the fact that most recorded music is not played at realistic performance levels - it is simply impractical for me to crank up my living room rig to even small ensemble levels without bugging half the neighborhood and myself as well. The spaces in which we listen to music are often utterly at odds with the original volume levels.

Compression, judiciously applied, allows music to be perceived as "full" and satisfying at substantially lower levels. Is it "real"? Well, no - but then, neither is a 7-piece band cranking away in my 15 x 25 ft living room being reproduced through 2 sound sources. At its best, compression is a compromise made to remedy the most basic compromise in the system - the fact that the music will NOT be played in original spaces at original levels. It will be coming out of 2 speakers in your home or car.

If anyone in this group is a musician, you can try some interest experiments. For example, I recorded a CD of 10 lullabies for my kids a while back. I did this on the cheap, just the way the kids hear me every night - 2 good condensors, a hard disk recorder, me and a Martin guitar, played live.

When the tracks were played back on good cans or speakers, they sounded great - just like the real thing. I trimmed up the beginnings and endings, burned a CD and popped it into the boom box in the kid's bedroom. My god, it was awful. At low volumes (the target use case) all detail was lost - only the vocals would jump out from the tiny box. The kids hated it.

So it pulled up the tracks again and applied global compression. I burned several discs to see which amount of compression would sound good for the kids. The answer: the one with a lot of compression. Listened to closely, the compression is obvious. But when heard at low volume on the boombox, it is great. Mission accomplished and lesson learned.

What was different from my live performances in the same room? A lot. Lower volume, limited frequency response, directional speakers, etc. The net result was that the music was better served by squishing it. Don't worry; I still play "live" most nights. Lately the kids like Gram Parsons songs.

The next thing is noise. The 70's saw the birth of real FM radio, 8-tracks and cassettes. This meant that the shift to "car listening" was underway, and the record companies obliged with increased compression. In a sense, they are right: uncompressed music in a noisy car is misery. The peaks take your head off while the music floor is inaudible. The trouble is, the cure is sometimes worse than the problem.

I agree that many recent pop recordings I have are excessively compressed, lifeless and grating. That to me is sacrificing one thing for another. But take heart; well recorded music is out there in spades as well. Lately I have been listening a lot to Sufjan Steven's excellent "Illinois" CD - wonderfully dynamic and spacious, not to mention GREAT songwriting and arranging.

And I'll bet that "Illinois" is using some compression - just not too much!

Current CD mastering practices
Reply #38
It's not that we are all the use of any compression... I use it all the time when I record but it's got to be used judicusly and musically.  What we see on many modern recordings is not even remotely artistic.
"You can fight without ever winning, but never win without a fight."  Neil Peart  'Resist'

  • Axon
  • [*][*][*][*][*]
  • Members (Donating)
Current CD mastering practices
Reply #39
Heh, I wouldn't be so sure about that. Allegedly the reason why Zwan's debut CD was mastered so freaking hot was because Billy Corgan specifically wanted it. He quite literally wanted that sound.

  • RockFan
  • [*][*][*][*]
Current CD mastering practices
Reply #40
Quote
Heh, I wouldn't be so sure about that. Allegedly the reason why Zwan's debut CD was mastered so freaking hot was because Billy Corgan specifically wanted it. He quite literally wanted that sound.
[a href="index.php?act=findpost&pid=320612"][{POST_SNAPBACK}][/a]


yeh, well that's rich and pampered a-holes for you.

R.

Current CD mastering practices
Reply #41
Quote
if i read that db-waveform correctly, then this track has 10db dynamic range? wtf?

- Lyx
[a href="index.php?act=findpost&pid=320405"][{POST_SNAPBACK}][/a]
Oh sorry, I think it was on the 36db scale view. Here's a 96db and 120db view of the wav.

Current CD mastering practices
Reply #42
Quote
Let me add something really quick and important here: totally uncompressed recorded music is relatively rare in the pop/jazz/folk world and has been for a long time.

[a href="index.php?act=findpost&pid=320609"][{POST_SNAPBACK}][/a]
Uncompressed music is also rare or nonexistent in metal too. In fact metal in general is more compressed than other genre's. The distortion of a guitar is a kind of compression itself, Am I right? But this amount of compression and the added fun of severe wave smashing/clipping is far beyond enhancement of the power or prescence of a song into the realm of ridiculous idiocy. The song of the waveform I posted is actually a good song, but it sounds like utter dogshit! It's massively irritating, in fact it actually becomes a form of punishment with the goal being how much of it can you endure. I have another bands re-issue from the same company with the exact same problem.
  • Last Edit: 16 August, 2005, 10:14:12 PM by Hermit-ically Sealed

  • WmAx
  • [*][*][*][*][*]
Current CD mastering practices
Reply #43
For examples of albums with simple microphone placement[2 or 3 mics placed at far and mid positions] recorded in real acoustic environments[as opposed to sound booth + effects] and no dynamic compression, refer to:
http://www.mapleshaderecords.com/index.php

Warning: Mapleshade is VERY heavy on the snake-oil[just try to overlook this aspect], but their recordings are still of superb quality.

-Chris
  • Last Edit: 16 August, 2005, 10:22:42 PM by WmAx

  • richms
  • [*][*]
Current CD mastering practices
Reply #44
Another issue I have is that a few of the recent CDs I have listened to have clearly had some form of "mega bass" processing done on them, there was virtually nothing under 60Hz with loads of harmonics, like you get from a mini system doing its best.

I like my bass to be felt, not to be made artifically louder by distorting it.

  • Lyx
  • [*][*][*][*][*]
Current CD mastering practices
Reply #45
Quote
But take heart; well recorded music is out there in spades as well. Lately I have been listening a lot to Sufjan Steven's excellent "Illinois" CD - wonderfully dynamic and spacious, not to mention GREAT songwriting and arranging.
[a href="index.php?act=findpost&pid=320609"][{POST_SNAPBACK}][/a]

That's right - i would even go on to say that the majority of the total of *different* albums released globally is not overcompressed. Compressed to increase volume, yes - but not in a destructive way. So, the majority of music released is very listenable. However, only a minority of *people* listen to this majority of *music*. Especially in experimental and non-mainstream genres you will find lots of good sounding records. So, the reason why people perceive the loudness-race to be omnipresent is because the majority of people listens to well known artists.

While the loudness race has definitelly had its impact on indy-music as well, it is not really a problem there. Thus, overcompression is definatelly a feature of the music-*industry* (and a mean "industry" literally here).


Quote
Uncompressed music is also rare or nonexistent in metal too. In fact metal in general is more compressed than other genre's. The distortion of a guitar is a kind of compression itself, Am I right?

Partially, yes. There are genres which use "instrument-specific" overcompression as an artistic tool. The main feature of shoegazer-music for example is to oversaturate a given space with effects-loaden e-guitars - and then paint on this "texture" with other instruments. Thus, the other instruments shall partially "melt" into the oversaturated ones, but still stay distinctive. Overcompression is not a mindless loudness-tool there, but instead is a very complex and difficult to master tool(difficult because of the inheritant problem of oversaturating a part of the sound, yet keeping other parts seperate and distintive). This kind of music often has high albumgain values (-6 to -7 being the average) - without being mastered hot.

Quote
Oh sorry, I think it was on the 36db scale view. Here's a 96db and 120db view of the wav.

Hmm, i think we've got a misunderstanding here - i was talking about the two shades of blue which are used for the db-waveform - i would asume that the brighter one represents the averaged "floor", while the darker shade represents the averaged "ceiling"?
  • Last Edit: 16 August, 2005, 11:22:54 PM by Lyx
I am arrogant and I can afford it because I deliver.

  • richms
  • [*][*]
Current CD mastering practices
Reply #46
Quote
The distortion of a guitar is a kind of compression itself, Am I right?
[a href="index.php?act=findpost&pid=320619"][{POST_SNAPBACK}][/a]


The compressed guitar is an instrument in its own right with the sound coming from the guitar+pickups, tone, pedals, amp and speaker all in unison. It is miked and mixed in just like any other instrument. On a well mastered metal CD you can hear the guitar as its own instrument placed in the soundfield with other stuff around it.

When a CD is mastered my a compulsive red line hitter, the guitar is usually the first to be affected I find, because its the most complex instrument, it will sound like its coming from all over the place long before the drums are sounding messed up (IMHO) - and is a good indicator of how bad the mixing is.

What they are doing is essentially negating the purchase of decent playback equipment.

  • GeSomeone
  • [*][*][*][*][*]
Current CD mastering practices
Reply #47
Quote
The practice of using large amounts of "global" compression [..] is a phenomenon with roots back into the 70's.[a href="index.php?act=findpost&pid=320609"][{POST_SNAPBACK}][/a]

Good point, all you said applies in general. Dynamics must be within a reasonable range for home enjoyment. Individual tracks may benefit from the right amount of compression (e.g. bass) or limiting (drums, vocals).

Ideally playback equipment should handle the issues like "car environment", low level listening, portable use etc. Not make every album to fit those.

The current issue here is more with the (re-)mastering practices of the last (10?) years.
Since digital there is an abrupt upper limit to a signal and a lot of times it seems the point is to get as close towards that as possible. It comes from the thinking "louder is better" and "in a CD changer our CD should at least be as loud as the others". This is referred to as "the loudness race".
In theory, there is no difference between theory and practice. In practice there is.

  • bug80
  • [*][*][*][*]
Current CD mastering practices
Reply #48
Quote
Quote
Oh sorry, I think it was on the 36db scale view. Here's a 96db and 120db view of the wav.

Hmm, i think we've got a misunderstanding here - i was talking about the two shades of blue which are used for the db-waveform - i would asume that the brighter one represents the averaged "floor", while the darker shade represents the averaged "ceiling"?
[a href="index.php?act=findpost&pid=320624"][{POST_SNAPBACK}][/a]

I can't find information on what the lighter color is in Audigy, maybe somebody else knows?

I think the darker shade is the "real" waveform and the lighter shade a moving avarage (The RMS value calculated over 50 ms or so).

  • Yaztromo
  • [*][*][*]
Current CD mastering practices
Reply #49
Sorry if this has been posted before but I noticed loudnessrace.net is back up, but now at http://www.loudnessrace.com