Audio Quality from CD's

[poulkirk]

This post isa placed in the hope to start healthy and controversial discussion

I am of the firm belief that well produced, recorder and mastered CD's are superior to LP's

Unfortunately, many CD's, particularly recently recorded and released ones are "Crappy"

This particularly applies to many "Pop" recordings which have obviously been monitored during final Mixdown and Mastering at an excessive sound level where they lose all subtleties when played back at ordinary sound levels

Others suffer from Digital Clipping

This includes the very popular Norah Jones "Come away with me" CD, presumably sold by the Million's

While quite pleasant to listen to, it does suffer from a cerrtain "Harshness" at the higher recording levels on it

Importing it as a wave file into a Professional Editing Software package, ot is painfully obvious that this particular recording, and many other recent "Pop" Recordings suffer from Digital Cipping

Digital Clipping sounds bad

For this to happen, is it that the "Recording or Mastering Engineers"

1; Lack the necessary skill to do a good recording ?

2; Would not know good sound if they "Fell over it" ?

3; Unable to produce a really good recording ?

4; Could not care less, "Just get it on the market" ?

I concede that Many LP's sound very good, but one does have to ignore Surface Noise

Many Re-Releases of LP's on CD are very good

Unfortunately many Re-Releases on CD are absolutely terrible

Maybe this is caused by lack of Skill, Care and Taste in the Re-Mastering

Maybe it is caused by "Lack of Care and Technical Ignorance"

Your comments are invited

[MWP]

Agreed, mastering quality is decreasing.

I have 3 albums by Faithless here... all three have mastering problems.
All three have very bad clipping, two of them have solid 12khz tones through some of the songs (obviously not meant to be there), and one has a track end/start points in the wrong place.

Faithless are one of my fave groups, so its damn annoying that i have to put up with these bad CDs.

I ended up ripping them to WAV files and editing them to fix the problems.

Im not sure who they are mastered or pressed by.

[joensd]

Very interesting thesis.
Actually it´s not as it happens over and over.(points 1 to 4)

A good friend of mine owns a recording studio
and so I got to know a few recording engineers myself.

Some of them just don´t have the ears to do this job properly.
They misjudge the sound, overpronounce certain instruments.

A lot of them don´t have the musicality to feel what is right.
The vocals are not in the foreground. The groove is missing.
Technical adds like effects are overdone, put in the wrong place
so the music is not reflected properly anymore.

And after all of this there comes another pair of potentially bad ears.
The ones from the mastering engineer. Same arguments as before.

After years being the managing director of the studio my buddy
found out
that he´s a much better engineer than most of those have worked there due to his musicality.
Being a professional musician (guitarist and songwriter) for years himself
he knows how to record songs.

Another important point IMO is the equipment that´s used.
My friend uses two very nice studer analog machines (2" 1/4")
still regarded as some of the best machines.

Why are they still so popular when you have "comfortable" digital suits?
Cause they do a 1:1 recording job without limits or colourations.

Another studio here in town uses digital equipment.
Due to a cold sound in their recordings
they started to use tube microphones only to get some life into their recordings!
The studio is not setup as professional as the analog one but I think you see
what I mean with colouration.
It doesn´t as easily do a 1:1 recording. It´s not the real sound.


Cheers
Jens



BTW: My friends studio is called "Real sound studio"

[mrfeedback]

Hi Poul,

Digital Clipping sounds bad

No dispute.
So do sloppy edits leaving a vertical transition in the waveform - makes a click.

For this to happen, is it that the "Recording or Mastering Engineers"

1; Lack the necessary skill to do a good recording ?

There are quite a number of skills to master - not all are intuitive, and not all are taught.

2; Would not know good sound if they "Fell over it" ?

Not a lot of home listeners have REALLY good gear to listen to, and not a lot of audio record and mixdown engineers have outstandingly good systems to monitor on, so not a lot of people get to hear really GOOD reproduced sound.
That said, I find that all listeners immediately recognise and prefer REALLY GOOD sound and desire it - even if it is out of financial reach.

3; Unable to produce a really good recording ?

That requires a bunch of elements like the appropriate listening and musicality skills, and the appropriate recording and monitoring equipment.
That also requires understanding of what that 'little extra' is that makes a recording a classic.

4; Could not care less, "Just get it on the market" ?

Modern high-end studio time is seriously expensive and limits most recording projects.
For modern 'disposable trash' music this kind of money is not available, and the ethos seems to be that if it sounds loud then that is good enough.

I concede that Many LP's sound very good, but one does have to ignore Surface Noise

Yeah, and dirty styluses too.
Vinyl sounds characteristically coloured to my ears, and according to the kind of vinyl used in addition to all the other variables in a vinyl record system.
I reckon REALLY GOOD cd is superior on all fronts.
There are arguments about >20kHz sounds adding 'air' and 'realism', but I suspect that much of this is sub 20kHz distortion components created.

Eric.

[stefanobilliani]

Originally posted by mrfeedback
Hi Poul,


There are arguments about >20kHz sounds adding 'air' and 'realism', but I suspect that much of this is sub 20kHz distortion components created.

Eric.

I studied music from when I was 6 ,and listened to real instruments (by real I mean not sintetics or electronics)like pianos , guitars , flutes , drums for at least 14 years, but before the "digital audio" I had never eard a bass guitar that tends to a sordid sinusoid the way many cd players plays it.

Having say that I find the non-os/reclocked dacs very interesting and able to let me listen to some Jazz records without too much pains.

[indoubt]

Unfortunately you seem to be right,

I have not always had concerns about the quality of CD's but as my system is improving it just seems that bad or mediocre recordings increase in numbers (and I don't have a high end system, just relatively good commercial stuff)

It seems that more and more CD's are put to the limit of loudness and beyond. The resulting clipping and distortions only seems to bugger a few and appearantly has no influence on sales (well, maybe it does, CD sales are dropping, isn't it)

The last trip to the music store resulted in purchase of 10 CD's of which only 1 sounded great, from a recording and distortionless point of view that is.

Unfortunately you cannot judge in the store whether a recording is good or bad, the machines at the store are often just to crappy, it's to noisy etc.

With DVD-A and SACD only being slightly better to my ears than a good recorded CD these formats are not the big leap forward either, although I must say that forums ad to their signifigance by reviewing and commenting on releases. It is therefore easier to discover whether a certain DVD-A or SACD is recorded wel than to discover the quality of a CD.

Maybe it's time for an independant third party to assess the quality of CD's and make an independeant rating. I think we would all benefit and the quality in general would increase.

[mrfeedback]

From December 1994 Stereophile

It's becoming incontrovertablethat CDs containing the same 1' and 0s produce varying levels of sound quality.
These sonic differences are not caused by data errors, as commonly assumed, but by some other mechanism.
The most likely culprit is jitter in the recovered HF signal, which varies greatly between CDs.
As we've seen, CD manufacturing isn't a black and white process of putting binary 1s and 0s on a piece of plastic, but a series of highly malleable procedures each of which introduces an analog-like variability in the signal read from the disc by your CD transport.
Unfortunately, there's no way for you to judge the quality of the discs you buy.
Until the record companies put pressure on CD manufacturers to make better-sounding discs, purchasing CDs will remain a crapshoot.

This article investigated error rates in the playback data-stream, and found that uncorrectable errors are very rare.
Their conclusion is that variations in the manufacturing produces discs with varying quality of HF signal and consequent jitter in the data causing sonic problems.

This is of course only one source of bad sound, and a secondary one perhaps.
In modern audio recording and mixdown the usual process involves digital process from start to saleable product.
Digital recording allows any manner of effects to be used in the digital domain, and each of these introduce rounding errors and these rounding errors add up, and the sonic errors imparted are strongly dependant on the algorithm used in the DAW softwares - IOW some effects are sonically bad due to bad error handling/rounding.

I have heard modern musos lament that digital has taken the 'warmth' out of recordings, and that it is common to put the mastered recordings through a 'finaliser' box (digital or analog) to restore the 'missing' colourations and distortions that are interpreted as 'warmth'.

I have also heard engineers comment that in the studio a high res/high sampling rate master can sound fantastic, but when down-converted to CD 16/44k, sound quality goes straight out the window and "What's the use".

I have heard playback of a master tape and directly compared it to DAT and CD, and yes the sound quality was decidedly inferior in terms dynamics and 'being there' feel and niceness.
I suspect this may be largely due to AD/DA convertor techniques, and I hear that big improvements are being made to the AD conversion process.

All of this does not explain faults like digital clipping in releases, and this comes down to operator error I expect.
It is normal for record companies to 'remaster' recordings before release, and if the source material has been 'normalised', then small amounts of eq or other effects can send the data into digital clipping, and this may or may not be listened for by these final mastering engineers.

For the sonic problems to be reduced, recording and mixdown operators need to understand that digital process has enough SNR that individual tracks and masters do not need to be recorded to peak max (0Vu) levels, and that some headroom should be left for the final release re-mastering and 'normalising'.

Bob Katz has a good article describing the importance of mastering at a fixed level (85 dBa) to ensure pleasant and consistent results.
With the cheap availability of modern recording equipment, 'every man and his dog' is involved in recording nowadays, and old techniques and knowledge are becoming lost or little known, and quality of analog equipment is becoming less also (Behringer etc).

I reckon 44k/16bit is adequate for replay, and modern 96k/24bit should be a help in the recording side to reduce digital errors and improve the quality of final masters.
All of this does not help idiot 'engineers' however.

Eric.

[fastcat95]

In reply to the question of CD playback quality:

There is no recording medium without coloration, but I will make
the following statement about CD playback:

I think that, especially for the sampling rate and 16 bit limit of
standard redbook CD playback, that it can be quite good! There are, however, many CD disks that suffer from all of the old problems that vinyl has suffered from, including:

1. Industry learning curve about technology
2. Poor recording techniques/ equipment
3. Biases of producers, engineers, record companies.
4. Poor editing and mastering
5. Average to poor disk production quality
6. inadequate playback systems/room interfaces

I have a pretty good system in an acoustically good room, and
a better than average transport and DAC. My DAC has been known for showing little mercy as to what is going on in a recording, and attained a bit of a reputation as being a little
on the bright side (with less than great CD recordings).

But.... when my system is well warmed up (48 hours), and I play
a CD that was recorded by someone like Keith Johnson, and produced by a good audiophile label... the sound is something quite remarkable. What I hear from redbook CD here is just about everything that I could hope for. Great air, detail resolution, dynamics, full soundstaging, natural upper mids and highs, very low noise levels. And very much like what I have heard from live performances of things like concert harp or flute
in settings where the room acoustics have made it easy to hear a very natural presentation.

Redbook CD is not perfect, but just try some of the great digital disks that have been produced on a fine system, and you may conclude that you like it better than what you hear from most vinyl recordings. If all my CD collection were like a few of my CD disks, I would truely be in heaven!!!

Fastcat95

[Lisandro_P]

Bad recordings are bad recordings, both in vinyl and CD. The thing with digital media it's that, besides all it's advantages, it's a LOT less forgiving than an analog equivalent. Too much ProTools "audio engenieers", and too much badly produced music are avaiable nowadays. I guess it also happened with vinyls back then.

[phase_accurate]

Regarding quality we don't have to forget bad taste (in terms of how someting should sound) of musicians, engineers and producers.

Regards

Charles