The death of dynamics

PAC

I've posted this elsewhere, but for those interested, its a really good article on dynamic compression:

http://www.cdmasteringservices.com/dynamicdeath.htm

uglymusic

Maybe, just maybe, as we see the death of the mega record company and Big Radio, we'll see a drift away from this sort of crap.

Alan

Gotta hope so. This is a very nice article Paul, thanks for sharing.

I was listening to 'Joshua Judges Ruth' by Lyle Lovett the other day; this is known to have almost no dynamic compression at all. It sounded superb, and I had the volume set around the 12 o clock position.

Then I went to the latest Wilco recording ('The Whole Love') - usually a very well recorded band - and it nearly blasted me back into the kitchen. Why?...  X_X

PAC

It may be an advert for a mastering studio, but what they say is pretty convincing IMHO.  I find it very difficult to get any decent uncompressed CD's these days as Red Book seems not to be so much of a standard as a ballpark target that's rarely achieved...through choice.  It's all about loudness and in-your-face music as that's apparently what the younger generation "want", especially with earphones and pop/rock.  Artists can and should take more of an interest in how their material is recorded or else be held at least partially responsible for the death of high fidelity.  If we cannot convince the younger generation now, then hifi as an industry and an interest will only become more marginalised until the concept dies out.

I have some nicely recorded CD's, and most of these are from the Opus 3 catalogue. 

uglymusic

I have loads of great recordings on CD, but they're not from the mainstream.

Docfoster

I suspect that with all but the most technically accomplished performers playing the most supreme instruments in the most perfect studios, some compression during the mastering process will often be inevitable, but this kind of post-production compression really is the work of the Satan.
He tortures us in the fires of hell with his non-existent dynamic range.

PAC

There's nothing wrong with some limited flat topped wave form mastering, it depends how many times during the performance that the peaks are reached.  In an ideal world, the ideal recording would only have one transient peak set and normalised to the 0dB reference without any compression. There's frequency clipping and dynamic audio compression.  Recordings that use both sound horrible.  Some recordings (in fact many) use dynamic compression below the 0dB reference, and this even without clipping will sound horrible.  Some musicians might argue that for distortion effect, flat top waveforms are essential which points to manufactured music sharing little with the live performance, but if that's how they want it to sound, then ok, just don't compress the hell out of the performance in an attempt to get it loud.

Docfoster

I'd agree with that. The little I've experienced in studios has suggested to me that "compression" is in fact a wide spectrum with numerous perameters and purposes. I doubt if there are many recordings that emerge from the cd or data file never having seen any compression at all, but equally as PAC says, there are pieces of recorded music that have seen a little too much, at a few too many moments for a few too many reasons to be enjoyable to the ears of the audiophile.

Alan

Is one of the main reasons why digital replay is seen as a poor cousin of good analogue to do with the mastering/compression issue? It seems to be observed often that vinyl vs CD releases differ greatly, perhaps because of the more mainstream audience that CD (now its downloads) has. Digital releases seem to be more compressed because that is what its main demographic tacitly requires.

Leaving aside the question as to HDD digital audio vs CD players (which is a significant factor IMO), which of the following identical recordings is likely to be preferable/worse?

• New recording, nasty radio/download friendly compression (vinyl)
  
• New recording, nasty radio/download friendly compression (Digital)

I like to think that given equal systems there would be nothing to tell between them.

Therefore, I feel differences between a mastering for vinyl and a mastering for CD can go some way towards explaining a general antipathy towards digital playback. That and the fact CD players generally suck too (unless you have something special).

The fact that some really well recorded redbook CDs sound simply stunning (once ripped) seems to add weight to this view.

PAC

You've nailed it Alan.  There's a few reasons why digital can be presumed as vinyl's poor cousin but it should in theory be the other way round.

Look at true dynamic range of each format:  Vinyl, at best, allows 50 to 60dB range whereas properly Mastered to Red Book standards CD should be capable of 90dB range.

The best mastered vinyl tends largely to be albums made in the 1970's and 1980s, although the advent of good virgin vinyl polymers has allowed the medium itself to present quieter noise floors today.

Many CD's, most especially pop and rock are mastered today with as little as 10dB dynamic range, some even less.  It doesn't matter what the source is, whether its CD, HD bitstream or whatever.  Its going to sound horrible.

Some degree of compression is arguably needed.  For example, in AV, if you had an action scene with a shotgun being fired (140dB) then comparing that with the comfortable vocal audible level of around 70dB would leave the audience deaf, so scenes like that are compressed to threshold plus around 20dB (say 90dBA).  That's understandable.

Next take a piece of music and compress it to 20 or even 10 dBA range with around half the peaks reaching 0dB reference.  It will sound lifeless, without dimension and get quite fatiguing.  Conversley, with no compression and badly mastered, if peaks are allowed to rise about the 0dB reference, it could run the D/A stage into clipping and digital clipping is horrible on the ear, so it's important that for any performance:

1.  peaks never exceed the 0dB reference

2.  range should be mastered with only one or at least a small handful of peaks reaching this reference

3.  Dynamic compression shouldn't be added post production to increase volume

sadly, that is exactly what happens as the marketing gits (sorry but its an apt description of the culprits) think that more air play time will result if their tune is louder than anyone elses.

For all intent purposes, 16 bit lightly oversampled CD playback with a true range of 90dB is as good as you;ll ever need.  This is why EARLY CD releases which rarely suffered from compression played on early 16Bits Phillips chipped DAC's still sound very fine today. All this progression to higher reolution adds nothing id the recordng isn't up to scratch in the first place and its partly that reason that's driven HD music servers and DAcs to supercede the humble 16Bit CD format, when really there's nothing at all wrong with the quality of a 16Bit CD D/A converter!

My own reference system still uses a player with 4 Phillips 16 bit chips and to my ears its a match for any digital source providing the recording is up to scratch.

Modern vinyl and modern CD recorded the same and to say 50dB dynamic range should sound the same but wont for a few simple reasons:

The CD uses an unbiased output source whilst the record player doesnt.  Every cartridge has different tracking abilities, damping factors, output impedances and capacitive loading requirements.  Also, many carts are able to dig out very high order harmonics which add to the sense of "realism" and timbre of music and its partly this that might just win some music fans over to the LP instead of the CD, but it would be a pretty close run thing IMHO. 

Alan

Preferences are understandable of course, I was impressed with Simon's self-awareness when he stated a preference for his new Kuzmi deck & arm because it's a bit more lively and coloured than his previous SME.

Early CDs are great though, I can get one or two of mine to the 12o'clock position volume wise, while newer ones (even from the same artist sadly) can't go past 8:30. I used to watch for AAD SPARS codes, though I have found plenty of ADD & even DDD that are as fine. It's often down to who does the mastering (I usually like anything that George Massenburg has been involved with).

There are/were some great vids on youtube by Bob Katz about good mastering, maybe that's an idea for another thread? They are most enlightening. This is Bob Katz's website, which is very interesting (it used to have really slow servers though). Under the 'Media' drop down menu there is an 'Honor Role' of well mastered CDs, with monitor gain information.

I must have another root around myself when I get more time...TTFN!