Quote:
Originally Posted by Meyvn /img/forum/go_quote.gif
Vinyl's frequency response will always be short of what digital is capable of; it's in their nature. The argument of quality is not about frequency bandwidth. It's about simple purity of what is recorded. Analog requires no conversion to and from digital, and is not done in the same sense of units that digital is, so its sheer resolution is fundamentally superior; it's arguable that we've come to the point where that fact is negligible, but I personally don't think it is just yet. At any rate, now that digital recording is so popular, obviously that advantage of vinyl is now gone on 99% of modern recordings, but on older recordings, or the few newer ones that are purely analog, it's just going to be superior in resolution, the same way that film is better than HD cam.
|
I agree with Mevyn's analysis. I am a declared analogue addict, so I am biassed, but there is certainly more low level detail on LP; and this in itself is quite strange, as it has to be extracted from the background 'wouussh' of the surface noise. This limitation on the LP data retrieval process, means that an equalization curve must be applied to the music or other sonic content prior to it being cut onto vinyl, so that in turn the RIAA curve at the pre-pre will return it to its original shape, while reducing background noise and sibilance.
This appears to be well implemented, as the result to my ears is better than CD. In fact I wonder whether something similar should not be tried with digital. A CD player's data retrieval has to contend with the background electronic "wooussh" produced by the large number of electronic components. Upgrading diodes, caps, and the PSU all bring massive improvements, but the last whisper always seems to be missing, due, I suppose, to the low sampling rate of standard CD, which just seems to limit its ability to retrieve low-level information.
On the other hand the actual analogue recording puts no such limit on information retrieval when upgrading the LP system. My latest Garrard plinth seems to have far more resolution than its predecessor :
http://www.theanalogdept.com/anthony_hind.htm
The problem with LP is not the recording itself, but the mechanical process in retrieving this information. I mentioned record diamond interface noise, but the play-back system also has to contend with noise feed-back from the motor to the cartridge and acoustical feed-back from the speakers. Lp upgrades are therefore usually mechanical rather than electronic. Although, of course upgrading the prep-pre, in the same way you would want to upgrade the CD player, will also bring enormous benefits.
The resulting well tweaked LP system seems to be superior to CD in capturing all phenomena related to low level detail : capturing the long decay of a piano note (with CD the note suddenly dies), capturing the micro-dynamics of a quiet but fast change, such as the release of a gut string on a lute.
If you listen to a bowed instrument, on a CD, you can hear the note that is played, but you rarely hear the texture of the bowing, so that chamber music always seems superior on LP.
It seems that whatever one does, the low standard sampling rate of CD limits the possibility of extracting low level detail.
Perhaps part of the problem concerns "brain overload" : when the brain has to "work hard", it tends to lose information. Notice that a number of analogue audiophiles prefer mono over stereo. I think that is because stereo is recorded to 'confuse' the brain into believing that the sound image is 3D; but the brain has to work harder to rebuild the stereo image.
A good mono recording, is therefore somehow less tiring to listen to. I think that in a slightly different way, the brain has to work even harder with digital. It is almost as though part of the digital to analogue conversion has to take place in the brain. In fact, it is rather that the brain has to cope with the rather unnatural and slightly grainy result of the digital conversion.
I think most of the basic variants (improvements) proposed for CD players are trying to respond to the obvious superiority of LP over CD.
Some digital engineers thought that the problem stemmed from the difference between the lightening reaction speed of the DAC and the relatively slow reaction of the analogue op-amps that followed.
This could obviously cause a form of distortion, and faster op-amps were added. There was a slight improvement.
Others thought it was simply the number of components in the CD's DAC and analogue chain, so they put in the best components available to limit the presence of electronic noise : ultra-fast diodes, low noise caps and resistors. There was a further slight improvement, but now the greater transparency also revealed more of the digital nasties.
It was obvious that the low sampling rate was involved, and SACD, in the best case, is clearly superior; however, for standard CD, instead of raising the sampling rate, upsampling was used; but this caused all sorts of aliasing problems, and the added detail is somehow far less "natural" than the standard 16 X 4, or even 16 x 2.
The response to this by some audiophile engineers was complete removal of over sampling, "NOS", and the removal of brick-wall filters; but the sound, though more "natural", is even less dynamic and detailed, while the removal of filters lets through a constant ultra-high frequency noise. This may not be audible, but I understand can damage some tweeters.
All these variations were made in comparison to analogue systems, and I think this, in itself, proves that digital engineers acknowledge the superiority of analogue in most areas, at least while the digital recording uses such low sampling rates.