"Would the Analog audiophile enjoy these "noisy" CD's more than a conventional CD mix with a very low noise floor?"
I don't think so. Those of us with analog preferences much prefer quiet records to noisy ones. Analog preference has to do with more actual music being stored on and retrievable from a vinyl groove than from a collection of digital pits on a Redbook CD, and to a lesser degree, perhaps the aesthetics involved. (If you disagree with me about the more music being retrievable thing, no sweat. I don't think we want to do the digital/vinyl debate again. I like both, and use them for different sorts of listening.)
Maybe, if the brain works the way a dac chip works. Some people have thought that the tape hiss that is (or used to be) audible on records enhanced the sound. I think Julian Hirsch wrote about this once.
No, the clicks and pops do.
Dither involves adding a small amount of noise to the signal. Vinyl reproduction involves (unintentionally, but inevitably) adding a much larger amount of noise to the signal. Whatever explains people's preference for one or the other, it isn't dither.
The other way arround:
An analogue audiophile would enjoy a vinyl playback as quiet as CD and it's nowdays possible to achieve.
Dither and analog noise are different. Dither enables a low level signal to be recorded instead of turned into distortion when the signal falls below the least significant bit (can't be quantisized). Added digital noise shaping doesn't make the ear more susceptible, it just increases the bandwidth of what can be recorded.
Analog recordings are preferable to many because the fidelity of the signal is more preserved than even the best digital. But the trade off (unavoidable in audio) of the analog hiss and pops are something we can ignore - much like the sound of our own breathing when we listen.
I agree with Marakanetz. The newer LPs I just bought on my first analog set up (Transrotor sirius, Jadis preamp DPMC, Jadis JOR, Dunlavy IV), seperate from digital set up does indeed have zero pops, low noise and sounds so great that I can't stop marvelling the sound even on temporary set up (just hooked up few weeks ago, roughed positioning, no isolation, no shelves, amps and preamps still on the floor) and sounds great out of box so to speak.It took years for me to get satisfactory sound from my main digital system set up w/even higher grade equipments. Of course, old uncleaned LPs are different story.
Not sure how this relates to the perceived enhancement of music by noise. Under certain circumstances, when noise is binaurally correlated and a signal is out of phase, detectibility of the signal can be enhanced. This phenomenon is called binaural masking level difference, BMLD or MLD in the literature (e.g. JASA). A demonstration of this effect is to imbed a sinusoid in binaural white noise, adjust the level so the sinusoid is just detectable, then reverse the phase of the sinusoid in one ear. Depending on the frequency, the out of phase sinusoid will become very detectable, and require 10 to 15 dB attenuation to again become just detectable. The noise can be shaped (e.g. pink), bandlimited and gated; greatest effect is found with low frequency sinusoids, i.e. under 1 kHz.
The argument that analog better preserves signal seems specious, unless the mechanically transduced sound is fed directly to the cutting lathe. I suspect current practice is to record to digital tape, given the problems of tape hiss and print through with analog tape. We used 1" tape running at 15" per second on Ampex consoles and never used fast rewind (just played the tape backwards). Nevertheless, you could hear hiss and print through in the lab. We were making tapes to be used to measure speech intelligibility of systems and speech perception of listeners.
Surface noise is not the only noise associated with analog playback, which is what I think Pabelson was really getting at. There is noise (rumble) associated with the table machinery itself, noise induced by record warps, RIAA equialization, inner-groove distortion, etc... That being said, a well set-up analog system playing clean, high-quality vinyl can sound very enjoyable, even lifelike as you approach S.O.T.A. systems. A good digital system can bring about the same level of performance with different qualities. Both can be great, as ylong as you don't compare too closely to real, live music...
I am not sure I made my self clear in the intial post;
As described by Gs5556, dither allows quantization of a low level signal to occur in the digital processing because it artifically raises the signal above the threshold of the least significant bit in a digital system.
I fully agree with replies from Gs5556 and others that a higher wideband noise floor is very different from dither, however, different though it may be, might a low level wideband noise raise the audibility of some low level musical details enough to take it above a human perception threshold? (above a kind of human quantization threshold)
An analogy might be a TV fed with a very low level of brightness in the input signal... so low that some low level details are lost in what are perceived as large completely dark parts of the screen. Adding a low level wideband noise signal to the same input signal of the TV would create a slight snowy appearance, just a shade or two above totaly dark, in the darkest parts of the screen. The question is whether this random low level wideband noise might allow a viewer to better discern some previsouly hidden detail in these dark parts of the screen?
Exactly right, Sdatch. I wasn't trying to be critical of vinyl. I was only trying to put the question of dither into perspective.
I am not an audio engineer, so I’m not sure I’m describing this correctly.
I thought the process of modern digital recording, mixing and mastering was done with very high resolution files, up until the music is encoded in the "distribution" format (on a CD being a PCM file with a sampling rate of 44.1kHz and 16 bit, bit depth).
To grossly oversimplify the issue, for argument’s sake, let’s say that the files in the studio were twice as large and detailed, compared to what the consumer purchased.
So the file in the studio might have samples like this:
0 1 2 3 4 5 6 7 8 9
But to distribute the music to a lower res CD, a "dithering algorithm" would need to be applied to that file, just for an example would look like this:
0 2 4 6 8
Sort of like taking a 12 megapixel JPG file and making a 256k thumbnail.
One of my engineer buddies says that the reason the more modern digital music sounds better is: 1) the analogue to digital converters, which take their input from say a microphone in the studio and convert its output into digital have become much more accurate (larger digital file sizes as well), and 2) with the newer CPU’s available of late, the "dithering algorithm" it can use is much more sophisticated, so it can figure out more accurately which samples to retain and which ones it can drop from the "distribution" copy.
I don’t do analog sources in my system, just isn’t my thing. I once attempted to do an "A-B" of a Mozart violin concerto, comparing a 44.1/16 source to a 192/24 source, after a lot of back and forth (getting the volumes the same, etc.) I thought the 192/24 file sounded just a bit cleaner and more open...then I discovered that the hi-res version had been "remastered"...probably the cause of my observation.
I’d enjoy responses to this posting from others, who probably know more about the subject!
The phenomenon shadorne alludes to in the first post from (only!) 10 years ago is generally known as "stochastic resonance." It’s a very real phenomenon that occurs in analog and natural biological sensory systems, including human audition. Broadly defined, "Stochastic resonance (SR) is a phenomenon where a signal that is normally too weak to be detected by a sensor, can be boosted by adding white noise to the signal, which contains a wide spectrum of frequencies." It applies to other sensory systems besides hearing, including taste and olfaction, as well as electro-mechanical sensing systems such as sonar. See:https://en.wikipedia.org/wiki/Stochastic_resonance
*and, if you REALLY want to dive in deep,*http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2660436/
So yes. it is possible that some people may indeed find their listening experience enhanced by at least a certain low but audible amount of surface noise from an LP. Dither might be considered in some senses to be the digital analog to analog stochastic resonance. In practice, however, they are functionally rather different, dither being applied at a very much lower level.
Let's turn it around and make it simpler than a+b = equation:
An analogue listener, as any listener of live music played in ANY venue, is exposed to ambient noise of live and natural performance while digital listener isn't no matter what kinda processing involved.
I like listening to music next to musicians in small venues, I like the way it sounds from orchestra hole or from the stage and I like to have company with musicians playing casual or simply rehearsing. There's no digital format can EVER get any closer than vinyl with any number of bits and/or megabits.
I know that lots of audiophiles don't support going to live venues because they're freaked that someone may sneeze or cough during... Imagine how these dummies would react ta any click'o'popie LMAO!
I prefer listening live. That's why my music collection is 95% analogue and the rest is few CDs and downloads. There's no CD or any digital format that can bring real live music.