Have ordered Vol. 26, as pictured. Expected delivery sometime in mid April. Will evaluate in PCM and DSD256. May I ask what dac you are using?
Hi George,
Yes I'm with you that 44.1/16 should be adequate and that compressed Vinyl sounds worse as an uncompressed CD.
And yes, there are many CD's that are more than worth to listen at.
But in fact the question was here, is digital after so many years able to be unnoticed or transparent when included in a vinyl reproduction chain.
If not, this doesn't disqualifies digital, but just that we haven't still found the full 100% solution..
Hans
Yes I'm with you that 44.1/16 should be adequate and that compressed Vinyl sounds worse as an uncompressed CD.
And yes, there are many CD's that are more than worth to listen at.
But in fact the question was here, is digital after so many years able to be unnoticed or transparent when included in a vinyl reproduction chain.
If not, this doesn't disqualifies digital, but just that we haven't still found the full 100% solution..
Hans
Enjoyable in part because it was the most hi-fi option available at the time?
Another issue for those seeking more realism is that the realism just isn't always on the recording. There aren't that many recordings that aim to mimic live performance (apart from live recordings) at least when it comes to classical music.
Last edited:
I set up a series of digital audio converter auditions for media plant mastering use in the early 1990s. These were dark days by Mark's metrics even though a lot of recordings were made then which are now being used as reference playback media, but I digress...
We used a very good vinyl source (Linn Sondek w/Linn arm and MC cart I forget the model of) playing Sheffield LPs as one of the sources to digitize, along with various excellent recordings on 15 and 30 ips 1/4" and 1/2" tape, reproduced using our modded Ampex ATR100. The converters we auditioned were a Denon 900, Sony PCM1610 and PCM 1630, Pygmy, Sonic Solutions and Apogee converters.This series of tests revealed so much about...ALL the equipment being used, but to keep it short(er) and relevant to bohrok2610's post: the nature of vinyl noise floor was extremely interesting compared to the others. The pseudo-random relationships between the R and L channels create a kind of 'acoustic image' which the static hiss of tape and digital dither did not. It was more like hearing hiss through an acoustic space than just hiss.
Since those tests I firmly believe one of the reasons people like vinyl is this 'live' background noise. It engages our innate sense of interpreting the physical space we are in using our hearing, it is a highly evolved instinct, especially in blind people. I do NOT think people prefer vinyl due to its many and various forms of distortion or dynamic non-linearity. I attribute this special noise to the fact that the stylus is responding to perturbations from both sides of the groove, creating L-R correlations which are only partly rejected by the 90° relationship. Since there is an L-R correlation our aural cortex interprets them as spatial data to be analyzed. This is in stark contrast to the (almost) entirely uncorrelated noise from a tape head or digital dither source. People describing why they like vinyl litter clues in their descriptions supporting this: "more engaging," "sounds more live," etc... I have been preaching this idea for decades, but people will believe what they want to about vinyl and it's allure...personally I like the two or four square feet of high-resolution graphics to look at while listening to LPs, but that is just me.
This attribute of vinyl noise falls squarely into the hypothesis Mark has been preaching of a technical specification not accurately nor fully describing some sonic attribute. None of the test equipment we used (ST1510A, HP 400EL, HP spectrum analyzers, etc) would tell us what was different about the noise other than perhaps a specific spectral distribution difference.
BTW, the Apogees won that converter shootout and we incorporated them in our mastering suite afterwards, working with them to further improve the dither spectra...
Cheers!
Howie
Once got an Apogee Rosetta based on its good reputation. Listened to it that day, and put if up for sale the next. It was replaced with a Lynx 2, which was better, yet pretty awful compared to the later Crane Song HEDD 192 which I still have. First ADC I ever heard that sounded just as good at 44.1 as it did at 192. However, hate the dac in the HEDD. The box is only useful as an ADC, IMHO.
Last edited:
U
Given that Vinyl obviously has a unique kind of correlated noise other than tape, why don't all the Vinyl anomalies stay intact when we let the signal successively pass through 192/24 A/D and 192/24 DAC having more than enough dynamic range and BW to theoretically catch everything being offered.
I would expect 100% transparency.
Hans
Howard, one question:I set up a series of digital audio converter auditions for media plant mastering use in the early 1990s. These were dark days by Mark's metrics even though a lot of recordings were made then which are now being used as reference playback media, but I digress...
This series of tests revealed so much about...ALL the equipment being used, but to keep it short(er) and relevant to bohrok2610's post: the nature of vinyl noise floor was extremely interesting compared to the others. The pseudo-random relationships between the R and L channels create a kind of 'acoustic image' which the static hiss of tape and digital dither did not. It was more like hearing hiss through an acoustic space than just hiss.
Since those tests I firmly believe one of the reasons people like vinyl is this 'live' background noise. It engages our innate sense of interpreting the physical space we are in using our hearing, it is a highly evolved instinct, especially in blind people. I do NOT think people prefer vinyl due to its many and various forms of distortion or dynamic non-linearity. I attribute this special noise to the fact that the stylus is responding to perturbations from both sides of the groove, creating L-R correlations which are only partly rejected by the 90° relationship. Since there is an L-R correlation our aural cortex interprets them as spatial data to be analyzed. This is in stark contrast to the (almost) entirely uncorrelated noise from a tape head or digital dither source. People describing why they like vinyl litter clues in their descriptions supporting this: "more engaging," "sounds more live," etc... I have been preaching this idea for decades, but people will believe what they want to about vinyl and it's allure...personally I like the two or four square feet of high-resolution graphics to look at while listening to LPs, but that is just me.
This attribute of vinyl noise falls squarely into the hypothesis Mark has been preaching of a technical specification not accurately nor fully describing some sonic attribute. None of the test equipment we used (ST1510A, HP 400EL, HP spectrum analyzers, etc) would tell us what was different about the noise other than perhaps a specific spectral distribution difference.
BTW, the Apogees won that converter shootout and we incorporated them in our mastering suite afterwards, working with them to further improve the dither spectra...
Cheers!
Howie
Given that Vinyl obviously has a unique kind of correlated noise other than tape, why don't all the Vinyl anomalies stay intact when we let the signal successively pass through 192/24 A/D and 192/24 DAC having more than enough dynamic range and BW to theoretically catch everything being offered.
I would expect 100% transparency.
Hans
The model is oversimplified and incomplete.
Are you familiar with the WYSIATI human bias principle?
Agree. Which part of the recording/reproduction chain do you blame the most? Or is it just an equally contributing downhill slippery slope to misery
//
Hi Hans!
In my tests that vinyl noise floor aspect is largely or completely preserved in the digital recording, but I cannot tell you what level it was recorded at. Perhaps dither hides much of it if the vinyl and preamp is quiet enough? It is a good question why some DACs wouldn't preserve it...! The (according to Mark) terrible and only 20-bit 1993 Apogee DACs we used which did preserve the vinyl sound has some aspects superior to others? Good question.
Howie
Don't know about Apogee DACs. The Rosetta was an ADC only, IIRC.
Thing is, some people still like some of old Burr-Brown dacs, and or something like AD1862. The best sounding of those old chips have been displaced by the superior measuring ESS and AKM parts. When that happened there was a subset of people who said they couldn't stand the ESS sound, that it was fatiguing and glaring. More recent work, such as by KSTR, captured the time-domain waveform involved in the so-called ESS 'hump' distortion.
Also IIRC, at around -20dBFS there is some tendency for some sigma-delta dacs to produce more modulator noise/distortion which is correlated with the audio signal. Maybe it has a dynamic masking effect?
Also noise skirts can be seen at the base of FFT spectral lines. The fact that they appear as skirts means they highly correlated with the audio signal. When there are lots and lots of frequencies present at once, some of which are low level reverb tails, maybe the skirts have some masking effect.
There there is issue of modulator state variable settling after volume transients.
There are also idle tones.
There are element matching problems in the dac output switches and resistors that are made to look like gaussian noise on an FFT. The elephant in the room may that what looks like noise on an FFT can be all sorts of non-PSS signal-correlated noise and or distortion.
And noise doesn't just mean white noise like from resistor. Many of us are familiar with frying noise, popcorn noise, etc. Well, sigma-delta can produce its own unpleasant noises which may mask vinyl noise.
Some of the possible issues are complicated, hard to measure, and even harder to pin down how as to how human perception may be affected.
There are also non-PSS distortions that can occur in output stage opamps. If non-PSS then the artifacts may not show up well on an FFT.
None of that stuff is well represented in the 'bandwidth and steady-state noise floor' model.
Other than the above, think I have said a few times that it appears the ADCs tend to be the least of the problem with digital. Most modern dacs I have heard, in particular the ones optimized to look really good on standard measurements are the ones don't reproduce low level details so well under dynamic conditions and or maintain good phase coherence between channels. If there is a channel coherence issue, maybe its because the modulator for each channel will produce artifacts unique the audio signal on that channel. The noise produced dynamically may mask or decohere recorded vinyl noise between the channels.
IOW, I'm sorta trying to make a laundry list of everything I can think of that could be a problem. Its an attempt to try to counter WYSIATI bias.
Last edited:
Howard,
There is another thing about noise that always surprised me:
A former Riaa MC preamp, first number one on Stereophiles A+ list , the ca. $40,000 Boulder 2008, had the worst ever a-weighted S/N of 58dBA ref 0.5mV@1Khz about the same noise as the vinyl’s surface noise, almost 20dB below the competition.
An other example:
The referenced Mark Levinson No.33 mono block monsters, have a 84dB S/N ref 1W@8R in the audio band, where many others have 20dB less noise.
It's almost as if "the more noise the better", although I would find this hard to interpret.
Hans
There is another thing about noise that always surprised me:
A former Riaa MC preamp, first number one on Stereophiles A+ list , the ca. $40,000 Boulder 2008, had the worst ever a-weighted S/N of 58dBA ref 0.5mV@1Khz about the same noise as the vinyl’s surface noise, almost 20dB below the competition.
An other example:
The referenced Mark Levinson No.33 mono block monsters, have a 84dB S/N ref 1W@8R in the audio band, where many others have 20dB less noise.
It's almost as if "the more noise the better", although I would find this hard to interpret.
Hans
There seems a misguided attention to a reduction in noise, as higher noise doesn't necessarily dictate a reduction in transparency nor that a reduction in noise necessarily promotes transparency, as transparency exists in noisy records. It is considered that the error relates to the condition that any finite resistance has inherent noise. This is to suggest that resistance and noise is correlated, leading to the possible conclusion that resistance reduction can be what is necessary to promote transparency, this in either clearly storage vessels of electrons or in neutralizing them wherever they exist.
What I am suggesting is that the correlation to psycho-acoustic phenomenon is related to impedance as mistakingly attributed to noise. This is to suggest that nothing can hold a charge or hide from a zero impedance environment.
What I am suggesting is that the correlation to psycho-acoustic phenomenon is related to impedance as mistakingly attributed to noise. This is to suggest that nothing can hold a charge or hide from a zero impedance environment.
Last edited:
Mic'ed how? Pianos radiate sound over a large surface area. Mic's are sometimes pretty close.