You can, but what you do if your final release is at different sample rate? Resample? Ouch!
DSD is "universal" 1 Bit stream at 2.82 MHz recorded directly after A to D modulator that allows decimation down to desired sample rate / release format at a later stage - 44.1; 48; 96; 384; whatever ...
http://www.positive-feedback.com/Issue11/meitnerinterview.htm
SACDs as such are very precious archives, because they do contain that 1 Bit data. Unfortunate thing is that we (users) do not have direct access to that data and can not choose ourselves the way we'd like to listen to it. Current methods to play back SACDs are somewhat compromised.
P.S. As far as I know, first Sony DSD systems were built around CS5390 ADC where 1 Bit stream could be accessed at one of the test pins.
Last edited:
And what's one good reason for using a 1-bit modulator in an ADC? Nowadays cheap multibit ADCs with great linearity are available so we can record direct to PCM and bypass the nasty noise modulation issues that come from not being able to control the pdf of the noise in the shaping loop. Or has that problem been solved now?
P.S. Cirrus Logic has gone over to multibit S-D in their later designs - ever wondered why?
Last edited:
elektroj,
Thanks for posting the article. I have to take my time and read the whole article but I get the jist of the concept and sampling accuracy and bit length. No real reason to go over the 96khz number for practical purposes with 20khz wide bandwidth and even going as high as 25khz which should be more than enough it appears that 50khz sampling would still meet the Nyquist ratio. I'll read more when I have time. This article address bandwidth requirements but I will have to look at the jitter situation and clock implementation later. Thanks.
Thanks for your response also Abraxalito. Kind of jumped in here at the end of this thread and I will have to go back and read more of the thread. Don't know where this was fitting into the original thread of preamplifier without going back and reading what happened here.
Thanks for posting the article. I have to take my time and read the whole article but I get the jist of the concept and sampling accuracy and bit length. No real reason to go over the 96khz number for practical purposes with 20khz wide bandwidth and even going as high as 25khz which should be more than enough it appears that 50khz sampling would still meet the Nyquist ratio. I'll read more when I have time. This article address bandwidth requirements but I will have to look at the jitter situation and clock implementation later. Thanks.
Thanks for your response also Abraxalito. Kind of jumped in here at the end of this thread and I will have to go back and read more of the thread. Don't know where this was fitting into the original thread of preamplifier without going back and reading what happened here.
You're right, Kindhornman, this has drifted way off topic. I'll stop now.
Apologies to John for hijacking his thread.
I speculate that your CD player doesn't have the right kind of DAC chip in it. Stanley Lipshitz says (more to the point, shows) DSD is technically flawed, Lynn Olson says the SACD sounds distinctly different from the mastertape. I can't see any reason to disbelieve either of them. But then maybe you value other aspects in the sound of SACD vs CD to those I do. Or perhaps I have too poor an SACD player.
Now that's beginning to sound more promising...So what does SACD give up to 24/192 to your ears?
You have something against the Benchmark DAC1? In comparisons with CD I can tell very little difference between it and our Oppo BDP83 Nuforce edition.
Good Bluray discs of some live music concerts sound more life like than anything I've heard. It's not just one aspect of the sound, it just sounds more real.
Last edited:
Other than the DAC chip used and the layout and grounding, no nothing
Being S-D it won't do justice to 44k1/16 material.
I wrote that to say that at least in that case I couldn’t blame the medium or even the playback equipment.
And this original LP/CD release comparison of live opera recording from 1950-60
is not the only one that I have come across.
(Unless the cartridge is of a true “mono” construction), when you play a mono record with a stereo cartridge and watch the signals on a phasescope (or through an oscilloscope in XY mode) you’ll see gross deviations from mono.
All these are due to artefacts atributed to cut/replay process and equipment.
The most upstream (at the cartridge output)one shorts the two channels, the most effective is the accidental stereo to mono conversion.
George
Other than the DAC chip used and the layout and grounding, no nothing
Being S-D it won't do justice to 44k1/16 material.
What do you mean by S-D?
The measurements I've seen look very good. I've read no complaints about how it measures in audio mags.
Anyway I've used a number of very good DACs from PS Audio, MSB and very good CD players. The Benchmark does a very good job.
Last edited:
Thanks Pavel
The measurements are not in any doubt - it was designed to deliver them and achieves its goal. Plenty of adverse comments about its SQ that I've read , even though lots of customers are very happy with it. The fact that the designer found NE5532s as I/V the best sounding says quite a lot about the surrounding circuitry, layout and grounding
The measurements are not in any doubt - it was designed to deliver them and achieves its goal. Plenty of adverse comments about its SQ that I've read , even though lots of customers are very happy with it. The fact that the designer found NE5532s as I/V the best sounding says quite a lot about the surrounding circuitry, layout and grounding
Maybe it's just me, but BluRay looks really weird to me, different than DVD but in no way better. Tang versus orange juice. The players and discs are inexpensive these days, but I can't justify them.
No, I haven't done a controlled test.
For me the stocastic resonance of old film stock is part of the experience easily captured on DVD, I suspect most classic films would not look better on Bluray without some kind of signal processing. I started a thread and no one answered concerning the film "Battle of Britain" where it had the appearance of being filmed yesterday in HD video. I want to figure out what's happening.
I'm glad that we have had some positive input on the differences between different digital formats and some specific info on advanced digital formats.
I am somewhat behind on this, because I chose, long ago, not to pursue digital design, except to attempt to understand it, when the requirement comes up. Now is the time for me to update, so useful input on digital, its problems and improvements, is just what I need, as I must use digital more in my audio playback system, now that I have a digital playback worth listening through.
I would like to make an overall commentary about the design of audio systems, that is seemingly overlooked. That is: ' increased displacement'.
This can apply to autos, telescopes, analog tape recording, and even digital systems.
In autos, this is 'cubic inches' or litres of displacement.
In telescopes, this is the diameter of the lens.
In analog tape recording, it is both the speed and the track width of the tape being recorded.
Finally, in digital, it is the number of bits and the sample rate.
I know this is obvious to anyone who is technically inclined, BUT many do not see the intrinsic aspects of what I am getting at.
An analog example:
Back about 30 years ago, actually starting about 1/3 of a century ago, I was commissioned to make an analog mastering recorder operating full track, or 2 channels on 1/2'' wide tape, and a speed of 30 ips. I know that this seems obvious today, BUT at that time, there was NO existing standard in the USA for it. Hard to believe, isn't it?
It had been left behind for years, since 15 ips, 1/2 track was considered adequate for audio mastering, given improvements in tape, etc.
In any case, we made a series of analog mastering recorders that only recorded and reproduced at 30 ips, full track. This gave me about 5 times the 'displacement' over 15 ips, 1/2 track mastering, more than that by far for consumer (at the time 7 1/2 ips, 1/4 track tapes), and even far more than a cassette tape.
Now what did it give us? Of course, somewhat lower noise, increased bandwidth, but also better transient response, 10KHz 10us risetime square waves were possible, lower wow and flutter, and INCREASED AVERAGING of the imperfect tape coating so that it appeared to be 'rock stable' when played.
Now what was the alternative? IF you did NOT want to use tape in terms of dollars/minute, you COULD, if you were sophisticated enough, make a SUPERIOR RECORDING SYSTEM, that would eliminate some of the problems with lower speed and even modest track width recording. This was done by an esteemed competitor of mine, who I initially asked to work with me on making a mastering recorder, and he declined.
Ultimately there was a contest between the two competing technologies, and 'brute displacement' won (perhaps slightly) over less displacement, but more sophistication. Think of a modest size auto engine with a turbo charger, against a V12 high displacement engine for comparison.
You see, the high displacement won, because it averaged over the ripples or errors in the medium, better.
In my opinion, higher speed, more bits will do the same thing, ALL ELSE BEING EQUAL.
So a $100 SACD player, MIGHT not sound as good as a $1000 CD only player, because they are NOT made to the same standard. This is normal and obvious. That is why we seem to be talking about apples vs oranges when in comes to digital listening comparisons.
I am somewhat behind on this, because I chose, long ago, not to pursue digital design, except to attempt to understand it, when the requirement comes up. Now is the time for me to update, so useful input on digital, its problems and improvements, is just what I need, as I must use digital more in my audio playback system, now that I have a digital playback worth listening through.
I would like to make an overall commentary about the design of audio systems, that is seemingly overlooked. That is: ' increased displacement'.
This can apply to autos, telescopes, analog tape recording, and even digital systems.
In autos, this is 'cubic inches' or litres of displacement.
In telescopes, this is the diameter of the lens.
In analog tape recording, it is both the speed and the track width of the tape being recorded.
Finally, in digital, it is the number of bits and the sample rate.
I know this is obvious to anyone who is technically inclined, BUT many do not see the intrinsic aspects of what I am getting at.
An analog example:
Back about 30 years ago, actually starting about 1/3 of a century ago, I was commissioned to make an analog mastering recorder operating full track, or 2 channels on 1/2'' wide tape, and a speed of 30 ips. I know that this seems obvious today, BUT at that time, there was NO existing standard in the USA for it. Hard to believe, isn't it?
It had been left behind for years, since 15 ips, 1/2 track was considered adequate for audio mastering, given improvements in tape, etc.
In any case, we made a series of analog mastering recorders that only recorded and reproduced at 30 ips, full track. This gave me about 5 times the 'displacement' over 15 ips, 1/2 track mastering, more than that by far for consumer (at the time 7 1/2 ips, 1/4 track tapes), and even far more than a cassette tape.
Now what did it give us? Of course, somewhat lower noise, increased bandwidth, but also better transient response, 10KHz 10us risetime square waves were possible, lower wow and flutter, and INCREASED AVERAGING of the imperfect tape coating so that it appeared to be 'rock stable' when played.
Now what was the alternative? IF you did NOT want to use tape in terms of dollars/minute, you COULD, if you were sophisticated enough, make a SUPERIOR RECORDING SYSTEM, that would eliminate some of the problems with lower speed and even modest track width recording. This was done by an esteemed competitor of mine, who I initially asked to work with me on making a mastering recorder, and he declined.
Ultimately there was a contest between the two competing technologies, and 'brute displacement' won (perhaps slightly) over less displacement, but more sophistication. Think of a modest size auto engine with a turbo charger, against a V12 high displacement engine for comparison.
You see, the high displacement won, because it averaged over the ripples or errors in the medium, better.
In my opinion, higher speed, more bits will do the same thing, ALL ELSE BEING EQUAL.
So a $100 SACD player, MIGHT not sound as good as a $1000 CD only player, because they are NOT made to the same standard. This is normal and obvious. That is why we seem to be talking about apples vs oranges when in comes to digital listening comparisons.
Good point. The notion that arose with early S-D in some quarters seemed to be a fondness for sending the people who knew how to do precision circuits packing up and being shown the door. A little like the mantra that all is being magically reduced to "digital" even when most don't know what that means.
Not to patronize Hegel too much here, but it does seem a lot of engineering follows the thesis, antithesis, synthesis process.
Reading Linear Audio Vol. 4* and Bordelon's article about a top-flight DAC system, he mentions considering designing the DAC itself but being deterred after reflection. But it sounds like an intriguing idea, taken in the spirit of diy and our enterprises being mostly unconstrained by size and power consumption. E.g., what's to prevent an MSB from being an ampere?
Or a full-scale range along the lines of ancient analog computers, +/- 100V?Brad
*from whom I receive no promotional consideration. No, really, I swear.
Last edited:
John, Since you have talked about full track tape and mono playback recently I have to point out my recent discovery that ties these together. And that is the RCA mono counterparts to the Living Stereo's. It seems that every session RCA did in the golden age had two tape recorders going. They were AFAIK the same machine, just one was running 1/2 track stereo and the other running full track mono. By serendipity once I was given a Heifetz Brahms Tchaikovsky Violin Concertos RCA mono. I thought "well what ever, I'll give it a try". I put it on the TT and was in SHOCK at the resolution, richness and as you say the TRANSIENT RESPONSE! Oh! THE TRANSIENT RESPONSE WAS TO DIE FOR! Since then I have picked up more RCA mono counterparts to the living stereo and not a stinker in the bunch yet. Chet Atkins! you gotta' hear Chet in mono. The other good bonus with these pressings is they are lower cost and usually less abused by past owners. Enjoythemusic.com has some write-ups featuring the RCA Mono's.
Did the mono Brahms still have the terrible overload problems at the end of the last movement?
Can't remember, I'll try to check later today since I have a phono preamp proto sitting on my bench.
I grew up with the Living Stereo version of that LP, and it led me to believe that all violinists played as well as Heifetz, until I started to hear some others. But it does sound as if the recording engineer was asleep at the controls toward the end of that amazing performance.
People seem to forget that we once had REAL recording engineers, who knew the musical score and changed the gain when anticipating too much level for the recording. Also, really quality analog recording is VERY forgiving as to overdrive, above 0Vu, better than most any other way of gentle overload, in lieu of better intrinsic dynamic range or S/N.
- Status
- Not open for further replies.