Head-Fi.org › Forums › Equipment Forums › Sound Science › SACDs and DSD vinyl
New Posts  All Forums:Forum Nav:

SACDs and DSD vinyl

post #1 of 24
Thread Starter 
I keep reading about sacd's but in the last day or so I've been reading about dsd vinyls. How do they compare? I ask because I know you need a special player for the sacd's and the dsd vinyls can be played on a normal turntable. Sorry if its a stupid question but I keep reading different things and I'm getting a little confused.
post #2 of 24

In terms of dynamic range the whole vinyl system is obviously gonna limit it to below CD (16 bit) performance.

High frequency content is a bit more complicated. There shouldn't be a roll-off at 20 to 21 kHz but most of the ultrasound content will probably be mostly noise due to DSD's aggressive noise shaping.

Frequency response at those frequencies can vary by a couple (!) of decibels.

Vinyl is also gonna add harmonic and intermodulation distortion.

 

 

So I guess clearly audible differences with vinyl depend the master, not its format.


Edited by xnor - 6/14/13 at 1:21pm
post #3 of 24

So, DSD Vinyl...that actually was a new one to me.  What I thought it might be is a DSD copy of an analog master used to create a new lacquer master for new vinyl pressings.  The concept makes no sense at all, so I googled a bit.  Info is sketch, but that seems to be exactly what it is.   There may be some valid argument to DSD being a little better than 16/44.1 PCM, but there's no reason not to go to a higher sample rate higher bit depth format for an intermediate.

 

Why would you go through a digital process at all if you're cutting a new lacquer?  I dunno.  Possibly the DSD intermediate is used for processing (again, makes no sense), or remastering (which is the same thing as processing).  And, it's likely that the processing tools need PCM of some kind to feed their DSPs, so the path is analog > DSD > PCM/DSP/PCM > DSD > lacquer.   

 

Pardon my ignorance.  DSD Vinyl...why?

post #4 of 24
Quote:
Originally Posted by jaddie View Post

 

Pardon my ignorance.  DSD Vinyl...why?

Oh come on now.  Many have been convinced the DSD format is higher resolution in ways that matter.  Early on it was sold as being more analog like in the way it works.  Then there are plenty who think LP is the reference against which to judge all others.  So sure, DSD re-mastered to PCM would cast that evil digital spell over the resulting sound.  While DSD re-mastered to the reference system of LP might preserve more of that DSD goodness vs hirez PCM.  (we will ignore your likely being right it will have been PCM somewhere in the middle, don't let details ruin a good thing).

 

So DSD vinyl....why?  Because there is a market for it.  

 

DISCLAIMER::  your mileage may vary, the opinions expressed above by me may not be my opinions, yadda, yadda, yadda.................... 

post #5 of 24

Ah.  Thanks for that.  I thought I detected a little Blue Smoke. 

 

I think it's a little funny that the vinyl guys today think it's such a "reference".  In the days when tape and vinyl was all we had, that 'reference" was a constant source of frustration.

post #6 of 24

CD, SACD, LP... all are high fidelity formats and are capable of sounding great. CD and SACD don't have surface noise or inner groove distortion, so they have an edge.

post #7 of 24
Quote:
Originally Posted by bigshot View Post

CD, SACD, LP... all are high fidelity formats and are capable of sounding great. CD and SACD don't have surface noise or inner groove distortion, so they have an edge.


I don't know.  Probably CD, SACD are capable of being audibly transparent to humans.  LP, can sound good, can be very pleasing, but it is demonstrably not audibly transparent.  It is somewhat colored even at its best.  So in answer to the OP, vinyl DSD will sound very much like an LP though perhaps a very good one.

post #8 of 24
Quote:
Originally Posted by jaddie View Post

Ah.  Thanks for that.  I thought I detected a little Blue Smoke. 

 

I think it's a little funny that the vinyl guys today think it's such a "reference".  In the days when tape and vinyl was all we had, that 'reference" was a constant source of frustration.


Yeah, my thoughts as well.  Back in the day, reel-to-reel was better.  It just cost too much for the average Joe.  I sort of think of LP as being to reel what MP3 is to CD these days. 

 

Once gathered several friends and we listened to multiple recordings that we had on reel, LP and CD.  Obviously the mastering may  not have been identical.  But in every case the reel and CD were much more similar than LP.  LP was always the odd man out. 

post #9 of 24
LPs have to be mastered differently for technical reasons.
post #10 of 24

Yeah, LP's have to be done differently for technical reasons related to the limitations of vinyl. 

 

Comparing reels, LP and CD it was obvious that primarily the LP was a very different frequency balance.  It actually surprised us how very close the EQ must have been on reels and CD.

post #11 of 24

The need for slightly different mastering for vinyl relates to the fact that the maximum level that can be recorded and reproduced on vinyl changes with frequency, and the desire to record as  high a level as possible.  There is a maximum velocity at which the cutter stylus can move before the trailing facet of the stylus collides with the groove wall it just cut. Low frequency L - R material  at high levels can create too much vertical groove modulation so that the resulting groove won't hold a stylus, or the cutter hits the aluminum substrate of the lacquer blank.  However, beyond the maximum limits, vinyl is a flat frequency response medium that has no specific frequency response balance (assuming proper playback EQ and alignment.  

 

Analog tape has yet another maximum modulation characteristic governed by entirely different factors, primarily tape saturation, track width and speed, which occurs following an entirely different response vs maximum level curve.  However, tape too, when machines are properly aligned and calibrated, is a flat frequency response medium with no characteristic frequency balance.  

 

PCM (without emphasis) is different in that its maximum level curve is flat, so as compared to tape or vinyl, you can record higher levels of high frequency material on PCM (CD) than either of the other media.

 

However, and this is the key, the application of different mastering processes is driven by the desire to record the highest possible levels, or some other subjective goal.  All three mediums, when operated below their maximum, are considered flat response media, and all three will produce fundamentally identical results as related to frequency response.  Noise floors and distortions will be different, and the analog mediums have plenty of time-base issues of course.

 

When comparing reels, LPs and CDs, you are comparing the entire chain including mastering (and in the case of pre-recorded reel to reel tapes, high-speed duplication too), and are not simply comparing the end media.  Results may seem similar, but can largely be attributed to artistic choices during mastering, and less about the technical limitations.

 

I knew this was sort of deja vu.  This post mentions an experiment I did a few years ago:

http://www.head-fi.org/t/643349/npr-covers-cd-vs-vinyl-the-music-experience/15#post_9017350

post #12 of 24

Yes, all so true.  But we had 3 dozen reels, 3 dozen CD, and 3 dozen LP to compare.  The reel was much more similar to CD.  LP just was very different.  We were using a MC cartridge with SOTA amp, Rega arm and table.  Good quality TT and phono with very good response.  Repeated the experiment with another completely different system using Shure MM cartridge, SOTA TT.  Same results.  Reel machine was a Studer Revox in good condition in both systems.  Yes, all very subjective.  And the number of listeners was 5.  Different expectations from each, and we were easily able to agree upon the result.  Not rigorous, not blind.  

 

As a recent addendum of sorts.  I took a 24/192 recording, used a reverse RIAA curve, played it at appropriate level through a good phono with a PC soundcard.  The phono was already tested for flat response to 25 khz, with appropriately low distortion.  Sounded great, no obvious coloration.  So everything except the LP/cartridge itself seemed quite capable.  Again all subjective impressions.  I believe LP's are quite colored by the whole process of pressing into vinyl.  

 

Once owned almost all the first couple dozen mono Mercury recordings.  Most were made with wires, bare wires on ceramic insulators, running a couple miles back to quality reel machines.  They were of very high quality.  Strange to hear about them using so much wire.  Later, "portable" high speed reels were put in vans to record just outside venues when Mercury did recordings.  Pretty capable, and at the same time 'enhanced' with triode electronics, and reel master, and LP mastering.  Quite interesting the various parts of this audio perception among humans. 


Edited by esldude - 6/16/13 at 8:22pm
post #13 of 24
Quote:
Originally Posted by esldude View Post

Yes, all so true.  But we had 3 dozen reels, 3 dozen CD, and 3 dozen LP to compare.  The reel was much more similar to CD.  LP just was very different.  We were using a MC cartridge with SOTA amp, Rega arm and table.  Good quality TT and phono with very good response.  Repeated the experiment with another completely different system using Shure MM cartridge, SOTA TT.  Same results.  Reel machine was a Studer Revox in good condition in both systems.  Yes, all very subjective.  And the number of listeners was 5.  Different expectations from each, and we were easily able to agree upon the result.  Not rigorous, not blind.  

Where did the reel to reel recordings come from?

 

How was the LP system's adherence to the RIAA curve confirmed?

 

How was the tape machine's performance confirmed?

 

And...no measurements?  All subjective,non-blind, a group of 5 listeners?

 

Are these older LPs or something cut more recently? LP mastering gear hit it's peak of fidelity in the mid 1980s, though there was some excellent lacquer cut during the entire microgroove era.   

 

Quote:

Originally Posted by esldude View Post

As a recent addendum of sorts.  I took a 24/192 recording, used a reverse RIAA curve, played it at appropriate level through a good phono with a PC soundcard.  The phono was already tested for flat response to 25 khz, with appropriately low distortion.  Sounded great, no obvious coloration.  So everything except the LP/cartridge itself seemed quite capable.  Again all subjective impressions.  

Confirmation of the RIAA curve by the total system is critical. You can't ignore it, it must be confirmed.  You have to use a test record recorded without RIAA equalization (Shure STR-100 is one), then plot the result against a theoretical RIAA curve to see where the response deviates from ideal.  Making a 24/192 recording with inverse RIAA curve is nice, but ignores the cartridge and it's load. You can still be way off the curve that way.

 

Quote:

Originally Posted by esldude View Post

 

  I believe LP's are quite colored by the whole process of pressing into vinyl.  

No doubt, but the "test" is seriously flawed, and in the end, you're still comparing the entire vinyl chain to the entire CD chain, and not comparing the media specifically.  There are very few cases where the CD and vinyl were made from identical masters.  In my test, that's precisely what we did.  The masters were identical. The vinyl play system was tested for perfect RIAA tracking from the stylus through the preamp. 

Quote:

Originally Posted by esldude View Post

Once owned almost all the first couple dozen mono Mercury recordings.  Most were made with wires, bare wires on ceramic insulators, running a couple miles back to quality reel machines.  They were of very high quality.  Strange to hear about them using so much wire.  Later, "portable" high speed reels were put in vans to record just outside venues when Mercury did recordings.  Pretty capable, and at the same time 'enhanced' with triode electronics, and reel master, and LP mastering.  Quite interesting the various parts of this audio perception among humans. 

You are referring to the 1951 Chicago Symphony sessions recorded at Orchestra Hall, and recorded at Universal Recording across town where the recordings were made by legendary Bill Putnam.  While well known that an equalized phone line was used, it is unlikely that it was bare wires on ceramic insulators.  It is, however, entirely possible to equalized what was then probably 22ga twisted pair for flat response to 15KHz or better over that distance, even including a trip to the central office first.  The equalizer would have been passive, followed by about 20 to 25dB of flat amplifier gain stage. The clue to the fact that they were standard equalized phone lines is that those recordings are noted for the slight phone line crackle. Phone companies discovered early in the 1900s that twisted wire pairs had much higher noise immunity, so that system was adopted and would have been standardized easily by 1951, but line noises was always a problem for high quality audio transmission.  Mercury also experimented with 3-track 35mm mag film with good results because the tracks were wider than tape, the base is thicker for less print through, and the speed was slightly faster at 18 inches per second (90 feet per minute is standard 35mm speed at 24fps).   Their vans included full-track mono Ampex 300 machines along with 35mm magnetic 3-track machines.  

post #14 of 24

As long as the three sources were playing at the same time, line level balanced and direct switchable, I think you can be pretty sure that the impression that vinyl was more different from CD than reel to reel could probably be determined... even without a double blind test.

post #15 of 24
Quote:
Originally Posted by bigshot View Post

As long as the three sources were playing at the same time, line level balanced and direct switchable, I think you can be pretty sure that the impression that vinyl was more different from CD than reel to reel could probably be determined... even without a double blind test.

Not disagreeing that the vinyl was different, just questioning the cause.  The observation that there is a difference points misleadingly to vinyl itself,  when as you know it's the entire chain from master to phono preamp output, and there's a lot to be different in there.  The vinyl itself isn't the largest cause for the difference at all.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › SACDs and DSD vinyl