Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

Return to Digital Drive

Red Book Riddle Solved

99.62.117.221
Posted on July 14, 2009 at 00:06:39
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 The advent of the Meridian 808.2 CD player and reviews in Stereophile and TAS of it prompted me to write this piece on standard resolution ("Red Book") digital audio. To be sure, advancements in playback have been very real over the past two decades. But with the advent of minimum-phase / "apodising" filters and the massive improvement they brought to CD, a lot of people will be wondering if 16 bits/44.1kHz sampling was enough on disc, for playback.

The answer, after decades of debate and endless argument amongst audiophiles and audio writers seems to be YES - to this big question. Beyond mike technique, venue, etc. an encoding format has three main areas of concern: capture (sample rate, etc.) production and playback. Many audiophiles think there are only *two* main areas concerning a format - recording and playback. This is wrong - there's one more in the middle !!

As it turns out, the areas affecting CD's sound were the last two - production and playback. The production problem was solved years ago and was an instant fix, as I show below. But first an overview: 16-bit digital encodes 95db of dynamic range - roughly 25db more than a large symphony orchestra produces. And orchestras produce far more "range" than a jazz combo or string quartet - both of which have ranges in the 50's db. This fact links up with important findings in recent years by audio researchers. One of them is Bob Katz - Bob says that for original sources, 16 bits/44.1kHz sampling IS ENOUGH for all kinds of music. By "original source", he means the encoded music before it was processed. This observation was derived from controlled listening experiments done over the years, that either Katz or others conducted.

The problem as (all) recording engineers know, were the intermediate calculations done on the signal after recording - the "DSP processing" done in production. With DSP, there were losses in a 16 bit recording - losses that eroded into the music - esp. symphonic-style and grand piano.

But in the early 1990's a technique came along that cured the problem - 20-BIT RECORDING. This would absorb all the production losses that hurt 16 bit recordings. Right after this, another technique called noise-shaping came out - another improvement - but this time for playback (on disc). As for the CD-standard, I quote Malcolm Hawksford from the March 1996 Stereophile article "Bits is Bits ?": "When correctly dithered using triangular PDF dither, a 16-bit digital audio signal possesses a dynamic range of 93.3db with zero distortion and zero noise modulation. The 16 bit format holds the possibility of even *higher* subjective dynamic range - up to 18db more - with minimally audible noise-shaping employed during CD mastering".

Knowing this, why would *anyone* desire a new audio format ? Zero distortion, zero noise and dynamic range of 110db - that's "not enough" ?? One of the arguments for more bits was that "it's not the number of bits available but the number used at one time" (during recording). Yes, but that was a *production* issue - not a capturing or playback-disc issue !! The same goes for the sample rate - *if* more than 44.1kHz was needed, we got it, in capture - starting 16 years ago. Sample rates of 48, 88, 176 kHz, you name it - it was used. The problem was that audio writers never explained the (true) virtue of 20-bit/high sample recording and that 16/44 might be enough, on disc.

All this wasn't enough, however, as advances in playback were needed as well. But the issues in playback seem to concern the laws of electrical current and audio-chasis design much more than the "band-aids" needed to improve CD sound. Yes, CD was "upsampled" etc...but let's not forget that some CD playback systems *didn't* upsample and produced great sound - like Zanden. In other words, most of the parameters concerning digital playback had to be improved anyway - no matter how many bits were on disc.

That said, it's not a surprise that CD sounds as good as it does today. Some audio writers are struggling to hear a difference between the Meridian 808.2 CD player and true Hi-Rez audio. In this month's TAS, Robert Harley (strangely) did not compare the 808.2 with his Hi-Rez files. That's probably because CD sounds a lot closer to Hi-Rez than he thought it could (or should). But it's right in the specs.......

These (new) observations are not a surprise to everyone. Below is a link which takes you to avguide.com. Simply scroll down (to the bottom of the first piece) and click on the links.


RE: Red Book Riddle Solved, posted on July 15, 2009 at 09:40:47
chris.redmond2@bushinternet.com
Reviewer

Posts: 10288
Location: Lancashire.
Joined: January 21, 2001
 "To be sure, advancements in playback have been very real over the past two decades."

Well, I'd say ye olde TDA1541 chips were better than anything around nowadays which is why the likes of Zanden and AMR are universally praised, and the older Philips swing-arm transports were a quality alternative to TEACs VRDS mechanism when now there is none, but I would agree that the technology has improved over the years so that the aforementioned players which use these chips now wring far more of the potential from these older chips than previously; shame there's no more CDM1 transports around.

As you note however, CD has all the resolution and dynamic range that 99% of recordings need, and in fact the few JVC XRCDs I own are so resolving and dynamic that they are exactly as fatiguing as the DVD-As I used to listen to.

Too much emphasis is put on resolution when even at a live, unamplified performance, unless you are sat next to the performer detail is lost and low level noise influences what reaches your ears.

CD done right is natural, involving, highly resolving and musical, and certainly leaves very little open to criticism to my ears at least.

SACD may be an option for those who listen to classical and there are some great players out there and also great DSD recordings, but all too often we see listeners comparing CD to SACD on the same player only to conclude that CD is poor, when in reality it is the CD replay of the SACD player which is poor.





Today is a gift - that's why it's called the Present.

Best Regards,
Chris Redmond.
Understand the big picture and everything is clear, posted on July 14, 2009 at 21:09:14
Adriel
Audiophile

Posts: 673
Joined: October 13, 2001
Contributor
  Since: January 18, 2010
 Redbook is an easy scapegoat for lack of listening enjoyment. But factors that create the "unenjoyable Redbook house sound" are legion, not just "Meridian" or "minimum-phase" or "20-bit".

Magazines are forever hawking pieces of digital gear as the final solution to perfect sound forever. All who believe the magazines in this regard are destined to be unsatisfied in the long run.

The answer to the Redbook riddle is The Holistic Approach.

We need to understand that there is a minimum threshold at which playback starts to become humanly enjoyable (in varying degrees of enjoyment). Beneath this threshold the playback is inhuman and difficult to enjoy, and rejected by the person.

The amount of leeway we have before sinking beneath the minimum threshold of enjoyment is the Margin of Error.

If everything is perfect from recording to production to playback, the Margin of Error is maintained at its initial level from the moment the sound wave hits the microphone. Within recording are many variable factors, as with production, as with playback. But I won't list them all.

With increasing data rates we increase the Margin of Error. That is the suffice summary.

Some production methods maximize the Margin of Error, and some do not. For example, this recording from Sony BMG:




As you can see, the Margin of Error in this case is already in the negative, as the digital format has already been exceeded. The rest of the playback chain will determine whether this is bearable or utter trash.

So, take the large view. It is not one small factor such as "minimum-phase" that drops Redbook down, but all factors.

Holistic allocation of time to address as many factors as possible, rather than on one single factor spit out by the latest audio fad machine, is the best way to maintain the Margin of Error, and maintain day to day enjoyment with more frequency.

The Holistic Approach, also explains why vinyl playback systems can be fatiguing and unenjoyable at times. Transient playback factors are going to influence all listeners; it just so happens that with Redbook listeners the Margin of Error is razor thin to begin with.

8-bit (or lower), 12khz (or lower): Why is that so hard to enjoy? It starts life with such a negative Margin of Error, rescuing it is very difficult and will probably be unsuccessful in broaching the minimum level of enjoyment.

RE: Understand the big picture and everything is clear, posted on July 15, 2009 at 14:19:40
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009

This "music" was trash when it was conceived in the minds of the musicians and producers. Were they not mentally and sonically deranged they would never have allowed such a travesty to be released.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

Not Really..........., posted on July 14, 2009 at 16:31:03
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "The advent of the Meridian 808.2 CD player and reviews in Stereophile and TAS of it prompted me to write this piece on standard resolution ('Red Book') digital audio. To be sure, advancements in playback have been very real over the past two decades. But with the advent of minimum-phase / 'apodising' filters and the massive improvement the brought to CD, a lot of people will be wondering if 16 bits/44.1kHz sampling was enough on disc, for playback."

This has been an ongoing debate..... In fact, the debate was "most-heated" in the early 1990s. (I remember debating this with the late Gabe Wiener.)

"The answer, after decades of debate and endless argument amongst audiophiles and audio writers seems to be YES - to this big question."

I think it's "no", based on what I've heard from CD relative to vinyl and DVD-A. (But DVD-A has other problems that I think are worse- very fatiguing.)

"Beyond mike technique, venue, etc. an encoding format has three main areas of concern: capture (sample rate, etc.) production and playback. Many audiophiles think there are only *two* main areas concerning a format - recording and playback. This is wrong - there's one more in the middle !!

"As it turns out, the areas affecting CD's sound were the last two - production and playback. The production problem was solved years ago and was an instant fix, as I show below. But first an overview: 16-bit digital encodes 95db of dynamic range - roughly 25db more than a large symphony orchestra produces. And orchestras produce far more 'range' than a rock band, jazz combo or string quartet - all of which have ranges in the 50's db. This fact links up with important findings in recent years by audio researchers. One of them is Bob Katz - Bob says that for original sources, 16 bits/44.1kHz sampling IS ENOUGH for all kinds of music. By 'original source', he means the encoded music before it was processed. This observation was derived from controlled listening experiments done over the years, that either Katz or others conducted."

I think the 44.1 kHz sample rate is more of a shortcoming for Redbook than the 16-bit wordlength........

"The problem as (all) recording engineers know, were the intermediate calculations done on the signal after recording - the 'DSP processing' done in production. With DSP, there were losses in a 16 bit recording - losses that eroded into the music - esp. symphonic-style and grand piano."

Everything with digital audio recording and playback involves "DSP".....

"But in the early 1990's a technique came along that cured the problem - 20-BIT RECORDING. This would absorb all the production losses that hurt 16 bit recordings."

The only thing 20-bit recording did was enable dithering off the master to the 16-bit media..... Dithering could have still been used in the original A to (16 bit) D process......... It IMO was more a marketing gimmick than anything else.

"Right after this, another technique called noise-shaping came out - another improvement - but this time for playback (on disc)."

Noise shaping was simply dither with a "frequency shaped" noise signal. (As opposed to plain white noise.) No big deal.

"As for the CD-standard, I quote Malcolm Hawksford from the March 1996 Stereophile article 'Bits is Bits ?': 'When correctly dithered using triangular PDF dither, a 16-bit digital audio signal possesses a dynamic range of 93.3db with zero distortion and zero noise modulation. The 16 bit format holds the possibility of even *higher* subjective dynamic range - up to 18db more - with minimally audible noise-shaping employed during CD mastering'."

The "zero distortion" is due to the harmonics being removed due to the 22 kHz bandwidth limitation. (It's only zero above 11 kHz.) It even removes distortion that was in the signal *prior* to digitization.

"Knowing this, why would *anyone* desire a new audio format ?"

Because a lot of the claims were more "spin" than reality. (There really is no such thing as "zero distortion".)

"Zero distortion, zero noise and dynamic range of 110db - that's 'not enough' ??"

If it were actually true...... The RFI from digital playback can be demonstrated by placing an AM radio near an active DAC. No system is totally immune to this noise. (This is why I prefer using optical connections to confine the RFI to the DAC portion of the system.)

"One of the arguments for more bits was that 'it's not the number of bits available but the number used at one time' (during recording). Yes, but that was a *production* issue - not a capturing or playback-disc issue !! The same goes for the sample rate - *if* more than 44.1kHz was needed, we got it, in capture - starting 16 years ago. The problem was that audio writers never explained the (true) virtue of 20-bit recording and that 16 bits might be enough, on disc."

Theoretically, 18 bits has been established as a minimum to encode recordings without dither. Dithering is not an exact science, and I would rather accurately encode of the signal at the lowest detectable levels using more bits than approximate the signal using dither and encoding to fewer bits.

"All this wasn't enough, however, as advances in playback were needed as well. But the issues in playback seem to concern the laws of electrical current and audio-chasis design much more than the 'band-aids' needed to improve CD sound. Yes, CD was 'upsampled' etc...but let's not forget that some CD playback systems *didn't* upsample and produced great sound
- like Zanden. In other words, most of the parameters concerning digital playback had to be improved anyway - no matter how many bits were on disc."

I'm afraid that while I agree the playback needed to be "improved", very few designers really have a good grasp regarding how to tangibly "improve" it.

"Tha said, it's not a surprise that CD sounds as good as it does today."

I think the playback sounds worse, by and large. A reason why there has been a new interest in vinyl playback. If the CD actually improved the way some have claimed, I really think vinyl playback would have been buried. Right next to 8-track.

"Some audio writers are struggling to hear a difference between the Meridian 808.2 CD player and true Hi-Rez audio. In this month's TAS, Robert Harley (strangely) did not compare the 808.2 with his Hi-Rez files. That's probably because CD sounds a lot closer to Hi-Rez than he thought it could (or should). But it's right in the specs......."

There are also people who have a hard time distinguishing CD from MP3.......

"These (new) observations are not a surprise to everyone. Below are four articles I dug up, starting with J. Gordon Holt in the mid-1980's and continuing forward. A little perspective and some solid answers is what we need."

I read the articles. I think the findings were more personal to his experiences than universally accepted.


RE: Not Really..........., posted on July 14, 2009 at 23:09:14
Werner
Audiophile

Posts: 1851
Joined: September 30, 1999
 "Noise shaping was simply dither with a "frequency shaped" noise signal. (As opposed to plain white noise.) No big deal."

Nope.

You are thinking of shaped dither.

Noise shaping is something a tad more complex.

It uses frequency-shaped feedback of the instantaneous (re)quantisation error in order to move the total quantisation noise out of the bands
where the ear is most sensitive. As a result, in these bands a dynamic
range is attained that exceeds the theoretical value for an N bit
system.

No big deal? Without noise shaping there would be no delta-sigma
ADCs and DACs, no DSD, ... These rely on exactly the same technique
to allow quantisation down to 1 bit with a reasonable (grin) dynamic
range in a limited band.



bring back dynamic range
RE: Not Really..........., posted on July 15, 2009 at 14:15:10
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "Without noise shaping there would be no delta-sigma ADCs and DACs, no DSD, ... These rely on exactly the same technique to allow quantisation down to 1 bit with a reasonable (grin) dynamic range in a limited band."

True enough, but with 1 bit it is not possible to eliminate noise modulation, and this may account for the "euphonic" sound some people hear with DSD. And for this reason, among others, all new high end converters have gone to multi-bit delta sigma.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

I Stand Corrected........, posted on July 15, 2009 at 00:40:36
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 My suspicion with noise shaping is I've never seen it explained to where it makes sense to me (aside from with the erroneous definition I've used). I am skeptical of its purported advantages, since I've never heard any positive sonic correlation with recordings known to have been encoded with it.


A very simple example of noise shaping, posted on July 17, 2009 at 14:54:57
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 Here's a simple example of noise shaping. Suppose that you quantize a number (say it was 0.5) but it was rounded down to 0. The result is noise in the amount of -0.5. If this error is fed back (negative feedback) into the next sample, which we will assume is also 0.5, the resulting sum, 1.0 will be rounded to 1. The residual error to carry forward will be 0. If the signal sits at DC, then the output will be a steady stream of 0101010101, and the result will be no noise at DC, but a lot of noise at period 2. The noise has been shaped.

Now this simplistic example shows one of the problems with noise shaping, namely limit-cycle behavior. This can be broken by adding randomness, i.e. dither, but now things get more complicated. The dither noise itself has to be shaped as well, unless it can be subtracted out at the very end. Proper dither linearizes the quantization and removes the limit cycles, which come from a non-linear system with feedback.

If you really have to understand how all this works, then read the attached link. It provides a comprehensive explanation, both to/from analog and from one digital bit depth to another. Caution: Tough Mathematics! Note that this thesis does not explain how 1 bit DSD works. Not surprising, because 1 bit delta-sigma modulation does not actually work properly, being sufficiently non-linear to preclude any rational analysis. (If you doubt this, read the user's manual for the Weiss Saracon conversion software, and how there is special klugery to fix "latch up".)

Noise shaping in systems with a 44.1 kHz sampling rate are of marginal value, as the noise is moved from one audible frequency to another slightly less audible frequency. Noise shaping works much better with higher sampling rates, where the noise can be moved to frequencies that are octaves above the audible range.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

+1, agree with most you said, Thks....................NT, posted on July 14, 2009 at 18:48:57
PS Yeo
Audiophile

Posts: 198
Joined: September 5, 2001
 NT

RE: Not Really..........., posted on July 14, 2009 at 17:47:09
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Why is there always someone in denial ? No matter what happens - with them, it doesn't. Now we have Mr. Krieger (again) poking his head out and in response, I'll say this: Anyone who thinks that CD *isn't* improving is someone I won't waste my time with. But quickly:

There are major "headroom" benefits to 20-bit recording. Ask any recording engineer. THESE ABSORB THE LOSSES THAN HURT 16-BIT RECORDINGS.

And the people/articles that I listed include professionals who have a lot more credibility than you do. And specs are specs - deny them all you want. CD is Hi-Rez !!!!!!
20-bit processing, posted on July 14, 2009 at 19:30:52
ephemere
Audiophile

Posts: 804
Joined: December 31, 2002
 I didn't know anything about this until your original post, but let me see if I understand what the 20-bit thing from the 1990s was all about.

If you do a sequence of arithmetic calculations and round to the nearest, say, 0.001 after each calculation, you'll eventually end up with accumulated error of 0.003, 0.004, etc. If you chain enough of these rounded calculations, you'll lose an arbitrary amount of precision in the result. But if you round to the nearest 0.000001 after every step and save the rounding to 0.001 for the final result, you'll almost certainly get the one and only right answer. This is why calculators have internal precisions that are several digits longer than the number of digits they can display. In your words, it's like "headroom".

I'm guessing this is what 20-bit processing is -- adding 4 extra bits for the number-crunching parts so that you can chain a bunch of processing steps, and when you round to 16 bits at the end of the day, you get the right answer.

digital photo analogy, posted on July 14, 2009 at 20:40:49
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 There is a general analogy to the use of jpg files for images.

One can display a beautiful, accurate photo with the jpg format, but if you want to do any image manipulation or editing, you don't want to do it with a jpg. You really want the original uncompressed lossless image if you need change the image.

That said, jpg files are a great way for images to be stored, transmitted and viewed, particularly at the consumer level.

Since the manipulation of individual audio tracks in a recording studio can involve a tremendous number of individual tracks that get individually processed and eventually combined to make the stereo playback mix, one wants a much healthier margin in the raw tracks for the processing and mixing that follows.

I've heard many consumer CDs that give absolutely stunning reproduction. It is certainly possible to put great sound on a CD because I've heard it. Unfortunately, I've also heard many CDs including recent releases that are downright insufferable. Of course, the same is also true in both directions for LP records.

For me, it is not the format. Its what the artists, engineers and producers do with it. New high rez formats will not fix the failure to use what we have now.

RE: digital photo analogy, posted on July 14, 2009 at 20:56:55
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 That was the beauty of 20-bit recording - it gave the "headroom" they needed to process. The 16-bit copy was finally an "access to the master".

The bad CDs you sight are probably not from independant-acoustic labels. I'd try labels like Chesky, Naxos, Hyperion and Harmonia Mundi.

recording fads and fashions, posted on July 15, 2009 at 06:43:37
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 >> The bad CDs you sight are probably not from independant-acoustic labels.
>> I'd try labels like Chesky, Naxos, Hyperion and Harmonia Mundi.

Actually, I have many great sounding CDs from "independent-acoustic" labels.

While the labels you mention often have excellent sound quality, I sometimes find their music not quite to my taste. (That was something that started with Sheffield back in the 70s. The music on a few of their releases was downright cringe-worthy.)

For me, I have to like the music first. It is then a bonus if the recording happens to be exceptional.

The sad thing is the music business is just as susceptible to fads and fashion as any other business. In the 1950s and 60s they worried about making the recordings stand out on AM radio.

Too-close miking is a perpetural audiophile favorite - hearing a singer breathe in may make you feel you have a personal relationship going with the artist, but it is never something you hear live unless you're doing a face-to-face duet with them. Spot or highlight miking is still a common fad even with classical music. Even the specialty labels you mention can fall prey to the current fads.

Today we have the "loudness wars" in which producers have willfully forgotten that CDs have a dynamic range that exceeds 90 dB.

So the technology is there right now for good recordings. It is just a question of whether they bother to take advantage of it.

RE: recording fads and fashions, posted on July 15, 2009 at 10:09:50
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 If you don't the music from these labels (are many other audiophile labels like DG, Pentatone, etc.) then you're in a bind. Because this means you like pop/rock and these simply are not going to be "distant miked". They never were. It's even questionable whether we should be playing this kind of music back through a high-resolution system.

Or maybe you like jazz - *now* you have a chance !!

I'm open to all kinds of music, posted on July 15, 2009 at 12:48:55
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 Don't get too presumptuous in telling me what I like and I don't. ;-)

I've been collecting music for about 40 years now and have well over 40,000 songs on my server. It runs the gamut from classical, jazz and Ghana funeral drumming to garage rock. Even have some 2nd gen open reel copies of studio masters.

There are times when fun and games in the manipulation and processing department are entertaining and appropriate, but there are lots of recordings where they just can't leave well enough alone.

And I didn't say I didn't like the music from the labels you mentioned. I just don't find them a guarantee of musical performance. And, while the audiophile labels may not be guilty of playing in the loudness wars, they are not completely immune from the current fads & fashions of the audiophile community.

RE: I'm open to all kinds of music, posted on July 15, 2009 at 17:43:16
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Don't know what the "current fads and fashions of the audiophile community" means....

The labels I mentioned sound about as good as it gets - for natural sound with proper perspective (very little, if any, close-miking).

an example, posted on July 15, 2009 at 20:43:41
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 >> Don't know what the "current fads and fashions of the audiophile community" means....

Fads and fashions can be difficult to spot when one is member of the group involved.

As an example, the other night I listened to the multi-disk set of the Emerson Quartet playing the Mendelssohn string quartets. This DG recording is quite well regarded.

It actually came with a short video. It was interesting to note (and they even commented on this) that 14 microphones were used. Eight "spot" mikes, two more group stereo mikes and others for hall ambiance. While the spot mikes were not directly on the instruments as one would find in pop or rock recordings, they were certainly far closer - perhaps three feet - than any position anyone outside of the musicians themselves would have been at a performance.

All of this for four players. Those 14 channels had to be mixed to create the final recording released to the public.

I think you'll find this is not uncommon for classical music these days and is a good illustration of the prevailing "fashion" in recording these days. I guess you could call it the "Kodachrome" effect where recordings are often a glossy, larger-than-life version even when their goal is capturing a live event.

RE: an example, posted on July 15, 2009 at 21:31:35
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 It's a rare audiophile label that uses 14 mikes !! The vast majority use far fewer. It's how *close* the mikes are that matters most........

Unfortunately..., posted on July 16, 2009 at 05:48:00
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 >> It's a rare audiophile label that uses 14 mikes !! The vast majority
>> use far fewer. It's how *close* the mikes are that matters most........

...the majority of classical recordings (or any other type for that matter) do not discuss the recording techniques used.

However, when one does see a session photo, it is a rare event to see only two mikes present. I just looked through some classical LPs (which have nice big photos.) The cover of Leonard Slatkin's recording of Prokofiev's 5th has a nice photo of the whole orchestra. I counted at least 10 mikes, including the two main stereo mikes which were only eight to ten feet in front of the stage from the nearest players.

I know there are exceptions, but frankly, multi-miking with highlighters in classical music is still more common than not.

Of course, we're pretty far from the original subject, but an interesting segue nonetheless.

RE: Unfortunately..., posted on July 16, 2009 at 08:22:44
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I just don't see multi-miking as the sin you make it out to be. In the end, it's the sound...and multi-miking sounds about as good (to me) as Blumlein. As long as they weren't *close* miking.

There is a small group of Blumlein junkies out there inc. audio writers Steven Stone and Robert Greene. And Chesky, Water Lily and Tony Faulkner (engineer) make them. I'd go this route for this unusual miking technique.....

RE: Unfortunately..., posted on July 16, 2009 at 20:23:39
mls-stl
Audiophile

Posts: 1702
Location: St. Louis
Joined: April 11, 2006
Contributor
  Since: January 14, 2010
 >> I just don't see multi-miking as the sin you make it out to be

Please don't put words in my mouth. I referred to multi-miking as a "fad and fashion" which is hardly the same thing as a sin, either mortal or venal.

Some degree of multi-miking is simply SOP for the vast majority of recordings. I accept that. It makes for generally pleasant recordings and offers much greater post-production flexibility than simpler miking arrangements. I find nothing surprising about its popularity in the recording industry. It simply is what it is.

However, if you look at the way humans hear in stereo (where transit arrival time, frequency response, high frequency shadowing, phase differences and other factors all play a role) the use of multi-miking really distorts the original sonic 3-D image. Having multiple sources for the same sound means that an artificial mix has replaced the original image.

One analogy would be a food manufacturer replacing a natural ingredient with an artificial one for the sake of improved consistency.

I'm not passing judgement one way or the other; there is a lot of wonderful music available and many enjoyable recordings. But I do marvel at times over the audiophile community's selective purism. In the reproduction chain, some departures from high standards of authenticity are unforgiveable while others are rarely even noticed. I guess audiophiles are human after all. ;-)

RE: Unfortunately..., posted on July 16, 2009 at 21:02:55
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I'm not putting words in your mouth - you obviously don't believe in multi-miking.

But this is a minority standpoint - most of the 9 and 10 star reviews (for sound) on Classics Today.com are multi-mike recordings.

And I have Blumlein's in my collection - but they do not sound better.

You are entitled to your opinion...but you're whipping-up your argument for one-point miking into a science which at this juncture, it is not....

RE: 20-bit processing, posted on July 14, 2009 at 19:47:20
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 If the quantizer rounds the LSB (16th bit) directly from the analog signal, it shouldn't make any difference.


RE: 20-bit processing, posted on July 14, 2009 at 21:25:22
ephemere
Audiophile

Posts: 804
Joined: December 31, 2002
 No, I don't think that's right. Imagine doing a sequence of arithmetic operations and rounding to a resolution of 0.001 every time:
  1. X = 4.001 (the original value)
  2. X = X / 4
  3. X = X - 1
  4. X = X * 1000
  5. X = X + 1
The right answer, rounded to the nearest 0.001, is 1.250. But if you round to the nearest 0.001 at each step then you'll end up with a value of 1.000 -- an error 250 times greater than your resolution of 0.001.

But if instead you rounded to the nearest 0.00001 at every intermediate step, and then rounded the final answer to the nearest 0.001, you'll get the right answer of 1.250.

This, apparently, is what the 20 bits were for, to provide an extra 4 bits of resolution for the intermediate values of the arithmetic computations. The starting value (the quantized analog sample) and the final value (the number put on the CD) may only be 16 bits, just like in the example above the starting and final values are rounded to the nearest 0.001. But you must have more resolution for internal arithmetic manipulation or else there could be a severe degree of error in the final numbers put on CD.

RE: 20-bit processing, posted on July 14, 2009 at 23:06:49
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 The mathematical operations don't really depict what takes place in A/D...... This was more an issue with early calculators, where limited bit depth in fractional calculations resulted in lost accuracy.

With a 20-bit calculation, the lower four bits might equal half the value of the 16-bit LSB, which in rounding would raise the LSB up a bit value..... But with a 16-bit A/D, if the analog signal below the LSB is at a level of half the LSB, it too could be rounded in the exact same manner.


RE: 20-bit processing, posted on July 15, 2009 at 01:03:22
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Krieger - there you go again - making unsubstantiated claims.

Ask any (acoustic) recording engineer and they'll tell you that 20-bit recordings sound better than 16, after processing. This, mainly for symphony and grand piano music.

Why is this ? Dither is not the answer !!
RE: 20-bit processing, posted on July 15, 2009 at 01:14:16
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "Ask any (acoustic) recording engineer and they'll tell you that 20-bit recordings sound better than 16, after processing. This, mainly for symphony and grand piano music."

We're talking about CDs..... All CDs are 16 bits.


RE: 20-bit processing, posted on July 15, 2009 at 01:37:18
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Wrong - we're talking how best to *get* sound on CD. A 16-bit recording does not sound as good as a 20-bit one, after processing. There's a reason - high-bit capture.

RE: 20-bit processing, posted on July 15, 2009 at 11:38:27
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 Enjoy your "20-bit" CDs...... ;-]


RE: 20-bit processing, posted on July 15, 2009 at 17:45:07
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 That's 16-bits. But with more "bits" in sound if dithered/noise-shaped....

RE: 20-bit processing, posted on July 15, 2009 at 23:56:02
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "That's 16-bits. But with more 'bits' in sound if dithered/noise-shaped...."

But whether it's dithered off a 20-bit master or directly from the original analog, it makes no tangible difference......

If anything, dithering directly from the original analog signal should be a little "truer" to the original analog signal than first encoding to an intermediate 20-bit master....... Because no intermediate digital conversion is perfectly transparent.


RE: 20-bit processing, posted on July 16, 2009 at 00:56:26
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Should...should...should. But might not.....

The bottom line is that we don't *have* to dither !! The Stereophile link (and common sense) shows why.

RE: 20-bit processing, posted on July 16, 2009 at 18:39:20
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "The bottom line is that we don't *have* to dither !! The Stereophile link (and common sense) shows why."

But even if you don't dither, there is still no advantage to an intermediate 20-bit master over a direct 16-bit transfer.

The dithering is what provides the "advantage" you purported, not the extra bits from an intermediate master.


RE: 20-bit processing, posted on July 16, 2009 at 10:10:05
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 Most 16 bit recordings sound better when dithered. I base this on my experience with many tape transfers to 44/16 format and a few original digital recordings as well, where the final stage in production involves reduction of 44/32 or 44/24 material down to 44/16. There are a variety of dither algorithms in my toolkit and I choose the one that sounds best for each recording.

Analog noise may mask the quantization distortion caused by lack of dither, but analog noise generally has a Gaussian probability distribution and hence is only an approximation to what is needed to eliminate all distortion and noise modulation, a triangular probability distribution precisely calibrated as to its amplitude. Even better would be subtractive dither, but this requires a decoder to subtract out the dither noise during playback and there is no standard non-proprietary way of doing this. (With subtractive dither one gets the benefit of dither with no noise penalty.)

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 16, 2009 at 12:39:52
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I don't doubt this. The issue I raised in my opening piece is that with *20-bit* recordings - dither/noise-shaping shouldn't be needed...

RE: 20-bit processing, posted on July 16, 2009 at 13:51:03
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 I have no problem with a 16 bit recording being marketing as having been made from a 20 bit production process or 20 bit master (today it would probably be 24 bits). However, it is inaccurate to call it a 20 or 24 bit recording. Back in the early 90's this might not have been deceptive, but today there true 24 bit recordings available to the consumer, so a claim that a RBCD has more than 16 bits constitutes deceptive advertising. In this sense, I agree with Todd.

For these improved recordings to benefit from the extra bits, then dither will be needed at the time of conversion into 16 bits. Without dither noise the extra bits are simply thrown away. Conversion software such as Izotope allows for control over the amount of dither noise added, perhaps to allow the extra noise to be reduced if the original recording has a certain amount of suitable noise to start with.

Incidentally, my experience has been pretty clear: most of the problems that I have heard when converting to 44/16 from higher resolution occur in the conversion stage from 176/24 or 88/24 down to 44/24, and not in the stage from 44/24 to 44/16. Since much of the material I have been working with comes from cassette tape, one would think that neither the sample rate nor bit depth are particularly critical. However, this is has not been the case. Some cassette masters sound very good when played on my Nak CR-7a deck, before and after digitization at 88/24 for eventual release in 44/16.



Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 16, 2009 at 15:31:50
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 No-one is saying that RBCD contains more than 16 bits. If dither/noise-shaping is applied - then beyond the 16-bit spec can be heard - subjectively. Many people (apparently) have a problem understanding this.

There is *no such thing* as 24-bit ADCs. You should know this. The last four bits on a (24-bit) ladder are not functional.

Finally, with intermediate calculations being done on a 20-bit - not 16 bit - capture, we shouldn't have to dither/noise-shape.

RE: 20-bit processing, posted on July 16, 2009 at 18:55:11
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "There is *no such thing* as 24-bit ADCs."

The Prism AD-2 is a 24-bit ADC...... One of many...... [-;


RE: 20-bit processing, posted on July 16, 2009 at 19:09:31
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 But the dynamic range only goes to the 20-bit level. Go to Bob Katz's website for more.

RE: 20-bit processing, posted on July 16, 2009 at 17:48:57
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 There is *no such thing* as 24-bit ADCs. You should know this. The last four bits on a (24-bit) ladder are not functional.

Most of the newest converters use delta-sigma modulators running at a very high oversampling rate and a bit depth around 4 to 8 bits. These have sufficient performance in the audio range that their performance is somewhat constrained by the 24 bit word length. Similarly on DACs. Accordingly, manufacturers are beginning to offer 32 bit converters. It is fair to say that the low order bits of a 24 bit converter are functional, albeit less functional and certainly less important than the higher order bits.

"Finally, with intermediate calculations being done on a 20-bit - not 16 bit - capture, we shouldn't have to dither/noise-shape.

This statement is simply false and demonstrates both practical and theoretical ignorance.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 17, 2009 at 01:21:17
Werner
Audiophile

Posts: 1851
Joined: September 30, 1999
 Tony,

true 24 bit performance over the audio range, at room temperature, is
virtually impossible due to thermal noise. Present-day's top-class ADCs
perform at 22-23 bits at the very best, which is rather good enough.

The emergence of 32 bit DACs is mind-bogglingly idiotic.


bring back dynamic range
Why 32 bit converters, posted on July 17, 2009 at 08:31:03
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 While thermal noise may be a problem, particularly with ADCs, its effect on audio quality at very low levels is benign, and in any event it is less of a problem in converters than with microphone preamplifiers.

There is a simple reason for going to 32 bit converters which use delta sigma modulators: guard bits. Take the case of a DAC, such as the ESS SABRE, which is being driven by the output of a sample rate converter that uses 64 bit floating point calculations. There are two conversions required, from 64 bit floating point to 24 bit fixed point in the computer and then from 24 bit PCM at the normal sample rate (e.g. 176.4 kHz) via a delta sigma modulator to something like 8 bits at 40 MHz in the DAC chip. Apart from the cost of I/O bandwidth there is little reason to skimp on the 24 bits, so a slightly better result will be obtained by using 32 bits. Note that even with TPDF dither there are high order correlations between the signal and the noise, and these correlations can be reduced by a lot by adding more bits to the output.

While there are other ways to achieve as good a final result, they will involve more processing in the conversions and the extra complexity allows for the possibility of various unwanted interactions between the algorithms used (e.g. noise shaping in the conversion to 24 bits interacting with modulator algorithms). If a few extra bits are affordable, they cause all of these problems to disappear, in accord with the KISS principle.

The greatest benefit of going to 32 bits is likely to be that it will remove the excuse the modulator designers presently have on what constitutes "good enough" performance. They will be forced to work on better algorithms. And once they have done this, the actual DAC circuit designers will be under pressure to improve the performance of their designs.

It is always easy when technology is progressing across a variety of bottlenecks for a lazy designer of one portion of a system to conclude that there is no need to improve, because other portions are the bottleneck. I've encountered people like this in the computer industry many times. These people may be individually correct, but when a group of people think this way progress is impeded. So, going to 32 bit converters may not be a big deal today, but it does represent progress.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 16, 2009 at 18:02:57
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Is this false or ignorant ?

RE: 20-bit processing, posted on July 16, 2009 at 19:18:11
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "Is this false or ignorant ?"

I don't know, but it could be true..........

This is the first article I've seen that expressly questions the benefits of dither for CD recordings..... The only way to find out its benefit is to compare two tracks from the same analog signal, one with dither, the other without. (Even dither apps sound different..... See link.)

I agree with Dave K on one point- The ambient noise is likely high enough to provide sufficient dither on its own. There are not many acoustic environments whose quietest levels relative to the peak levels of the music performed are lower than -90 dB. (The only environment that is void of ambient noise is where the music is 100 percent digitally synthesized. The next best thing is an anechoic chamber.)


RE: 20-bit processing, posted on July 16, 2009 at 18:27:01
CarlyBoy
Is this false or ignorant?


No, but the conclusions you're drawing from it are. In Keith's example, the background noise in the recording is already high enough to provide effective dithering on its own, without the need for additive dithering when decimating down to 16 bits. If you look at Figure 1, it's >10 dB above the 16-bit noise floor.

But in cases where the background noise floor is low enough that you'll actually benefit from using 20 bits, you will need to add dither when decimating down to 16 bits for delivery. Otherwise, you'll lose any information content below the 16-bit quantization noise floor.

It's contradictory to say that 20 bits are required in the production process but you don't need dither when decimating down to 16 bits.
RE: 20-bit processing, posted on July 16, 2009 at 19:05:48
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I support - and believe in - dithering/noise-shaping. But I thought that there isn't much information below the 16th bit - when processing from a 20-bit capture. In another words - RBCD's specs are beyond what we can hear OR what playback systems can reproduce.

If we can dither a 16-bit recording, why did the whole world go to *20 bit* recordings in the early 1990's ? Weren't there "headroom" advantages ?

RE: 20-bit processing, posted on July 16, 2009 at 20:39:33
CarlyBoy
 Playback systems can reproduce better than 16-bit dynamic range, but it's doubtful we can hear it except with headphones or in an anechoic chamber.

I should backpedal a little bit. You're right that there's headroom advantages to recording with 20 bits - it's easier to avoid clipping while still capturing the low level stuff. And there are benefits to doing non-linear intermediate processing like compression & effects with a longer word length.

I guess my point is that if the production process is 20-bit or 24-bit, there's no advantage to throwing away some of the information content by truncating to 16 bits without dither.
RE: 20-bit processing, posted on July 16, 2009 at 21:11:17
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 So - if we're recording at 16/44, then process the signal, will there be any loss in sound quality ? If you say yes, then that explains why we needed 20-bit recording !!!

*And* why we might not need dither, as Keith Howard showed in Stereophile.

RE: 20-bit processing, posted on July 16, 2009 at 21:39:57
CarlyBoy
 One thing to be clear on is that Keith isn't saying we don't need dither. His point is that the background noise in the recording is often sufficient to provide the required dither.

And I'm not going to argue that we don't need 20-bit recording. I think we'd be best served if the playback format was the same as the recording, with as little processing in between as possible. But if we're stuck with 16/44.1 as the delivery format, we may as well make the most of it.
RE: 20-bit processing, posted on July 16, 2009 at 22:20:39
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 But we must process anyway, no matter where we start.

I was simply positing that by capturing at 20-bit/88kHz, etc., any losses in production are moved away from the 16/44 level on disc.

I am still not convinced that dithering/noise-shaping is needed, knowing this fact. I guess it depends on the venue's acoustics.....

RE: 20-bit processing, posted on July 17, 2009 at 19:55:26
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "I guess it depends on the venue's acoustics....."

Yes, that and the technical details of the original recording plus any subsequent processing.

I have yet to notice a large gain with the material I work with when doing noise shaping. But on at least one occasion there was a big loss in the form of added harshness. This in this case the original recording had come from a 44/16 live recording and there had been some gain changes and other processing to create a 44/24 master. It may be that the original already had some noise shaping, and a "double dose" is definitely not a good thing to have.

I have never found any benefit from disabling dither Since I hate all forms of distortion and don't mind low level noise, I can't see any reason for avoiding dither, except in the case where there are no extra bits present that will be discarded (example being a 16 bit recording that has been padded out to 24 bits and that is going back to 16 without any further processing).

Of course having some headroom is great to prevent clipping on the original recording or losing too much resolution. Before going to 16 bits one can normalize the levels and so take full advantage of the 16 bit format. This makes up for a lot of experience and luck on the part of the original engineer. However, it is possible with experience to get the levels right. For example I did this one time when filling in for the regular broadcast engineer from WGBH-FM for a mono broadcast of a Boston Symphony concert in Sanders Theater. He told me to use the usual microphone placement above the podium and to set the levels during the rehearsal. Then he told me a magic number of dB to turn down my gain. This would make up for the musicians playing louder in front of an audience. Sure enough, when the concert came off the levels were perfect!

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 17, 2009 at 20:17:18
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I respect your experience...but I am still not convinced that dither is needed in all situations. Actually, I'm shocked it's needed at all - being that any production losses are now moved away from the 16/44 spec. Wasn't that the benefit of recording at 20-bit/88-96kHz ?

I would like to see more engineers describe their experience with high-bit/high sample with dither. I think engineers are so used to using dither that it's automatic. But is it really needed ? If it is - I'd like to hear why.

What's wrong with a format that's zero in distortion, zero in noise and has a dynamic range of 95db (at full resolution) ??

RE: 20-bit processing, posted on July 16, 2009 at 18:47:12
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 A further point to be made is that the entire problem is somewhat different at the higher sampling rate, because there is more noise to start with because of the greater bandwidth. Indeed, it is possible to get good sound at 176.4 KHz sampling rate with as little as 12 bits, or possibly even 8 bits. (In the limit, good sound is possible with only one bit at a sampling rate of 2.8 mHz.)

Bit depth is not a particularly good measure of digital audio resolution. If a single measure is needed, then it should be total bit rate, which at least has the support of Shannon's information theory behind it.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 16, 2009 at 19:12:41
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 The bit rate of SACD is *four times* that of CD. Yet, to my ears...and in the opinion of many others, it doesn't sound better.....

Many others ?, posted on July 17, 2009 at 04:26:22
Metralla
Audiophile

Posts: 6502
Location: San Jose, California
Joined: January 30, 2001
 Barry Diament has gone on record as saying he doesn't like the top end of DSD as much as other hi-rez recording techniques.

And I recall others posting that they were not enamoured of the higher frequencies of SACD compared to DVD-Audio.

But that is comparing different media whose resolution is greater than Redbook.

When you say (comparing SACD to CD):

Yet, to my ears...and in the opinion of many others, it doesn't sound better.....

you need to at least provide a couple of references. You and one other is not MANY.


Regards,
Geoff
RE: Many others ?, posted on July 17, 2009 at 05:45:14
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 For one, reviewers are struggling to hear a difference between the formats - or prefer RBCD. One is Sam Tellig, who said recently "he doesn't like SACD as much as CD". Another is Roy Gregory, editor of HiFi Plus (U.K.). Roy said that listeners could go "either way" concerning CD vs. SACD - via the multi-format player he was reviewing.

I could go on and on...but my impressions of the Meridian 808.2 CD player compared to SACD (via Esoteric) had me preferring the Meridian. One reason why was its more organic presentation - SACD has a glossy and "gleamy" effect that I found annoying. And always did, BTW....

RE: Many others ?, posted on July 17, 2009 at 20:49:38
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "SACD has a glossy and "gleamy" effect that I found annoying. And always did, BTW...."

You may have heard the problems with DSD. This format is theoretically flawed because it can not be properly dithered and this may account for the glossy effect that others have called euphonic. Or it may just be an artifact of the particular player you were using.

I suggest auditioning higher rate PCM, 88/24 or 176/24. If you have a computer audio setup there are some download sites that offer test samples at various rates.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: 20-bit processing, posted on July 14, 2009 at 19:34:08
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Exactly !!!! I wish more people would realize this.

Very interesting post., posted on July 14, 2009 at 10:44:06
DSG
Audiophile

Posts: 245
Joined: March 9, 2005
 Unfortunately, I am not an engineer so I am not qualified to pass judgement on almost all of the technical information presented in this forum. However, I do read it with interest and this post is certainly interesting.

I grew up with vinyl and since the advent of CD I preferred its convenience, such as no cleaning, no pops and ticks, small form and easy storage, portability and ability to program tracks and skip tracks. However, at its inception, the sound of red book audio did not seem right to me. It was not until about ten years ago that I began enjoying red book music playback and completely abandoned vinyl with no regrets.

I was not aware until the statements made by the OP that there is a technical reason why CD, and now for me also computer playback of red book music, is capable of sounding really good. My ears told me this long ago.

"Some audio writers are struggling to hear a difference between the Meridian 808.2 CD player and true Hi-Rez", posted on July 14, 2009 at 07:38:39
Metralla
Audiophile

Posts: 6502
Location: San Jose, California
Joined: January 30, 2001
 As much as I am happy to have an improvement in CD sound, I would be very surprised if the above were true for a seasoned listener.

Regards,
Geoff
a matter of expectations ..., posted on July 14, 2009 at 16:44:02
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 I truly believe, in an absolute world (best software/hardware) SACD is superior to CD. The problem is, however, we don't live in an absolute world.

Although I realize this is an older 808 model, I supplied a review of the original Meridian 808 in comparison to EMM/SACD.

I found this review interesting, simply because I could relate to the writers position ...

"I am surprised to be writing these words, since SACD has never fully convinced me before now. Sure, the bass is exceptional, and I like the detail, but I've often found the treble unconvincing and the music to have lost some of its excitement. I've blamed the medium and remained happy with my vast collection of CDs."

That's very near my exact reflection with the majority of SACD players I've auditioned.

"The EMM throws a larger image where each instrument is more precisely located, and the level of detail is even higher. Perhaps it is just easier to make out fine details because the acoustic space is so large that the sounds do not run into each other."

That's exactly right, to my ears, soundstage width & depth, the "image", is a prerequisite to hi-end sound quality - with any medium. I've heard far to many players, both CD & SACD, collapse the image, becoming tonally bright in the process, as the player struggles to cope with difficult peak passages. The exact same can be said for analog reproduction, the best 'tables offer spectacularly wide & deep images, even during peak moments, more a kin to what you'd hear in a real world environment.

"The SACD layer is much more palpable and present than the CD layer. Intimacy is increased, the bass line is firmer and more defined, the top more open, the detail greater, the image more realistic. Every aspect of performance is improved, and improved markedly. Now it is easy to understand why people fuss over their inconvenient analog turntables rather than switch to the ubiquitous CD. CD is an approximation to the analog signal and so of course is SACD. But SACD is a much better approximation and therefore more faithful to the original sound. Good enough to challenge vinyl head on and miles ahead in convenience. I'm not saying the CDSA puts vinyl to shame or competes with analog master tape. Let's just say it is much harder to dismiss digital on principle when it sounds like this."

When he said ...

"Let's just say it is much harder to dismiss digital on principle when it sounds like this."

Exactly, and that's the key to any medium, when music is reproduced with emotional impact, when sounds like real-music, you stop caring about the medium, and simply enjoy the music as if it was real.

I thought the exact same thing when I finally (after near two decades of disappointment) heard CD replay sound like real music. I was in shock! Although I still considered analog far superior, I was surprised how "musical" CD could sound.

So, even though I own a far superior analog system, I also own a CD that allows me to listen for hours on end without feeling like I'm missing something special (which in reality, I am). The same goes for SACD, it can sound convincing enough to be enjoyable for hours on end, and this despite the fact that the best SACD I've heard, although clearly better than "ubiquitous CD", was certainly not nearly as wonderful, big & dynamic as my analog system.

It's a matter of experience, perspective, and expectations.

TB1
RE: a matter of expectations ..., posted on July 14, 2009 at 17:29:54
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 T-Bone: There is *no way* that SACD sounds better than a (new) minimum-phase filtering CD device - esp. the Meridian 808.2. Expensive - yes, but it shows what could be done....

"*no way* that SACD sounds better than a (new) minimum-phase filtering CD device - esp. the Meridian 808.2", posted on July 14, 2009 at 21:13:05
Metralla
Audiophile

Posts: 6502
Location: San Jose, California
Joined: January 30, 2001
 Haven't heard the Meridian unit for myself, so I can't comment.

Describe the discs auditioned and the SACD players used that enabled you to make this generalization.

Regards,
Geoff
RE: "*no way* that SACD sounds better than a (new) minimum-phase filtering CD device - esp. the Meridian 808.2", posted on July 14, 2009 at 21:42:27
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Hybrids from Pentatone, BIS, Telarc and others...

The players was Ayre and EMM - for SACD and Meridian's 808.2 for the CD layer. Comparisons were level-matched to .1db.

One interesting thing was that BIS, an SACD label, has been recording in high-bit *PCM* since 2005 !! That means if you loved SACD through this label, you love PCM.....

RE: "*no way* that SACD sounds better than a (new) minimum-phase filtering CD device - esp. the Meridian 808.2", posted on July 17, 2009 at 09:58:49
David Pritchard
Audiophile

Posts: 30
Location: New Mexico
Joined: February 5, 2008
 If you go to the SaCd.net forums you will find many people who are not happy with the sound qualities of the BIS recordings that are PCM recorded. I believe there is a significant change in BIS sound from their older recordings done in DSD to their recent PCM products.
David Pritchard

well ... I've heard this all before ..., posted on July 14, 2009 at 17:38:25
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 In fact, the 808.2, if indeed it is as advertised, won't be the first CDP that "showed what could be done" with CD.

In my experience with relatively expensive equipment, especially CD players, only a rare few have met my expectations regarding hi-end sound quality. That said, the same can be said for much of the SACD players I've heard ... so ...

Anyway, until I hear otherwise, I stand pat ... BTW ... have you heard the 808.2 compared to the EMM, or other expensive SACD players?

TB1

RE: well ... I've heard this all before ..., posted on July 14, 2009 at 17:52:36
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Yes, I have heard the 808.2 - at length. John Atkinson's (and now Robert Harley's) comments on this unit are *no fluke*. Read the reviews, go listen to the player.....

I did read the review ... and ..., posted on July 14, 2009 at 18:23:29
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 I'm certain that the Ayre C5 is an excellent SACD player, but I wish JA had compared the Meridian to a equally priced SACD player.

Still, it's interesting that he still preferred SACD via the $6K Ayre compared to CD via the $16K Meridian ...

"Tonally, spatially, the presentations were virtually identical. In the end, there was a difference, though not one I was expecting: The hi-rez playback had slightly better defined and slightly more extended low- and mid-bass with the bass guitar, tom toms, and kick drum on Live at Merkin Hall, which in turn led to a very slightly less congested lower midrange.

Ultimately, my preference was still for the hi-rez versions of the music, but it was a closer-run thing than I had expected."

TB1

Need to compare against an Ayre w/ the MP Upgrade, posted on July 16, 2009 at 07:51:13
BobM
Audiophile

Posts: 1192
Location: Long Island, NY
Joined: September 24, 1999
 Ayre is addressing this same issue with their MP upgrade. Now it would really be interesting if they compared the Meridian to a C5Xe-MP on the CD layer to see how the 2 slightly different approaches correlate.

Bob

A gentleman is best defined as someone who knows how to play the accordion ... and doesn't.
RE: Need to compare against an Ayre w/ the MP Upgrade, posted on July 16, 2009 at 17:57:38
RLF
Audiophile

Posts: 460
Joined: July 22, 2000
 "Now it would really be interesting if they compared the Meridian to a C5Xe-MP on the CD layer to see how the 2 slightly different approaches correlate."

According to John Atkinson he is going to explore the sonic differences between the Ayre and the Meridian in a future Follow-Up (page 73 in the July 2009 Stereophile review of the Boulder player)

RE: I did read the review ... and ..., posted on July 14, 2009 at 18:32:35
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 "This led to a very slightly less congested lower midrange". *That* was the main advantage.

"Very slightly" means what it says - and only in one sound category, after extended listening !!

Harley's comments in the current TAS may go even further. If CD was "low-rez", then why is this such a horse race ??!!

good point ... but, posted on July 14, 2009 at 21:35:35
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 >>If CD was "low-rez", then why is this such a horse race ??!!<<

Good point.

I've thought exactly that a number times, and when you consider the limited software, the cost of SACD playback has always been a hard sell for me.

>>"Very slightly" means what it says - and only in one sound category, after extended listening !!<<

OK, perhaps the difference were negligible, but you're still comparing a $6K player to a $16K player, and still, the $6K player was "slightly" better. You could argue that the Ayre was better simply because of DSD, but you'd be trumping that format, and this agenda seemed never to be your intent.

But think about it, if anyone owned a large SACD collection, which player would offer them the best - bang for the buck - sound wise?

A mega-buck great sounding CD player may indeed be all that it is, but it doesn't really represent progress until the sound quality filters down to realistic prices. Until then, who cares if a $16K meridian was totally awesome. It's been done before at those price levels.

I'm more impressed when a reasonably priced CD player can compete with a $6K SACD player. Perhaps, the Meridian G08 can, that would be a fairer test.

TB1
RE: good point ... but, posted on July 14, 2009 at 22:00:46
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 The bottom line is that CD sounds virtually as good as Hi-Rez - no matter the price. And it's still improving.......

This *will* trickle-down - give it time. Bryston's CD player & DAC and Audio Research's DAC-7 (all reviewed in TAS) are closing the gap - even as we speak.

RE: good point ... but, posted on July 14, 2009 at 22:19:18
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 >>The bottom line is that CD sounds virtually as good as Hi-Rez - no matter the price.<<

A vast generalization, it really depends on the individual players, and price will always remain a considerable consideration when justifying performance, but I understand your intent.

>>This *will* trickle-down - give it time. Bryston's CD player //DAC and Audio Research's DAC 7 (all reviewed in TAS) are closing the gap - even as we speak.<<

A few excellent mid-priced players have long existed which, I thought, gave mid-priced SACD players a run for their money, including the Meridian 588.

I've auditioned the Bryston, very good value for the $.

The Linn Majik sounds even better.

TB1

RE: good point ... but, posted on July 14, 2009 at 22:37:04
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Let me clarify: CD through the Meridian 808.2 and (probably) through the Boulder CD player (currently reviewed in Stereophile).

in the grand scheme of digital, the problems remain ..., posted on July 16, 2009 at 20:13:31
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 and little has been "solved".

In general terms, who really benefits if a $20K product has the potential to transform 16/44 into what was originally promised with CD, when the vast majority of players still fail miserably.

And I say "potential", because I've read far too many reviews of products that have claimed the same thing, and yet, they still sounded compromised.

Also, since the vast majority of software is recorded like crap, I doubt that any MP based player will transform them into something special.

Hell, even SACD can't transform a bad mastering job into music.

And, although I'm not dismissing MP by any means, I've already heard a few relatively affordable CD players that reproduced the hi-frequencies properly, naturally, with both power and delicacy, given the proper system/software. I'm not suggesting that these players are as good as a 808 or a Bolder, but certainly, to the average audiophile, they represent real world attainable products/prices, and therefore, it's that majority of audiophiles who'll decide if digitals problems are now "solved".

Damn, I hate digital "hash" with a passion, because the wider the bandwidth, the more transparent your system becomes, the more the hash becomes a sonic showstopper.

IMO, that's one of the main reasons why many analog devotees, those with true hi-end systems, never warmed to digital reproduction. Digital hash, once heard, becomes a noise that permeates the entire system and listening experience, it pollutes the music with an grain that's both obvious and offensive to my ears.

Therefore, despite the format, if a player's noise floor fogs over low level details, or it flattens the soundstage into 2 dimensions, if the eb & flow of the music sounds reconstructed instead of natural, when everything becomes grainy, and the hi-frequencies grate at your ears, compresses and changes the tonality of the music in the process, these are the MAIN digital problems that have taken years to solve.

To a degree, over the last decade, a number of players arrived that solved many of the above issues, but not on a universal basis, and certainly, not at the affordable prices that CD promised EVERYONE in the first place!!!

TB1
RE: in the grand scheme of digital, the problems remain ..., posted on July 16, 2009 at 20:37:53
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 It's true that CD sound has suffered...but does no more. And it didn't take all that long to "solve".

LP playback began improving in the early 1960's(!) - with the AR turntable. And it hasn't stopped since. CD's improvements began in the late 1980's - barely 20 years ago.

And it's a lot cheaper to obtain great sound from CD. Great LP cost as much as a house did in the 1970's.

New CD players (like the $2500 Bryston) are finally showing what CD can do at "lower" price points.....

RE: Red Book Riddle Solved, posted on July 14, 2009 at 06:45:02
coolhand
Audiophile

Posts: 233
Location: BBQ Area
Joined: June 5, 2006
 Where's an annoying woman to stomp her feet and pour water on your BBQ when you need one..?!!
RE: Red Book Riddle Solved, posted on July 14, 2009 at 13:04:24
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 The main point is that CD's specs are high resolution and CD sound has been improving *rapidly* since the early 1990's. Knowing these two things, why did any audio expert and/or theorist ask for a whole new format ? SACD was created to replace the royality profits of a big corporation and confused the parameters for decoding on playback. Even worse, it did *not* sound better than CD, could not be downloaded or ripped to hard-drive (in native form) and had few titles to choose from.

When I say "not better" than CD - it means compared to CD systems costing over 2K. Now that threshold is even lower. Let the truth be known !!!
RE: Red Book Riddle Solved, posted on July 14, 2009 at 15:46:14
Todd Krieger
Audiophile

Posts: 22458
Location: SW United States
Joined: November 2, 2000
 "The main point is that CD's specs are high resolution and CD sound has been improving *rapidly* since the early 1990's."

I could not disagree more..... I think there are a lot more awful recordings than ever (loudness wars, MP3-quality masters, substandard A/D), and I think aside from the new MP technology, the playback has also gotten a lot worse.


NOT, posted on July 14, 2009 at 14:33:02
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 Sorry. This is wishful thinking that reflects the inadequacies of tests, commercial bias, and/or lack of familiarity with high resolution digital.

Mastering engineers who specialize in classical music know better, as does anyone who has played around with high quality recordings in resolutions at least 88.2 kHz and 24 bits. These can be downsampled to 44.1 kHz using the best pro-sample rate converters, and the results will not be as good as the original, as will be readily apparent to most listeners in a non-blind test. One explanation for this is poor playback capability at 44.1/16, but this can be countered by using the same SRC to convert the 44/16 back to its original format and played through the identical playback chain. The difference will be clear and can only be attributed to the sample rate conversions, not the playback gear.

With many recordings the differences are subtle, but it is a sad situation to have made a hi-res recording and listen to it degrade when it is shrunk into the obsolete 44/16 format.

It is true that these modern converters can produce better CDs and when 44/16 recordings are made with loving care they can sound good, better than previously. But you are fooling yourself if you believe they sound as good as they could have had they been released in a higher resolution format. IMO the problem is primarily with the 44 and not with the 16, at least not when the best dither techniques are used. Fitting hi-res into 44/16 is a two step process, e.g. 88/24 to 44/24 and then 44/24 to 44/16. The separate degradations at each stage can be observed.

I agree with your other comments about SACD. It deserves to go out of existence, if only because of its copy protection.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: NOT, posted on July 14, 2009 at 18:01:53
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Tony: Those comparisons in the studio are flawed and you know it. CD's specs are high-resolution - period.

And I, nor others whose ears I trust, are hearing *any* difference between the Meridian 808.2 CD player and downloaded 24/96 HD Tracks. This mirror's John Atkinson's remarks in Stereophile when performing the same comparison.

We could not have said this two years ago - CD is improving. Error-correction (on disc) could be one of the culprits limiting its sound, compared to 24/96.......

RE: NOT, posted on July 14, 2009 at 18:51:59
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "Tony: Those comparisons in the studio are flawed and you know it. CD's specs are high-resolution - period."

I have no reason to believe that the studio comparisons are flawed. They are made by engineers who produce high quality recordings and have no reason to fool themselves. They reflect what manufacturers of high end players and DACs report. I would believe these people before I would believe anything written by salesmen or magazine writers. More to the point, they reflect what I hear on my own equipment, using the best available sample rate converters (such as Izotope and Sox) at a variety of settings.

Those who fail to hear these differences are either not looking for them, lack the experience needed to detect them, or are using recordings that aren't sufficiently revealing. No expensive playback equipment is needed, just the ability to play WAV files at a variety of sample rates at 24 bits. Although the Izotope 64 bit SRC is not cheap ($1199 with the RX restoration software) the SOX SRC is free. Best to compare music with a fair amount of high frequency content, such as symphonic music.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: NOT, posted on July 14, 2009 at 19:57:13
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 The studio converters are not minimum-phase filtering devices. Cutting-edge playback gear - like the Meridian 808.2 - is the best way to hear the 16/44 version(s).

RE: NOT, posted on July 14, 2009 at 20:15:02
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "The studio converters are not minimum-phase filtering devices."

I will say nothing more to you, because you simply do not know what you are talking about.

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: NOT, posted on July 14, 2009 at 20:20:22
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I know what I hear through the Meridian 808.2 CD player (vs. full-band 24/96). It's a tie.

I am *not alone* in this assessment. What's your problem ?

Who else is making minimum-phase filtering devices, besides Meridian? *, posted on July 14, 2009 at 20:00:51
gme109
Audiophile

Posts: 11344
Joined: April 3, 2001
 *

RE: Who else is making minimum-phase filtering devices, besides Meridian? *, posted on July 14, 2009 at 21:02:41
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Berkeley, Ayre and Spectral - for now....

Where did you come up with that list???, posted on July 14, 2009 at 23:18:49
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 Ayre, yes. The information is on a white paper on our "What's New" page.

But where did you come up with Berkeley and Spectral? Fine products, and fine sounding, but absolutely no mention of minimum phase digital filters on either company's website.

I'm open to correction, but the only other companies I've heard of using any minimum phase filters are Cambridge Audio in the DAC Magic and PS Audio in their new Perfect Wave series.

RE: Where did you come up with that list???, posted on July 14, 2009 at 23:50:12
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 They might not call it "minimum-phase" but if we read between the lines, it looks like they're doing it. A direct question to these folks is the only way to find out....

I think you need some new glasses., posted on July 16, 2009 at 00:14:10
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 >> if we read between the lines, it looks like they're doing it. <<

Please provide us with any examples that might even hint about minimum phase.

RE: I think you need some new glasses., posted on July 16, 2009 at 00:47:34
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 I can see fine. Here's a "hint":

Google-type "Goodwins high end spectral long filter" and click on the links.

It wouldn't load with this _ from here.

Not even close, posted on July 16, 2009 at 19:09:07
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 First there is a blatant error:

"Since its introduction in 1995 the original PMD-200 HDCD digital filter software has been the choice of high-end digital audio designers for their most ambitious processors and players. This was no accident. Digital designers in most high-end companies recognized the unrivaled performance of the PMD-200."

This is utter nonsense, which is why it isn't on the Spectral website.

The truth is that the PMD-200 never made it to production. It was a special version of one of the Motorola DSP chips that came pre-programmed with both HDCD decoding and the digital filters developed by the HDCD team.

The reason for its existence was because Pacific Microsonics had invested heavily in making an ASIC (custom IC) for the PMD-100, the original HDCD decoding digital filter. It costs around $500,000 to have a custom IC made. But it was made on a 0.6 micron process. After about five or six years, the foundry shut down that line as it wasn't making enough money for them. Pacific Microsonics was never as profitable as the founders had hoped for and they didn't have another $500,000 to make another custom IC.

So instead they used the fact that general purpose audio DSP chips had gotten cheaper to migrate the whole thing into a Motorola DSP platform (I forget which one now.)

But the whole thing unraveled pretty quickly after that. Motorola spun off the DSP chip division as "Freescale" to concentrate on cell phones (one of the perils of being a publicly-held company that has to have 10% growth per quarter or face the wrath of the stock market). Pacific Microsonics wasn't profitable enough to survive and were probably relieved to find a buyer -- Microsoft.

Microsoft didn't want to dick around selling chips to nutty high-end companies. So the PMD-200 never made it to full production. There were a handful of companies that used it. The most obvious example was the Assemblage DAC (a division of Sonic Frontiers, that also went belly-up shortly thereafter). There are very few of those DACs in existence.

The only other model that I am aware of is one of the Cary disc players. They claimed to use a PMD-200 years after it was discontinued. I don't know if they found a stash of chips somewhere or were copies. (All of the Cary disc players were made in China according to one fairly recent post.) It would be tough to copy the PMD-200 without the source code. But you can still buy it from Microsoft for $10,000 if you want to.

Bottom line is that the article starts out on the wrong foot.

The rest of the article basically says that the Spectral filter has more taps than most filters do. Nothing about minimum phase or pre-ringing.

well ..., posted on July 16, 2009 at 20:55:36
TBone
Audiophile

Posts: 2678
Joined: April 5, 2000
 Actually, IIRC, the PMD-200 was introduced in 1999, not 1995. Therefore, it can be claimed that in 1995 ... "Since its introduction in 1995 the original PMD-100 HDCD digital filter software has been the choice of high-end digital audio designers for their most ambitious processors and players. This was no accident. Digital designers in most high-end companies recognized the unrivaled performance of the PMD-100."

I can think of a number of players that used the 100, including Linn, Sim, ML, Cary, Spectral, EAD, Sonic Frontiers. At that time, many of those 100 based machines represented the best sounding digital players.

As for the 200, Linn, Sim & Cary used them sparingly. The original Sim Nova and the second gen. Eclipse used the 200, both truly excellent sounding players, with excellent transient response and natural hi-freq. content. Perhaps the most famous cutting-edge and amazing sounding CDP, the Linn CD12, started with the 100, but could also be purchased or upgraded to the 200 for a short time prior to its eventual retirement.

EDIT: Chord had something called the "Watts Transient Aligned Filter" and Linn used a 2-D-DSP algorithm to "shape" the sound. Don't now if these proprietary systems are related to the MP.

EDIT2: Correction - The otherwise excellent sounding Sim Nova did NOT use the PMD-200, I remember this because it was not an HDCD player, unlike the first and second gen. Eclipse.

TB1

This is probably the best explanation, posted on July 17, 2009 at 19:47:10
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 >> Actually, IIRC, the PMD-200 was introduced in 1999, not 1995 <<

Yes, then this would make sense. The webpage in question must have meant the PMD-100, not the PMD-200. In that case, many, many of the best CD players *did* use the PMD-100.

Two things killed it. First was that it would not process the new higher resolution 96/24 data rates of DVD-Video (and later DVD-Audio). The second was that they discontinued it when the foundry shut down the 0.6 micron fab process.

Remember that this is a *dealer* webpage and not the manufacturer's webpage. It is much more likely to contain technical errors.

RE: Not even close, posted on July 16, 2009 at 19:35:33
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Thank you, Charlie H.

Spectral still *might* be minimum-phasing - they're famous for their secrecy.

But their statements from the past would make me question anything they say now....

Spectral has never made any misleading claims, posted on July 17, 2009 at 19:57:04
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 >> Spectral still *might* be minimum-phasing - they're famous for their secrecy. <<

Yes, they are famous for their secrecy. But I'm quite confident that they are not using a minimum phase digital filter.

>> But their statements from the past would make me question anything they say now.... <<

I think that is not the correct response. The page you linked was a *dealer's* webpage and not the Spectral webpage. Another poster pointed out that the major mistake was actually probably just a typo where they meant to type "PMD-100" instead of "PMD-200". Then everything else they said would have made perfect sense.

I have never known Spectral to make any misleading claims about any of their products. They are clearly world-class designs offering outstanding performance and build quality.

Pacific Microsonics (inventors of HDCD and now out of business) walked a fine line between secrecy and (deceptive?) misinformation. But that is *not* the case with Spectral. While they are not forthcoming with an abundance of technical information, neither have they ever made any misleading or deceptive claims. I have the highest respect for their company and products.

And I'm also sure that they aren't using minimum phase filters. However, since their current CD player *does* use custom DSP chips to implement their digital filters, it would be very easy for them to perform a simple firmware update that would change the filters to just about anything they wanted to offer.

RE: I think you need some new glasses., posted on July 16, 2009 at 10:50:41
Werner
Audiophile

Posts: 1851
Joined: September 30, 1999
 These blurbs look, to me, like big nets
full of red herrings.


bring back dynamic range
RE: I think you need some new glasses., posted on July 16, 2009 at 18:58:23
CarlyBoy
 I believe this is the relevant bit of marketing speak. Spectral design bulletin:... These advanced correction algorithms improve timing accuracy at all frequencies we hear. By anticipating time shift and producing an opposing response program, time dispersion is greatly corrected adding the third dimension to filter math that reveals more life, sparkle and dimensionality. ...

Since the bulletin is describing a higher sampling rate, 32-bit floating point implementation of the HDCD filter, we can probably conclude it's not talking about a MP filter. Instead, it's probably trying to refer to the HDCD transient filter selection. HDCD encoders have two decimation filters, a sharp roll-off normal filter and a slow roll-off transient filter. The encoder performs a continuous FFT looking ahead over a sliding window of data, feeds it through a perceptual model, and selects which filter to use. It also encodes the filter selection bit in the dither, but there is some debate about whether anything actually uses that bit upon playback.
Not a lot of debate, actually, posted on July 16, 2009 at 19:20:27
Charles Hansen
Manufacturer

Posts: 4355
Joined: August 1, 2001
 >> It also encodes the filter selection bit in the dither, but there is some debate about whether anything actually uses that bit upon playback. <<

Many people have decoded the HDCD codes hidden in the LSB of the audio material. They have found the three things mentioned in the patent -- low-level extension, peak expanding, and filter selection.

But neither the data sheets for the PMD-100 and PMD-200 mention anything about two different *playback* filters. So in March of this year I asked one of the HDCD principals about this. He replied that the playback filter was fixed.

I would have to assume that he would have mentioned if the Spectral 4000 was an exception to this, as he would be in a position to know.

I don't know why the information was included at all. Presumably they had plans to use that to switch playback filters at some point, but I guess they never got around to it. Even with the Spectral 4000 (a *very* fine player, I'm sure).

RE: Where did you come up with that list???, posted on July 15, 2009 at 09:32:21
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "They might not call it "minimum-phase" but if we read between the lines, it looks like they're doing it. A direct question to these folks is the only way to find out...."

Or put a test CD with an impulse into the player and scope the output. Avoids talking to marketing people.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

Thanks, posted on July 14, 2009 at 21:29:43
gme109
Audiophile

Posts: 11344
Joined: April 3, 2001
 I found this review on the Berkeley. It seems as good as standard red book sounded on this unit, when fed higher resolution material, it was in a different league. This seems to contradict what you were saying about devices that use minimum-phase filtering, leveling the playing field. Not trying to start a debate, or question your personal experience, I'm just curious about these newer filters and their potential for greater sound.



The Berkeley Audio Design Alpha DAC is the best-sounding outboard digital-to-analog converter I’ve heard. The fact that it decodes high-resolution sources of any sampling rate and word length is icing on the cake. Although the Alpha DAC is spectacularly great on CD, this converter really shows its prodigious resolution, dynamic agility, and soundstaging capabilities when fed 176.4kHz/24-bit digital audio. Moreover, the Alpha DAC’s feature set, operation (instant locking to changing sampling rates, for example), and ability to drive a power amplifier directly expands the product’s utility and makes it ideal for the next generation of high-resolution music servers that is just around the corner.

RE: Thanks, posted on July 14, 2009 at 21:52:01
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 So I guess you already knew what systems use minimum-phase !!

Your experience is incomplete because you need to hear a (cutting-edge) CD-only player. Multi-format players always sound better with the "new" format........

RE: Thanks, posted on July 14, 2009 at 21:57:03
gme109
Audiophile

Posts: 11344
Joined: April 3, 2001
 "So I guess you already knew what systems use minimum-phase !! "

No I didn't, I Googled it right after I read your post. And then found the review. Isn't the information age great!

"Your experience is incomplete because you need to hear a (cutting-edge) CD-only player."

Agreed.

Re: "...this forum does not allow editing." ..., posted on July 14, 2009 at 14:17:05
Neil49
Audiophile

Posts: 7991
Location: Northern Virginia
Joined: June 13, 2003
 Sure it does! If you wish to change some text,there is a link,"Edit", which can be found at the very bottom of the page when you click on your own message.There will be notation that the post has been edited after the revised mesage has been re-posted.

RE: Re: "...this forum does not allow editing." ..., posted on July 14, 2009 at 18:48:07
J. Phelan
Audiophile

Posts: 73
Joined: May 12, 2009
 Got it - but how do I highlight a URL link so it loads directly ?

RE: Re: "...this forum does not allow editing." ..., posted on July 15, 2009 at 14:29:10
Tony Lauck
Audiophile

Posts: 3323
Location: Vermont
Joined: November 12, 2007
Contributor
  Since: February 24, 2009
 "Got it - but how do I highlight a URL link so it loads directly ?"

If you want an inline link to appear in your post, just type in the necessary html coding for the link with the a tag and the /a tag. (Be sure and use the "Preview Message" button to make sure that you got this right.)

If you don't want to mess with the html coding, you can just paste the URL into the "Optional Link URL" field. Again, it is a good idea to preview the post and follow the link through to make sure that it works. If you navigate away from your post before hitting the "post message" button take care not to lose your creation from your browser.

If you don't know html, you can use your browser to look at the page source for some web pages and figure out what is needed for a workable link.


Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

RE: Re: "...this forum does not allow editing." ..., posted on July 14, 2009 at 18:58:52
Neil49
Audiophile

Posts: 7991
Location: Northern Virginia
Joined: June 13, 2003
 Not sure if this is what you're asking, but to place a link to a URL, simply copy and paste the address into the "Optional Link URL" box beneath the message field. You can also re-name that link ("Optional Link Title"), to make it more comprehensible to readers.

Page processed in 0.180 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2010, Audio Asylum, all rights reserved ] [ Take5 ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]