DSD (SACD) vs PCM

[KeithL]

I'm starting a thread here to discuss the differences (and similarities) between DSD and PCM.

The DSD format (on SACD discs) has been widely considered to be a commercial failure for several years, yet some people remain convinced of its technical superiority (and, yes, since obtaining the best possible sound quality is the goal for digital audio,if you insist that "it sounds better", then you are claiming that it is technically superior).

Well, here's a place to discuss the actual theory behind DSD, to explain why you think it's worth the extra effort required to support it (or not), and ask questions about the actual technology and capabilities of the DSD digital audio format. (I titled it this way because PCM is easier to play, and much more widely supported, so nobody is going to be buying or playing DSD files unless they believe DSD is somehow better than PCM.)

[ejn1111]

I have a dedicated 2 channel system with many HD downloads. I can say that I do like the sound of DSD downloads a lot. It's an impressive format. I have some 24/96/192 that sound great also but some of my best recordings are in the DSD format.

[geebo]

I have some very good DSD / SACD examples and some not so good. And I also have some redbook CDs that are as good as anything else I've heard. So how much of the good sound on our SACDs is due the the technology and how much is due to the care, mastering and production? CDs are often compressed and recorded at such high levels causing peaks to clip badly just to to make it louder than other CDs. SACDs are generally not compressed so heavily and are recorded at a lower levels for more headroom for peaks. So is it the vehicle or the driver?

[bootman]

Sept 28, 2014 9:09:42 GMT -5 ejn1111 said:

I have a dedicated 2 channel system with many HD downloads. I can say that I do like the sound of DSD downloads a lot. It's an impressive format. I have some 24/96/192 that sound great also but some of my best recordings are in the DSD format.

But is the SQ due to the mix or the bits?

I'll bet it is the mix.

[LCSeminole]

I've said before that in general I hear no difference. I do however agree with "geebo" that the mastering and production is the key factor in the quality of the sound. I do also believe that if I were forced to choose in a blind testing which was which, I couldn't reliably choose which was which, so from that I've concluded that DSD, in my mind, is redundant. Just my opinion of course.

[geebo]

Sept 28, 2014 10:44:48 GMT -5 bootman said:

Sept 28, 2014 9:09:42 GMT -5 ejn1111 said:

I have a dedicated 2 channel system with many HD downloads. I can say that I do like the sound of DSD downloads a lot. It's an impressive format. I have some 24/96/192 that sound great also but some of my best recordings are in the DSD format.

But is the SQ due to the mix or the bits?

I'll bet it is the mix.

I agree. Now if the mere act of releasing something in DSD / SACD makes the record companies take more care in mixing and production then I'll buy it over the compressed and clipped CD version.

[sct]

LCS hit the nail on the head. I can honestly not hear any difference between well mastered DSD and well mastered PCM. Bot sound absolutely fantastic to me. And both sound so much better than the stuff I cut my teeth on, vinyl LPs.

And I will also say that well mastered CDs can also sound fantastic. I have some Creedence Clearwater, Grateful Dead, and Doors CDs (actually K2 and HDCDs) that sound terrific. And this is due, IMHO, to the fact that a great deal of care was used in the mastering and cutting processes.

We are currently the recipients of very, very good sound quality from all of the digital formats. Providing, of course, that proper care was exhibited in the mastering and production stages...

SCT

[KeithL]

At this point I'd like to say that I'm one of those who DOESN'T believe that DSD has any INHERENT superiority. While I would agree that SACD versions of particular albums do indeed often sound slightly different than their PCM equivalents, I don't credit the technology itself for being responsible.

I've read any number of articles purporting to "explain" why DSD is "better" - and they've all ranged between "specious" and "downright silly". 

The fact that the DSD pulse train VISUALLY happens to look more like the analog signal than the string of binary numbers that result from PCM encoding is meaningless - since both must be converted to analog audio anyway. (Believe it or not, this was a major argument in some of the original SACD sales literature.) And the claim that "you can convert DSD to analog audio with just a few components and no complicated filtering" is specious; if you do so, you end up with a reasonable facsimile of the analog signal accompanied by a MASSIVE amount of ultrasonic noise. You must then apply a strong ultrasonic filter if you want to avoid damaging your equipment and get back a clean signal. (This is not a huge drawback; it just means that doing a proper conversion from DSD requires circuitry that is similar in complexity to what you need for PCM - so the "simplicity claim" is bogus.)

The fact that the DSD format ("single speed DSD" aka DSD64) has a higher bandwidth than 44.1k PCM is also unimportant. While it's true that DSD64 has frequency response that extends to about 100 kHz, it is also true that the S/N ratio drops very low at high frequencies - which means that you need "aggressive noise shaping" to eliminate this noise - which reduces the practical frequency response to something similar to 24/88 encoded PCM audio (or slightly less). Again, not a condemnation of DSD as a viable format; it just invalidates THAT claim for superiority.

Here are a few more things you should always keep in mind when comparing DSD to PCM:

1) It is difficult-to-impossible to edit DSD digital recordings. There used to be only one brand of recording console that was able to do so. (All others were limited to basically trimming the ends of songs; even adjusting levels is tricky, and you can forget about mixing and EQ.) There are now a few ways around this, but all are a nuisance, and none is really standard. As a result, virtually all recordings that are "true DSD end-to-end" are the equivalent of a direct-to-disc album; with no "engineering" or "mixing" involved. In contrast, with the huge assortment of processing available in even "low end" PCM editing suites, very few PCM-mastered albums are "lightly engineered". It is distinctly possible that this is the main difference you hear between "DSD" and "PCM" recordings. It also means that many so-called "DSD mastered" recordings have been converted to PCM for editing, then converted back afterwards. Unfortunately for DSD fans, this means that the resulting audio has been "damaged" by any and all problems caused by BOTH formats. (I'm not saying that this is a problem either, but it renders the idea that you've avoided all sorts of nasty problems introduced by PCM rather moot.)

2) The conversion process between DSD and PCM is NOT lossless. The conversion process is complex enough that converting in either direction is going to result in (minor) audible differences. I've heard comparisons of conversions - from the same DSD original to 24/96 PCM - using the two most popular high-end commercial converters (Weiss Saracon and Korg Audiogate); even they sound slightly different. This means that you can NEVER do a direct comparison between the two. Whichever format you start out with as your master, converting to the other WILL make them sound slightly different due to the conversion process itself. The best you could do would be to make simultaneous recordings in both formats, from the same LIVE source, end even then you would only be comparing the way both formats sound on the specific recording hardware you used. (If you use a single unit, then it is always going to start with one format or the other, and convert to the second one. If you use different recorders, then the other differences between them will probably be far more audible than the differences between DSD and PCM. Don't confuse the fact that a specific DSD recorder happens to sound very good with the idea that DSD itself is superior.)

3) But what about DSD128 - and other possible "better" DSD formats? Well, without arguing whether you can hear a difference, DSD64 is arguably equivalent to 24/88 (which does make it slightly better than a plain CD), so DSD128 is even better (maybe as good as 24/192 PCM audio). But a LOT of modern DACs support 24/192k PCM, while very few support "double DSD".... and a few DACs already support 24/384k PCM (double rate 192k). Back when the arms race was between CD discs at 16/44 and SACD discs, you might be willing to declare SACD the winner, but that isn't true today. Whatever sample rate is used for next year's DSD files, there will always be a PCM sample rate that is equivalent to it.

4) So what if I have a disc that was mastered on DSD? OK, you've just described the ONE legitimate reason to want to play DSD in native DSD format. Assuming you have a recording that was mastered in DSD, or even if you just happen to really like the way a specific DSD disc sounds - regardless of how it was processed along the way - converting it to PCM at this point probably will introduce a small difference in the way it sounds. (This is the same "dilemma" you face if you have a DAC that only plays up to 24/96 and you happen to have a 24/192 file that sounds really good. Whether there is any difference inherent in the formats, performing the conversion process will introduce a difference.) Personally, I find the difference introduced by a good converter to me minimal (as in "barely audible"), and I've never found it to be "clearly negative". I've heard any number of SACD recordings that sound very good; and all of them have sounded equally good whether I played them directly in DSD or converted them to PCM. They didn't suddenly sound awful, or even significantly different, after I converted them. I've found the differences in how various DACs sound - for other reasons - to far exceed the differences caused by the conversion. (Now, you may hear - or care about - things that I don't... and, for equivalent performance, you probably should convert DSD64 content to 24/88, or the more standard 24/96, rather than the 16/44 quality used by CDs. Luckily, since you're producing a digital file and not a CD disc, this isn't a problem.) ANd, if you want to avoid all this with download files you purchase, then buy them in PCM WAV or the EXACTLY equivalent FLAC format - and not in DSD.

Incidentally, if you want a top-quality converter, Saracon costs about two grand, but Korg Audiogate (the other of the top two commercial ones) is now FREE. (Although you do have to agree to let it Tweet "I'm using Korg Audiogate" every time you use it... yeah, really.)

5) And what if I have actual SACD discs? OK, NOW you've got a problem. SACD discs include an aggressive copy-protection scheme. BY LEGAL REQUIREMENT, no SADC player is ALLOWED to give you access to a full quality digital version of the digital audio contained on the disc. Obviously, they can give you an analog decoded output, which is also obviously not a bit-perfect rendition of the original content. They can give you the DSD audio signal, or a high quality PCM conversion from it, via HDMI - because the HDMI link is encrypted and so "secure" - and because nobody makes (or is allowed to make) recorders that accept an HDMI input and does bit-perfect recordings. However, they are not allowed to give you a full quality output from any of those digital outputs. (Some devices will give you a reduced quality "mix down" of the SACD content, while most will simply mute their digital outputs when you're playing an SACD.)

Since computer DVD players CANNOT play SACDs, and there are no SACD players that can be connected to a computer as an external data drive for purposes of playing SACDs, there is no way to RIP SACDs using only software.

There are only two ways to RIP an SACD "perfectly" - and both are illegal. You can have a hardware modification performed on a player - which means "tapping off" the digital audio signal from the internal circuitry right before the DAC. This can be tricky and expensive, recording the resulting real-time digital stream perfectly can also be tricky, and you will of course void the warranty on your player.

The other way is to purchase, and then "hack" a vintage Playstation 3 player. The hardware in the early models can play SACDs and, by installing the proper software and hacks, you can get them to allow you to record an SACD to a file. The process, however, is complicated, tricky, and prone to failure.

If you REALLY want to go that route, you're better off letting someone else do the work and simply looking for an (illegal) download site where you can download SACDs that have been ripped by other people. (There are a LOT of them out there.... but I'm not going to discuss the details. Let's just say that, as far as I know, you can't put a virus INSIDE a ripped SACD ISO, at least as long as it's really an ISO and not just another type of file that's been renamed, so any you find are PROBABLY safe to play. And, if you just happen to like a certain SACD recording, but aren't especially obsessed with its being DSD, you will also find SACD rips that have been converted to FLAC.)

[garbulky]

Just a small observation keith. The better sounding solution isn't necessarily a technically superior one or a more transparent one. Just a better sounding one.

[KeithL]

I disagree - sort of.
(Garbulky... I'm NOT beating on you here... you've just given me a good excuse to liven up the discussion a bit more )

By definition, the "technically superior" solution is the one that best meets your stated goals the best.
(To me, it's just a fancy way of saying "it does what I said I wanted it to do" and excluding the possibility of "but I like the other one better anyway".)
If your stated goal is "The most pleasant pen to use - at any price", then a Monte Blanc may be a technically superior.
If your goal is instead "To make the most reliable pen that can be sold for under $2", then a Bic might well be technically superior.

My goal is "perfectly accurate reproduction of the original". (If the original didn't sound good, then my stated goal may NOT equate to the best sounding reproduction, but that is MY choice.) Now, while accuracy CAN be defined in terms of specifications, even that is subject to individual interpretation. (Which is more accurate, a recording with a 60 dB S/N and 0.1% THD, or one with an 80 dB S/N and 1% THD? The answer would depend on whether you consider THD or S/N more important, and how the two correlate. It's sort of like asking whether the color accuracy or the perspective is more important in a painting.)

Unfortunately, when you start with subjective terms like "sounds best", it becomes even more complicated - because that is almost purely a matter of personal opinion.... because everybody places different values on different criteria. (And those values can even be negative - some tube lovers actually prefer MORE second harmonic distortion... just as lovers of impressionist art like the fact that impressionist paintings DON'T look like accurate photographs.)  As with many other debates in the audiophile world, many SACD fans seem to conflate "accurate" with "I like the way it sounds". This would be like a lover of impressionist art declaring that a painting by his favorite painter is "more accurate than a photograph". (This could arguably be true - in a very subjective sense - but, by most normal definitions of accurate, it is absurd.)

The purveyors and supporters of DSD would certainly like for you to equate "accurate sound" with "good sound" - and equate both with their DSD format. Since they've already decided that they know what the original - and so an accurate reproduction of it - SHOULD sound like, they're convinced that the sound they prefer must be the most accurate. I've honestly never heard anyone say something like: "I think SACDs have a highly colored sound, but I prefer it anyway." (I also do wonder how many "DSD lovers" are simply equating "different" with "better" - which is common tendency of humans.)

In all fairness, both DSD and PCM are very accurate, so we're all splitting hairs here.

And, since every conversion process imposes some slight change in sound, don't we really need to compare ALL the alternatives.

If you think "a certain file mastered in DSD sounds better than the PCM version"....

... do you think it sounds better than the same recording mastered in PCM and played in PCM?
... do you think it sounds better than that same file mastered in DSD and then converted to 24/96 PCM using Korg Audiogate?
... do you think it sounds better than that same file mastered in DSD and then converted to 24/96 PCM using Weiss Saracon (they DO sound different)?
... and how does it sound compared to that same file recorded in PCM, THEN converted to DSD using Audiogate?
... and how does it sound compared to that same thing, only converted using Saracon?
... and how about all these options, except now with DSD128, or with PCM 24/192, or converted using dBPowerAmp?

They're ALL going to sound a tiny bit different (and that difference may be greater than the difference you originally heard "between DSD and PCM").

And so far I've avoided mentioning that Saracon (and probably Audiogate) has different filter settings, each of which produces slightly different results...
so multiply that list of possible options by the number of filter choices in Saracon and Audiogate...
and, while you're at it, throw in the different DSD and PCM recorder options.

I think you'll agree that comparing a few of each type of files makes about as much sense as comparing two car manufacturers based on the fact that you've spend ten minutes driving one model from each....

There are really two interdependent questions here:

1) Does a DSD recording actually sound audibly different than a PCM recording of similar technical capabilities (DSD64 vs PCM 24/96)?
(And we mean SPECIFICALLY because one is DSD and the other PCM. And that question is subject to all the annoying details I outlined above.)

2) IF there is indeed a difference, then which is "superior"?

That second question is even more widely open to interpretation. As I described above, there are any number of ways of defining "better". However, if the answer to the first question is "nobody can hear any difference", or "the difference isn't significant" or even "I can't hear the difference" , then the second question is entirely moot. If they don't sound audibly different, then neither can possibly sound better - by ANYONE'S chosen criteria.

[I'll interject here that I had an opportunity to hear sample files, mastered in DSD64, then converted to 24/96 PCM, using both Audiogate and Saracon. It was intended as an example of the conversion quality of the two converters. I did indeed hear a slight difference... and, no, I didn't really prefer one or the other - they were just slightly different. ]

If you read a lot of reviews and comments about DSD vs PCM, you'll find that most approach the whole subject sort of sideways. The "reviewer" has heard some DSD/SACD recordings that they really like, and has become convinced that the difference is due to some inherent difference between PCM and DSD rather than a simple difference in mastering, or conversion software, or filter choices. They then use this valid, but quite limited, experience as a justification to believe the (mostly specious) technical arguments in favor of DSD being more accurate. They then look for some good sounding DSD discs to purchase to support their assumption.

The logic goes like this:

"I like this SACD disc"
"Such and such expert says SACD sounds better because x,y,z (even though I really don't understand the details, it seems to make sense)"
"He was right about this disc sounding better so I guess his argument about SACD being technically superior is valid".

Unfortunately, the logic is flawed.

Imagine my claiming that "wine tastes better than beer".
I then hand you two bottles (hidden in brown bags), one wine and the other beer, and the wine does indeed taste better than the beer.
I hope you're smart enough to realize that all I've proven is that one particular wine tastes better to you than one particular beer.

This same exact logic currently seems to have a lot of people convinced that "DSD is better than PCM".

*** NOW, HERE'S A VERY SERIOUS SUGGESTION FOR THOSE WHO LIKE TO TINKER.....

If you really do like the way DSD files sound better than PCM ones, or vice versa, or think you do, then maybe you should try converting some PCM files to DSD using Audiogate, or converting some DSD files to PCM using Foobar2000. On the PCM side, try converting them to 24/88, and 24/96, and maybe even 24/192. Or, if your pre/pro supports DSD (like the XMC-1),  you could try switching your Oppo between "convert DSD to PCM" and "bitstream". Of course, if you have an Oppo, it can play any files of either format off a disc or USB stick, to your XMC-1. Listen for yourself to the subtle differences between different files and conversions.

Wouldn't it be cool if it turned out that all of your 24/96 PCM files sounded better (to you) if you converted them to DSD, or if all your DSD files sounded better if you converted them to PCM, at a certain sample rate, using this or that conversion software... In fact, it's one of the few options you can explore for little or no cost.

Now, bear in mind that I personally consider PCM and DSD to be audibly equivalent, and I find this whole discussion somewhat silly.... but that's just me. However, I could be wrong, so I've provided you with a whole slew of options to explore that could prove that one way or the other, and they are all a whole lot cheaper than buying or upgrading a whole bunch of hardware "just in case".

Hey, maybe it will turn out that the DSD downloads from Blue Coast sound better because they do a really great mastering job on them. If so, maybe it WOULD make sense to buy their DSD downloads, then convert them to PCM after you download them, and play them on a DC-1.....

Hmmmm.......  now there's a thought....
 
Hopefully, if you like this sort of thing, I've given you a bit more to think about. 

Kiss another weekend goodbye...

Sept 28, 2014 12:52:10 GMT -5 garbulky said:

Just a small observation keith. The better sounding solution isn't necessarily a technically superior one or a more transparent one. Just a better sounding one.

[KeithL]

For anyone who takes my suggestion seriously - to try the different formats yourself (at the end of my last rather long post), here is a rather extensive and well written article about how well Weiss Saracon, Korg Audiogate, and jRiver Media Center perform the DSD <-> PCM conversion process. (Yes, it seems that jRiver can do it too... and a lot of you already have that.)

Isn't it interesting how the DSD performance is WORSE than PCM in ALL of the examples

archimago.blogspot.com/2014/04/analysis-comparison-of-dsd-encoders.html

Also, incidentally, if you happen to "come across" any SACD ISO rips, Foobar2000 can play them, and can convert them to PCM on the fly. Foobar2000 can also use the information in the accompanying CUE file to convert an SACD ISO to other formats like FLAC and can, at the same time, break them up into individual songs (and even label the tracks).

[geebo]

For anyone that wants to try some, David Elias has some free samples available on the Oppo web site. These were recorded in DSD and have never seen PCM.

[garbulky]

I think you make some great points Keith! Also, I'm not going to deny what the specs say!

But....well....that's where the thing gets hazy. Accuracy or not.
First: how do you know what you're hearing is accurate? Is it because a certain type of measurement tells you that? How do you know that that measurement (or combination) of measurement gets you actual accuracy in the sound when you hear it?

I don't think accuracy really exists by the time it gets to our brains. The room, the speakers, the way the amps react to the signal, all alter it from whatever box you considered accurate it was.
Then we get to our hearing. If our hearing is deficient in a certain frequency range. What we hear is NOT accurate. In that way an innacurate device may end up sounding closer or more accurate to the original if it boosts the sound in sync with our hearing loss.

Hence I go with which sounds better. And yes, all it does is prove which sounds better to that person. But it is no reason to ignore it as that's the whole point isn't it? Otherwise our world will be audio boxes hooked up to AP machines - with nobody around to hear it.

Sounds better doesn't have to mean
- it's louder
- it has more bass
- this person has terrible taste in reproduced music as he listens to bose only.
- we've tossed out any measurements that disagree with the observation.

Colored also doesn't mean
- it's less clear.
- it's got a response that sounds less realistic than the original signal (which btw is not what we end up with anyway).
-It sounds colored and just not realistic.

It could mean:
- it sounds more realistic
- it's clearer
- it sounds MORE accurate (as you've noted).
etc.

The reason I say that is
People like their vinyl, they like their tubes etc. Does that make them technically superior. Well hard to define what that means right?
Would technically superior involve - NOT having to have a digital conversion from analog into digital or would it simply involve THD values? What about better headroom, when does that outweight THD values. Blah blah.

I agree that comparing them side by side is a good idea. I haven't heard real DSD as my DAC can't do it. But I would like to. Schiit makes a really cheap DSD unit. I'm not sure how it would be any good at that price but apparently schiit thinks so....

[sct]

A Mr. G Bulky offered: "People like their vinyl, they like their tubes etc. Does that make them technically superior. Well hard to define what that means right? "

I disagree. Vinyl and tubes are demonstrably technically inferior at reproducing the original sonic event. They are, by definition, technically inferior. Now folks may *like* the distortions that these devices impart to the signal, but that has nothing to do with technical superiority...

SCT

[Gary Cook]

A question I have asked many people many times and I've not yet received an answer that I can relate to, probably my failure to understand rather than an inability to explain. This seems as good a place as any to ask, you never know I might understand the answers this time.

Is it possible to have so much information in a piece of music that DSD or PCM can't handle (process) all of it in the required time frame? For example if I have a large symphony orchestra playing a classis rock song is it possible that digital limitations are such that not every instrument can be heard by the listener? Simply because they are there, there's not enough room for them? Or they have been deliberately mixed out because the sound engineer knows that they won't fit?

In the case of say a 7,223 instrument orchestra (a world record last year in Brisbane) would DSD or PCM be able to handle all of them?


Cheers
Gary

[garbulky]

Sept 28, 2014 19:32:00 GMT -5 sct said:

A Mr. G Bulky offered: "People like their vinyl, they like their tubes etc. Does that make them technically superior. Well hard to define what that means right? "

I disagree. Vinyl and tubes are demonstrably technically inferior at reproducing the original sonic event. They are, by definition, technically inferior. Now folks may *like* the distortions that these devices impart to the signal, but that has nothing to do with technical superiority...

SCT

A blanket statement like technically superior can't be applied as they are very different technologies. According to Keith's original post, he was saying if the sound is better then it is technically superior (with the caveat being that better = more accurate to the source).
But....
For instance: on a vinyl disc - you do not have to have any analog to digital conversions. While a CD has a higher SNR. Which of these is more valuable. Etc.
Keith mentioned and rightfully that vinyl has a TON of things that distort it away from the original sound. But if people end up preferring vinyl then which is more important?
I'm afraid I might have gone off on a tangent from the original subject. Sorry!

[garbulky]

Sept 28, 2014 19:58:10 GMT -5 Gary Cook said:

A question I have asked many people many times and I've not yet received an answer that I can relate to, probably my failure to understand rather than an inability to explain. This seems as good a place as any to ask, you never know I might understand the answers this time.

Is it possible to have so much information in a piece of music that DSD or PCM can't handle (process) all of it in the required time frame? For example if I have a large symphony orchestra playing a classis rock song is it possible that digital limitations are such that not every instrument can be heard by the listener? Simply because they are there, there's not enough room for them? Or they have been deliberately mixed out because the sound engineer knows that they won't fit?

In the case of say a 7,223 instrument orchestra (a world record last year in Brisbane) would DSD or PCM be able to handle all of them?


Cheers
Gary

I've wondered this myself. It's been explained how CD"s can represent the entire frequency spectrum. But what about multiple frequencies playing at or nearly at the same time. There MUST be a limitation right? (Same goes for vinyl imo)

[geebo]

Sept 28, 2014 19:58:10 GMT -5 Gary Cook said:

A question I have asked many people many times and I've not yet received an answer that I can relate to, probably my failure to understand rather than an inability to explain. This seems as good a place as any to ask, you never know I might understand the answers this time.

Is it possible to have so much information in a piece of music that DSD or PCM can't handle (process) all of it in the required time frame? For example if I have a large symphony orchestra playing a classis rock song is it possible that digital limitations are such that not every instrument can be heard by the listener? Simply because they are there, there's not enough room for them? Or they have been deliberately mixed out because the sound engineer knows that they won't fit?

In the case of say a 7,223 instrument orchestra (a world record last year in Brisbane) would DSD or PCM be able to handle all of them?


Cheers
Gary

I've always thought that at any point in time, there is but one value to measure regardless of how many frequencies are present or how many instruments are playing. There is only one waveform that is the sum of all sounds in the music.

[badong]

I buy Hybrid SACDs because they are cheaper than online retailers of high-rez files and the red-label layer is backward compatible with just about any device.

[Gary Cook]

Sept 28, 2014 20:47:32 GMT -5 geebo said:

Sept 28, 2014 19:58:10 GMT -5 Gary Cook said:

A question I have asked many people many times and I've not yet received an answer that I can relate to, probably my failure to understand rather than an inability to explain. This seems as good a place as any to ask, you never know I might understand the answers this time.

Is it possible to have so much information in a piece of music that DSD or PCM can't handle (process) all of it in the required time frame? For example if I have a large symphony orchestra playing a classis rock song is it possible that digital limitations are such that not every instrument can be heard by the listener? Simply because they are there, there's not enough room for them? Or they have been deliberately mixed out because the sound engineer knows that they won't fit?

In the case of say a 7,223 instrument orchestra (a world record last year in Brisbane) would DSD or PCM be able to handle all of them?

I've always thought that at any point in time, there is but one value to measure regardless of how many frequencies are present or how many instruments are playing. There is only one waveform that is the sum of all sounds in the music.

Simplistically, because I'm a simple guy, if one (or a number of instruments) change their notes faster than the sample rate then that change in note could be missed? If that's actually the case, then logically more instruments means more chances of that occurring. For example on the S&M album there are 26 violins in the SF Symphony Orchestra. If the sound mixer/engineer knows that say DSD can only handle perhaps 6 violins does he deliberately mix out the other 20? Or if he is using say PCM, can he fit in another 6 and only mix out 14 violins?


Cheers
Gary