Why do high up-sampling/ over-sampling rates (DSD,) kill PRAT and aliveness of music? Any ideas?

caesar

Well-Known Member
May 30, 2010
4,290
767
1,698
Personally, I enjoy rock, blues, and jazz, and can't stand how DSD makes the vast majority of these recordings so dull. DSD sure does make things smoother, but without the snappiness PRAT, DSD sounds very artificial to me.

Personal tastes aside, does anybody understand or have any theories why the high up-sampling/ oversampling rates of DSD, and such, kill the snappiness, aliveness, and PRAT of music?
 

Don Hills

Well-Known Member
Jun 20, 2013
366
1
323
Wellington, New Zealand
Personally, I enjoy rock, blues, and jazz, and can't stand how DSD makes the vast majority of these recordings so dull. DSD sure does make things smoother, but without the snappiness PRAT, DSD sounds very artificial to me.

Personal tastes aside, does anybody understand or have any theories why the high up-sampling/ oversampling rates of DSD, and such, kill the snappiness, aliveness, and PRAT of music?

This is, of course, merely your personal taste.

But seriously, the majority opinion is that DSD doesn't sound worse than PCM, and many prefer it. This assumes, of course, that both DSD and PCM come from the same master. This is often not the case, especially for the genres you favour.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Yeah sorry I also can't relate to your findings. DSD in my system sounds very vivid and alive with great presence. Mostly (although recording dependent) it sounds just a fraction closer to alive for me vs pcm.

I get that pcm has a bite but to me that bite makes it less realistic.

Which medium sounds closer to great analogue? For me this is dsd but again I must add that recording quality is the key here.

Another experiment worth doing is recording a great TT output using the same recorder in pcm and then dsd. I did this with my Tascam and settled on double dsd for archival purposes based on it sounding a smidge more like the analogue source (vinyl pressing).
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
Personal tastes aside, does anybody understand or have any theories why the high up-sampling/ oversampling rates of DSD, and such, kill the snappiness, aliveness, and PRAT of music?

Its not the high up-sampling rates doing the damage its that a 1bit DAC cannot be correctly dithered. There's a whole paper on this by Lipshitz and Vanderkooy dating back a decade and a half. The result of incorrect dither is noise modulation and that's very likely the reason for the loss of aliveness.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Its not the high up-sampling rates doing the damage its that a 1bit DAC cannot be correctly dithered. There's a whole paper on this by Lipshitz and Vanderkooy dating back a decade and a half. The result of incorrect dither is noise modulation and that's very likely the reason for the loss of aliveness.

Please can you post a link to the research because I find it nigh on impossible to believe any credible scientist would make the link to "aliveness" at all. To do that would mean being fundamentally certain that one can define "aliveness" in the digital / noise context, which I find hard to believe. That said happy to be wrong and be shown evidence.
 

bonzo75

Member Sponsor
Feb 26, 2014
22,434
13,467
2,710
London
Personally, I enjoy rock, blues, and jazz, and can't stand how DSD makes the vast majority of these recordings so dull. DSD sure does make things smoother, but without the snappiness PRAT, DSD sounds very artificial to me.

Personal tastes aside, does anybody understand or have any theories why the high up-sampling/ oversampling rates of DSD, and such, kill the snappiness, aliveness, and PRAT of music?

What streamers/Software did you use for DSD upsampling and which dacs were they feeding into? And did you do proper compares?
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
Please can you post a link to the research because I find it nigh on impossible to believe any credible scientist would make the link to "aliveness" at all.

Indeed the authors of the paper I referred to (being mathematicians) did not make a link to 'aliveness'. That link has been made by myself, given the format's shown therein to be mathematically flawed.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Indeed the authors of the paper I referred to (being mathematicians) did not make a link to 'aliveness'. That link has been made by myself, given the format's shown therein to be mathematically flawed.

Being mathematically flawed has no bearing on whether it would sound alive or not. This is the problem with pseudoscientific conclusions like this.
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
Being mathematically flawed has no bearing on whether it would sound alive or not.

How would you know this for sure? Do please set out your evidence and/or reasoning in support of your claim. Then once that's out in the open we can have a real scientific dialogue.

The mathematical flaws that L&V have demonstrated show conclusively that DSD isn't transparent. So then its up to DSD's supporters to demonstrate that this loss of transparency has no bearing on what's heard.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
How would you know this for sure? Do please set out your evidence and/or reasoning in support of your claim. Then once that's out in the open we can have a real scientific dialogue.

The mathematical flaws that L&V have demonstrated show conclusively that DSD isn't transparent. So then its up to DSD's supporters to demonstrate that this loss of transparency has no bearing on what's heard.

Sounding "alive" is a purely subjective term with no scientific construct. If you mean accurate / faithful to recording then that is different but not the definition of "alive" that I personally have in mind.

**But - the point I am making is that it is dangerous, indeed highly unscientific, to make pseudoscientific pontifications that link such a paper to whether something sounds alive or not and present it to people, quote you verbatim "very likely the reason..." for this phenomena. This kind of conclusion is bad science. No further debate needed from me.
 

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,455
29
405
Sounding "alive" is a purely subjective term with no scientific construct. If you mean accurate / faithful to recording then that is different but not the definition of "alive" that I personally have in mind.

**But - the point I am making is that it is dangerous, indeed highly unscientific, to make pseudoscientific pontifications that link such a paper to whether something sounds alive or not and present it to people, quote you verbatim "very likely the reason..." for this phenomena. This kind of conclusion is bad science. No further debate needed from me.

+1

And no further from me either :)

And as mostly an RBCD person with the odd >48 KHz recordings here and there ... I frankly don't understand what Caesar is talking about. What the heck is "kill PRAT"??? Really???
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
Sounding "alive" is a purely subjective term with no scientific construct.

Just because it currently has no measurement behind it does not mean it cannot have in future. Several people have remarked on the 'softness' of DSD, even you yourself notice a difference between PCM and DSD sound, with (your words) PCM having more 'bite'.

If you mean accurate / faithful to recording then that is different but not the definition of "alive" that I personally have in mind.

The math is saying that DSD cannot be truly faithful to the recording. If you wish to dispute that math, by all means do so I shall be interested to learn of the errors that L&V may have made.

**But - the point I am making is that it is dangerous, indeed highly unscientific, to make pseudoscientific pontifications that link such a paper to whether something sounds alive or not and present it to people, quote you verbatim "very likely the reason..." for this phenomena. This kind of conclusion is bad science. No further debate needed from me.

You're tilting at windmills here. 'Very likely the reason' is saying its probable, but that's far from being a conclusion. Why do you make a choice to interpret something I've claimed as probable as a 'conclusion' ?

Going back a few posts you claimed 'Being mathematically flawed has no bearing on whether it would sound alive or not. ' Notice no qualification, an absolute statement. What makes your claim 'good science' and mine 'bad science' ?
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Just because it currently has no measurement behind it does not mean it cannot have in future. Several people have remarked on the 'softness' of DSD, even you yourself notice a difference between PCM and DSD sound, with (your words) PCM having more 'bite'.

>> at no point have I said it could not in the future

The math is saying that DSD cannot be truly faithful to the recording. If you wish to dispute that math, by all means do so I shall be interested to learn of the errors that L&V may have made.

>> again you entirely missed the point - this is not about maths and I have not disputed the maths in any of my statements

You're tilting at windmills here. 'Very likely the reason' is saying its probable

>> as a fellow scientist "probable" carries statistical meaning and if your are a Fisherian frequentist then 95% would be that threshold. You have no estimation whatsoever about your probability - you can't even estimate it because "alive" doesn't have a measurement scale as explained

, but that's far from being a conclusion. Why do you make a choice to interpret something I've claimed as probable as a 'conclusion' ?

When someone says "very likely" most people would read that as someone is concluding as such.

Going back a few posts you claimed 'Being mathematically flawed has no bearing on whether it would sound alive or not. ' Notice no qualification, an absolute statement.

>> an absolute statement is appropriate here because at the present time alive or not can't be measured.

What makes your claim 'good science' and mine 'bad science' ?

>> I have no claim other than a desire as a fellow scientist to point out bad science when report as fact, which imho you have done. But whether you think "very likely" is not conclusive is semantics.

I will gracefully bow out of this exchange now as I have said all that I feel I need to and wish to spend next few hours listening to tunes.
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
I'll bow out too, not out of desire to listen to tunes - I can quite happily listen and type - but rather because the only statement left quoted when I hit 'reply with quote' is your last one which is not one in dispute.
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
The math is saying it isn't lossless (i.e. conversion to DSD is not fully reversible). Have you read the paper I've been referring to?
 

Yuri Korzunov

Member
Jul 30, 2015
138
0
16
The math is saying it isn't lossless (i.e. conversion to DSD is not fully reversible). Have you read the paper I've been referring to?

Conversion DSD to PCM and back for lossless definition is not matter.

Analog signal converted to PCM and DSD with losses anyway. If use this point of view both formats are lossy: PCM and DSD.
 

jfrech

VIP/Donor
Sep 3, 2012
2,152
749
1,160
Austin
Personally, I enjoy rock, blues, and jazz, and can't stand how DSD makes the vast majority of these recordings so dull. DSD sure does make things smoother, but without the snappiness PRAT, DSD sounds very artificial to me.

Personal tastes aside, does anybody understand or have any theories why the high up-sampling/ oversampling rates of DSD, and such, kill the snappiness, aliveness, and PRAT of music?

My experiences are opposite for Jazz and blues. I don't listen to much rock. No way DSD sounds artificial to me...
 

opus112

Well-Known Member
Feb 24, 2016
462
4
148
Zhejiang
Analog signal converted to PCM and DSD with losses anyway. If use this point of view both formats are lossy: PCM and DSD.

Sure, coming from analog of course all digitization is somewhat lossy. However with PCM the losses are quantified, we can select any bit length we can afford to get losses as low as we dare. The same is not true for DSD.
 

Yuri Korzunov

Member
Jul 30, 2015
138
0
16
Sure, coming from analog of course all digitization is somewhat lossy. However with PCM the losses are quantified, we can select any bit length we can afford to get losses as low as we dare. The same is not true for DSD.

In ideal case DSD go to zero quantization error/noise in low part (useful for audio applications) of spectrum.

It is good observed at DSD256 and above.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing