High Resolution Audio: Does It Matter?

[solderdude]

It’s as if you disregard the message in the opening post, where @amirm concludes thus:

«As I mentioned at the outset, high resolution audio makes a difference. And a huge one at that in the way it gives us access to stereo masters prior to re-mastering for the CD. That path means the music can be free of loudness compression which will have clear benefit, putting aside any additional sonic fidelity due to use of higher bit depths and sampling. Given the fact that CD has no choice but to go away in the future, we as enthusiasts better get behind high resolution audio distribution. Nothing but goodness comes from having more choices of formats for our music».

Let's have a look at Amir's conclusion... the part below is something I don't agree with 100%.

"high resolution audio makes a difference. And a huge one at that in the way it gives us access to stereo masters prior to re-mastering for the CD"

Of all the high res offerings, what percentage would you estimate are made directly from stereo masters and have not been remastered and compressed to hell to give the illusion of a higher dynamic range than CD.
Keep in mind: A compressed recording has the differences between softer sounds and the loudest made smaller. In this case you can hear the softer sounds more clearly (as they are louder in comparison) and 'people' say... wow I can hear the softer details much better now so the dynamic range is 'higher'.

Also, I think a CD can be free of loudness compression and loudness compression stands loose from the CD format and is forced upon us by record companies that want to accommodate people playing music in the car and directly from phone or BT speakers. For that category of listeners (the biggest chunk of the music consumers) compressed to hell music is ideal.

Compression is not needed for the CD format.
It is needed for Vinyl which has a smaller dynamic range. So the notion that one needs hires to get non-loudness compressed offerings is not valid. They just need to master a CD without loudness compression.
Masters that already are compressed on individual tracks can not be un-compressed, so higher res won't give us better sound in that case.

I do agree with 'aside any additional sonic fidelity due to use of higher bit depths and sampling' as arguably a higher bit depth and perhaps a somewhat wider FR won't hurt the fidelity and is indeed 'Nothing but goodness comes from having more choices of formats for our music'.

The discussion here is whether for consumers 96/24 is already more than enough 'hi-res' and the need forhome reproduction in DXD, DSDx8 or 768/32 and whatever next will come will be giving us even better sound quality free of loudness compression.

At some point the resolution and needed bandwidth/memory is enough.
Loudness wars compression is not related to any format.

 

[Blumlein 88]

snippage.........

At some point the resolution and needed bandwidth/memory is enough.
Loudness wars compression is not related to any format.

What rubbish is this? Enough is never enough. Only too much is enough. DXD is just barely getting us to the threshold of nearly having the capability to sometimes acceptable sound quality. Once someone hears the huge improvement this makes then they see the way clear to getting good sound with even higher rates. 1 biggagigahertz sample rates are what will be needed in the future. We aren't there yet. People with your attitude that 96 khz is more than enough..................people like that are holding us back. Get with the program partner. Come on in for the big win.

 

[Thomas savage]

What rubbish is this? Enough is never enough. Only too much is enough. DXD is just barely getting us to the threshold of nearly having the capability to sometimes acceptable sound quality. Once someone hears the huge improvement this makes then they see the way clear to getting good sound with even higher rates. 1 biggagigahertz sample rates are what will be needed in the future. We aren't there yet. People with your attitude that 96 khz is more than enough..................people like that are holding us back. Get with the program partner. Come on in for the big win.

No mater what the resolution you can’t beat a vinyl record for fidelity so you are all arguing about what’s second best #fact

 

[RayDunzl]

It’s another example of subjective indulgence trumping logical thought, “ man I really love this new tool” yea right jog on mate lol

I bought a new tool... I've fondled it a bit, but not put it to its (probably) one-time use yet.

The individual parts ring like a tuning fork... Must be some hardcore steel in it.

 

[tomelex]

Now you done it Ray, here goes the cooks giving us better quality ground clamps, maybe cryo treated or perhaps made from 99% ofc or the sky is the limit. blacker blacks, higher highs, lower lows, as everyone knows....

 

[Sal1950]

I bought a new tool... I've fondled it a bit, but not put it to its (probably) one-time use yet.

So you gonna leave us in suspense? What's it do?
Inquiring minds want to know!

 

[Blumlein 88]

Now you done it Ray, here goes the cooks giving us better quality ground clamps, maybe cryo treated or perhaps made from 99% ofc or the sky is the limit. blacker blacks, higher highs, lower lows, as everyone knows....

No that is not it. It is the blessing by Bhuddist monks. Simply read the writing on the side facing us in the picture.

 

[Guermantes]

No that is not it. It is the blessing by Bhuddist monks. Simply read the writing on the side facing us in the picture.

"Wear safety glasses" (?)

 

[Mark S.]

Could I just make the observation that hi-res transfers of old, analogue tapes often sound far better than they ever could at 16/44 for the simple reason that it better deals with the obvious flaws in them. In particular 'scrape flutter', which itself adds HF spuriae to HF (harmonic) content of the music, which the steep digital filtering during 'reconstruction' of said frequencies at 16/44 then makes a mess of. This wasn't such a problem when the tapes were cut to laquer for LP stamping, and is the reason so many CD's which SHOULD better the LP versions don't - often by a long chalk.

ETA >> check out https://www.plangentprocesses.com/ for de-facto proof of this.

 

[tomelex]

Could I just make the observation that hi-res transfers of old, analogue tapes often sound far better than they ever could at 16/44 for the simple reason that it better deals with the obvious flaws in them. In particular 'scrape flutter', which itself adds HF spuriae to HF (harmonic) content of the music, which the steep digital filtering during 'reconstruction' of said frequencies at 16/44 then makes a mess of. This wasn't such a problem when the tapes were cut to laquer for LP stamping, and is the reason so many CD's which SHOULD better the LP versions don't - often by a long chalk.

ETA >> check out https://www.plangentprocesses.com/ for de-facto proof of this.

oh, oooohhhh, ooooo hh boy, don't touch those master tapes, they were perfect, digital interbreeding with pristine tape outputs can only lead to this, thump, thump, the sound of audiophiles fainting to the floor. Any how, that is interesting stuff, cool, thanks for the information.

 

[Blumlein 88]

Could I just make the observation that hi-res transfers of old, analogue tapes often sound far better than they ever could at 16/44 for the simple reason that it better deals with the obvious flaws in them. In particular 'scrape flutter', which itself adds HF spuriae to HF (harmonic) content of the music, which the steep digital filtering during 'reconstruction' of said frequencies at 16/44 then makes a mess of. This wasn't such a problem when the tapes were cut to laquer for LP stamping, and is the reason so many CD's which SHOULD better the LP versions don't - often by a long chalk.

ETA >> check out https://www.plangentprocesses.com/ for de-facto proof of this.

Firstly, as a relatively new member welcome to the forum.

I read the AES paper from the link you posted. I'm not sure I agree with your view of it exactly.

They use 96 khz to better see and adjust or de-modulate frequency modulated shifts in speed. They are even down-modulating the higher tape bias frequencies to do this. In a sense they take a very high frequency and down convert to fit between audio and the extra bandwidth of 96 khz so they can use that signal to make adjustments in speed and flutter. So 44 khz would fail to catch that or give them the somewhat dead band to work with. The difference in the sound at the end however isn't because 44 khz sampling made a mess of scrape flutter. It is that their process let them fix the scrape flutter.

Also interesting in that they describe flutter as blunting transients while the tape sound with warmer tone is a big part of what tape is all about. While all the time audiophiles complain about digital transients.

 

[Wombat]

Could I just make the observation that hi-res transfers of old, analogue tapes often sound far better than they ever could at 16/44 for the simple reason that it better deals with the obvious flaws in them. In particular 'scrape flutter', which itself adds HF spuriae to HF (harmonic) content of the music, which the steep digital filtering during 'reconstruction' of said frequencies at 16/44 then makes a mess of. This wasn't such a problem when the tapes were cut to laquer for LP stamping, and is the reason so many CD's which SHOULD better the LP versions don't - often by a long chalk.

ETA >> check out https://www.plangentprocesses.com/ for de-facto proof of this.

If there are transport errors on master tape recordings then surely they will be present on both vinyl and digital copies.

 

[Mark S.]

The various types of digital filtering/reconstruction all require certain compromises to be made - you trade off aliasing against phase-shifting against pre/post ringing. In 16/44 that steep filter slope has always been perilously close to the audio band, and I don't think it's arguable that this hasn't caused problems with 'imperfect' (sometimes practically defective) analogue sources, notably the aforementioned scrape-flutter, which while still present on vinyl transfers, wasn't then exacerbated by that filtering. Does anyone really not remember being bitterly disappointed by the sound of CD's that they replaced their LP versions with?

Incidentally, what Plangent are doing is analogous to de-juddering and de-weaving film stock, which is the first process implemented during transfer and restoration, and makes restored movies so much more watchable. Check out any decent, modern DVD or Bluray transfer, or even digital terrestrial or streaming broadcasts of old(er) movies - the image is almost always rock solid (watch a scene with no camera movement) in a way the film reels they're taken from weren't even close to.

ETA >> this used to be a real problem - I worked in one of the first DVD authoring facilities in Europe back in 1998, they were using one of about seven rack-mount Panasonic MPEG2 encoders in the world and the results were often, frankly, pretty dire. Even with an experienced operator running multi-pass encodes you'd sometimes see artifacts like people's features moving about on their faces slightly as the encoder moved macro-blocks about trying to deal with the frame judder on the raw Digi-Beta transfers.

 

[Wombat]

The various types of digital filtering/reconstruction all require certain compromises to be made - you trade off aliasing against phase-shifting against pre/post ringing. In 16/44 that steep filter slope has always been perilously close to the audio band, and I don't think it's arguable that this hasn't caused problems with 'imperfect' (sometimes practically defective) analogue sources, notably the aforementioned scrape-flutter, which while still present on vinyl transfers, wasn't then exacerbated by that filtering. Does anyone really not remember being bitterly disappointed by the sound of CD's that they replaced their LP versions with?

Incidentally, what Plangent are doing is analogous to de-juddering and de-weaving film stock, which is the first process implemented during transfer and restoration, and makes restored movies so much more watchable. Check out any decent, modern DVD or Bluray transfer, or even digital terrestrial or streaming broadcasts of old(er) movies - the image is almost always rock solid (watch a scene with no camera movement) in a way the film reels they're taken from weren't even close to.

ETA >> this used to be a real problem - I worked in one of the first DVD authoring facilities in Europe back in 1998, they were using one of about seven rack-mount Panasonic MPEG2 encoders in the world and the results were often, frankly, pretty dire. Even with an experienced operator running multi-pass encodes you'd sometimes see artifacts like people's features moving about on their faces slightly as the encoder moved macro-blocks about trying to deal with the frame judder on the raw Digi-Beta transfers.

I hadn't noticed the first paragraph stuff in my LPs and CDs.

The LP vs early CD opinions have been well covered in more objective terms.

I am open to new evidence when it is substantiated.

Unsubstantiated subjective opinions re audibility backed up by unsubstantiated subjective opinions re audibility don't cut it with me.

 

[Mark S.]

Firstly, as a relatively new member welcome to the forum.

I read the AES paper from the link you posted. I'm not sure I agree with your view of it exactly.

They use 96 khz to better see and adjust or de-modulate frequency modulated shifts in speed. They are even down-modulating the higher tape bias frequencies to do this. In a sense they take a very high frequency and down convert to fit between audio and the extra bandwidth of 96 khz so they can use that signal to make adjustments in speed and flutter. So 44 khz would fail to catch that or give them the somewhat dead band to work with. The difference in the sound at the end however isn't because 44 khz sampling made a mess of scrape flutter. It is that their process let them fix the scrape flutter.

Also interesting in that they describe flutter as blunting transients while the tape sound with warmer tone is a big part of what tape is all about. While all the time audiophiles complain about digital transients.

I believe that they often start with a transfer at 384KHz. And yes, IMO the effects of the absolutely minute timing errors which is basically what flutter consists of, goes far beyond affecting only HF (such as 'ambience cues', harmonics, whatever).

*Audiophile Anecdote Alert* - I spent an evening about 16 years ago with Max Townshend at his home, listening to his own 'development' system - vinyl, CD, DVD-Audio and SACD (the latter on his modified Pioneer 747 universal player). One of the most fascinating things I've ever experienced was the effect of switching his super-tweeters in and out - there was a quite noticeable improvement in the apparent dynamics of instruments like stand-up double-bass - there was no mistaking it. I asked to hear *only* the super-tweeters and could *just* make out what sounded like a pair of ear-buds playing at low volume at the other side of the room.

 

[solderdude]

Does anyone really not remember being bitterly disappointed by the sound of CD's that they replaced their LP versions with?

Actually I remember the first CD's to be quite well sounding.
At least when played louder and with some tone control added.

Instead I am bitterly disappointed by the sound of current and re-releases that have been 'remastered' and compressed to death.

In all cases (video and audio) I think the quality of encoders has come a long way since they were invented and applied in early stages.

 

[Wombat]

I believe that they often start with a transfer at 384KHz. And yes, IMO the effects of the absolutely minute timing errors which is basically what flutter consists of, goes far beyond affecting only HF (such as 'ambience cues', harmonics, whatever).

*Audiophile Anecdote Alert* - I spent an evening about 16 years ago with Max Townshend at his home, listening to his own 'development' system - vinyl, CD, DVD-Audio and SACD (the latter on his modified Pioneer 747 universal player). One of the most fascinating things I've ever experienced was the effect of switching his super-tweeters in and out - there was a quite noticeable improvement in the apparent dynamics of instruments like stand-up double-bass - there was no mistaking it. I asked to hear *only* the super-tweeters and could *just* make out what sounded like a pair of ear-buds playing at low volume at the other side of the room.

So? Where does that take us?

 

[solderdude]

I asked to hear *only* the super-tweeters and could *just* make out what sounded like a pair of ear-buds playing at low volume at the other side of the room.

One has to keep in mind that most supertweeters cross-over in or just above the audible range and, more often than not, still produce sound in the audible range as they are often filtered 6dB/oct. Switching this on and off while acoustically interfering (phase) with the 'normal' tweeters may well be audible and is no 'evidence' that one can hear above the audible range.

 

[Mark S.]

Firstly, as a relatively new member welcome to the forum.

thanks, BTW.

... The difference in the sound at the end however isn't because 44 khz sampling made a mess of scrape flutter. It is that their process let them fix the scrape flutter.
...

Well, there is a consensus (for what ever that's worth) that the Red Book releases of almost everything they've done are night-and-day, order-of-magnitude better than previous ones. I guess the only way of confirming my argument (that 16/44 cannot deal with the flutter benignly) would be to compare these with 'raw' hi-res and 16/44 transfers the tapes. Easy!

 

[Wombat]

thanks, BTW.

Well, there is a consensus (for what ever that's worth) that the Red Book releases of almost everything they've done are night-and-day, order-of-magnitude better than previous ones. I guess the only way of confirming my argument (that 16/44 cannot deal with the flutter benignly) would be to compare these with 'raw' hi-res and 16/44 transfers the tapes. Easy!

And whom do you think may do that to support your view when you don't propitiate it yourself?

Seriously unsupported conjecture put forward in audio forums is a weakness of the medium. Let's do better.