XXHighEnd

High Resolution Audio: Is the future really low resolution?
Article By Ryan Mintz of Core Audio Technology


http://enjoythemusic.com/magazine/manufacture/0512/

This is one comment of interest:

This article starts out nice but gets off track at the end. 1-bit (or multi single bit like dcs and ess) conversion is a can of worms that for shure does not adress the mentioned a-d conversion problems and has loads of issues on its own (like hf noise and increased jitter susceptabily). Most stuff said about sampling also hold true to 1-bit sampling, more importantly practically all pcm hires material is sigma delta "1-bit" sampled and decimated to the desired output (like 24/96). Too bad sigma delta conversion cannot be explained in laymans terms like lpcm since it is based on a highly abstract and mathematical process. Simple in excecution but very tough to grasp.

I agree that our da process, recording and playback chains aren't at all up to getting a resolution of 24bits. But for our xx playback the 24 bit DA is darn handy to attenuate digitally!

Fwiw I am in the camp that for most material I cannot observe a difference that is only due to the sampling rate/bit length. At best it sounds as good as 16/44.1, but that may be Peters' accomplishment.

Regards, Coen

Agree.

The article includes several interesting technical points. But he is cherry picking. And why does he post this biased article?

My guess is than Ryan Mintz, the owner of Core Audio Technology (manufacturer of DACs and digital preamps), is about to launch new products with 1-bit converters.

For the record:

Digital recording of 2012 has become "perfect".
Music producers and recording engineers are no longer able to discern between the live feed they hear in the control room and the digital recording of it.

Morten Lindberg of 2L (state of the art recordings of live classical music utilizing the very best microphones and recording equipment available) says he reaches this level of perfection when recording at 24/352.8kHz (DXD).
Source: http://morten.lindberg.no/category/Audio+Engineering/

Bill Schnee, when recording jazz/rock in his studio, says it happens at 24/192 PCM, using a custom build DAC.
Source: http://bravurarecords.com/n_hist.html

Buying a Blu-ray disc or a hi-rez download of such a recording gives you the live feed. You hear what they heard in the monitoring room. If your playback system is better than theirs (doesn’t take much) you will hear the music with higher fidelity than the musicians did. This is the blessing of digital.

-------

Digital recording has become so transparent that it can be considered “perfect”. Digital mixing and postproduction is near perfect, although some engineers say mixing software like ProTool has a signature of its own. My guess is that this is anyhow less harming than good old (bad) analog mixing hardware.

Important note: Some artists (rock and pop genre) prefer recording to analog. But that is because they want the “color” of magnetic tape recording (tape saturation distortion). For such artists the desired “sound” is more important than absolute transparency. -Often they choose mixing and post production in digital domain. Example: Steely Dan, Everything Must Go.
Source: http://mixonline.com/recording/interviews/audio_steely_dan_everything/

Recording trivia: In the 60s when the Rolling Stones recorded in the States, they drove directly from the studio to the local radio station and gave the DJ a copy of their latest recording session. Then they sat out in the car, in the middle of the night, stoned on drugs and booze, listening through AM mono, and deciding if the mix was ok or not, before returning to the studio.

Keith Richards’s attitude towards sound quality hasn’t changed much. In his home studio of today he prefers directing the microphones away from the instruments.
Source: "Life", his resent autobiography.


-You might ask yourself why bother to buy the old Stones albums in hi-rez? Well, there are several valid reasons for that, but that’s food for another thread than this one.

Well,

I stubled upon this:  http://www.ultrahighendreview.com/are-9624-downloads-better-than-cds/ (http://www.ultrahighendreview.com/are-9624-downloads-better-than-cds/). Mastering is a better explanation for the percieved sq differences than the inherent technicalities of the formats and conversion

But we came to that conclusion some time ago, didn't we?!

Regards, Coen

Yes we did :)

Hi all,

I wass wondering if in the light of the latest developments (music industry, XX developments) this view on highres has changed a bit? It seems that there are more and more digital masters in 24bit/192 available, is it worth revisiting this topic? Or is there a definite technical answer to all of this?

Could you point me to this discussion here in the forum?

Thanks Carsten

Hi there Carsten,

I'm not sure whether this is worth more than 2c, but I'll give it a small try anyway :

Me, myself and I may be known as someone who is explicitly against Hires. Not from this forum, but from others sure. I think I was also the very first one to come up with this explicitly being against, because I should have been the first one who could explicitly test it with a DAC that 100% does nothing to the music data - contrary to everything else out there. And, this DAC (which was the predecessor of the current Phasure NOS1 USB) was even explicitly made to check this out. Check this yout by a means of apples to apples because it behaves electrically 100% the same with 16/44.1 upsampled to 24/705.6 (or 24/176.4 if you like) and native 24/176.4.

Today all NOS1 customers can check this of course.

Now, not thinking about the rare posts like this in the Phasure forums (but you might find half a dozen others) it goes by complete nature that you won't find any others about this in the Phasure forum. So, with "by nature" I mean something like : if this weren't so, there would at least be dozens and dozens of topics about it, like it goes in other forums. And, I don't think people will hold back because they know my point of view. As said, it is hardly known.

Point is : Hires should technically be better, but no good masters in Hires exist; I have some 500 under my hands by now and there's just no reason to even start looking for the good ones. Not with the thousands and thousands of Redbook I can choose from.

Current recordings in Hires virtually don't exist, and when they do it's produced by the 2 or 3 annually (by the company concerned) and "obviously" it will not be my genre or otherwise I won't like the artists. And this all in the realm of Redbook not sounding worse than the best Hires in the first place.

This latter won't be true for everyone, but for NOS1 users it sure will. And anyway, I am one of them.
Key will be the Arc Prediction filtering (for Redbook) and as we know that works out for virtually any DAC anyway. So, it's not even really dedicated to the NOS1 although it was designed for that.

You talk about 24/192 (must be DVD-A) which has a chance. Why ? because it should have been a dedicated 24/192 2ch recording back at the time. Not so with any 24/96 which undoubtedly will have been a multi channel recording which today is turned into 2ch of that. And this can't work (needs 12dB of compression). I know, DVD-A's may come with separate 24/96 2ch tracks in the first place, but still it will have been "flawed" in the first place because a 2ch recording requires a totally different set up (in the studio / on stage) compared to multi channel. Only if the take (the recording) itself is from a different version, yes *then* 24/96 has a chance too. Well, be happy to find those.

The story is inifinitely more long and the reasons for Hires in good shape being inexistent are numerous. But, in my view all can be summarized under "back then no means were applied to make that good hires recording". As said, only 24/192 has a chance (because that will never have been multi channel).

The mere point will be that no digital mastering engineers from today seem to know how to ever get it done. So, when there would be master tapes available those engineers all know how to improve on that and next make a mess of it. But now something else comes into play : they don't even have the proper means to listen to the result. They use DACs which destroy in the first place, and painfully enough those DACs will let sound Hires better than Redbook on those same DACs. Not because the DAC is good at Hires but because it's poor on Redbook. for you, Carsten, this may come across as blahblah, but NOS1 customers will know. Only those who compare apples to oranges may conclude that some Hires is better than the redbook version, but what they don't "see" is that it's remastered in the same time, there's bass added, volume added and what not to make it sound better than the original.

The other day I visited the one and only person in this forum who is keen on Hires and so for once and for all we compared. A few settings in XXHighEnd needed to change according to me, and then we listened to Let it Bleed which is one of the very best recordings ever in my view. Not that you would have guessed that, but it is so when all is arranged for in well fashion. The result ?
No conclusion because the Hires version of Let it Bleed is mixed so much different that no comparison is possible; when you like bassy sound, you'd choose for the Hires. But when you're into definition more you'd go for the Redbook.
Did I say "no conclusion" ? Well of course for me it was clear. Buy to some extend we all listen subjectively and when you're into slam instead of accuracy you could have choosen that mix used for the Hires. So, you choose for a mix and not for Redbook vs Hires ...

Don't we all know that "best" provocer of Hires : Neil Young ? well, listen to his 24/176.4. It is unlistenable. Ear hurting. Okay, for me it is. Now take Greendale - Redbook. So open, so clear, so full of dynamics, so spatious. So gorgeous ... (through my system)

Maybe 3c,
Peter

Well, maybe a couple of other people keen on at least the *potential* of high res; I'd be one of those.

You're so right (of course) to focus on the recording/production end of things.  I started a thread over at CA about the notion of "bit perfect," but not just into the DAC inputs as most people think of it - all the way through the process from A-D and through the DAC until conversion back to analog.  I wonder how a well recorded 24/192 original would sound through the NOS1 with no sample rate conversion applied in the computer; versus the 192 interpolated to 768 with Arc Prediction; versus the 192 decimated to 44.1 by the producer then interpolated to 705.6 with Arc Prediction.

My experience for what it is worth. I have 24/96 and 24/192 hires files (and they are good recording mastered directly in those formats with no intermediate conversion) but unfortunately I do not have any music that is an identical recording both in hires and 16/44 (does anyone?) so I cannot make a meaningful A/B comparison.

What I can say is that the very best recordings I have happen to be from Redbook 16/44 CD's and they tend to be (but not always) older (early 60's to 70's Jazz recordings). By the way I did not think they were that special until I got NOS1. But through NOS1 they are very special. They are typically very simple recordings and I suppose very simple digital transfer. I cannot really comment on hires but if I am going to play demonstration tracks to someone they will all be 16/44 (both Nick & Mani have heard my system but I only played redbook to them).

Having said the above some of the 24/96 files I have are very good SQ because a lot of care was taken in the recording and mastering process (small company who *care*). But still the very best of the very best for me happen to be on 16/44.

From where I am listening it would be difficult to imagine better than I get from some of those 16/44 recordings. But I do not think that it is necessarily because 16/44 is inherently better (but who knows?) it just seems to me that the recording quality (simple and no digital engineering) is overwhelmingly the most important thing. From what I have heard so far (with the perspective of listening through NOS1) hires seems irrelevant compared with the recording quality.

What a shame the music industry cannot focus on improving sound quality. But as there is generation of recording engineers out there that do not have any experience of really good sound quality and there are all those buttons, knobs and levers on the recording console (all implying DSP) well they just can't help themselves and that is a great shame when someone like Peter has gone to a great deal of trouble in avoiding it in XX and NOS1.

P



 

But I take up the cudgels for the recording engineers. I got some of them in the circle of friends and one of them with a long experience told me that there is only 10% to 20% of the budget available for a standard production today than 30 years ago. Need I say more?

Georg

I do.

I owned a number of native hires albums produced by Chesky and Reference Recordings. Those albums are exact copy of their master hires wave files. RR uses 176.4 kHz (4 times that of red book) resolution while Chesky is 192 kHz. Chesky only delivers their hires album to US address.  I also bought the redbook version of those albums too. Example is Rebecca Pidgeon's Raven.

The popular track in Raven that I used to AB test the difference in SQ is "Spanish Harlem". With XXHE's AP and NOS1 USB,  I must admit that the difference in music presentation between the two resolutions is very subtle. With 24/192 playing in native format without up sampling (as compared to 4x AP redbook), I can feel a slightly deeper sense of stage dimension. This can be identified with only critical listening session. However, there seems to be a magic in 4xAP which delivers more musicality presentation than the native hires. Rebecca voice has a richer body blending better with instruments around her and finally give me a better sense of involvement in this piece of music.

Note that I tested this again today with similar view. PA must be off since it produces continuous ticks on hires materials.

To me, well recorded redbook can deliver stunning and excellent music presentation. My reference albums are mostly produced by Winston Ma of FIM (First Impression Music) such as the TBM Sounds which was remastered from 1970's Japanese analog master tape by Grammy award winning mastering engineers. Another reference album that Jim Smith who wrote "Get Better Sound" used extensively in audio show set up is Tears of Stone : The Chieftains. One of the best recordings I ever heard. These are all REDBOOKs.

Best,
Praphan

Thanks for sharing.

Well, these observations are not exclusive to this forum. Others have also noticed that hires is mostly "old wine in new bags" to milk the old cows a little further. Ever noticed how many versions of "Kind of Blue" are in your jazz collection?
Very little if any original hires material is available

With respect to the playback gear, last month steroephile contained an interesting article and comments about the msb platinum full option dac. The reviewer fwiw clearly preferred this discrete R-2R dac to a (also discrete) Sigma Delta DCS stack. The manufacturer indicated that meticulous execution of all dac aspecs improve the sound of "red book". Even so that on demos people where surprised to learn that no hires material was used. Shurely this can be interpreted as marketing mumbojumbo, but we know better by now. I think this article in a way supports Peters view on the adverse effect of The "wrong" (sigma delta LP FIR filtered) dacs on 44.1 vs hires.

I for shure have no special appetite for hires anymore since the nos1 got into 24/705.8 mode with the many gigs of 16/44.1 music I have collected.

The hype is starting to wear off.

Regards, Coen

I only have the 24/192 wav files but Soundkeeper Recordings are available in redbook, 24/96 and 24/192 formats. The original recordings are made as 24/192 files but Barry Diament, a highly regarded engineer and frequent contributor to the Computer Audiophile website. If you're looking for true hi res recordings to compare with redbook obtained from the same source, well, look no further.

Here's an example:
http://www.soundkeeperrecordings.com/americas.htm

Esau

True.
So, heard me ? True.
But :

His recordings are the typicle example of the 2 or 3 productions per year I mentioned (stick to 1-2 in his case) and then what.
But merely (and this is a dangerous field of course) :

It is beyond me where this Hires shows off. I'd go even further by saying that this is quite unlistenable for *me* which must be related to the means of recording (placing of microphones and such). And, no I don't like this music (not the artists at least). So, recording so "diffuse" and distant like this will not show off the resolution of an upright bass. In a for recording decorated room perhaps, but this just wasn't the idea (of Barry).

So there we go. I bought all the albums and all the versions just because it would be genuine Hires and I could compare. But there is nothing to compare because I too much don't like the music AND not the recordings in this case.


So yes, the value should be in that. But now from another angle : and how was that 44.1 created ? Right at the moment I have this problem myself; I have "recorded" a bunch of master tapes into 24/192. This was done by machinery up to the task (PMII). Now I want that on CD. Great. So, what program to use to decimate it ?
It is not much our world (as audiophiles) to decimate. We upsample, but not downsample. But, from upsampling we know how different the results can be (with Arc Prediction as an example). So, all "sound" different. Okay, all *are* different - that's the problem.
And so I really feel the safest to re-record those tapes, this time directly into 16/44.1 (HDCD encoded).

Well, I won't work it out in front of you, but I only wanted to indicate that there's also the process of how we get it into Redbook format. The differences should be as large as DACs can sound because the process needed is about the same. So, when we think Redbook can sound very good, how much of that is contributed by the A/D process and how will the older recordings sound generally so much better and open while no 96 or 192 A/D existed and thus no decimating software can have been in order (for albums digitized in, say, the 80's).

Look how difficult it is. Here "downsamplers" are compared : http://src.infinitewave.ca/
See ? all different. They could be representatives of the A/D converters used for recording analogue into digital. Different brand, different filter (which is what this again all is about). When the recording was in Hires, well, then *reallly* one of these will have been used to create that particular Redbook CD.

And trust me : all those various graphs from the various decimators will for sure sound different.

Peter

For me, this is really hitting the nail on the head again, as far as what I wanted to get to with my "bit perfect" thread on CA.  With the challenges of decimation on the one hand and interpolation on the other, and *knowing* all the converters will sound different (and therefore none can be ultimately "right," though it is possible to get very, very close), why do this down and up conversion? 

There were plainly reasons once upon a time, and to a certain extent these procedures have become set in stone now in the recording industry even though the technology has progressed to the point where we should no longer need them.

A couple of other comments:

- Though Peter has referred to decimation and interpolation as the same sort of process, I wonder if decimation might even be potentially more destructive.  With interpolation there may be the possibility of retaining all the original samples, while with decimation you know you will be throwing some away, a higher proportion the higher the sample rate of the original.  So is it better to sample in the A/D step at 44.1, or to sample at 176.4 and decimate?

- Very interesting comment from praphan that he finds himself preferring the interpolated 44.1 version of a recording to the "native" resolution, though the two are quite close.  With something as good as Peter's software, and so customizable, the dividing line between more accurate and more pleasing has got to be *very* difficult to ascertain.  My usual test (how different various recordings sound from each other, or to flip it around, how much of its own sound the hardware or software you're listening to imparts to the music) might not be enough in such circumstances. 

Really, for such a task one would ultimately like to be able to compare "native" and decimated-then-interpolated versions from a session where one was present for the recording.

:offtopic: (sort of)

LOL Jud. I know you mentioned it before but I forgot to look (or was busy maybe :)). I just did at light speed but it has almost become impossible to say something with sense.
Maybe later, when I can really spend some time on it.

Btw, I also saw some nice matches between what I said here and what was suggested by some (which is never SP...etc. :))
But I can promise you, I never read it until just now. Nor it can have been read by others what I wrote today. Funny. And it wasn't even you motivating it. Still you did overthere ...

Sorry to be cryptic. I think it's funny ...
One day we will be there I guess.

Peter

While I acknowledge that Barry Diament's recordings are not everyone's musical taste, the same can also be said about any recording ever made. Most of the music that any of us listen to is recorded with close miking and that goes for the overwhelming majority of your and my favorites. At the same time, I agree with the notion that musicians and engineers have a right to use nontraditional means of recordings to fit their own vision of how the music should sound. I have AIX, 2L and Naxos discs that bravely go that route producing interesting musical experiments that I also enjoy. Though part of a niche market, I don't find those recordings to be unlistenable and I also don't think it impossile to compare sound quality when listening to differently formatted recordings that are not exactly one's cup of tea.

Still, I have to admit that the bulk of music recordings that most of us(myself included) want to hear are available as popular CD's in redbook format and, short of a flood of legacy recordings by companies like WMG becoming downloadable for Neil Young's Ponos device(and I don't even want to begin to get into a debate right now about that enterprise which has yet to surface in the marketplace) and usable by folks using non-Ponos playback systems, our favorite recordings will continue to only be available in redbook format.  I find it interesting that some folks on the CA website have dissed the HDTracks 24/96 download of Crosby, Stills and Nash while Steve Hoffman's clearly superior remaster of the original recording is available as a redbook CD. I've listened to the Hoffman remaster as an XXHE upsampled file playing through my NOS1 and have not been left wanting.

The point that Peter makes about downsampling a 24/192 wav file to 16/44.1, effectively decimating, if you will, the original recording, is an excellent one. While not the same thing as, say, downsampling a 24/44.1 or 24/192 Beatles recording to an mp3, it is still a move in the same direction. I agree with Jud's proposition that the ultimate test is to listen to and compare recordings of performances where you were present at the actual performance.

Esau

Isn't downsampling more of a problem than upsampling?. Upsampling moves aliasing images outside the audio band the more upsampling the further outside the audio band (a key reason to upsample). Whereas downsampling moves the aliasing images inside the audio band (yuk) so downsampling has to involve a brick wall filter which is the only way the effect of those aliasing images can be mitigated.

So why not take an analogue recording and create one digital master at say 24/192 and another at 16/44 and then compare them? Peter - can you do that with the PMII? I guess maybe it does not which would be a great shame.

P

Synchronous downsampling is easy (ie from 192 to 96 or to 48). Compare this to Peter's AP algorithm: allways exponents of 2 (2-4-8-16). Downsampling is the same, just drop half the bits. I am shure it will make a difference to the reconstruction however which bits you drop.

Async interpolation needs interpolation (ie from 96 to 88.4). This is done by finite bandwidth filters. Here you go. Options and implementations are wide open. Many ways to mess it up.

The only way to truely compare is indeed to record 44.1 vs 192. But beware, adcs that can record 44.1 ' native' have not been made for many years (if ever). With 44.1 coming out of an adc there is allways a high risc that some dsp going on since professional recording standard is 48 kHz (or multiples). The best way to create an 44.1 kHz master is to start out at ie 176.4kHz and decemate from there. No dsp involved to do this!

Regards, Coen

Coen ... No way downsampling works like that !
Sorry ...

(with Upsampled Arc Prediction, yes because that is a lossless means; but that's quite another beast)

Ok I've havent been stating which adc process I assumed, but sigma delta adc chips are doing just that: dropping samples for synchronous conversion (but this also has the prupose to establish the sample value in this process).
Halving the number of samples in sync conversion is mathematically correct because in the higher sampling rate all the bandwidth of the halved rate is allready present and no interpolation is necessary to obtain the samples within that reduced bandwidth (the aliasing is a reconstruction problem in this case). That is the same reason why you need an AA filter for asynchronous conversion of sample rates (otherwise you would interpolate aliasses into the lower rate).

Then there is also the unmentioned issue of reducing the bits from 24 to 16....

Am I missing a point somewhere?

Regards, Coen

Added: from a theoretical vieuwpoint it does not matter which samples you drop in sync conversion. Both are perfect descriptions of the same signal PROVIDED that the conversion back to analog is perfectly limited to the nyquist frequency. You can find some flash on the net illustrating this counter intuitive phenomenon.

Coen, not to debate it, but to give some counterweight to people who might like to pick up something : still not true.
I'd say : try Linear Interpolation and hook up an analyser (if you don't hear it in the first place). That's just injecting samples like you describe ...
(I should remove that stupid option :))

Peter

I chewed a little on my post over dishwashing and you are right (as allways :))! What I posted is not true and contains a flaw and you will end up with aliasing into the target bandwidth.

My post only makes sense if you filtered out all frequencies above the target nyquist freqency in advance. Only then you will (at least theoretically) end up with samplesets that all represent the same waveform (with synchrounous downsampling).

My synchronous 'experiment':
So take a 176.4 kHz original (88.4 kHz audio bandwidth). Filter it to 22.05kHz audio bandwidth digitally with a brick wall filter and then you will end up with four sets of 44.1 kHz samples all representing the same waveform. So no new samples on the time axis, but recalculated sample amplitudes.

So no way to downsample without filtering it first. This is the basis of Shannon!

The adc chips contain many of these filters. You will have to down sample from 24 MHz or something to 192k...


Well this is not what I had in mind of course. With asynchronous conversion you also change the timebase of the samples (only once every x samples you end up with a sample timed on the original timebase). The filter will determine the amplitude of all samples, so no one in his right mind will use Lin Interp for that.

I have allways wondered what the above synchronous experiment will result in. Do the four 44.1kHz sub files all sound indentical after proper filtering of the 176.4kHz 'motherfile', like they theoretically should?
This would be quite illuminating. Actually my idea is that they will NOT...

Back to the sq of hirez material.
Theoretically the hires in native samplerate contains much more original information than the downsampled 44.1 red book variant. You will have to filter and reduce bits by some inherently lossy process to get from the master to red book format whatever you record with these days. I don't buy 'native' 44.1k.
In practice you seem to have ameliorated this loss to something insignificant, like we experience with xx and the NOS1.
Thanks for that!

Regards, Coen