[Geardaddy]

According to a 2008 Stereophile survey, "a whopping 32% of you already have a server of some kind set up, and 44% are ready to jump in. We received more votes in this poll than any other in 2007!"

Computer audio is here to stay.  Given that fact, I was curious as to people's experiences with hi rez formats (particularly downloads) and any comparisons to older media such as CDs, LPs, and master tape?  Does it live up to the hype?

I have sound engineering types who claim that Redbook has all the necessary resolution as a media format, and that the crux of the issue is not one of newer formats but execution.  Bad recordings are bad recordings, and upsampling and other trickery in the digital domain cannot fix that.  A related claim is that the hype is driven by the desire for new markets.

[skunark]

There's been discussions on both sides of the front.   Redbook focuses on frequencies that we can hear, but does the fact that it chops off the higher frequencies remove any harmonics?  You could argue that records have there own issues with the RIAA equalization, historically, this isn't a new battle.   

I think moving from 16-bit to 24-bit is advantageous to improve the overall dynamic range of your system, but can most systems playback 120db?   Probably not...   

Also increasing the sampling frequency will allow you playback frequencies higher than 20kHz SACD/24-bit files can playback around 30-50 kHz.  192/24 will allow you to playback 80kHz.  Also very doubtful that most systems can playback higher than 25kHz and most of the older "ears" probably don't hear past 18kHz.   

Most manufactures are still producing specifications slanted towards 20-20kHz, so that is contradicting any "high-resolution" claims they make, either they just haven't tested it or perhaps based on the circuit or cross-over designs they filter out the higher frequencies.   Does it matter?

In the end, I think it's too early to discount the benefits of having a file direct from the studios even with the above issues noted.  I actually would prefer the final "mastered" file vs having it down-converted to 44.1/16, one less conversion process.  Your "sound engineering types" should at least see that benefit with that as they use Pro Tools.

What I find interesting along these lines are all the "32-bit" DACs coming out, we barely have any high-resolution sources, let along does S/PDIF or HDMI support that.  So without any material or source components to drive the signal to the DAC, what are they after?   Hopefully their marketing department isn't getting confused by referring to CPU word length vs bit depth supported for a given file.

[werd]

To me its a do or die.. If i am going to pay for files i want good recordings. Whether they are hi rez or not they better be sounding good. They can change the name to - This is a dam good recording - for all i care.
I think it would help if the people who remastered also included a methology to their remastering. Little tidbits like - we wanted to bring out the fillers a little bit more since they reinforce the beat of the music.... that kind of stuff. Probably dreaming there but it would sure be nice.

[mcgsxr]

For me it comes down to format wars.  I have close to 1TB of FLAC of redbook.  I currently have 0 of anything hi rez.  I won't buy anything I already own in hi rez, and am at a point where I no longer frequently buy music.

So, for me, it is about optimizing the playback of the material I already own.

That does include a dedicated PC, and a modded SB3, and I am in the market for a USB DAC for my headphone setup, but nothing hi rez.

For the record, I never bought any DVD-A, or SACD, so I guess I am a luddite after all!

Also for the record, I do understand that folks currently playing hi rez are very happy with it, and I do read threads about it.

[HT cOz]

Can someone who has listened to master tapes on a good system say if any digital can compare to great analogue?

If it doesn't then we still have room for improvement.

[Elizabeth]

I'm in the 'not yet' camp. Primarily due to buying used music exclusively. Why would I buy expensive stuff when i can find far more music I want locally on used, cheap LPs and CDs.
i mean $1, $2 a pop albums... Locally. And not trash/trashed either. The good news for cheapskates like me is everyone is dumping CDs. The mountains of bargain bin stuff is HUGE. So they just dump first rate stuff into them due to having more than they can handle in the regular used bins. (I am good at searching, i can sit and go through the bins for hours.. being retired and all)
I can get 15 really great albums (All Music Guide rated five star) for $25. nearly any time I go look.
So until I find a need for the costly hiRez downloads, it is secondhand music for me. (Also, I had been there and done that with rebuying all my stuff on CD back in the 1980's, when they came out. I am not going to ever do that again. Ditto reason for sticking with DVD rather than Blue Ray)
Between the Lps, and the CDs, Then Amazon for specific items.. Buying only stuff that is cheap there too. (eventually I will find it for less, in perfect condition, so why buy it for more?
I already have 6,000 Lps and about 3,000 CDs. My LP playback equipment is better than my CD playback stuff.
With the hard copy on hand, why bother filling up a hard drive. IMO.
So eventually I am certain I will add a server type of music in addition to my current setup. But that is still a ways in the future.
=========
finally, I am not in the music must sound perfect camp. I have a good system, but if am album has good music, I do not get 'bent' over poor sonics. So the improvement of HiRez, while nice, is not alone enough for me to get into it yet. When it is the norm, and most stuff is in high rez, yeah, i will do it. But i am currently still having plenty of fun with the Lp and Cd formats.

[Geardaddy]

Agreed to all of the above.

I have close to 700 CDs all on my HD, and I will say, after being to thumb through your entire collection via an Iphone, its hard to go back to a physical medium.  Furthermore, while some of the higher rez I have heard sounds good, I am not yet chomping at the bit to replace my library.

skunark, very good points.  There are indeed technical limitations of the media and playback technology that no one seems to talk about.  I too would rather have the master file than Redbook.  On the flip side, if most of the music I own and love was recorded in one particular format, why would I want an upsampled version?  Not sure about the merits of that approach.  Someone more technical needs to chime in.

I have friends who are tape heads, and they all say that master tape still wins the day with "saturation" and an organic, bewitching quality.  I am contemplating getting a analog deck just to experiment. 

Digital still seems to have a ways to go.  Its as if the brain/ear can detect the seams in digital reconstruction of bits. 

[firedog]

I have a few hi-res files and all sound very good. Several sound better than the equivalent in 16/44.1.

There are good arguments that say we "can't tell" the difference and at least one well done DB test that confirms it.

However, there is no shortage of recording professionals who swear by hi-res, and say that hi-res is indistinguishable from their live mic feeds, etc (and that this isn't true for 16/44).

Barry Diament (producer) made a good argument for why 24 bit is significant over at computeraudiophile: http://www.computeraudiophile.com/content/debate-24-bit-it-just-audiophoolery#comment-72930

Of course it may be that hi-res files sound better b/c they generally are more carefully produced and mastered, aren't volume compressed, and not b/c of any intrinsic superiority. In the end, I don't really care as long as I get something with high quality sound.

[Bigfish]

I have a few hi-res files and all sound very good. Several sound better than the equivalent in 16/44.1.


Of course it may be that hi-res files sound better b/c they generally are more carefully produced and mastered, aren't volume compressed, and not b/c of any intrinsic superiority. In the end, I don't really care as long as I get something with high quality sound.

Recently I have been purchasing and downloading music directly from on-line sites both in 16/44.1 and some Hi-Rez Files.  My experiences, thus far, with downloading the digital files has been stellar as I have received excellent quality recordings that I believe sometime exceed purchase of a CD/ripping.

I tend to agree with the above comment that hi-res files may sound better because of the special handling.  I also believe that most audiophiles are willing to pay more for a recording that has had special treatment. 

[Vincent Kars]

Excellent points by Skunark.
Another one is if our DAC is capable of resolving the bit depth.
Cheap ones probably do 11/12 bits accurately, the best ones up to 20.


Here you see one doing 70/6=11.6 correctly

[skunark]

skunark, very good points.  There are indeed technical limitations of the media and playback technology that no one seems to talk about.  I too would rather have the master file than Redbook.  On the flip side, if most of the music I own and love was recorded in one particular format, why would I want an upsampled version?  Not sure about the merits of that approach.  Someone more technical needs to chime in.

So why would you want the "down-sampled" version?      It all boils down to the algorithm used to up-sample or down-sample a file, but the down-sampled will have a little bit of intrinsic jitter when compared to the original sampled signal, which it will too have quantization noise when compared to the analog waveform.  The higher the sampling frequency will produce a waveform that is closer to the analog signal when compared to a slower sampling frequency.   If someone had a Microphone, A/D and D/A converters along with an oscilloscope they could show you the difference in the waveforms at the input of the A/D vs the output of the D/A with the various sampling frequencies..    Probably something googlable.

[DanH]

Like all things, it just depends. I download albums from hdtracks.com. Some albums are difficult to say that they sound better than a Redbook cd. Some can be worse. I downloaded a Elton John and Leon Russel album in Hi rez that has no dynamics and no high frequencies. It would have sounded just as good as an mp3. Then there are the gems such as Rebecca Pidgeon's The Raven and Steely Dan's Gaucho and Art Peppers Moder Jazz Classics that just sound exquisite in hi rez.

[joeriz]

In my opinion -- based on my experience with the high rez titles that I own -- the care given in recording, mixing, and mastering is much more important than whether or not a given title is delivered in 'high rez' to the consumer.  (When I say 'mastering' here I'm thinking mainly of things like EQ and how compression is applied.)

With the high rez titles that I own I am hard pressed to hear a clear difference in quality that I can attirbute to it being high rez.  However, I own some standard 'redbook' recordings that make me completely forget that there is such a thing as high rez when I listen to them.  I'll cite the MFSL remaster of Beck's "Sea Change" as one such title...absolutely exquisite...and I own the SACD version as well.  Also, something like Steve Hoffman's remaster of "McCartney" is just so good that I can't really imagine it sounding any better.

Granted, you could say that my system and/or my ears are not up to the task, but so be it.  I'm just relating my experience.

Joe

[Geardaddy]

So why would you want the "down-sampled" version?      It all boils down to the algorithm used to up-sample or down-sample a file, but the down-sampled will have a little bit of intrinsic jitter when compared to the original sampled signal, which it will too have quantization noise when compared to the analog waveform.  The higher the sampling frequency will produce a waveform that is closer to the analog signal when compared to a slower sampling frequency.

Very good question.  I did not complete my thoughts since it was early in the am.  The issue I was trying to address is:  what does "up-sampling" add to the equation?  Is there anything deleterious about up-sampling just as there is with compression to 16/44.1?  I know the higher frequency better mimics the analog waveform, but I have toyed with up-sampling using Pure Music recently, and I do not care for the effects.  I blinded my wife to the tracks, and she also preferred NOS.  I know this is a whole other potential debate and it may not correspond to the topic at hand due to issues of software and hardware (Protools) implemented. 

Still, most music, at least in recent digital past, was recorded in 24/96, correct?  HD tracks now offers Rebecca Pidgeons "The Raven" in a 176kHz/24bit format.  That album was recorded in 1999 and presumably not in 176kHz/24bit.  It would be interesting to do a blinded listening session of all three formats (CD; 24/96/24/176).  I have the Redbook version of that CD and it is good.  I also have a friend who recently got the HD tracks hi rez equivalent and also said it was spectacular. 

I think hi rez downloads are obviously the future.  I am mostly curious about the technical logic behind this.  I too have heard "hi rez" that sounds worse than the CD and vis versa.  Very curious.  A matter of recording implementation obviously and not format.

As a final note, can anyone address, from experience, the difference between master tape and hi rez?  There are purists out there who state that most modern recordings rely on Protools which is a sonic compromise and has its own "sound."  One word that constantly comes up to describe the difference is "saturation."  Hi rez is dry and lacks "saturation," which I intuit as organic richness and fluidity.  Digital, even hi rez digital, still sounds like digital to me if you listen closely enough.  Its almost as if our brain/ear interface can discern the artificial, reconstructed waveform that it is....

[bdiament]

Hi Geardaddy,

According to a 2008 Stereophile survey, "a whopping 32% of you already have a server of some kind set up, and 44% are ready to jump in. We received more votes in this poll than any other in 2007!"

Computer audio is here to stay.  Given that fact, I was curious as to people's experiences with hi rez formats (particularly downloads) and any comparisons to older media such as CDs, LPs, and master tape?  Does it live up to the hype?

I have sound engineering types who claim that Redbook has all the necessary resolution as a media format, and that the crux of the issue is not one of newer formats but execution.  Bad recordings are bad recordings, and upsampling and other trickery in the digital domain cannot fix that.  A related claim is that the hype is driven by the desire for new markets.

Those who claim Redbook is all that is required are, in my estimation, thinking theoretically, not practically.  They are also thinking in terms of sine waves and not music, which does not resemble sine waves.

A few points from a post I put up elsewhere, which Firedog linked to:

The problem for many listeners is that different formats are issued from different masterings, often done by different engineers in different rooms, sometimes with different source tapes.

The only way to conduct a fair comparison is to use different formats created from the same mastering.

Another problem arises when folks look at 16-bit vs. 24-bit only in terms of "dynamic range". I put that last in quotes because they are really confusing this with signal to noise ratio.

Even if we consider dynamic range, from this standpoint, having the noise floor 96 dB down from max level, as it is (in theory) with 16-bits, sounds more than adequate, particularly for most pop recordings where the real dynamic range of the final result rarely gets much past 12 dB and if often (scarily!) less.

But that is only a tiny part of the story. Much of the information in music or in a recording of music (and here I'm speaking of a good recording) is well below the "top" of the envelope we see in a musical waveform. That is, it is much lower in level than the foremost sounds we hear.

Instrumental harmonics, for example, the parts of the sound that differentiate a C played on a guitar from a C played on a piano, or a C played on a Baldwin piano from a C played on a Steinway, are much lower in level than the fundamental (the C itself).

Spatial information too - what tells us about the space the players are in, the size of the room, its "character", is much lower in level than the loudest sounds we hear at a given moment.

If this information is say, 12 dB lower in level, it will be quantized using approximately 2 bits fewer than the total word length (i.e. 14 bits in a 16-bit encoding, 22 bits in a 24-bit encoding). If it is say, 36 dB lower in level, it will be quantized usings approximately 6 bits fewer than the total word length (i.e. 10 bits in a 16-bit encoding, 18 bits in a 24-bit encoding). Some information, such as the end of a reverb tail such as in a recording made in a large room, where the music ends suddenly, can be well more than that 36 dB lower in level than the loudest sounds and will be encoded with correspondingly fewer bits.

This manifests itself in the thinned, bleached and coarsened instrumental harmonics in even the best 16-bit recordings, as compared to a good 24-bit recording (or of course, the original sound in real life). It also manifests itself in the defocusing of the spatial information in the 16-bit recording compared to a good 24-bit recording (and real life).

How audible this is, will depend on the listener and on the system they're listening to. I find that on a good system, the differences are quite far from subtle, with the 16-bit (good as it might otherwise be when heard in isolation) sounding coarse and quite out of focus compared to the 24-bit version.

I have created a Web page that I hope will help folks make such comparisons. (See www.soundkeeperrecordings.com/format.htm )
This page has versions that vary in word length and sample rate - because these are the final, release versions that are available. (Perhaps it would be even more useful for the purpose of this conversation, to have versions at the same sample rate and where only the word length varies.)


To answer your question: "does it live up to the hype and why?"
it depends.
It depends on the hardware and software used for playback.

A great deal of the software I've tested is not capable of playing 24-bit files cleanly (regardless of its specifications).

A great deal of the hardware I've tested is not capable of playing 4x rates (i.e. 176.4k and 192k files) properly (regardless of its specifications).  This, I attribute to the significantly increased demands on clocking accuracy at the higher rates and similarly increased demands on the analog stages at wide bandwidth.  Indeed, many converters, even some pro units, actually sound worse at the 4x rates than they do at 2x rates (i.e. 88.2k and 96k).

Lastly, many of the so-called "high res" downloads available today - from some big name "audiophile" sources - are in fact upsampled Redbook and not genuine high res.

All that said, with good software (even the free iTunes can do it), good hardware and real high res, how does it compare?  Does it live up to the "hype"? 

For my ears, it isn't "hype" (i.e. exaggerated claims).  It is real.  For example, with my Metric Halo ULN-8 recording at 24/192, for the very first time in my experience, I'm hearing the sound of my mic feeds.  It no longer sounds like "great digital".  It no longer sounds like "great analog".  It sounds like what is coming from my mics.  I've never heard that before from any analog device and certainly not from any digital device in my experience, regardless of price or design.

Why?  In my view, when the 4x rates are properly done, a threshold is crossed and the promise of digital is at last fulfilled -- to an extent I would not have believed had I not experienced it myself, over and over again.

The Web page I linked to above contains some samples.  Within each of the two groups, they are all from the same mastering session so the ONLY difference is the final format.  (The samples from "Lift" only go to 24/96 because when I recorded "Lift" that was the highest the gear could go.  I approached 192 very cautiously and did not use it for my work until I had put it to several tests.  Now, it is the rate at which I do all my work.)

Proud as I am of the Redbook version, what I hear on my system is an "opening up" when one goes to the 24/96 version.  But I find the jump from 24/96 to 24/192 a larger one than from Redbook to 24/96.  Listening in the reverse direction, starting with the 24/192, I hear what my mics captured.  Going to the 24/96 starts to sound to me like a very good recording, as opposed to hearing past the recording to the event itself.  Going from the 24/96 to the Redbook is yet another step down, with even less apparent fine detail and overall focus --- a cloud seems to have descended on the soundstage and it is much harder to hear the space around the players.

All just my perspective, of course.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

[bdiament]

Hi Geardaddy,

...Still, most music, at least in recent digital past, was recorded in 24/96, correct?....

Actually, no.  Most of the recordings I've encountered are recorded originally at 24/44.1k.  This is because many computer systems simply cannot handle dozens of tracks (on the multitrack) at higher res.  Especially once the engineer decides to add "plug-ins" for EQ and other processing, most systems will simply choke at higher rates.

Of course, among the audiophile, minimalist type recordings, the originals - generally done direct to stereo, even if they use many mics - are higher res.

For Soundkeeper, I'm recording everything at 24/192.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

[Geardaddy]

Thanks for the data Barry.  That's great.

Your opinions on analog decks versus newer hi rez?  Is this simply romanticism (and placebo effect)?

[bdiament]

Hi Geardaddy,

Thanks for the data Barry.  That's great.

Your opinions on analog decks versus newer hi rez?  Is this simply romanticism (and placebo effect)?

To my ears, it has nothing to do with "placebo", it is all about what the event itself sounds like.  At one time, I'd have selected a good analog deck over any digital device.  Today, I no longer think it is close. 

All the reservations I've held for years, where I felt the best analog does a number of things better than the best digital, have evaporated with my experience with the ULN-8.

For me, no analog deck in my experience can capture as much information as a ULN-8 at 24/192.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

[Mike Nomad]

Can someone who has listened to master tapes on a good system say if any digital can compare to great analogue?

If it doesn't then we still have room for improvement.

Yes, digital recordings can sound as good, or better, than analogue. Also, digitized analogue recordings can sound great. Easy examples to illustrate the point(s) are the Cowboy Junkies' Trinity Sessions, and Mercury Rev's Deserter's Songs (recorded on 35mm Mag).

The idea of holding Redbook recordings as the digital standard, and comparing them to a contemporary vinyl release is for chumps. Over a hundred years of evolution has gone into the making of recordings for release on vinyl, while we have been getting digital releases for about 25 years. To make a meaningful comparison, compare Redbook to 78RPM shellac recordings, and see which one you prefer...

The digital recording (and release) process continues to evolve. A properly done SACD or 24/192 recording played back properly will leave any current consumer analogue format eating sonic dust.

[woodsyi]

I have Rebecca Pidgeon's Retrospective on LP, CD and 24/96 FLAC.  I think Chesky records everything in hi rez digital.  With my equipment, the order of excellence is LP first, 24/96 second and CD last.