Computer Based Audio vs. Physical Media

[LenWhite]

I do hope CA succeeds in making most music available in a high resolution format; consensus the format is matured with wide agreement on the best implementation method; good backup and recovery procedures built in; album art and literature equivalent or exceeding LP's; and significant computer knowledge to implement correctly is not a requirement.

While it's true SACD selection is primarily oriented towards classical music, I'm not so sure it's a "dying" format. Even though the audio media is furiously promoting CA as the new frontier, so far CA itself has a rather "thin" RBCD or higher resolution catalog. I don't believe it's currently even equivalent to SACD except for pop/rock music.

My passion for music is the main purpose for having a high quality stereo system. But I want the playback to be as close to an intimate live acoustic performance as possible, and I believe I've largely achieved that except for the scale of large classical orchestral music. Currently I and other audiophiles are not convinced this level of playback has been achieved when comparing CA to physical media.

I do agree CA has a storage and portability advantage over physical media that is appealing, together with a "potential" to make much more high resolution music available. But currently that comes at a cost of royalty issues (this may well be the "elephant in the room"), limited availability, questionable sonic superiority compared to the best implemented CD/SACD, complexity, and a loss of most album art and literature.

It took almost 20 years for RBCD to sonically mature, and many audiophiles simply stayed with LP's - some still consider this 60 year old technology the best music format. If it wasn't for the "hobby" nature of LP's I may have done the same, but I'd rather listen to music than play with equipment.

I do truly hope CA becomes the de facto medium allowing audiophiles access to the equivalent of master files for most music produced. But until that and my other pre-conditions occur, I'll continue enjoying my physical media.

I do understand the allure of CA and applaud those "pioneers" of this new experience. I just hope it doesn't take CA as long as CD's to mature, or stay a primarily low resolution format.

 

[Steve Williams]

Len

I believe you will be proven correct on all counts and much sooner than later.

 

[jriver]

Audiophiles will sometimes dispute this, but the bytes read off a CD are exactly the same as the bytes that result from decoding a lossless track from the same original audio CD. No difference if you're giving those bytes to the same DAC, etc. Zero difference in sound quality.

Availability of digital files for commerce is a different subject, but we're also pretty far along there. 320Kbps MP3 and 256Kbps AAC files are widely available for more than 10,000,000 tracks. They aren't theoretically as good as a lossless file or a WAV file, but I doubt that they can be distinguished by most people. I'm not arguing for MP3. I'm just saying that it can be essentially the same experience.

Higher resolution files are becoming more widely available, but the primary obstacle to having them widely available is probably still the commercial terms available from the record labels and music publishers.

More music is now sold digitally than on physical media. That just happened in the last year. So we're past the tipping point.

 

[microstrip]

(...) More music is now sold digitally than on physical media. That just happened in the last year. So we're past the tipping point.

Do you know what are the quota percentages of each digital media?

 

[Thomas.Dennehy]

Do you know what are the quota percentages of each digital media?

Some representative data for 2010 music data was reported by NPR.

An excerpt:

Another name for [people who buy physical media] might be older purchasers, but it's hard to know and we wouldn't want to make anyone upset, since they're basically single-handedly keeping the music industry afloat. Whatever the demographic of her audience, [Susan] Boyle sold 1.852 million copies of her Christmas-themed album The Gift in 2010. Of those sales, 1.820 M were [physical media, namely CD]. That leaves just 32,000 copies sold digitally, a tiny 1.7% of the album's total sales. Compare that with 488,000 digital copies (16.5% of the total sales) of Taylor Swift's Speak Now or 852,000 (25%) for Eminem's Recovery.

 

[bdiament]

My experience here may be somewhat different.

Ever since early 1983, when I created my first CD master and heard the results that came back from different replication facilities, I've been saying that CDs from different plants all sound different from each other and *none* sounds indistinguishable from the master used to create it.

Zoom forward 29 years to today and I still find exactly the same thing. The only change is that the very best replicators can create results that sound *very* close, requiring me to set up a direct, synchronized A/B in order to discern the differences. Most other plants still create discs my clients, many colleagues, associates and myself find do *not* sound like the master I send them. The differences can range from a loss of focus and fine detail to "what did they do?!" This has been the case, without a single exception, for almost three decades, regardless of the CD player used or the transport and DAC used.

As an aside, not too long ago, I compared a bunch of SHM and HQCD pressings with their plain vanilla CD counterparts. The differences were immediate and obvious - to a degree that had me quite sure I was listening to two different masterings with different EQ. Extraction to hard disk and some null testing revealed the mastering was identical in both instances, i.e., both very different *sounding* discs were created from the same master.

What I discovered a while back is that taking these CDs that sound different from each other and different from the master and extracting them to the computer eradicated all the sonic differences between discs and between the discs and the master. Once played from the computer, all the extractions sounded indistinguishable from the master used to create them.

What I take from this is this, aside from using the computer for most of my own pleasure listening, is that something in playback from disc is changing the audio. Clearly the bits are the same. What's different? The first thing that comes to mind is that playback from disc involves tracking the spiral pattern made by the pits in order to read the disc, operation of a tracking servo to keep the read laser in the correct position and properly focused, redundant reads when necessary, decoding the 8:14 modulation back into binary code (there are no "ones and zeros" on a CD but instead, 9 different length "pits") and performing error correction when necessary. All of these must be performed in "real time" as the disc plays. With playback from a computer, all of these functions have already been performed (when the file has been extracted from a disc) or are not necessary (when the music has been delivered as a computer file).

Whether this accounts for the differences, I can't say for sure. I do know that extracted files can be played against the CD master and no differences can be discerned. They can also be nulled against the CD master and the result is dead silence, all the way down, to the sample.

None of this is to suggest that everyone will find computer playback "better". It is merely to say that if one wants the sound of the master used to create the disc, I know of no better way to get it, regardless of the budget.

Lastly, to Len's desire for "playback to be as close to an intimate live acoustic performance as possible", I think this is a function of the recording itself. The best our systems can do is provide what the recording has to offer. In my experience, there can be no suggestion of reality that is not contained in the source itself.

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

P.S. To be clear, when I refer to playback from the computer, I'm referring to raw PCM files, i.e. .aif or .wav. To my ears, no other format has yet achieved the same thing. (The so-called "lossless" formats, to my ears, are more like playback from a not-so-great CD player.)

 

[microstrip]

My experience here may be somewhat different. (...)

Bdiament,

Thanks for posting your opinions - I have read something similar things from other professionals who referred the importance of the last phase of mastering - preparing the digital master to be sent to the data stamper.

Before I ask another naive question, can I ask exactly what you are calling "the CD master"?

 

[bdiament]

Hi microstrip,

Bdiament,

Thanks for posting your opinions - I have read something similar things from other professionals who referred the importance of the last phase of mastering - preparing the digital master to be sent to the data stamper.

Before I ask another naive question, can I ask exactly what you are calling "the CD master"?

The CD master is what the replication facility receives and what they use to create the CD "pressings".
In the old days, it took the form of a 3/4" U-Matic tape (where all the video "real estate" was used to hold the digital audio along side SPMTE time code).
In the late 80's to early '90s, plants started accepting masters on audio CD-Rs.
Not long afterward, came the DDP (or Disc Description Protocol) file set, usually sent to the plant burned onto a DVD-R. I find these, which are data files and not formatted audio files, to result in the most faithful CDs. Part of the reason is that any errors contained in an audio CD-R will be copied to the replicated discs that come back from the plant.

The CD master contains the CD program after final mastering and any changes (level adjustments, EQ, etc.) that have occurred in the mastering process. The individual songs are sequenced and the spacing between songs already adjusted to where the producer/engineer/artist want it to be. There is also a "table of contents" the plant uses and which is at the front of every CD (the first thing your player reads), so the playback device "knows" for example, where to find song #3.

In other words, the CD master is the production-ready source the plant uses to create the final CDs. The first step at the plant is burning the glass master, which is the first step in the injection molding process -- i.e, the glass is electroplated to create a "negative", similar in some ways to how vinyl is made.

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

 

[Phelonious Ponk]

Barry has more sophisticated ears than I, and he has master tapes, which puts him at a distinct advantage . But what I have noticed that is significant is that in the transfer from CD to my MacBook, CDs which have skipped or otherwise erred for years are restored. These are just the obvious errors. I assume more subtle ones, that didn't literally make Tom Petty sound like he was being scratched by an 80s DJ, are being corrected as well. In the vast world of possibilities, I imagine it's pretty possible that has an impact on playback quality. I'd even put it above things like a few extra picoseconds of jitter. I've also noticed that the system sounds very good playing from the computer, but I changed DACs and speakers, and headphone electronics (DAC/amp), at about the same time I converted to a server, so who knows what I'm hearing? But the best thing? Bar none? I find myself listening to music I haven't listened to in years because I can search for it and have it right there, right now, ready to push play. And I can search lots of ways - by album, artist, genre, song - and save things in playlists, and let the software create playlists for me. The convenince is impossible to argue with. And I've had no trouble finding 16/44.1 material, because I ripped my entire CD library and since, the used CD store has become my friend. Computer audio has been wonderful for me.

Tim

 

[bdiament]

Hi Tim,

Agreed 100%. One of the things I find most wonderful about having set up a music server (aside from the audio) is the unprecedented access to my collection and the fact that I'm hearing musical gems I haven't heard in a long time.

I'll often let the server app run in "shuffle" mode, letting it select the next track to play. Sometimes the segues can be quite amusing - going from Hendrix to Erik Satie to Son Volt to a Bach cantata. Other times, shuffle mode finds a track that makes me want to hear the whole album. Just happened last week when we were relaxing to the music and the first track from "A Love Supreme" came on. We ended up listening to the whole album, enraptured by the sheer energy those guys created.

I also like the idea of getting an urge to hear a particular piece and having the ability to hear it in less time than it would take me to walk to the CD shelf (much less find the disc, open the jewel box, place it in the player and press the button). And other times, hearing a great piece makes you want to hear more by the same artist or same composer. With the server, as you know, a few keystrokes are all it takes.

As soon as I started building and enjoying the server (took me over 11 months to get all of our music into it, properly tagged, etc.) I felt "This is the future." And it is here now. ;-}

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

 

[microstrip]

Hi microstrip,



The CD master is what the replication facility receives and what they use to create the CD "pressings".
(...)

Re-reading your posts it seems to me that you just saying that the CD players are all sounding different from your computer system and that its is not due to bit content - just that and nothing else! Am I wrong?

 

[bdiament]

Hi microstrip,

Re-reading your posts it seems to me that you just saying that the CD players are all sounding different from your computer system and that its is not due to bit content - just that and nothing else! Am I wrong?

Yes, you misunderstood me - or I didn't state it clearly enough.
There are several aspects to this. First, sending the same CD master to different plants results in discs that sound different from each other when played on the same CD player or when played on the same CD transport feeding the same DAC. It doesn't matter what CD player, transport or DAC.
The CD can even be played in the computer drive and fed to the same DAC (in my case, via Firewire) as the files played from the computer. From the disc, it never sounds indistinguishable from the CD master. From the hard drive, it always does.

Also note, there are no ones and zeros on a CD. They must be *decoded* from the 8:14 modulation that is used to create the different length pits read by the laser of the transport or player.

In addition, if I slow burn a CD-R from the CD master, then play it in the same CD player I use to play the pressing (whether using the internal D-A or taking a digital output and feeding the studio D-A), they still don't sound like each other. The slow burned CD-R, in my experience sounds closer to the master, every time. (This is why Soundkeeper offers CD-R versions of our releases. For folks using a transport or player, I find the results sound more like the master than even the best CD pressings.)

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

 

[Phelonious Ponk]

Hi Tim,

Agreed 100%. One of the things I find most wonderful about having set up a music server (aside from the audio) is the unprecedented access to my collection and the fact that I'm hearing musical gems I haven't heard in a long time.

I'll often let the server app run in "shuffle" mode, letting it select the next track to play. Sometimes the segues can be quite amusing - going from Hendrix to Erik Satie to Son Volt to a Bach cantata. Other times, shuffle mode finds a track that makes me want to hear the whole album. Just happened last week when we were relaxing to the music and the first track from "A Love Supreme" came on. We ended up listening to the whole album, enraptured by the sheer energy those guys created.

I also like the idea of getting an urge to hear a particular piece and having the ability to hear it in less time than it would take me to walk to the CD shelf (much less find the disc, open the jewel box, place it in the player and press the button). And other times, hearing a great piece makes you want to hear more by the same artist or same composer. With the server, as you know, a few keystrokes are all it takes.

As soon as I started building and enjoying the server (took me over 11 months to get all of our music into it, properly tagged, etc.) I felt "This is the future." And it is here now. ;-}

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

Yeah, it's the present for a lot of us. I think the future is streaming music with really good VBR codecs. Actually that's pretty close to the present too. The best way to discover new music I've found is to listen to genre lists I create on Pandora, then when I hit something I like, go listen to whole albums on Spotify. If I still like, I buy. It's a little lo-fi but it sure is fun.

Tim

 

[amirm]

One of the reasons a CD can sound different than a computer file is the clock. CD bit stream has the clock embedded in it that is ultimately used to drive the DAC. Mechanical variations can therefore lead to clock jitter and with it, increased distortion.

 

[bdiament]

Hi Amir,

Some DACs (such as the ULN-8 I use) allow the user to select the clock source. In my case, I use the internal clock. Even with audio from the CD player, it provides better results.

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

 

[microstrip]

One of the reasons a CD can sound different than a computer file is the clock. CD bit stream has the clock embedded in it that is ultimately used to drive the DAC. Mechanical variations can therefore lead to clock jitter and with it, increased distortion.

Amir,
Can I ask what are the other reasons?
I find curious that we must accept that all manufacturers were not able to suppress the interference due to the mechanical variations, even using FIFOs and similar storing systems, even synchronized to external clocks.

 

[microstrip]

Hi microstrip,


Yes, you misunderstood me - or I didn't state it clearly enough.
There are several aspects to this. First, sending the same CD master to different plants results in discs that sound different from each other when played on the same CD player or when played on the same CD transport feeding the same DAC. It doesn't matter what CD player, transport or DAC.
The CD can even be played in the computer drive and fed to the same DAC (in my case, via Firewire) as the files played from the computer. From the disc, it never sounds indistinguishable from the CD master. From the hard drive, it always does.

Also note, there are no ones and zeros on a CD. They must be *decoded* from the 8:14 modulation that is used to create the different length pits read by the laser of the transport or player. (...)

Thanks, but I need to ask for some more details. Isn't the CD master stored in an hard drive? When you compare it with what you call "playing from the hard drive" in both cases you play from the same type of physical media.
And it seems that we can conclude that, considering current technology limitations in CD players, EFM is not perfect, as it is audible. And may be the Firewire interface can also be blamed.

 

[bdiament]

Hi microstrip,

I would not draw those conclusions myself.
In my opinion, there is no evidence to suggest EFM is the fault and in my experience, Firewire is quite easily the best digital interface for computer audio.

Besides, the same Firewire interface is at play when the disc is played from the computer's disc drive as when the file is played from the hard drive.
So I believe it is quite safe to say it has nothing whatsoever to do with Firewire. In my experience, it has to do with both the manufacture of the disc and the playback of the disc - it all comes down to various aspects of the disc itself.

And yes, the CD master (as I create them) is stored on the hard drive. However, exactly the same thing was true all the way back to the early '80s when the CD master existed on a U-matic cartridge.

Most replication facilities will tell you their CDs sound indistinguishable from the masters. I can only conclude that they don't actually listen very carefully or they just don't hear it. Every experienced mastering engineer I know hears it. So far, only one plant of the dozens I've spoken with has said to me "The pressing will never sound indistinguishable from the master" (this with no prompting or hints from me). Perhaps coincidentally, though I think not, this plant happens to make CDs that are closer to the master than any in my experience. They do certain things differently than most other plants, in order to get better quality. Their work tells me they are in a small minority who are actually listening.

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

 

[amirm]

Amir,
Can I ask what are the other reasons?
I find curious that we must accept that all manufacturers were not able to suppress the interference due to the mechanical variations, even using FIFOs and similar storing systems, even synchronized to external clocks.

The other theory I have heard is what Barry mentioned. The disc is spinning without proper support of even the hole being in center. So the OPU (optical pick up) servo is constantly trying to focus the laser as the thing wobbles. That power causes fluctuations that could lead either through power supply or other means into the clock circuit and the DAC if it is in the player. This is one of the reasons behind independent power supplies. Buffering can't help this because the problem occurs when the data is pulled out.

 

[jap]

Hi microstrip,



Yes, you misunderstood me - or I didn't state it clearly enough.
There are several aspects to this. First, sending the same CD master to different plants results in discs that sound different from each other when played on the same CD player or when played on the same CD transport feeding the same DAC. It doesn't matter what CD player, transport or DAC.
The CD can even be played in the computer drive and fed to the same DAC (in my case, via Firewire) as the files played from the computer. From the disc, it never sounds indistinguishable from the CD master. From the hard drive, it always does.

Also note, there are no ones and zeros on a CD. They must be *decoded* from the 8:14 modulation that is used to create the different length pits read by the laser of the transport or player.

In addition, if I slow burn a CD-R from the CD master, then play it in the same CD player I use to play the pressing (whether using the internal D-A or taking a digital output and feeding the studio D-A), they still don't sound like each other. The slow burned CD-R, in my experience sounds closer to the master, every time. (This is why Soundkeeper offers CD-R versions of our releases. For folks using a transport or player, I find the results sound more like the master than even the best CD pressings.)

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

Since your slow burned CD-Rs still sound different from your master and you don't offer downloads from your website, why don't you distribute your music on USB flash drive sticks or SD cards.

Greenleaf Music has done this, see:
http://www.greenleafmusic.com/greenleaf-flash-drive