What's the deal with CD players used as transports?

[nicholtl]

From my understanding, CD players/multichangers can be used 1 of 2 ways.

1) As a transport.

2) As a player utilizing it's own internal DAC's.

As such, this basically means that by using digital interconnects (optical or coaxial) between the CD player and the receiver/preamp, the CD player is thus qualified as a "transport." If, on the other hand, the CD player is connected via analog interconnects (RCA or XLR), then you will be utlizing the CD's own internal DAC's.

This being true, and I don't know if it is, but wouldn't that mean no matter what level a CD player is ($60-$25,000), connecting it with a digital cable would make it all sound the same, since all signal processing would be taken charge of by the receiver/preamp? Becaues if that's the case, why would ANYONE with a high-end CD player use it as a transport? Wouldn't they simply be better off using the CD player's own, internal circuitry to produce sound?

And along these same lines, wouldn't that mean any DVD player, SACD player, or DVD-A player connected with a digital cable produce the same results? Identical sound compared to another player connected to the same receiver/preamp??

I am really quite confused on this subject, and if someone can shed a little light on it I'd be much obliged.

[marksdad]

to me its very simple, use the dacs from the machine that sound the best to your ears

[edwinr]

Good question, nicholtl. I know several low volume, high end CD manufacturers use so called off the shelf CD and DVD ROM drives purchased from OEM suppliers. These drives are then placed in very expensive boxes. But in saying that, the high end manufacturers put a lot of effort into the peripherals like power supplies, internal DAC's etc.

I would imagine that the outsourced drives are modified somewhat from the original to improve isolation and tracking and reduce jitter. I recall some high end manufacturers saying they went with CD ROM drives because of their inherent reliability. I've got a sneaking suspicion though, that the reason CD ROM drives are used is because they are cheap to buy in bulk.

Whether the end result from the high end market is worth $1000's more than the less expensive options from the likes of quality mass-market manufacturers like Sony and Philips is another matter. Sony and Philips probably use DAC's in their upmarket players as good as, or better than, the DAC's in most receivers. In relation to the down-market CD/DVD players from other manufacturers, I would agree that there may be considerable benefits in using receiver based DAC's. I don't know if I would go to the trouble, however, of a component DAC upgrade and using a cheaper player as a transport only. It would be more cost effective to buy the better quality Sony or Philips player in the first place.

[nicholtl]

Interesting, Edwinr. Here is a post made from a member named Randy Bey back on March 17th, 2001 on these forums.

--------------------------------------------

Guys, - especially those of you with the 300+1 disc changers -

I have to tell you about this. I became convinced of the potential for audible changes via a transport change while at a friends house. He switched between a high end Rotel CD player as transport and a mid-range DVD player as transport for a little A/B comparison.

There was a difference that could be noted in the first 20 seconds!

I squirreled this info away in my cheek for quite a while. I have been using a NAD C520 CD player as transport to my MSB Link III DAC.

Finally I got a chance to get a dedicated transport, a used Theta Data Basic, which is pretty high end. I put it in my system last night.

Oh my Lord.

It replaced the NAD and a Monarchy DIP. Without the DIP the NAD had sibilence problems, female vocals had a touch of graininess, a white noise component to the higher ranges. With the Monarchy, which I was very happy with BTW, the sibilence smoothed out.

The Theta crisped the sibilence, took out all traces of white noise, and did not give any impression of "smoothing". There is definition now in these upper registers that wasn't there before.

The other immediately observable difference is the sound of percussive instruments. They have a complex waveform, asymmetric not symmetric like non-percussive instruments, and I did not notice the "glossing-over" effect that was there on the NAD. The Theta brings out details on those percussions that I didn't know were there.

Oh, and the imaging has improved. This is something to say since imaging was a strong suite with my rig all along. Didn't think it could get better, but surely there is more spacial focus now than before.

For you guys with the mondo changers, I'm not saying give them up; they are still quite good for most of the time, non-intensive listening.

When you want to really immerse yourself, perhaps you could pop the CD out of the changer and put it in a dedicated transport.

It was money well spent. This is not something I say casually.

--------------------------------------------

It was primarily posts like these, as well as others similar on other forums, that made me question the term "transport." What exactly constitutes a "transport?" And if it theoretically means to simply "transport" the 0's and 1's of a CD to the processor, how on earth could it matter what the means is, if the ends are the same regardless? From what I gather, I would just assume any CD player costing $300 or less, and ANY multichanger, should be automatically used as a transport as they probably do not utilize ultra high-quality DAC's?

And yet, when playing the exact same CD on my 400 Sony ES megachanger and then on my Denon 3800 DVD player, both connected with a high quality digital connection, they sound VERY DIFFERENT!

[unimorpheus]

I am somewhat confused on this transport thing as well. While I understand some users are experiencing audible differences in transports, my question is why. Unless the lower cost transports are generating read errors that are being mathematically corrected in error, I cannot see why the transports would generate differing data streams. The artifacts and errors the users are hearing can only be the result of transport read errors since all other decompression/conversion is taking place at the receiver or outboard processor. Any explanation of this phenomenon would be greatly appreciated.

[dougdrake]

I'll speculate (it's what I'm best at). This is real time audio, which is great affected by jitter and latency and drop-outs. There's no time for error recovery, and because it's real time it doesn't tolerate jitter very well (fluctuations in the latency rate). It may be these factors that are minimized with a higher-end transport that may have better lasers/lens/signal paths, etc...

[WMcD]

I'm in the same camp as Nicholti as far as suspicions.

There may well be an issue of DAC and how they work and to what benefit. DAC conversion is a long standing issue. That is something else.

However, the audiophile camp seems to argue that there are shortcomings in inexpensive computer type technology which impacts on the transfer or copying of digital data.

At the hardware level, you open up your computer CD drive and take a look; then you open up your audio CD drive and take a look. They look about the same. They both came out some factory in the Orient. It seems ludicrous that one or the other is optimized for any form of digital. I.e. audio versus computer data in general.

I wonder about data integrity and claims there is an issue. For example: You duplicate a data CD with 100s of megabites of program code. It runs without error. Yet people contend that an audio CD dupicated on the same system is lacking something. That can't be. Our programs would crash because of errors in transcription. They don't.

There seems to be a fall back argument by audiophiles that crummy old (but let me say current) off the shelf CD-ROM drives suffer some timing errors. This doesn't make sense either.

I say that because we typically see that we can duplicate a music CD at 48 times the normal rate without error. Roughly, we pop an audio CD in the drive and then write off to a hard drive at 48 times the normal rate of what we'd listen to, and then copy back to burn a CD-R in a minute or less.

You can see my argument. The currrent off the shelf CD-ROM drive can transfer data off at 48 times the normal play back rate without error. Yet people contend that when it is running at 1/48 of that high speed (normal listening of a CD) it is flawed in performance. Gosh, hard to believe.

Best.

Gil

[meuge]

Actually, CDROMs don't transfer data without error, it's just that the Input/Output systems on a computer have really advanced error correction technology that also interacts with the error correction code written on the CD.

The best way to transport audio without errors to a good DAC is to rip it off a CD at low speed (8X or less) using a good drive, and verify the bit-for-bit precision of the conversion (some software will do it automatically either along the way, or at the end). Once you've ripped the music using an uncompressed format such as FLAC or WAV you can stream the music from the hard drive to an external DAC using digital out on your sound card. That seems to be a much more precise way than a sequential read off a CD, and that's the way I do it. The only problem I can see with this approach is the fact that the computer is a huge source of EMI and the sound card is not shielded. A simple way of eliminating that problem is to have an external, shielded sound card, which is what professionals, who actually master the music we listen to, do.

[cwm]

----------------

On 9/26/2004 11:10:58 PM William F. Gil McDermott wrote:

I'm in the same camp as Nicholti as far as suspicions.

There may well be an issue of DAC and how they work and to what benefit. DAC conversion is a long standing issue. That is something else.

However, the audiophile camp seems to argue that there are shortcomings in inexpensive computer type technology which impacts on the transfer or copying of digital data.

At the hardware level, you open up your computer CD drive and take a look; then you open up your audio CD drive and take a look. They look about the same. They both came out some factory in the Orient. It seems ludicrous that one or the other is optimized for any form of digital. I.e. audio versus computer data in general.

I wonder about data integrity and claims there is an issue. For example: You duplicate a data CD with 100s of megabites of program code. It runs without error. Yet people contend that an audio CD dupicated on the same system is lacking something. That can't be. Our programs would crash because of errors in transcription. They don't.

There seems to be a fall back argument by audiophiles that crummy old (but let me say current) off the shelf CD-ROM drives suffer some timing errors. This doesn't make sense either.

I say that because we typically see that we can duplicate a music CD at 48 times the normal rate without error. Roughly, we pop an audio CD in the drive and then write off to a hard drive at 48 times the normal rate of what we'd listen to, and then copy back to burn a CD-R in a minute or less.

You can see my argument. The currrent off the shelf CD-ROM drive can transfer data off at 48 times the normal play back rate without error. Yet people contend that when it is running at 1/48 of that high speed (normal listening of a CD) it is flawed in performance. Gosh, hard to believe.

Best.

Gil

----------------

[/blockquote

Now there is what I would call impeccable logic and reasoning. But I can see where it would be "fighting words" to some........!

[edwinr]

Just another thought on this topic. I would imagine that independant DAC manufacturers would like you (and me) to believe that their DAC's are purely a technical upgrade to those DAC's inside an off the shelf CD player. By technical upgrade I mean better design, components and perceived build quality, which in turn leads to improved frequency response, reduced errors, etc, etc. I am not going to argue that these DAC's aren't a technical upgrade. But what I will argue is that I, with my 'golden' cloth ears probably wouldn't HEAR any audible differences if the upgrade follows a purely technical course. So why would I part with my hard earned cash?

Which leads me on to my next point.

I have read some reviews in the past where upgrade DAC's were tested and the results showed a 'tailored' frequency response. Some DAC's had a small hump in the upper bass, and others had a slowly falling treble response. Other DAC's had a slightly recessed midrange. These variations from the expected super flat frequency response somewhat surprised me. But on reflection, I can see that the DAC manufacturers were trying to put their sonic signature on their product.

Why this sonic signature? Well, this signature may well have a two fold effect. Firstly, some may argue that this is an opportunity for the DAC manufacturer to demonstrate their expertise in rectifying perceived audible difficiencies in digital sound. Secondly, some more cynical people would argue that this sonic signature is aimed at people like me. I would be able to hear an audible difference. I would then part with my hard earned cash and buy the product.

Does this hurt anybody? No, not really. After all, we are all in the market for products with their own sonic signature. I guess that's what makes us buy them. As long as the technical aspects of the product are okay...... now, where's my credit card?

[dougdrake]

Gil - I don't take issue with what you said. It's true for file copies, which is essentially what ripping a CD or duplicating a CD is - copying a file from one place to another. During that process, bits can be error-checked and if found in error, can be corrected by re-reading that disk segment until it get its right.

In real-time audio (and video), which is what you encounter when listening to a CD or watching a DVD, there isn't time to go back and re-read a disk - you have to take what comes and make the best of it (and sometimes the DAC has to interpolate based on the preceeding and following bits).

Again, I don't know if this affects the sound quality to the extent it may be partially responsible for the improved sound some listeners claim from "high end" transports. I mention it only as a possibility.

[meuge]

Here's another thought. I am not sure if high end players do this, but a computer will never read a CD or DVD at 1X speed. What usually happens is that the disk will be read at higher speed, and some amount of the content will be buffered ahead (configurable on some software) into memory. While doing this, the integrity of the buffered data will be checked and errors re-read and corrected. If regular players simply read a continuous stream, rather than buffer and error-check, I could see how you could benefit from a more advanced player.

[formica]

I too have an issue with high end Transports. Having read endless discussions about jitter and error correction... it seems to me the solution of producing a perfect transport I quite simple.

If there is a problem due to the "real time" nature of digital audio repro, why not eliminate it by adding a simple memory buffer with an error correction algorithm. It doesn't have to be large, just a couple of MB given how quick the data can be read and reread from disk. This would add perhaps a second or two delay to the access time which already varies from machine to machine.

I figure that it isn't as much of an issue as the discussions make it out to be, since my idea is far from revolutionary. Most inexpensive portable players have this integrated into them, which would make them better transports if this were true.

Perhaps the differences come from our misunderstanding what tasks are being performed by the transport and the DAC.

----------------

On 9/26/2004 1:09:20 PM nicholtl wrote:

And yet, when playing the exact same CD on my 400 Sony ES megachanger and then on my Denon 3800 DVD player, both connected with a high quality digital connection, they sound VERY DIFFERENT!

----------------

I think this is still an interesting observation... which may point to that division of tasks?

I'm assuming that they are both connected as two channel units in order to use the same DAC conversion in your processor, and both disks are of the same type/reflectivity? (both originals or both CDRs?)

Rob

EDIT: meuge, we were thinking along the same lines.

[D-MAN]

CD data format and CD audio format are 2 different things.

Whereas you can read/write on a computer both ways, the medium is different vs. a commercial "glass-pressed" copy.

Thats the difference between "burning" and "pressing".

Here's the proof in the pudding:

1) take a familiar commercial audio cd and burn a copy of it anyway you can.

2) a/b it on your high-end audio system - maybe a slight difference, but pretty darn close...

3) burn a copy of the copy.

4) a/b that copy (gen 2) against the original commercial one- more difference?

Repeat until your convinced...it takes me only one generation. The first thing to go are the micro-transients.

There is a high degree of loss associated with the "burnable" media. It is a matter of physics. The idea that you get a bit-to-bit copy is erroneous but "close-enough" where data (computer) does not have amplitude information involved in it. The computer data is much simpler in complexity than an audio signal.

DM

[nicholtl]

Edwin, that's definitely news to me. Although I don't doubt for a second that some companies put their own sonic signature into their products. I've never had the pleasure of auditioning anything from Meridian, although I know they are famous/infamous for tweaking circuitry like crazy. I'll bet they tweak their gear to produce a sound that they believe to be more pleasing.

How strange though, because from what I gathered from articles and forum speak, it was always the units that had a "lack of sound," so to speak, that were the true audiophile must-haves.

Rob, yes, they were both connected as 2-channel players, and I even tried switching between different DSP modes on my preamp. Each time, the sound replication was quite unique to each player. The only non-controlled element in my controlled experiment, however, was that while both the CD megachanger and the DVD player were hooked up using digital cables, one was a toslink, the other a coaxial.

I guess a good test will be when my new dedicated Classe CD player arrives, at which time I will plug in both balanced XLR analog cables, AND a digital cable, to my preamps inputs, and then see just how different the sound is from analog to digital, and how it's digital sound compares to my other 2 players'.

D-man, you must have extremely, inhumanly acute ears. Maybe a dog's ears! It took me 2 generations of CD burns before I can notice a loss of dynamics.

[D-MAN]

----------------

On 9/27/2004 6:42:44 PM nicholtl wrote:

D-man, you must have extremely, inhumanly acute ears. Maybe a dog's ears! It took me 2 generations of CD burns before I can notice a loss of dynamics.

----------------

I don't think so, but I can tell. The nuance defining the soundstage goes first (the small transients). It's subtle, but a degration for sure.

By-the-way, I have a Meridian cd player - Ironic!

DM

[Mario NJ]

----------------

On 9/27/2004 11:33:19 AM formica wrote:

I too have an issue with high end Transports. Having read endless discussions about jitter and error correction... it seems to me the solution of producing a perfect transport I quite simple.

If there is a problem due to the "real time" nature of digital audio repro, why not eliminate it by adding a simple memory buffer with an error correction algorithm. It doesn't have to be large, just a couple of MB given how quick the data can be read and reread from disk. This would add perhaps a second or two delay to the access time which already varies from machine to machine.

Unfortunately, jitter cannot be eliminated simply by using a memory buffer as you describe. Jitter is not read error, it is frequency distortion caused by random, or worse, systemic variations in the clock frequency of the digital chips used to move data around a binary system. So using a memory buffer would only pass along the source of jitter to the clock of the chip that is reading data out of the memory buffer and onto the digital output cable.

Of course you could put a memory buffer in your receiver/processor to buffer this data as it comes in, and then the source of jitter in the system would be the jitter in the chip that reads this data and passes it on to the DAC inside your processor.

Lucky for me, it appears my ears aren't good enough to detect jitter. When I upgraded my old CD player to my new universal player (both used only as transports) I immediately thought, "Wow, what a great upgrade in treble clarity and soundstaging." But then I had a friend come over and I convinced him to help me do a blind (single blind) test between the two players and I honestly couldn't tell the difference between them.

Best,

Mario

[thebes]

Great topic Nicholti and one I've wondered about and scratched my head over as well. The replies have also been very thoughtful and detailed.

Right now I'm using an 12 year old Yamnaha high end-dual power supply ($1000 new) cd player (18 bit)in analog to my tube amp and have often wondered if I should use it as a transport to my Denon 2803 because it is only a year old and probably has a better DAC. Or if I should just use my Pioneer 563a DVD universal player direct for 2 channel to the Denon (SACD and DVD-A are via analog). Or, of course use the Pioneer's DAC to run the sound.

This stuff gets pretty confusing.

[nicholtl]

Thebes, it sure does! Well my Classe CDP-10 player arrived today, but it's just sitting pretty in the brown box right now, as I have to wait for the rest of my "upgrade" gear (PS Audio P1000 Power Plant, 15amp Ultimate Outlet, 3 PS Audio power cables, a 250ft spool of Z1-series monster cable, 2 15ft lengths of Z2-series monster cable, a center channel 10ft length of Z2-series monster cable, all the appropriate banana plug terminations, and a pair of PS Audio XLR cables).

SO! Once all that crap arrives at the doorstep, I can finally take apart the whole system, and find out once and for all how different all these players sound as transports, and then how they sound utilizing their own internal DAC's. So I will find out the how's. But the why's are still yet to be determined...

[thebes]

All right Nicholti-new toys to play with! Can't wait to hear a review/report.

Oh PS I'll be happy to take that tired out old gear off your hands. Can't have it cluttering up the place now, can we?