Are expensive digital transports useless?
May 17, 2013 at 6:00 PM Thread Starter Post #1 of 10

akart

100+ Head-Fier
Joined
Nov 22, 2010
Posts
139
Likes
13
First of all, by "transport" I mean a unit that just spins the discs and outputs a DIGITAL signal to an external DAC/AMP combination. The question I am wondering about is whether a $20,000 transport is even *in principle* (forget reality) better than the CD player on my IBM Thinkpad when it comes to putting out a digital signal. Considering that all computer CD players have to be bit-accurate when reading out CDs (otherwise you couldn't do anything with the software on the CD), I am looking for a convincing explanation.
 
I am not convinced by the often abused "jitter" phenomenon. Jitter only matters when an analog signal is being quantized. Even if there is substantial jitter/noise when reading out a CD, ECC ensures that the bits read out are 100% accurate (otherwise simple CD read tests would fail). Let the wars begin!
 
May 17, 2013 at 7:22 PM Post #2 of 10
Quote:
First of all, by "transport" I mean a unit that just spins the discs and outputs a DIGITAL signal to an external DAC/AMP combination. The question I am wondering about is whether a $20,000 transport is even *in principle* (forget reality) better than the CD player on my IBM Thinkpad when it comes to putting out a digital signal. Considering that all computer CD players have to be bit-accurate when reading out CDs (otherwise you couldn't do anything with the software on the CD), I am looking for a convincing explanation.
 
I am not convinced by the often abused "jitter" phenomenon. Jitter only matters when an analog signal is being quantized. Even if there is substantial jitter/noise when reading out a CD, ECC ensures that the bits read out are 100% accurate (otherwise simple CD read tests would fail). Let the wars begin!

 
While jitter also affects digital signals, an expensive "transport" is in principle no better than a well-build but inexpensive "player". If the CD spins, a laser reads the bits, and transmits them properly to the next stage, the "machine" has done all it can do.
 
ECC for an audio CD (CDDA) is limited. It's possible for uncorrected bit errors to occur. If they're few enough and far enough apart, the errors are inaudible. Data CD's, on the other hand, have substantial, additional ECC because they do require "bit perfect" accuracy.
 
May 17, 2013 at 7:39 PM Post #3 of 10
Quote:
 
While jitter also affects digital signals, an expensive "transport" is in principle no better than a well-build but inexpensive "player". If the CD spins, a laser reads the bits, and transmits them properly to the next stage, the "machine" has done all it can do.
 
ECC for an audio CD (CDDA) is limited. It's possible for uncorrected bit errors to occur. If they're few enough and far enough apart, the errors are inaudible. Data CD's, on the other hand, have substantial, additional ECC because they do require "bit perfect" accuracy.


Thanks. Your explanation does at least permit the possibility of a non bit-perfect stream being extracted out of a CD (Redbook CDDA) due to various (electro)mechanical reasons. As to whether those inaccuracies are audible or not, is, I guess a completely different question. I would imagine that if the uncorrected error occurs in a higher bit position, the error is more likely to be audible. Also, there is the question of what errors does CDDA ECC correct, and whether those are audible in the bitstream after applying CDDA ECC.
 
May 17, 2013 at 10:45 PM Post #4 of 10
If you rip with software like dBpoweramp which uses Secure Ripping, you will get a bit-accurate copy of the disc unless it is damaged, in which case an error will be reported.
 
But a computer can read and re-read a section of a track as many times as it needs to extract it accurately. Once ripped, playback is perfect every time.
A CD player reading the disc in realtime only has one chance to get it right.
 
CD players are a relic of the past in my opinion. That's not to say they can't sound good, but you can get verifiably bit-perfect data off your CDs with a computer, and then you never have to worry about it again.
 
Each time you use a CD player you may or may not have had bit-perfect playback, and have no way of knowing unless the error is audible. But I suppose if it's an inaudible error, it doesn't really matter anyway.
 
May 18, 2013 at 11:17 AM Post #5 of 10
What you'll find is that the better CD rippers will be reading the disk twice and checking the data against each other so there should be no skipping or jitter, anything like that. I was in my local audio shop yesterday trying the HDVD800 with a CD transport, I asked the shop staff if the CD player made any improvements over properly encoded tracks from your computer and they said no so I'm not really sure what you're paying for in a 20K digital transport.
 
May 18, 2013 at 12:15 PM Post #6 of 10
Quote:
Thanks. Your explanation does at least permit the possibility of a non bit-perfect stream being extracted out of a CD (Redbook CDDA) due to various (electro)mechanical reasons. As to whether those inaccuracies are audible or not, is, I guess a completely different question. I would imagine that if the uncorrected error occurs in a higher bit position, the error is more likely to be audible. Also, there is the question of what errors does CDDA ECC correct, and whether those are audible in the bitstream after applying CDDA ECC.

 
Seems to be both inner and outer Reed Solomon codes with 8 total bytes of redundancy for every 24 bytes of audio data, making use of interleaving as well as erasures correction capability of RS codes:
http://makbit.com/articles/cd-overview.pdf
http://en.wikipedia.org/wiki/Reed-Solomon_code#Data_storage
 
Raw error rate for a CD is quoted to be 10^-5 to 10^-6 before error correction, but I'm presuming that's before scratches, dirt, etc.:
http://books.google.com/books?id=Ed388jKoOfQC
 
And that's probably assuming a standard player and not some super hardcore expensive transport—even assuming that those are superior.  But yeah, a more expensive player might have a lower raw bit error rate prior to decoding?  If the CD is in decent enough condition to be decoded all correctly, then there's not much to improve on there.
 
As you point out, electromechanical issues, lasers, jitter, etc. on the CD reading part are not related to the jitter or any considerations of the D/A process (which is what determines the output signal quality).
 
May 18, 2013 at 12:20 PM Post #7 of 10
I consider that CD transports players will some day be obsolete but as long as I have 2000+ cd´s I want to have them spinning properly. I am the lucky owner of a Sony XA7ES considered to be to date one of the best built transports made. It is feeding a top end Dac and the sound it puts out is just awsome. Would I spend 20 Grand to get better sound from a new transport. NO WAY! Chances are it will sound different but no necessarily better.  I suggest to stick to a quility spinner and forget about it.  DACS are whole different story.
 
May 18, 2013 at 3:33 PM Post #8 of 10
Years ago, when megabuck transports were being sold, there were a number of posts on Audiogon declaring the importance of quality transports. Being in my impressionable years, I invested in a Mephisto X from Audiomeca (a French firm, now out of digital manufacturing). It still sounds wonderful out of a number of DACs I've tried with it including the Enkianthus, also from the above firm. However, I would never invest in another pricey DAC or Transport. Great fidelity is now available for very reasonable prices if one really listens and doesn't get fixated with the price=quality urban legend.
 
May 18, 2013 at 7:43 PM Post #9 of 10
Quote:
Years ago, when megabuck transports were being sold, there were a number of posts on Audiogon declaring the importance of quality transports. Being in my impressionable years, I invested in a Mephisto X from Audiomeca (a French firm, now out of digital manufacturing). It still sounds wonderful out of a number of DACs I've tried with it including the Enkianthus, also from the above firm. However, I would never invest in another pricey DAC or Transport. Great fidelity is now available for very reasonable prices if one really listens and doesn't get fixated with the price=quality urban legend.

 
Q: How do manufacturers justify $1,000-$50,000 CD players?
A: Their marketers pretend they're actually "transports."
 
I'm another who wasted real money buying audiophile equipment until I recognized audiophile superstitions for what they are. You're exactly right when you say great fidelity is available at very reasonable prices.
 
Oct 24, 2019 at 5:03 AM Post #10 of 10
I am about to give up my Aurender N100H for a reasonable offer. I used it literally twice since purchase a few years ago and decided to use my DAP with optical out as a transport instead. It’s essentially brand new!

Please PM me if you have any interest.
 

Users who are viewing this thread

Back
Top