New decade, end of the CD?

 

What I do know, is that for my 57 year old ears, which experienced Sgt. Pepper in its pristine condition on the day it was released, and then endured more extremely loud concerts, back when concerts were loud, than I care to count, I seriously doubt I could even possibly tell the difference between a $5 cable and one costing $300.  Really, when I get out my old vinyl LP's and put on the headphones, I'm not even sure I hear the pops, clicks, and scratches any longer. 

 

It was what we were using those fine gatefold sleeves for, in addition to holding the LP, that provided the additional expansiveness.  There's a very solid reason that ITCOTCK sounded much better having been packaged in a gatefold rather than in a single sleeve .  Not to mention that the particular gatefold in this case had a certain roughness to it.  I suppose it would still be relevant today, though decidedly less so than in the late 60's, if anyone actually bought LP's anymore.

 

S/PDIF cables are co-axial with a 75 ohm impedance and carry data as a multiplexed PCM stream at around 2MHz

USB 2.0 cables are twisted pair 17mA loop and carries data as 8-bit serial words in packets at 480MHz

 

So as you can see they are not the same, so whatever pseudo-scientific mantra is used for explaining the audiophile misinformation for one cannot be applied to the other.

 

Let's take a little look at jitter as this is something that gets highlighted by audiophilists as an *explanation* of why digital cables *sound* different. And let's try a slightly different approach to explaining it because it is clear I didn't explain it clearly enough the last time...

 

Jitter is when a transition from a "1" to a "0" (or "0" to a "1") occurs in a digital wave a little later (or earlier) than expected. This is a known phenomena and we (engineers) can see it (on an oscilloscope) and we (engineers) can measure it and we (engineers) can tell you where it comes from and why it is there and all those wonderful things that we (engineers) get paid a lot of money for knowing all about. we (engineers) also know how to deal with it, and we (engineers) do deal with it, because we (engineers) get paid a lot of money for knowing how to deal with it and for coming up with really smart ways of eliminating it or overcoming it - seriously, a lot of money, more money than it costs to buy a magic USB cable from a snake-oil salesman. We (engineers) like that because more than just liking the lot of money they pay us (engineers) we (engineers) like to solve problems like that, that's why we (engineers) got into engineering in the first place. Now, snake-oil sales men don't like solving problems, they like inventing problems so they can sell more snake-oil because they really do like money,

 

Now, if you imagine that a change in data from a "0" to a "1" is like sunrise and a change from a "1" to a "0" is like sunset it would be unwise for us to go looking for the "1" at sunrise because we cannot be sure when it will rise exactly, however we can be sure that it will have risen at noon, so that's when we go looking for it, similarily if we go looking for the "1" at midnight we can be sure it will not be there (and we will call it a "0").

 

All well and good.

 

The catch is knowing when sunrise and sunset are each day so we can work out when noon and midnight are, and that is actually quite simple and uses things like phase discriminators and voltage controlled oscillators and stuff like that to do something called clock-recovery, which is a way of getting the clock out of the data (which is coded into the data stream by exclusive-or'ing the original clock with the data ... a process called Manchester Encoding).

 

Now it stands to reason that if the data has jitter then the clock will also have jitter, but that isn't nearly as important as it sounds because if we look for the "1" at a few minutes before noon or a few minutes after noon, the "1" will be there. Now if we store these "1"s that we found at noon (± a few minutes) and the "0"s we found at midnight (± a few minutes) into a buffer using the "noon/midnight"-clock we recovered from the data stream then any jitter is eliminated because we can read this data out of the buffer using a different, much more stable crystal controlled clock from the DAC converter itself.

 

Now in S/PDIF system this is all done "in real time" and requires special circuitry to make it happen (called a SPDIF FIFO that uses a device called a Dual Port RAM or DPRAM... see, we're so amazingly good at this game we've even given these things special names). In USB audio this isn't done in real time, the digitally encoded audio data is arriving in packets at much faster bit rates than the DAC needs them in, so it also has to buffer them in some pretty normal everyday memory, which again, eliminates all the jitter that occurs in the twisted pair of the USB cable because again, the DAC reads this memory in it's own time using its own nice and ultra stable clock. Simples.

 

(time for another paragraph begining with the word "Now")

 

Now, once the data is stored in the SPDIF FIFO or the USB buffer memory it is a verbatim (ie exact) copy of the digital data that was sent there by the CD player or the PC or whatever sent the SPDIF or USB data - all the jitter has been eliminated, all the missing packets have been recovered and all that crap, so from the DAC point of view this data is the good stuff, indistinguishable from the PCM data that was created in the studio before it was put onto the CD master. Seriously, it's the same - the same bits, words, and everything. So when the DAC converts that PCM digital data back into analogue voltages they are exact replicas of the analogue waveforms that were encoded into digital data in the first place.

 

Now, the choice is yours - you can believe the explanation of the nice audio cable man who would really like you to buy his nice USB cable for $300 (and not the horribly inexpensive one from Belkin for $5) or you can believe the explanation of this not very nice electronics engineer who couldn't give a damn whether you bought the nice USB cable from the nice audio snake-oil salesman cable man or not.

 

Your avatar can.... So there..nahhh

 

Favorite hairspray from the 80's.