Some digital comparisons: cables,transports&jitter


A friend and I spent a couple afternoons comparing various digital cables, transports and jitter/resolution enhancement boxes in my Tact/Talon system.

Cables on hand: HMS Il Primo II, Argent Jaden, Verastarr Silver, PS Audio XStream, Illuminati DX-50, Cardas AES, Legend Chinchilla XLR, Green Hornet, and 2 homemade AES/EBU.

Transports included: Micromega T-Drive, Modwright Sony 7700, Tascam RW2000, Modded Marantz, and 2 Toshibas and a Panasonic DVD player.

Jitter boxes were the Camelot Dragon Pro2 Mk 2 and the Assemblage D2D w/& w/o upsampling (the Tact upsamples if the boxes don't). As you can see the variables were pretty extensive, but it was fun and enlightening, and here are some of my impressions:

First, anyone who doesn't believe/hear differences between different digital cables/transports please start your own thread. Differences were clearly obvious to us, even to the person sitting behind the speakers switching the cables/settings (my system uses no analog interconnects).

Surprisingly, I found the differences between type of cable generally more pronounced than the make of cable. Also surprising (to me) was that the coax generally sounded fuller, richer, more laid back and deeper in staging (with the possible exception of the Legend) than the AES/EBU.
The AES provided more immediacy, more presence and impact, and a wider more detailed stage. This was as true INTO the jitter box as after it.

Also surprising was that the degree of change with cables was as pronounced before the jitter box as after (though the box was a nice improvement--blacker background and better bass). My favorite cables were the Chinchilla (which is actually an analog XLR) and the HMS. The Chinchilla is open, rich and natural sounding with no loss of detail. The HMS is also quite rich with better dynamics and drive than the other coaxes.

Regarding the dejitterers, they were slightly different, but more alike than different when just dejittering. I generally preferred the upsampling done in the Tact when using the D2D; with the Camelot I liked using the 20 bit mode when using a Coax into it--more dynamic; otherwise liked the Tact doing all the upsampling (BTW, I always prefer the signal upsampled in the Tact to 24/192 when using the Tact 2.2x with the 2150 amp).

For transports, I didn't find the Toshibas and especially the Panny in the same league as the Micromega or Sony (or the Tascam either). With all the talk about the Toshibas I was disappointed. The 4960 was closer than the 3960 though. And my friend is hoping that some mods will improve them (we'll try again after they're done).

I liked the Sony better when not using the jitter boxes, but with them I preferred the Micromega (smoother and more dimensional, but also somewhat lacking in the bass).

Well that's about it...more general impressions than a real shootout. But, again, the differences were clearly obvious in most cases. Surprised that so many don't hear them. Maybe it's all system dependent ;-))
richards
The article I wrote for PFOnline explains some of this:
http://www.positive-feedback.com/Issue14/spdif.htm
Jea48,
I did try a Nordost glass Toslink (sorry, forgot to mention). It was a little bright and 'tipped up' for my taste in this system (the Toslink input in the Tact only upsamples to 96; the others do 192). While quite airy and detailed, it didn't convey enough power in the lower mids/upper bass.

Re: RFI, I use Shaktis on a couple of my PCs (both ends) including those on an Audio Magic Clairvoyant. The amp gets a Foundation Research LC2 and all other digital goes into a custom AM Digital Stealth via cords from Wolff, Silent Source, BMI (the differences here are another story) . Also use Quietlines.

Dmitrydr,
'fuller and richer' to me imply a better sense of instrumental decay and hall ambiance, with some added (?) warmth in the lower mids. I agree that we all use descriptive words slightly differently.

Probably the simplest way to convey my feelings was that the coax cables seemed to move me further from the performers and made the venue space seem larger and less intimate. Which I preferred was very recording-dependent. But within each category, each cable also sounded different, and some were clearly better (vague word, but I'm trying to avoid the word 'accurate'--let's just say more more real, and more enjoyable to listen to, easier to believe that the performers are really there).

Another thing I didn't mention that we tried just for a kick. We combined some of the cables together (using XLR ends only) and could still tell which ones we were using. For example, when we put the Cardas and Chinchilla together, it definitely sounded more like the Chinchilla than the Cardas. Weird stuff, but it was fun (I know, only an audiophile would find this fun, eh?)
Interesting to perform the same comparison if all the digital components were "normalised" as described in Audioengr's article (very educative, BTW - thanks!)