Digital Cable Choice; how critical is it?

You said:

"considering that the digital information that runs from the DAC to the Preamp is 1's & 0's or bits & bytes or whatever they are called;"

didn't you mean the (single) cable that runs from the CD transport (or digital output of a CD player) to the DAC?
And yes it does make a difference.

It's also important what kind of analog IC's you use from the DAC to the preamp and whether they're single ended (RCA) or balanced (XLR). Balanced seem to work better if your preamp has balanced inputs.

It's also important what kind of power cord is used with DACs, CD players, or basically anything with digital processing circuitry in it (meaning even preamps with digital volume controls). It has to be shielded, and it seems that with most digital-to-analog equipment, bigger (wire size) is better for sonics -- although I still can't explain why, because DACs aren't power hungry like most amps.

I also prefer fiberoptic cable (glass AT&T, not plastic Toslink) to metal coax (S/PDIF, RCA, BNC) or AES/EBU (XLR) for a number of reasons, not the least of which it isolates the DAC from the transport electrically. Unfortunately, there's only one really good one and it's a thousand dollars. Three thousand if you own EMM gear, and have enough money left over ;--)
.

Nsgarch, you are correct! I obviously had a brain freeze. I am questionning the importance of the single cable that runs from the transport to the DAC.

Regarding the analog cables, I have always run balanced XLR. However, I was considering going single ended, from the DAC to the pre, because my new pre has a single ended (only) "CD Direct" input that supposedly offers better sound than the XLR inputs.

All of my cabling is Kubala-Sosna, which I am VERY pleased with & intend to stay with.

I don't know if Kubala makes digital (75 ohm) ICs, but if they do, stay with them (for RCA or BNC which is the 75 ohm I'm talking about). Otherwise, if you have AES/EBU connectors on your DAC and CD transport/player, those are always preferred over coax S/PDIF (RCA) or BNC (bayonet) IMO. Kubala has those I'm sure.

I don't know about that "CD direct" business. If your new preamp also has balanced inputs, I'd try it both ways. Betcha a new CD you like the XLR better ;--)
.

Nsgarch, Toslink cable is now available with GLASS stranding. Glass Toslink made a HUGE improvement over the cheapo plastic Toslink. And this was in a relatively modest system!

Hi Mike -- That's not a surprise. Glass Toslink has almost twice the bandwidth as plastic. But its practical length is still only 5 meters, while AT&T ST cable can go up to 100 Km between repeaters (for those REALLY large listening rooms ;--)

The real issue with Toslink (for audio, that is) the poor quality of the little transceiver units. They cost about $5 wholesale (STs cost about $120 wholesale) and the quality of their signal output is affected by end-to-end reflections in the cables (another reason to use glass if you can) and transmission jitter (glass won't help with that.)

It can't ever beat AT&T, but whether a glass Toslink connection would outperform a high quality coax (RCA) or aes/ebu (XLR) I really don't know. It would be easy enough to try though, because today, almost all stuff that has Toslink has coax inputs/outputs well.

Here's a little Toslink history if you're interested:

http://www.audioholics.com/techtips/audioprinciples/interconnects/toslink.php
.

Barrelchief, digital cable sonic difference = placebo effect = no measurable difference on recorded waveforms downstream of your dac. But they do look cool!

Oh forgot something I thought might be useful. Signal recording and analysis tools are getting reasonable in cost, very powerful and somewhat user friendly, certainly user friendly enough for a lot of serious audiophiles. For around $3000 you can get a complete data recording system (sans computer which you already have)and you can progress from wondering & parts replacing to knowing and making informed decisions. I recently bought a Dataq instruments model 730 for work, (I'm sure there are others that are equivalent, I have no tie to the vendor, just wanted to give an example). Three grand is a big chunck of change, but for folks who are heavy into the hobby I think it can save a whole lot of money and frustration in the long run and allow a better final end result. Data is good!

Some say that PC's make no difference.
Some say that digital IC's make no difference.
Like Dalton said; "Opinions Vary!"

In my experience, in moving (in a fairly resolving system with a good transport and DAC) from a well received but modestly priced IC (Apogee Wyde-Eye) to a much more expensive and widely acclaimed IC (Kimber Orchid), the comparative difference was very distinct and clearly distinguishable. In simple terms, I found the difference to be comparable to looking thru recently cleaned windows. Prior to cleaning they were not really dirty, but afterward they were REALLY clear.

It has also been carefully detailed by an AG member with impeccable credentials, that IC length can play a significant role.

As I have stated before, if you have not made the comparison yourself all you have is an uninformed opinion.

My experience is a lot like Jeffcott's. I have had Wyde Eye to Wireworld Gold Starlight and in between and my conclusion is the same as his. I would only want to add that digital cable _length_ made a big difference to my system and an audio pal's. We compared a 1m length of Wyde Eye to 1.5m and the longer length was incontestably better.

Are all of you (Tobias, jeffcott, Jeff_jones) referring to a lack of difference only between 75 ohm cables (either RCA or BNC), or ALL digital data cable including 110 ohm XLR. ST glass, and Toslink (plastic and glass)?
.

First of all, difference between digital cables depends on a dac you're going to get. When I had a Bel Canto DAC2, I tried a lot of different digital cables varying in price from $50 to $1000: WireWorld SuperNova 3+ and 5+ Glass Toslink, Acoustic Zen Silver Bytes, Acoustic Zen MC2, Synergistic Resolution Reference, Virtual Dynamics NiteII and Master, Stereovox, and even some Monster. Difference between these cables was alway apparent. I chose Virtual Dynamics NiteII which was the best of the bunch, then I upgraded to Virtual Dynamics Master, which was in a league of its own. There are some dacs though, that don't react to the digital cable changes, so you'll have to try and find out.

If two digital (let's say S/PDIF) cables measure the same, they will sound identical regardless of price, material (copper, silver, gold, palladium, amorphous metal), dielectric or wire configuration. This might be a dangerous statement but I personally don't see why these two cables would sound differently. Do those measurements have limited value with regard to how good the cable will perform?

Chris

Nsgarch - fair question, but my better angel is telling me to just give up, take my toys and go home. If anybody is interested and has a fairly fast internet connection an insight into what is possible to take audiophile concerns (ac line effect, vibration tweaks, digital & power cables, why does my system sound different than it used to, etc.) more toward science is at http://www.dataq.com/applicat/presentation/I_A_hardware-overview.html
Happy listening!

Jeff, thanks for the link. I enjoyed the presentation.

Have you used this device to measure the things you mentioned in your audio system, or it it strictly a work tool?

I've never used anything but AT&T glass. For one thing, the Wadia DAC I used to have just had one glass input! (My new one has several different kinds.) I did try a lot of different AT&T cables though without any impressive differences UNTIL I sprung for an Aural Symphonics Optimism V2 and WOW! What an ear opening experience that was, with slightly different tonal balance depending on which way the cable was oriented. I know, according to your measurements, that should have no effect -- and it didn't with all my other ST cables.

As for 75ohm jobbies, all I know is what I've heard about connectors -- that BNC connections maintain the 75ohm continuity better than RCA connectors and therefore provide a more accurate datastream to the chips. But I don't use metal, so I haven't ever compared.
.

Nsgarch, I'm only referring to 75-ohm metal cables, I have no experience with any other kind.

My experiences have shown that there are dramatic sonic differences between digital cables. I cannot explain these differences in scientific terms but I am sure they exist, zeros and ones notwithstanding. One thing of which I'm sure, the differences are not "placebo" effect. Does everything that can't be explained by science fail to exist?

Nsgarch - Just work to date for a recording device. The one o these days plan is to get the meaningful measurables on my audio system recorded to hard disc for reference and troubleshooting, and then too start digging into things like does the dc buss on my pre amp wiggle when the fridge kicks on & does this create a measurable difference to a constant frequency wave output measured between amp & speakers.

VERY... I've owned the i2digital X-60 for the previous two years. (This cable is no slouch) http://www.i2digital.com/splash.html#

I just purchased a Harmonic Technology Cyberlight. Very obvious differences that even my girlfriend easily hears.

(1) Very faint sounds that were never heard before come thru easily.
(2) When the system volume is high, instead of the music sounding loud, it sounds like real instruments in your room, all very relaxed sounding.
(3) Lower noise floor
(4) Top to bottom improvement in reproduction.

I could go on and on, but will stop here.

Neil (Nsgarch): My remarks could have been more clear....Please let me clarify. My first ICs were Apogee Wyde-Eye and I used both 75 Ohm and 110 Ohm, with a CEC TL2 and an Audio Logic M34. Between the two Apogee ICs I really did not discern much difference, with only a very slight preference for the 110 Ohm. Both cables were 1 Meter in length. I then moved from the 110 Ohm Wyde-Eye to 110 Ohm Kimber Orchid, which was longer at 1.5 Meter. I would not choose to state that the length was a significant factor, but the improvement in clarity and detail in moving to the Orchid was very revealing and very suprising. Until that time I was in the camp of 'can it really make that much difference?'.

I have had only minor exposure to ST-Optical as my transport does not currently have a working ST-Optical output. My only experience with ST-Optical in my system was with another transport and that experience was that ST-Optical was suprisingly good.

Jeffcott,
Although the Cyberlight mentioned above is fiberoptic it needs RCA or XLR connections on your transport and DAC. It sends and receives light signals at its terminations.

I really am a music guy & not a computer/techie guy. So, considering that the digital information that runs from the DAC to the Preamp is 1's & 0's or bits & bytes or whatever they are called; how critical is it to the sonic perfromance that I use an elaborate digital cable, as I have done with my analog interconnects?

Apart from clocking issues...I really can't see how it can make a difference. Either the bits all arrive or some of them don't.....you will soon know if some don't as you may get some nasty pops or clicks.

As for clocking.....a cable, being a passive device, physically cannot add "jitter". Length and electrical properties will of course modify the absolute timing and shape of the clock signal but a constant shift is NOT of concern, as audible jitter issues come from VARIATIONS in the clock signal between ticks. Furthermore, some jitter is not a problem provided it is small enough and random....i.e NOT influenced by some highly correlated influence such as oscillations in a power supply with bass response or induced correlated RF noise from somewhere.

Jitter is more component related than cabling. Changing from optical to SP/DIF might make a difference but the cause is most likely due to RF interference or the accuracy with which components either generate or interpret these two different signals.

If your clock signal is accurately re-clocked in the DAC (rather than a slave to the transport) then you have nothing to worry about (except your DAC)

....just my two cents.

My experience has been that most digital cables do sound different and that it's not necessarily a matter of cost. Please remember that a bit stream (0's and 1's) is NOT traveling down the cable, but an analog waveform with peaks and valleys that gets sampled and turned back into 0's and 1's. So anything that can effect an analog cable can effect a digital cable. In fact, some of the best digital cables I have heard (at least metal S/PDIF cables) are really analog cables at heart. So keep that thought in mind when auditioning cables. FWIW, the best performing S/PDIF cables I have heard are the Audio Note Pallas, the Locus Design Core, the Apogee Wyde Eye, and the Kondo Sounds Labs KSL-LP, most of which are terminated with RCA's (in my case).

"As for clocking.....a cable, being a passive device, physically cannot add "jitter". Length and electrical properties will of course modify the absolute timing and shape of the clock signal but a constant shift is NOT of concern, as audible jitter issues come from VARIATIONS in the clock signal between ticks."

Everyone thinks they are an expert. You are incorrect. Read this white-paper:

http://www.positive-feedback.com/Issue14/spdif.htm

Important aspects for a S/PDIF coax cable:

1) length
2) characteristic impedance
3) terminations
4) conductor sizes
5) geometry
6) metallurgy
7) dielectric materials

Steve N.
Empirical Audio

"Please remember that a bit stream (0's and 1's) is NOT traveling down the cable, but an analog waveform with peaks and valleys that gets sampled and turned back into 0's and 1's."

This is certainly true, however the analog nature alone of ALL signals is NOT the main reason why digital cables often sound different. This has specifically to do with digital circuits and how they treat the signal versus analog circuits.

Digital circuits detect transitions in signals and make decisions based on this. The detection of these transitions is not an absolute thing. There is uncertainly involved because of the imperfect voltage references, ground noise and voltage noise. Slow risetimes in the signal can increase the uncertainty, making it much worse. Reflections can superimpose on the signal shifting it in time.

Steve N.
Empirical Audio