Digital Interconnect Cables - What's Your Experience?

deronb1 wrote: »

The file is perfect, or it is un-readable. Once it is processed by a DAC, and on it's way to another source, whether 're-read' or ended in a signal determines it's eventual sound signature.

This is a popular misconception. However, a "perfect" digital file, meaning one in which the differences between high and low are clearly distinguishable, will still have varying amounts of noise corruption in every bit.

As I stated previously in this thread, there is no such thing as a "digital" signal that consists only of high and low pulses. A "digital" signal is actually an analog signal with quasi-digital waveform characteristics. Since a digital signal is actually an analog signal, it is subject to the same noise corruption mechanisms as any other analog signal. However, a digital signal will be less affected by some noise mechanisms that affect signal amplitude.

deronb1 wrote: »

To get back to the original question. Can one USB cable transfer binary information with a better sound quality than another? Have not seen anything to make me think otherwise.

I started this thread with the same question. So far, I have not heard a difference among the USB cables I have auditioned, but it is theoretically possible. The key thing to grasp is that real world "binary information" does not consist of just high (1) and low (0) square wave pulses. There are deformations in the waveform which are analog in nature and there is noise being transmitted along with the digital signal. That noise can be transmitted all the way through the DAC and have an audible effect on sound quality.

Nightfall wrote: »

BlueFox wrote: »

At this point, a self-proclaimed expert needs to step-in and assure you guys that what you are hearing is impossible. In fact, it is because you are delusional that you think you hear a difference. .

Usually an electrical engineer.

I think most electrical engineers who have some knowledge of signal theory understand that a signal is going to be changed by passing through a conductor and that the change may or may not be audible.

There is actually an audiophile community within the Institute of Electrical and Electronics Engineers whose members have published many papers relating to stereophonic performance and perception.

DSkip wrote: »

Dk, as far as your original statement of not hearing differences in cables, I read someone say that once they invested significantly in improving the power aspect, differences in cabling became less apparent. This could have been what you were experiencing.

Yes, I made the same observation and comment. I also wondered in this thread if the differences in USB cables that others so easily hear is due to them using (electrically) noisy computers.

Nightfall wrote: »

DSkip wrote: »

I read someone say that once they invested significantly in improving the power aspect, differences in cabling became less apparent.

As in this?



...or amplifier power?

Yes, as in power quality treatment. In my case, in my two channel system, I have:

1. A PS Audio P10 ac regenerator for source components and preamplifiers.
2. Three dedicated 20 amp power circuits, each terminated with a passive power conditioner.
3. Low noise power cables.
4. Audio grade fuses.

Further Study: USB Cables And System Master Clock

I was getting ready for another round of USB cable trials, but the arrival of the Revelation Audio Labs Prophecy Cryo-Silver Reference DualConduit USB cables caused a "slight" change of plans. I have both the dual-A plug and single-A plug versions of the RAL USB cable. Both versions have the signal and power wires in a separate cable. The dual-A plug version allows the disconnection of the power leg after the DAC has set up the USB connection. Removing the power leg is claimed to improve performance by lowering the noise floor. That claim was immediately confirmed.

I wasn't going to do another USB cable shootout until my dCS Debussy DAC arrived. My current DAC, the DAC section of a Cary Audio CD 306 Pro Version SACD player, does not have a USB input. With the assistance of dCS's North American general manager, I recently found a good deal on a used recertified current version dCS Puccini U-Clock master clock that came with full three year warranty.

The U-Clock acts as both a system master clock, for up to four digital components, and as a USB-to-SPDIF converter. The USB-to-SPDIF converter is particularly useful to me since the USB output of the BDP-2 is the lowest in sound quality among its output types (SPDIF is best, then AES)

I had never used an external master clock before, but I had modified a Sony CD player (CDP-XA7ES) and a Denon CD player (DCD-1650AR) with upgrade master clock kits. There were substantial performance improvements in both cases.

I can't currently get the full benefit (significant jitter reduction) of the U-Clock because all digital components in my system would have to be connected to it. The CD 306 does not have a word clock input. Out of curiosity, I wanted to see if the U-Clock alone would facilitate sonic differentiation in USB cables. It did.

The U-Clock conforms to the "Grade 1" AES11-2003 standard for master clocks and has a frequency deviation of no more than +/- 1 part per million. Grade 2 clocks have a frequency deviation of no more than 10 parts per million. The class of master clocks above AES Grade 1 are "atomic frequency standard" clocks, such as the Esoteric G-01 ($22,900). The G-01 has a frequency deviation of +/- 0.05 parts per billion. In cost comparison, the U-Clock's MSRP is $5,500.


Figure 1. RAL USB cables: single-A plug version in foreground, dual-A plug version in rear.

Prior to this test, I had tried the following USB cables with a Cary Audio DMC-600SE DAC:

1. Generic computer grade USB cable #1 (free).
2. Generic computer grade USB cable #2 (free).
3. AudioQuest Coffee ($279).
4. AudioQuest Carbon ($129).
5. AudioQuest Forest ($35).
6. Pangea USB-PC ($35).

I didn't hear a difference among any of the above in the previous test.


Figure 2. Center bottom to top: PS Audio PowerBase isolation platform, dCS Puccini U-Clock, Bryston BDP-2, granite slab for vibration damping, Dell M1330 laptop computer running JRiver Media Center.

The current test included the following USB cables:

1. Generic computer grade USB cable #1 (free).
2. Generic computer grade USB cable #2 (free).
3. Pangea USB-PC cable ($35).
4. RAL USB single-A plug cable.
5. RAL USB dual-A plugs cable.

Digital sources were a Dell M1330 laptop computer running JRiver Media Center and a Bryston BDP-2 digital file player. Either the BDP-2 or the M1330 were connected to the U-Clock's USB input. One of the U-Clock's two coaxial outputs was connected to the coaxial input of the CD 306's DAC section.

The most surprising result was the tremendous improvement in sound quality of files played from the laptop, which were previously "unlistenable" compared to files played from the BDP-2. I could also discern a moderate improvement in clarity, detail, and image weight when I went from the generic USB cables to the Pangea. The RAL cables were an order of magnitude better than the Pangea and generic cables. In addition to sounding clearer and more detailed than the Pangea and generic USB cables, the RAL USB cables also sounded apparently louder.

The RAL single-A plug cable sounded identical to the RAL dual-A plug cable, until I pulled out the power leg (the leg with the name label). Pulling out the power leg resulted in a further apparent increase in sound level, and a further increase in overall clarity, detail, and image weight. I had planned on ordering a few more USB cables to evaluate, but they were all single-A plug designs, and I am not going back to that.


Figure 3. dCS Puccini U-Clock front. The curved fascia is styled to match the Puccini SACD player.


Figure 4. dCS Puccini U-Clock rear.


Figure 5. dCS Puccini U-Clock inside. Like an atom, the interior of a U-Clock consists of mostly empty space.

I hope that the most expensive component of the U-Clock isn't its sculpted and milled aluminum case. The U-Clock uses an oven controlled crystal oscillator (OCXO) to help maintain frequency stability. The crystal is isolated from thermal effects in an electrically heated enclosure or "oven" . OCXOs range in cost from a few hundred to a few thousand dollars. The OCXO in figure 5 is the orange and white box.

Assuming things work out with the dCS Debussy DAC, the U-Clock and Debussy will be stacked and located in the location currently occupied by the Cary CD 306 Pro version SACD player. The Cary will be retired and retained for research/testing/other unspecified purposes. I went into some detail with dCS explaining the difficulties I had with two other DACs. I was assured that I would not encounter such problems with the Debussy and that there weren't other interface issues lurking.

For one micro-nanosecond I thought about replacing the Cary with a dCS Puccini SACD player, but I realized that would have been a waste of money. I don't currently use the Cary, not even to audition newly purchased CDs and SACDs. New disc purchases are ripped and copied to the BDP-2's solid state drive, then auditioned. I definitely can't imagine ever going back to spinning digital discs one-by-one in my two channel system-not as long as I can sit in my listening chair and pull up high quality rips of any and every CD and SACD that I own-not as long as my rips sound equivalent to, and in most cases much better than, the physical media they were ripped from.

References

1. Word Clock Synchronization - When and Why You Need It (PDF Download Link)

2. Does Your Studio Need A Digital Master Clock?

3. Oven Controlled Crystal Oscillator

4. AES Standard AES11-2003 (PDF Download Link)