Poll: Can I hear differences between cables?
This poll is closed.
Yes
77.14%
162 77.14%
No
22.86%
48 22.86%
Total 210 vote(s) 100%
* You voted for this item. [Show Results]

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
CABLE SURVEY
#11
(18-Jul-2014, 10:32)Mohmm Wrote: Sorry, but this choice between "yes" and "no" for such a critical and contested issue is far too global and not enough discriminating. I cannot answer this.

Sorry Mohmm the idea was just to get a very quick snapshot of what fellow Devialet users felt regarding cables. I agree this is a very complex area and it might make sense to do a more detailed survey, possibly using something like SurveyMonkey. Let me know your thoughts on this.

Guillaume
Industry disclosure: UK distributor for Shunyata Research

O d'A no.63, Absolare Integrated Signature, totaldac d1 server with additional external power supply, totaldac d1-six, Wilson Audio Sasha 2, Transparent Ref XL digital cables, Transparent XL Gen 5 speaker cables, Ansuz Diamond Darkz, Shunyata mains cable prototypes, Shunyata Denali D6000/S, Shunyata Venom RCA, Audioquest Diamond Ethernet, Various Entreq ground boxes and cables, AQVOX Audiophile High-End Network Switch, sonicTransporter w/ 1TB SSD & MCRU LPS
UK
Reply
#12
Just my 2c, anything that relates in analog realm is more sensitive to types of able used. Shielding plays a big part, So is gauge.
However, when in digital realm, things gets shady. Some says data transferred in cables are in fact still analog(cause you need max distance for Ethernet cat5 cables or coax to work within specs). Cross talk between multiple cables (4 for usb) does play a part too...
It goes downhill from there on


Sent from my iPhone using Tapatalk
Reply
#13
(26-Dec-2014, 06:49)madumm Wrote: Just my 2c, anything that relates in analog realm is more sensitive to types of able used. Shielding plays a big part, So is gauge.
However, when in digital realm, things gets shady. Some says data transferred in cables are in fact still analog(cause you need max distance for Ethernet cat5 cables or coax to work within specs). Cross talk between multiple cables (4 for usb) does play a part too...
It goes downhill from there on


Sent from my iPhone using Tapatalk

Here is an analogy that might help to understand how digital is actually an analog signal and how it gets influenced (or not)
As some of you already know I work on a drilling rig. We use logging tools that are in the drill string and send up down hole data. It's my job to run those tools and extract the data. I will try to explain without getting too much into detail. We use mud to pump through the drill string and then around out of the hole to clean the hole out. The tool restricts the flow of that mud ever so slightly at a given rate. The result is a difference in pressure on surface of about 15-30 psi over a pressure of 1500-3500psi. That little pressure change is our 'digital' pulse. The surface equipment converts the pressure change into an electrical pulse and the computer works out through and algorithm what the tool is trying to say.

And this is how I believe it relates to cable; The pulses are not all the same strength and they don't all come in at the right time either.
For the difference in strength it is quite simple to ensure you get it right; the computer sets a threshold pulse height and everything above it is a 1 and everything below is a 0. Simple enough and in digital cables it works the same way. So if there is a difference in Voltage height because the cable isn't perfect there still is no difference in calculation of 1's and 0's on the other end.

As for the timing it gets more complicated; That's where the algorithm needs to be smart enough to work out which 1's and 0's are in the right place and which ones are not. In my job that's easy; there are only so many values available so all the computer needs to do is work out what value is closest to what it 'sees'. It takes a pick out of a bunch, literally.
I could have 0.9993 and the next one up is 0.9998. There are no 0.999(4)(5)(6)or(7)s in the algorithm.

And that's where with music it gets more complicated. Music is 'organic' (for a better word) and there are no known values to pick from. Al the DA converter can do is 'see' 1's and 0's at a certain time and convert them into a Voltage change. If the timing is wrong coming in then the timing of that Voltage change is out. And therefore the timing of your music is out.

Timing of pulses can be influenced by (reflection of) magnetic fields generated by the current going through the cable and generating a current in other parts of the cable.
The same way a transformer builds a magnetic field in one coil by running a current through it and then it converts that magnetic field into a current in the other coil.
That's why in Cat 5 cable they twist the paired wires; the magnetic fields are countering each other out by being set up in opposed directions resulting in less of a chance to create a magnetic field from one wire large enough to create a current in the other wire (called Cross Talk and in Cat 5 cables it is where the twist gets un twisted at the connection where that Cross Talk is the biggest worry. This is called Near End Cross Talk. The less un-twisting, the better the connection)

Magnetic interference can also come from outside; Never have a power cable lie alongside a signal cable!

Conclusion; a bad cable will have a negative influence on music because the timing can be out. The question remaining (I'm sorry but this is where you'll be disappointed because I don't have the answer) is how bad does a cable need to be before the timing goes out or how good does a cable need to be to get the timing 100%?

As I said, I don't have the answer to that. I guess that will be in the listening Wink

Maybe someone in the chat room knows a bit more about how data transfer gets 'checked' between sender and receiver. I would be very interested in that.

Cheers,

Pim
Lifetime Roon, Mac mini, int. SSD, ext. HDD, tv as monitor, key board and track pad on bean bag as remote,Devialet 200, O d'A #097, Blue jeans speaker cable, Dynaudio C1 MkII.
Jim Smith's GBS.
Northern NSW Australia.
Reply
#14
I started digital data logging in 1982 IIRC. I had done lots of analogue measuring previously but no analogue recorder was good enough and small enough to put on a F1 car. The nearest was a modified Uher cassette machine with a 4 channel head and using FM recording techniques to get good low frequency results. One of the channels was dedicated to cancel speed variation effects. So we only got 3 channels of useful data.
The digital recorder was better than analogue in terms of data accuracy from day 1 but the data analysis was slow and the hardware very expensive.
Eventually we made our own controller/recorder and things got faster and less expensive but the quality of the data we got hardly improved. It was already hugely better than anything analogue.
I am not sure your analogy is correct. Looking for a 15psi difference in a 1500psi signal is nothing like what we are doing with a digital interconnect.
Digital devices normally work on a nominal 0 to 5 volts. It doesn't matter if these values are accurate since the trigger is the change of state, not knowing what state it is at. It would be stupid, frankly, and abandoning one of the benefits of digital, to try to decide whether the signal was at 5 volts or zero. So nobody does. And nobody bothers too much how close the nominal 5V line is 5 volts is, for example, either because it doesn't matter at all.

In all my 30+ years of digital recording I have never known of a data error due to a non-broken cable.
A timing error may have been there, but it was not important for what I was doing, so I would not have been aware.

Personally I hear no difference between SPDIF cables. But that could be expectation bias.
Devialet Original d'Atelier 44, Goldmund Reference/T3f /Ortofon A90, Goldmund Mimesis 36+ & Chord Blu, iMac/Air, Meridian Control 15, Lynx Theta, Tune Audio Anima, Goldmund Epilog 1&2, REL Studio. Dialog, Silver Phantoms, Branch stands, copper cables (mainly).

Reply
#15
I haven't done extensive testing myself yet, and won't really do that untill I am able to buy a new house.
Anyway, what I have been able to test so far is that I can hear a difference in speaker cables.

Last year I had Vicol speaker cables myself (now a cheap second hand MIT Terminator 2, which sounds exactly the same as the Vicol), and tried a set of Siltech 330L.
There was a very noticeable difference. The Siltech had less deeper bass, but was faster.

The difference in power cables is something I need to check more.
In my current situation, where almost everything is on the same power group, I heard a very, very slight difference between a standard computer type cable, and a Crystal cable of €600.

Interconnects I want to try out someday. Especially coaxial cables, as I one between the CD player/transport and the D200.
At the moment I use a €50 Via Blue cable which sounds great. Really wonder what a different/expensive cable can do. For better or worse.

So, to me there can be a difference in cables. But I need to hear it immediately.
I'm not going to change a cable, listen to it for a week, and then change it back to hear a difference.
I really do not see the point in that. Either I hear a difference, or it don't.
Devialet 200 | Marten Django L | Modified Philips CD940 | Technics SL1210 Mk2
Reply
#16
(26-Dec-2014, 11:22)Pim van Vliet Wrote: Conclusion; a bad cable will have a negative influence on music because the timing can be out. The question remaining (I'm sorry but this is where you'll be disappointed because I don't have the answer) is how bad does a cable need to be before the timing goes out or how good does a cable need to be to get the timing 100%?

Pim

that's where I think there is misunderstanding, the accuracy of the 1 and 0 signal levels is unimportant in this discussion, as f1eng points out. The trigger thresholds have so much margin (from each other and from full scale 0.0 and full scale x.0) that the actual signal level fluctuations introduce no errors concerning bit change detection. For sure there is a rise time associated with every data bit change, but this is controlled by standardized interface specification of the driver and detector circuitry, embedded in every single device communicating digitally.

More importantly, even if there is an effect due to the rise time or clock out differences at the transmitter end, in most cases the clock errors etc are recovered further down the chain. If it wasn't like this then digital transfer just couldn't work reliably for any applications..... in the digital audio domain, the only aspect of audio data transfer that matters is where the final recovery of the clock associated with the audio data is made, the later the better (there are other discussions about power/noise transfer but I only talk about digital data transfer and recovery here)

So it is wrong to generalise about the effect of different cables on the final integrity and timing of digital data. Feel free to blame cables, wires, analogue circuitry etc on timing problems AFTER the final digital clock recovery. Before that, the choice of cable (assuming it meets the standard specification) can make no discernible difference to either digital integrity or final timing.

Taking asynchronous DACs as an example, the data is transferred asynchronously & buffered, with timing regenerated by the clock in the DAC, and the DAC outputs an analogue signal. There is nowhere in such a system where a normal digital cable can make any difference due to either timing or bit recovery integrity (though it is theoretically possible for such cables to introduce other problems, noise/power etc)
 _________________________________________________________________________
Aurender X100, Audiophilleo, Devialet 200, Verity Audio Parsifal Ovation Monitors
Leiden, the Netherlands
Reply
#17
The absolute voltage levels do not matter for digital signals but the slopes between 0 and 1 votage levels matter as the 1 phase is longer at slow slope rates compared to fast ones at the threshold voltage. That plays a role if the timing = jitter is important like with SPDIF or AES. Not so much with USB or ethernet but still they can also transmit noise independent from the digital data stream.
The problem with cables is that there is little correlation between price and performance.
Reply
#18
I'm adding to this thread, as I recently changed cables on my system (twice) and the differences in sound were dramatic. My equipment list is below, and there is no Devialet product there- yet. Those familiar with Naim equipment may get more from my descriptions as well as those whose equipment is modest, as is my own.

I had been using solid-core copper, 12 gauge, for a long time, as the hifi set up was 'temporary'. For the first change, removed the solid core copper and installed Distek cable. This is a directional cable made up of multiple strands of silver, each teflon-wrapped. It is yellow and grey inside a clear cover. The sound was very different; the upper mids and highs were clear, with a nice, easy sense of space around the notes. The mids and lows, however, became hollow and dull. it took me a while to recognize this lack in the the lower half, as the upper end clarity improvement was appealing.

A few days later, I removed the Distek cable and installed some Wireworld Polaris cable. I think this is a multi-strand copper cable. It came to me from a friend. The difference in sound quality was immediate and dramatic. I could tell, within moments, that the missing lower half was back. The midrange tones as well as the bass notes were back, with a fullness and 'roundness' that was apparent without any effort at listening for a difference. The 'front and back' of the notes were there and I could here this difference day after day of listening.

I know nothing about the Wireworld cables, but I think they are fairly old. There is absolutely no doubt, however, that they made a remarkable difference in the music that my system makes. Hence my vote for 'yes', cables do make a difference. I haven't used the Naim NAC cable in a while so I won't make any inferences about the quality of that cable.

BTW, if anyone knows anything about the Wireworld Polaris (V1, I think) I'd welcome any information or opinions.
Damon
W4S Sonos Connect & VPI HW19 mkIII into a D220 Pro, thence made expansive by Lawrence Audio Violin SE speakers, capably carried by Analysis Plus Oval 9 speaker cable, and juiced by power cords and coax cables in decorator colors
Vancouver, Canada
At the office: NuPrime IDA-16 into Amphion Argon 1, QED cable, ordinary Sonos Connect
Reply
#19
(29-Dec-2014, 05:49)Damon Wrote: I'm adding to this thread, as I recently changed cables on my system (twice) and the differences in sound were dramatic. My equipment list is below, and there is no Devialet product there- yet. Those familiar with Naim equipment may get more from my descriptions as well as those whose equipment is modest, as is my own.

I had been using solid-core copper, 12 gauge, for a long time, as the hifi set up was 'temporary'. For the first change, removed the solid core copper and installed Distek cable. This is a directional cable made up of multiple strands of silver, each teflon-wrapped. It is yellow and grey inside a clear cover. The sound was very different; the upper mids and highs were clear, with a nice, easy sense of space around the notes. The mids and lows, however, became hollow and dull. it took me a while to recognize this lack in the the lower half, as the upper end clarity improvement was appealing.

A few days later, I removed the Distek cable and installed some Wireworld Polaris cable. I think this is a multi-strand copper cable. It came to me from a friend. The difference in sound quality was immediate and dramatic. I could tell, within moments, that the missing lower half was back. The midrange tones as well as the bass notes were back, with a fullness and 'roundness' that was apparent without any effort at listening for a difference. The 'front and back' of the notes were there and I could here this difference day after day of listening.

I know nothing about the Wireworld cables, but I think they are fairly old. There is absolutely no doubt, however, that they made a remarkable difference in the music that my system makes. Hence my vote for 'yes', cables do make a difference. I haven't used the Naim NAC cable in a while so I won't make any inferences about the quality of that cable.

BTW, if anyone knows anything about the Wireworld Polaris (V1, I think) I'd welcome any information or opinions.

Naim amps do not have an output zobel network iirc. This is a choice which was made for SQ grounds, by the founder, decades ago.
What it does mean, though, is that without the correct loudspeaker cable load the amp can go unstable at supersonic frequencies and whilst not directly audible itself this overload can cause audible effects since the amp is probably no longer working in its linear region.
There are other manufacturers with a similar philosophy, Spectral springs to mind, and they are also speaker cable sensitive.
Personally I can't understand how a Devialet, with its unconditional stability and ultra low output impedance, would be effected by a properly terminated cable of adequate gauge, but I completely understand why Naim and Spectral could and do sound different with different cables.
Devialet Original d'Atelier 44, Goldmund Reference/T3f /Ortofon A90, Goldmund Mimesis 36+ & Chord Blu, iMac/Air, Meridian Control 15, Lynx Theta, Tune Audio Anima, Goldmund Epilog 1&2, REL Studio. Dialog, Silver Phantoms, Branch stands, copper cables (mainly).

Reply
#20
(29-Dec-2014, 09:17)f1eng Wrote: Naim amps do not have an output zobel network iirc. This is a choice which was made for SQ grounds, by the founder, decades ago.
What it does mean, though, is that without the correct loudspeaker cable load the amp can go unstable at supersonic frequencies .......
Personally I can't understand how a Devialet, with its unconditional stability and ultra low output impedance, would be effected by a properly terminated cable of adequate gauge, but I completely understand why Naim and Spectral could and do sound different with different cables.

Ah, that must be a 'no' vote. I was simply explaining my experience with speaker cables and the change in sound quality. Other cables in the system may or may not make any difference; others, on other forums, do believe the quality of say, ethernet cables makes an audible difference. As to that, I have no idea. No idea about the Devialet, either. But I suspect that most owners are using 'decent' lamp cord at least. My local dealer is demonstrating the Devialet with Nordost speaker cables that cost about $400.00, so not a bank-breaking recommendation from him.
Damon
W4S Sonos Connect & VPI HW19 mkIII into a D220 Pro, thence made expansive by Lawrence Audio Violin SE speakers, capably carried by Analysis Plus Oval 9 speaker cable, and juiced by power cords and coax cables in decorator colors
Vancouver, Canada
At the office: NuPrime IDA-16 into Amphion Argon 1, QED cable, ordinary Sonos Connect
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)