Quote Guy Johnson:

---

I don't care either way at those prices.

---

Indeed, it would be interesting to know from the current "testers" so far if they can see any justification in the construction? Or is it just "What the traffic can bear"?

The high cost of high end mic pre amps for example can be seen* in the quality of construction and in many cases the need for component selection, individual FETs, transistors, valves. The market is very small and the process cannot be economically automated. The same is not true to anything like the same extent for cable, it is all machine made.

*Tho' there are some devices I think that are exctracting the urine a bit!

Dave.

Quote ef37a:

---

The same is not true to anything like the same extent for cable, it is all machine made.

---

To be fair, I don't think the physical construction of Vovox cables (from individual screened wires) lends itself very well to total machine manufacture.

hugh

--------------------
Technical Editor, Sound On Sound

Quote Hugh Robjohns:

---

Quote ef37a:

---

The same is not true to anything like the same extent for cable, it is all machine made.

---

To be fair, I don't think the physical construction of Vovox cables (from individual screened wires) lends itself very well to total machine manufacture.

hugh

---

Ah! As I said, has anyone taken some apart and checked it out?
Are there links to decent close up photos?

Dave.

I have. I don't know of links to photos, but the brochures on the Vovox side include accurate cross-section images:

http://www.vovox.ch/index.php?option=com_docman&task=doc_download&gid=2&It emid=6&lang=en



The 'link protect S' version of the Vovox cable has two individual screened cores for the hot and cold connections, with a third ground wire. I can't remember for sure, but I think the screen wires are connected at one end only. The lay length is pretty long. As a direct result, rejection of EM interference is pretty poor, and RFI isn't much better.

The 'direct S' versions have three unscreened wires with a long lay length. There is no significant rejection of EM or RFI at all.

hugh

--------------------
Technical Editor, Sound On Sound

This thread is interesting to see as I had missed it the first time around.

1. I'm interested in knowing the conclusions of James, Hugh, and others who have tried the cable. Does it still seem to sound different, and if so, in what conditions?

2. Im my own experience, this different "sound" can happen with cables at all price points. The bulk Mogami quad always sounds different to me, and I use this as line interconnects. It has a larger and more focussed lower end, more body, and a much smoother upper end. The source sounds bigger. The cost? Room ambience. It was (is) definitely different, but not necessarily better nor worse.

It was an "effect," as a couple of meters in the middle of everything else would make the same difference.

In spite of this, I didn't run out and replace my mic cables, nor the rest of my interconnects.I could have, since the cable is the same or only a little more than Belden, Canare, and a dozen others. I did find myself preferring the 8 or so lengths of the cable to the rest of the various line level connects I had.

More things moved in to the box, and now I haven't nearly as much use for a myriad of line level cables.

I chose to view the Mogami quad as a "colour," only because it was opposite of how most other cables sounded, not because it was not neutral.

So Hugh, James, and others, what's your current opinion of the cables, now that it's years later? Though there is no significant rejection of RF, were RF problems encountered? And did third party analysis turn up anything interesting on the "control" or Vovox cables?

Star quad inherently has much higher capacitance between the hit and cold wires, and between them and the screen. That would tend to lower the hf roll off and affect the phase response, which would tend to reduce the sense of room and accentuate the impression of the low end. The strength of the effect would depend on interface impedances.

My experiences with vovox cables have been variable. On some systems I've heard a clear difference relative to other cables, and on other systems I've heard no difference at all. I've tried various subtractive comparisons and found nothing reliably repeatable, other than the poor rejection of common mode interference.

Hugh

Thanks for that, Hugh. Interesting.

Regarding the quad, true, but Canare quad sounds pretty much like everything else non-quad, yet Mogami has this very different sound. I haven't looked at the cap figures, though, and it may be that Mogami is much higher.

However, if that were the operative factor, would only a meter or three have such an impact (with Mogami, and on live level sources? Caveat acknowledged about the devices on each end of the cable.

Ogami W2549 lists a core to screen capacitance as 76pf/mtr

A similar Cannare product is nearly twice that at 120pf and for comparison a Belden mic cable showed just 42pf/mtr but none of these values should have the least effect upon any competently designed line output amp* at sensible lengths and even 100mtrs hung on a 300RZout mic only gives a turnover frequency of about 44khZ?

But in any case if capacitance were the problem here surely the effect would have been noticed between standard and star quad cables?

The TL07* series of amp might be embarrassed by high capacitive loads but the very common NE5532 should not be and it is in any case a trivial matter to arrange for the latter to have a virtually zero output Z and so not even a low pass filter is formed at even the daftest audiophool frequencies. I have a theory, be back after the wrap!
Dave.

Capacitance cont..

It is known that certain capacitor dielectrics, polyester give more distortion than other materials such as polystyrene and so is it just possible that the dielectric in cable could produce distortion and "colour" the sound?

The theory has a big weakness tho' there needs to be a significant voltage across the capacitor for even a tiny level of distortion to be detected (i.e. not much out of noise), in the order of 10Vrms (+20dBV).

Dave.

Quote Jeraldo:

---

I haven't looked at the cap figures, though, and it may be that Mogami is much higher.

---

I haven't either -- but capacitance is an obvious first port of call. There are numerous other factors that affect cable 'sound' though, including the nature and diameter of the dialectrics between the core and screens. And then there is the issue of RFI and how the cable deals with that... and how the equipment it is connected to deals with it.

There is so much RF flying around these days, and a surprising amount of equipment really doesn't deal with it very well. I am quite convinced that this plays a big part in the impresion that different cables sound different -- even over short lengths (which could suffer RFI issues even if the capacitance/dielectric issues can be discounted).

However, gaining access to a suitable faraday cage to test the hypothesis isn't easy...

hugh

--------------------
Technical Editor, Sound On Sound

Container Studio!

--------------------
PA stuff on FB

Surely the acid test of any "improving" hardware is a third party one?

If the supposed improvements cannot be carried forward into recording and folks can say "Yes THAT recording with X sounds better than that one without" We is all hissing in the wind?

Give me the .wavs! (well. not me personally! I have long since left the quality critique domain!)

Dave.

Thanks, Hugh. It's a rare thing to have an engineer give an open minded (at least to my mind) assessment of cables, and the identification of important variables beyond capacitance has been interesting and informative.

Wow! I pop back to SOS for the first time in eons and someone's dug up an old thread of mine from 2005!

Quote Jeraldo:

---

So Hugh, James, and others, what's your current opinion of the cables, now that it's years later?

---

Well, I still think there is a repeatable difference and I still think I like it. It's not like I've gone crazy and rewired the entire studio or anything, but I have purchased a 5m VoVox pair for mission-critical stereo work to go with the original VoVox mic cable I got for free.

It was remiss of me not to post some WAVs with this thread originally. Fact is I did do some basic testing, but it was a few years later, so I'm happy to post the files again, as there still seems to be some residual interest in this.

I'm a voice-over artist - this is a spoken word only test...

Vovox Cable WAV
'Normal' (Belden/Neutrik) Cable WAV

First some preemptory Q & A...

Do the files null?
I don't know - I haven't tested this but I doubt it since the sonic differences (at least to my ears) are so apparent. Feel free to try it and report back! Of course some obvious reasons these particular files probably won't is that however good manufacturing tolerances are these days I suspect there's no such thing as a perfectly cloned microphone. And also that I cannot say my positioning was accurate between them to less than a millimeter. Did I swap around the mics and rerun the test? Yes. Did I swap around the pre-amp channels and rerun the test? Yes. Were the results and sonic differences pretty much identical and thus attributable to the principal variable of swapping the cable? Yes. Will I be re-testing this one day with a single microphone and a splitter to eliminate at least some of these outstanding variables? Yes - watch this space!

Why didn't you make this test blind and ask folks to choose?
Because it's irrelevant which one you 'prefer'; for the purposes of this argument the only interest is in establishing whether or not there is a sonic difference between the two WAV's.

Can we trust you not to have messed with the files?
I have no connection or financial interest in Belden, Neutrik, Vovox or any of the other cable/equipment-makers here. I'm just a regular gear-head like you are. I have absolutely nothing to gain from posting the results of my test and it's not even like I need to justify a large outlay on the so-called 'boutique' Vovox cable as it was included free with my Brauner Phantom AE microphone. So why on earth would I bother skewing the results?

Do you honestly expect this to be the last word on this?
Of course not - I'm not that naive, this is a recording forum after all and we're all here to exchange views and share expertise! However, I'm an empiricist so I'm firmly of the view that if at all possible, in audio as in anything, it's very helpful to provide evidence to back up one's POV. So with respect to this thread I think it's reasonable to ask anyone wishing to dispute my opinion (which is that switching cables does make a difference to the sound) to do so either in reference to the WAV's I've provided or to ante up WAV's of their own test.

Testing equipment was as follows:
Mics: Matched pair of Microtech Gefell M300 SDC's
Pre-amps & Conversion: Metric Halo ULN-2
DAW: Logic 7
Format: 24-bit/44.1k WAVs
Processing/Normalising etc: None
Room Type: Living room, ie no special acoustic treatment
Pop shield: I honestly don't recall if I used one or not, I think not but if I did it's a Sennheiser
Tracking Cans: Sony MDR-7509


--------------------
I used to be a rocker, but now I've gone off it and just sit in one. (JL)

Edited by James Lehmann (14/03/12 06:50 PM)

James:
Many thanks for coming back with your commentary and your comparison files.

The difference is very apparent, although depending on market and tastes, I'm not sure one would be consistently preferred. No doubt, though, there is certainly a difference.

While voice is always the ultimate test of naturalness, I can't help but wonder what an orchestra recording might sound like with a pair of these vs a pair of something else.

To me, the difference is solely in the mid frequencies that determine the quality (hardness, softness, etc.) of the voice. I felt the recording a bit sibilant and a bit heavy in the lower end (proximity effect?), and those qualities may reduce the apparent differences of the cables. That upper and lower end is no bad thing, however, as it will be preferred by as many as not.

In my own experimentation with Mogami quad used as line level interconnects, I find the difference substantially more pronounced than evidenced here.

Oh, is that you doing the VO? Very nicely done!

Once again, thanks for comments and files.

Edit: if I had to listen to long portions-say a book or documentary- of this VO, I would prefer the VoVox! I suspect the quality of cables would be less important with lesser talent and lesser recording technique than you have here.

Yes. Don't forget that that the mics I used were Gefell M300 SDCs - very fine mics, but not what I would normally reach for when recording V/O (that would be SM7B or M930), but they're the only matched pair I have and that suited the needs of this particular test.

Despite what I stated earlier about planning to repeat the test with a mic splitter, I am still scratching my head as to how this would work because of course the mic needs to be connected to the splitter with a single cable before it can be split into two pre-amp feeds; this somewhat negates the point of pinpointing the cable variables, although IIRC Hugh has said that he's observed the VoVox effect simply by adding a length of it to another cable, which only serves to deepen the mystery!

I think the close-proximity matched pair of mics method like the one I used is probably the best realworld test of microphone cable effects outside of a laboratory - ideally one would provide a second set of tests with two identical cables as a control so folks could evaluate the differences between the mics themselves, and even the pre-amps. As I said, I did test all this for myself at the time and rapidly concluded that changing cables was having a far greater effect on the sound that either of the other variables, so I just posted those WAVs.

--------------------
I used to be a rocker, but now I've gone off it and just sit in one. (JL)

Edited by James Lehmann (16/03/12 03:11 PM)

This thread, 7 years after it began, has still not resolved the controversy of whether Vovox or usual high quality balanced cables sound better. The controversy will probably not be solved by physics or technical arguments alone. Most of us would agree that the laws of classical physics and the electronics and acoustical principles derived from those laws adequately describe and predict the audio phenomena we are observing. The difficulty is that many different, sometimes conflicting, principles can be invoked to explain an observed phenomenon, and changes in physical parameters do not necessarily correlate in a simple way with perceived changes in sound. The bottom line here is whether the audio engineers and the listening public with the most discerning ears can perceive a difference and whether there is a consistent and reproducible preference for one or the other. Being a believer in both science and reproducible subjective excellence at minimum cost, I don't see much role for gorgeous new audio equipment that incorporates some impressive sounding engineering innovation but performs no better and costs a fortune.

So I propose a publicly observable, transparent, objective randomized trial method to compare two products, say Cable 1 and Cable 2, in a controlled manner while addressing readers' perceptions of conflicts of interest and potential biases. This design is based on medical randomized clinical trials, which address some of the same problems we have here including subjective outcome measures, subtle differences, variability among judges, strong industry financial interests and potential for conflicts of interest among investigators.

1) Conduct a blinded experiment in which the participants do not know which cable is which. A third party not involved in any other aspect of the experiment prepares the cables so they are indistinguishable even to the person doing the switching, for example Cable 1 is Vovox and Cable 2 is Belden. This key code is hidden from all others involved in the experiment and the code is not broken until the end. For each trial, 2 scenarios, A then B, are compared. The switcher can choose any of 4 possible permutations:
Cable 1 as A and Cable 2 as B
Cable 1 as A and Cable 1 as B
Cable 2 as A and Cable 1 as B
Cable 2 as A and Cable 2 as B

2) For each trial, the switching person randomly (using a random numbers table) connects one of these 4 permutations and records which permutation was used in a database but does not otherwise participate in the experiment.

3) Invite "Golden Ears" listeners from various settings, independents, cable manufacturers, SOS staff and readers, etc. For each trial, have everyone choose and submit their preference, A or B or no perceived difference, as well as describing qualitatively what they perceive as the differences. Tell them that on average, in a randomized trial, half the A/B combinations will compare the same cable as both A and B.

4) Let everyone discuss, even argue, all they want about each trial and the entire experiment, WITHOUT BREAKING THE CODE.

5) A separate person, blinded to the code key, does the statistical analyses and hypothesis testing and tabulates the results in real time as the experiment progresses. Possible results are cable 1 is preferred to cable 2, cable 2 is preferred to cable 1, or neither is preferred to the other. He or she tells the switcher to stop once there is a consistent statistically significant difference or it becomes clear that the results are not converging to a difference. Appropriate statistical procedures for tests of difference or noninferiority are available, as are procedures to adjust p-values for repeated looks at the data. The statistical and experimental methods would make a nice AES paper, if this same idea has not already been published (I have not looked).

6) Everyone eats a large lunch with no alcohol and comes back satiated and happy. You can promise alcohol with dinner if anyone objects.

7) Show everyone the tabulated results in terms of Cable 1 and Cable 2 (but still not identified which is which). Repeat the discussion #4, still before breaking the code. Many participants may have their own biases, but no one will know which cable his or her bias favors. To the extent possible, try to get consensus on as many results as possible. When consensus cannot be achieved, characterize the differences qualitatively.

8) Only after discussion has concluded, break the code publicly so all can see.

9) Record the controversy that erupts after the code is broken and see if the conflicts of interest that then become apparent can explain some previously unsuspected bias in the results.

10) Publish your results in SOS. You could even post a set of .wav files for SOS readers to vote on A vs B vs same and see if your readers replicate the consensus results of the Golden Ears crew. Obviously, variability among readers in listening devices, ears, listening skills, etc will be large, but that is the real world and you can get data on what you expect will be the major variables.

Good post Jorge, and I'm sure we all agree some more rigorous tests (ie more than the very rudimentary one I've done here) would be very informative.

I think the key though is not to confuse establishing whether or not there is a difference between cables vs establishing which one a given individual might prefer.

For me, establishing that there is an identifiable difference between cables should come before trying to select a preference. Preferences are highly subjective - I like more LF, you like more HF, ad infinitum - which like anything in audio often just comes down to taste. I like apples, you like oranges. Preferences don't really prove anything in absolute terms other than being useful for statistical patterns and predictions.

I think it would be very worthwhile to first assemble more convincing evidence to support or refute the following (over-simplified) assertion: "changing cables changes the sound". There appear to be plenty of folks out there who (quite surprisingly) feel able to support or refute this without reference to any solid evidence at all, as well as those who are (quite reasonably) more cautious and remain on the fence in attendance of a wider body of evidence that would help them make up their mind either way. I'm probably still on the fence myself, albeit clearly leaning one way based on what limited evidence I have been able to assemble.

So it's a two stage process. First an A-B-X to establish differences, then an A/B to establish preferences.

A proper A-B-X listening test is useful mainly to establish whether or not people can repeatedly and blindly identify a random sample (X) as being file type A or B. The available responses are: X is an orange, X is an orange, X is an apple... etc. "I like apples" or "I don't like oranges" doesn't come into it! A-B-X can prove or not (well make statistically likely or not) that there are differences between the two files. I am very interested in the results of this.

After that a simple A/B is all that is required to express a preference and for this there is no right or wrong and really very little scope for 'argument' in the strictly logical sense; otherwise once again we end up with, "The apple I like is better than the orange you like", although of course internet Forums are full of exactly this kind of essentially futile discussion that we all get carried away with sometimes! I would be less interested in the results of this - it's already clear from the many replies to my threads that some folks prefer the Belden and some prefer the VoVox cable, just as in other threads some folks might prefer a GAP-73 while others prefer a Forssell or Gordon pre-amp.

--------------------
I used to be a rocker, but now I've gone off it and just sit in one. (JL)

Quote James Lehmann:

---

I think it would be very worthwhile to first assemble more convincing evidence to support or refute the following (over-simplified) assertion: "changing cables changes the sound".

---

It's an utterly pointless exercise in my view. It is perfectly well understood that cables can change the sound if the interface design -- whether deliberate or through incompetence or cost-cutting -- permits that state of affairs.

No one will argue, for example, that different cable constructions can have a significant effect on the tone of electric guitars when playing through a guitar amplifier, and the primary reason is that the interface operates in a high impedance domain which readily exposes variations in cable capacitance and other parameters.

However, what I would argue is that if you can hear changes in sound quality when changing cables between line level equipment, then you are using poorly designed equipment. The interface should be designed such that any adequate cable of reasonable length should not interfere with the transmission of information in any detectable way.

If it does then something is fundamentally wrong with the design. So if you want to buy duff gear, and then spend a fortune trying to make it sound 'better' with exotic cables, carry on. I have no problem with that.... but I won't be doing the same thing, nor wasting my time with double blind experiments of the bleedin obvious!

However, I would point out that you can not 'test' a cable in isolation. The characteristics of the source and of the receiver MUST be closely defined and controlled since they too are essential elements of the entire interface.

Hugh



--------------------
Technical Editor, Sound On Sound

You all have thoroughly confused this newbie (as measured by total hours behind a DAW)!

After 5 years or more of reading (and SOS is my audio bible!), I had come to believe the Holy Grail of audio recording and monitoring equipment was accurate reproduction, with any effects to be added after the fact and totally at the discretion of the engineer. For example, SOS reviews of preamps will often have as a "Con" - some phrase like, "not the flattest/cleanest response, but the effect can be useful in certain circumstances". Similarly that's why, for example (I believed!), we don't use hi-fi speakers to mix with - dispassionate (and sometimes cruel) accurate sound reproduction is the goal, regardless of whether it sounds "better", or flatters the sound.

Now I read that some highly esteemed and respected (by me!) SOS-ers are saying they could see buying these cables because they "sound better".

How do I reconcile these two points of view? If (as an absurd example to try and make a point) a frequency analysis of the cable showed it magically added high frequency harmonics like an exciter, wouldn't that make it unsuitable for use, even though it "sounds better"? I thought "better" wasn't even a relevant concept when discussing recording equipment ... only "accurate" was ...

Thanks for sorting me out - my terra firma of basic audio recording goals and concepts seems a little shaky to me right now!

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

Agree absolutely Hugh.
Can I just point out tho' that it is the middling value (about 7-10k but VERY complex)of the source impedance of the guitar/lead/amp setup that is the root cause of the sound changes?

The high load impedance comes about more as an historical hangover from the days when all guitar amps were valved. Loads down to 200k or so tend to go unnoticed unless an A/B is done.

Just trying to keep the myths and misinformation to a minimum!

Dave.

James and Alexis, I would think that in some cases the question of whether a change in cables produces a change in sound could be answered for a given hardware configuration by subtracting signals and looking at and listening to the difference signal. Only if the difference is not negligible relatively to the source signal should we even entertain the possibility of audible differences. What is "negligible", ie, how much and what kind of differences are too small to be audible to trained/untrained listeners? I don't know the psychoacoustics literature to answer that. Depending on the answer to that, you may be able to rule out audible differences by this sort of simple electronics method in some cases. More simply, from the perspective of faithful transmission of a source signal, if the difference signal between the original source signal and the signal after transmission over the cable being tested is negligible, the transmission is nearly exact. I agree with Alexis that this is a main goal for our cables.

Hugh, you are essentially saying that whether or not changing cables changes the signal/sound is dependent on how robust the hardware is to variability in cable parameters. This is a very important consideration for line level signals and I agree completely. With microphone cables, however, it seems to me that the output circuitry of different microphones is much more variable than that of line level devices, and the signals are orders of magnitude lower, resulting in lower S/N ratios than for line levels. Does your point carry over to microphone cables? In other words, do you think that the sound quality of high quality microphones should not vary audibly with changes in cables from normal studio quality cables to novel innovative cables?

Hugh your point about not testing a cable in isolation is key and shows that the simple test procedure I proposed is probably overly simplified and impractical, at least for microphones. It is easy to imagine significantly different results for different microphones, microphone / preamp combinations, or even different voices.

Quote Jorge:

---

I would think that in some cases the question of whether a change in cables produces a change in sound could be answered for a given hardware configuration by subtracting signals and looking at and listening to the difference signal.

---

Been there, done that. No significant difference in audio performance found. The only difference I've found consistently is the poor rejection of external interference.

Here are a couple of plots I've just done using an Audio Precision test set. I've plugged a Vovox XLR cable into channel one and a cheap Chinese ready-made XLR cable into channel two. Both are approximately 10m cables.

The AP only tests up to 80kHz, but there is a very subtle difference in the frequency response above 20kHz. I have expanded the scale enormously of this plot to +/-1dB extremes, with 0.1dB increments. The Chinese cable (purple plot) shows a fractional reduction in level, although we are talking about much less than 0.01dB which I really don't think is in any way significant or even humanly detectable as far as the audio quality is concerned. If the HF response continues significantly into the stratosphere it might have implications in terms of ultrasonic interference in susceptible receiving equipment.



I performed a range of other tests, all with insignificant differences. For example, the signal-noise ratio measured 116.535dB in the Vovox cable and a better (but insignificantly so) 116.924dB for the cheap Chinese cable.

However, the one test that did provide dramatic differences was an examination of the averaged frequency spectrum with no signal present:



The Vovox cable is the brown trace and the cheap Chinese cable is the purple trace. As you can see the Vovox cable has much worse interference from induced mains hum -- over 20dB worse at 50Hz -- and with a lot of strong higher harmonics extending up to 2kHz that are absent from the Chinese cable. And this is with the two cables deliberately looped next to each other and as far away from the mains wiring of my test bench to minimse EM induction.

I think it is entirely possible that some if not all the perceived differences are due to this kind of interference, either through its directly audible contribution, or through its influence on the receiving circuitry. Bear in mind this spectrum plot doesn't reveal the state of RF interference which may also be significant.

Quote:

---

With microphone cables, however, it seems to me that the output circuitry of different microphones is much more variable than that of line level devices, and the signals are orders of magnitude lower, resulting in lower S/N ratios than for line levels. Does your point carry over to microphone cables?

---

Yes. There isn't that much difference between mic outputs, really. Some use transformers, most use electronic output stages and of those, some drive the line differentially and some single-sided in impedance-balanced configurations. But they all have source impedances of around 150 ohms and feed receovers in the order of 1500-2400 ohms. It's not a difficult interface to design.

However, we pick and choose mics for their specific sound characters and increasingly we are being encouraged to change the mic preamp input impedance specifically to change the sound of the mic... so why not change it with a 'characterful cable' as well?

Quote:

---

In other words, do you think that the sound quality of high quality microphones should not vary audibly with changes in cables from normal studio quality cables to novel innovative cables?

---

As with mic preamp impedances, some mics -- especially those with output transformers -- do suffer from audible changes in character. Most transformerless designs either don't at all, or are far less variable.

hugh

--------------------
Technical Editor, Sound On Sound

Alexis, If I catch your drift you are confused about "high fidelity" and accuracy and recording?

A hi-fi system should indeed reproduce the signal it is fed without adding or subtracting from the original. Put another way, an output signal, attenuated to exactly the same voltage as the input should be indistiguishable from said input in distortion, noise, F response or sound and decent power amplifiers at least have been able to do this for several decades.

Loudspeakers are still a problem, put two top end jobs side by side and they will sound different (so one or both must be "wrong"!)Same goes for microphones.
In the recording arena take the simple (!) task of recording an acoustic jazz trio with a co-i pair. Received wisdom is to find a spot in the room that sounds best and plonk the mics there. But who decides where is best?

Dave.

Wow, Hugh. That was an impressively clear and thorough response!
Your noise finding is not entirely surprising. It seems to me that a 3 conductor twisted cable should inherently have less complete common mode rejection than a 2 conductor twisted pair cable, based on the less symmetrical geometry of the hot and cold helical balanced signal conductors when a 3rd conductor (ground) is twisted in with the other 2.
I can't read the vertical scale on the noise plot.

Quote ef37a:

---

Alexis, If I catch your drift you are confused about "high fidelity" and accuracy and recording?

A hi-fi system should indeed reproduce the signal it is fed without adding or subtracting from the original. Put another way, an output signal, attenuated to exactly the same voltage as the input should be indistiguishable from said input in distortion, noise, F response or sound ...

Dave.

---

Thanks for that, Dave! What was confusing me is that I thought it would be anathema to choose a cable because it sounds "better", without at least confirming that it doesn't "corrupt" the input signal ("distort"? ... sorry for the probable incorrect verbage).

I guess what Hugh has shown in his 1st graph is that, except for what he suggests is likely insignificant handling of the highest frequencies, it doesn't distort the input signal, at least relative to the Chinese cable.

Regarding his second graph, I'm not sure how there can be such a large difference in output between the cables when no signal is present - but still have the results plotted in his 1st graph ... unless maybe the scale in the 2nd graph is an order of magnitude or more expanded/more sensitive than the 1st?

So, Hugh's testing suggests there is negligible difference in frequency response between the two cables when presented with a signal, but nevertheless he and others "hear" an "improvement" using the Vovox.

I'm not sure how that would work ... would that mean there is another quantifiable parameter, as yet untested in the comparison between the two cables, that affects what we hear (I assume any difference in voltage at the end of the cables is adjusted, so that the differences in "loudness" are eliminated as a variable in comparing how they sound)? Or is it possible that even a miniscule amount of additional distortion in the Vovox cable from mains hum interference (Hugh's 2nd graph), when added to signal, acts as some sort of acoustic enhancer that subjectively makes it sound better? I assume these possibilities, or others involving repeatable measurements, are more likely than one that the Vovox simply sounds "better", but for a reason that cannot be measured ...

Hugh - did you say in an earlier post that you heard the improvement using the Vovox cable in some locations but not others? Did you generate your graphs in a location where the same Vovox cable sounded significantly different/better than the Chinese one?

Finally - if someone made an unshielded cable from scratch and A/B/C'd it with the Vovox and the Chinese cable, would a blinded test show the Vovox and home-made unshielded cable sound similarly "better" than the shielded Chinese cable? In other words, could the difference in how it sounds just come down to a difference in shielding?

And really finally now - why the emphasis on the country of origin on the shielded, non-Vovox cable in Hugh's post?




I'm sorry if I'm asking questions that have been answered already. I did read all the above posts, but much of it was over my head, and I think it's not at all unlikely that I missed the answers to my questions up there - apologies ahead of time!

Thank you!

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

The freq response plot was measured with a test signal at 0dBu. I also ran tests at -40 and -60dBu to replicate mic leve signals, but there was no difference... So it's not a level dependent change.

The spectrum plot was taken with no input signal, but the cable still properly terminated and the residual noise floor shown is around -160dBu, which I presume to be the AP noise floor. The electromagnetic interference from local mains wiring peaks to around -130dBu for the Vovox cable. Sorry you can't read the scale; the image zooming function seems to be broken!

I ran a variety of distortion and intermodulation tests, at a variety of levels, none of which revealed any significant difference between the Vovox and Chinese cable performance.

As I have said, I have heard differences in various Vovox demonstrations, but I'm not sure I'd say it was a definite improvement... Just different. Moreover, I've never been able to repeat the differences reliably on my own equipment.

I didn't mean to emphasise the origin of the comparative cable specifically. The point was that I was comparing the expensive special Vovox cable with a very cheap mass produced cable using ordinary unpretentious wire. It was something that I had lying around which happened to be the same length as the Vovox.

Vovox claim the sonic benefits of their design are due to a combination of the very non-standard construction, the shielding (or lack of it) and the insulation materials and conductive cores used.

I did once make a comparison cable using single core screened wires twisted together and observed the same poor interference rejection properties, so I think it's fair to say that most of that aspect is down to the specific construction.

Hugh

--------------------
Technical Editor, Sound On Sound

Thank you, Hugh! I'm sorry I said you had posted it sounded "better" instead of "different" - I got you mixed up with someone else way up in the thread!

And thank you immensely for the detailed response. It helps me understand these ideas and concepts so much better. At this point, I'm kind of wondering if Vovox is charging premium prices for essentially eliminating shielding from their cables?!

I believe I have some residual questions from your plots and testing, and your kind response. Hopefully they'll make sense to me as I think about this more, but if you or someone else wanted to spend even more time on this and give some further insight, I'd of course be grateful for that as well!

1) Hugh concluded that it wasn't a level-dependent change because the test signals were at various dBu. What I was wondering is if whether the difference in construction between Vovox and other cables would result in a different/perhaps higher signal level being outputted from the Vovox cable (for a given equal input), and thus sound louder (and possibly better?). Has anyone measured how much signal is lost from traversing a Vovox cable vs. other?

2) Hugh, you mentioned you'd never been able to reliable hear a difference on your own system, just other systems. I wonder if the comprehensive testing you did (and showed above) would have had different (and less equal) results if they were done on a system that the Vovox cable sounded "different" on?
3) Finally, and this is probably a silly thought (so I stand ready to receive ) - if one wanted to determine whether the Vovox cable's greater interference induced from the mains hum (low in amplitude though it may be) could be responsible for the audible differences between the cables - could the Figure 2 cable signals be mathematically "differenced", the resulting plot used to generate a "differenced" signal from some sort of generator, which could then be "added" to the non-Vovox cable ... all to see if the resulting sound made the normal cable sound more "Vovox-like"?

Thanks again, Hugh and everyone else!

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1

Quote alexis:

---

At this point, I'm kind of wondering if Vovox is charging premium prices for essentially eliminating shielding from their cables?!

---

To be fair, the man behind Vovox is a very intelligent scientist and is very earnest about the sonic advantages he believes his cable designs offer. He is aware of the interference issues, but believes the arrangement gains more than it loses... My view is different and others can decide for themselves.

1) What I was wondering is if whether the difference in construction between Vovox and other cables would result in a different/perhaps higher signal level being outputted from the Vovox cable (for a given equal input), and thus sound louder (and possibly better?). Has anyone measured how much signal is lost from traversing a Vovox cable vs. other?

---

There is no significant loss and cables can't make an audio signal louder. The frequency response plot above reveals clearly that when a 0dBu signal is fed in to one end, it comes out the other end at the same level for all audio frequencies, and is no different in loss from the cheap cable I used for comparison on the second channel of the AP test set.

Quote:

---

2) Hugh, you mentioned you'd never been able to reliable hear a difference on your own system, just other systems.

---

Yes, in organised demonstrations, both private and public.

Quote:

---

I wonder if the comprehensive testing you did (and showed above) would have had different (and less equal) results if they were done on a system that the Vovox cable sounded "different" on?

---

I inserted the cable into a signal path around an AP test set. That was then 'the system' and revealed no significant measurable changes. I suppose it would be possible to test a demonstration system, but that's not practical here and now.

Quote:

---

3) ...could the Figure 2 cable signals be mathematically "differenced", the resulting plot used to generate a "differenced" signal from some sort of generator, which could then be "added" to the non-Vovox cable ... all to see if the resulting sound made the normal cable sound more "Vovox-like"?

---

Yes in theory, but very difficult to do in practice because the EM interference varies with proximity to the cable amongst other things and you'd have to match the phasing precisely, which would be difficult.. It would be easier to run the cable through some form of external mumetal shielding.. but that would be impractical and hugely expensive... and it could probably be argued that the proximity of other metals might equally affect the sound...

hugh

--------------------
Technical Editor, Sound On Sound

Thank you, Hugh! I'm very grateful for your taking time to help understand this. As an ex-engineer (though not in the sound/music field), I'm so intrigued why some property of the Vovox would make it sound "different", but isn't able to be quantified/graphed, etc. Sounds like the answer will have to wait for another day!

Thanks again -

--------------------
Alexis -Cubase 6.5.0/SX3.1.1.944, XP SP2, 4GB RAM (1GB not accessible, but used just to balance the computer so it doesn't tip over); Delta 66 in Omni i/O Studio; Motif8; UAD-1