Wednesday, 2 November 2016

MEASUREMENTS: Crystal Cable CrystalConnect Micro 1m Interconnects... Differences vs. Generics???

Alright guys... As you know I've measured cables in the past - why audiophile magazines never seem to get around to this is anyone's guess :-). And as I summarized here, there really hasn't been evidence of a meaningful difference between various cables of all sorts.

Over the years, I have tried to measure cables which are considered "audiophile quality" when I can get my hands on them. You know, things like the Synergistic "Tesla" power cables, or these Kimber 8TC speaker cables. Note that obviously if I really wanted to, it's not like there aren't a selection of audiophile stores locally in Vancouver. Without evidence in my own subjective experience or reasonable theory that major differences are even possible, I'm obviously not going to go out of my way to test these things when there are so many other topics to learn about and explore! When I had the opportunity from a friend to hang on to these cables for awhile and test them out, well, why not? Especially since he's had these for years and they're well "broken in"! :-)

Behold, the Crystal Cable Micro; we're often reminded in reviews that Crystal Cable is based in The Netherlands (parent company International Audio Holdings which also owns Siltech):
Notice the small round "silver billets", all part of the typical Crystal Cable esthetic. New models have a more ovoid shape.

No, it's not the most expensive cable out there by a wide margin. But it was positively reviewed in 6Moons in 2005 with an asking price of around US$600; supposedly "very open, dynamic sound with exceptional resolution, transparency and speed... largely neutral and uncolored but fill of life... Zing...". More recently (why don't they date this?!), a newer version called the Micro Diamond was reviewed in TONEAudio and was said to impart "a spectacular level of inner detail... without compromising musicality in any way". Well then, looks like we're in for a treat!

Now, before I go further, I of course need to tell you about the special construction. As described in the 6Moons review, this cable's conductor is made with an annealed gold-injected silver. It's then "covered by a triple-layer of helically wrapped ultra-thin Kapton film before a 99.99% pure silver braid shield is applied". Nice and hi-tech. Then for the outer insulator, we have a "Teflon jacket". Though not shown here, they did come with a nice presentation case according to my friend (he apparently packed it up somewhere but you can see a picture of it on that 6Moons review).

In terms of look and feel, yeah, they feel like "premium" products. Surprisingly heavy given the gauge of the wires (silver and gold after all). The connectors are heavy and solid feel. Well machined. The cable is stiffer than the inexpensive RCA's I'll be using for comparison. The connection is tight but not too tight to worry about ripping out RCA sockets. The silver color below the clear outer insulation is attractive and adds to the uniqueness of the look.

I. Setup

Here's the measurement chain:
Battery powered Raspberry Pi 3 --> 6' shielded USB --> TEAC UD-501 DAC --> RCA cable tested --> Focusrite Forte --> Shielded USB --> Win10 computer for analysis 
I had my Raspberry Pi 3 connected to my Logitech Media Server using piCorePlayer software for streaming the test signals over wired ethernet in the basement. Notice that the measurement chain is all battery powered from the Pi 3 to the Windows 10 computer (as per the battery powered tests previously). To minimize possible electrical noise, test was run around 11:30PM when kids asleep, no heavy electrical appliances running, and the power grid hopefully quieter as well.

Here are the 3 cables I'll be using today:

There's of course the 1m (3.28') silver-colored Crystal Cable Micro. In the middle is a "generic" but good quality Radio Shack RCA cable I've had for maybe a decade. Then you see the 6' truly unbranded generic but decent gold-plated RCA cable. I can't remember where this 6' cable came from; I vaguely recall acquiring this before 2000 when I bought a Harman Kardon HDCD CD player. Note that the Crystal Cable consists of 2 individual lengths whereas the Radio Shack and 6' generic are of zip cord construction.

I. 16/44 Resolution

As much as CD resolution may be passé these days, it is important to measure at this setting I think for completeness. Unless your listening habits are very different from mine, easily >75% of what I listen to is in regular 16-bits 44.1kHz. Furthermore cable reviewers often claim they can detect differences even with a CD-quality source, so let's have a look.

Here's the RightMark summary:

Mmmmm hmmmm. Clearly we don't have any hints here that the Crystal Cable Micro differentiates itself from the 3' and 6' inexpensive RCA cables. How about the individual graphs?

Nothing to see here folks. At "standard" resolution, the cables are indistinguishable even objectively. I trust this is absolutely no surprise.

II. 24/96 Resolution

Okay then, time for high-resolution. Although we might wonder if there could be differences with ultrasonic frequencies, I suspect most of us would recognize that it's the 24-bits of resolution that might reveal something regarding noise floor and sensitivity to interference.

As usual, here are the summary numerical results to start with:

As with the 16/44 results, we're really not seeing much here. However notice that the Crystal Cable Micro has a better result for stereo crosstalk of up to a whole 1dB difference compared to the 6' generic RCA.

Let's now look at the graphs themselves... First the frequency response:
Frequency Response
Nothing. Nor would I expect any difference of course! Cables unless of anomalous or inadequate construction perhaps purposely created to have coloration should not mess around with frequency response. So often we hear of cables having "better" bass or "better" high frequency extension, yet never objective evidence for such a claim.

How about the noise floor?
Noise Level
Interesting. I'm enlarging the graph a bit compared to the usual and I've highlighted a few regions of interest. This is looking quite good for the Crystal Cable. It actually is less susceptible to noise compared to the other cables. Less 60Hz mains hum pickup. Likewise less noise around 180Hz and ~300Hz. You see that the 3' Radio Shack is intermediate between the Crystal Cable and the generic 6'. They're all about equivalent with the noise peak around 120Hz.

Suffice it to say that we are looking at differences down below -120dB!

As commented above, the stereo crosstalk measurement showed the most difference in the numerical data. Again, this is looking pretty good for the Crystal Cable up to about 2kHz:
Stereo Crosstalk
Again, remember the practical aspect of this, on average, we're looking at crosstalk of <-100dB for all cables!

Finally just a look at the IMD+N sweep graph across frequencies:
IMD+N Sweep
No difference worth commenting about. These are simple, passive lengths of metal wires after all, carrying line-level voltage with low current; nothing here that's going to influence specific frequencies with obvious distortion.

Note that although I only showed one single measurement for each cable, the measurements were done 3 times with each cable and the differences I pointed out are actually consistent across the runs; not just inter-measurement variation.

III. Conclusions

Okay. There you go folks. Measurement of a cable that costs quite a bit of change (price all relative of course); 600 bucks when new about a decade ago for 1m (3.3'). Apparently there's been some inflation and they are closer to US$750 these days at The Cable Company. Well... Silver and gold have gone up in price.

As I noted above, they're fancy cables and one could buy them for the non-utilitarian reasons one buys "luxury" items. If you're wondering how they "sound"; they sound great. But a generic set of 6' RCA sounds great with great music too! Implicit in that statement is of course my belief that subjective "greatness" lies more in the artistry and quality of recording compared to a passive length of metal. It also goes without saying that there's no way I would be able to differentiate these cables in a blind test.

I trust it's interesting to see that there are objective differences I can measure with this cable compared to the inexpensive zip cord RCA's. It speaks to the sensitivity of measurements even without using expensive 'pro' measurement gear. The Crystal Cable Micro does seem to pick up less hum and other noise, and the stereo crosstalk measurement is better. Nice to see that indeed it does measure better; kudos to Crystal Cables!

Remember that we must put this all into context of course... We are looking at differences well below -100dB, with noise differences affecting a few frequencies only. Consider this into the context of the cost and how you define value. I think it's fair to say that no human being would likely be able to "hear" the difference when playing music in any event (when blinded to control expectation bias of course, not the kind of "test" that typical subjectivist reviewers "do").

For completeness, remember in 2013, I made some measurements of various RCA interconnects as well. Back then I noted that stereo crosstalk can be easily demonstrated to be different which is what we see here again. This time around, with the completely battery powered setup using the laptop and Focusrite Forte, I can show a difference at the 60Hz AC hum frequency (among a couple other frequencies as well); cool.


Greetings from overseas. Happy November everyone! Obviously I got these measurements done before leaving Canada... What else is more productive than to edit a blog post during a transpacific flight? :-)

Album of the week for me is the octogenarian troubadour Leonard Cohen's latest You Want It Darker (DR13, 2016). As expected, this is presented in his deep baritone speaking/singing voice with variable accompaniment from orchestral, choral to simple acoustic instrumentation. Also as expected, lots of allusions to religious themes, emotional ups and downs, relational conflicts, sexuality/sensuality, and of course mortality. Lots of content to dig into as usual.

Going to be busy at work for the next while in any event. Going "dark" probably for the next couple weeks :-).

Hope you're all enjoying the music!


  1. For interlinks the screening properties are the most important properties IMO.
    Also IRL the biggest actual differences will be measured here as well, aside from capacitance that is.
    Above 1MHz there will be more measurable differences, but the last time I checked I couldn't hear that high up nor could my transducers reproduce it.

    Have measured and opened up a few cheapies in the past and some weren't even shielded. These are thus quite susceptible for picking up mains nasties or radio frequencies. This can't be seen from the outside b.t.w., those cables looked 'decent'.
    Stay away from nicely braided non-shielded 'super cables' is my advise for that reason alone (not being shielded).

    Using silver isn't going to increase the higher frequencies (people claim they sound brighter !).
    Yes, silver does conduct better than copper or gold but it conducts the same for all frequencies, it can't favor any frequency bands (in and well outside the audible band).

    I have some info on cables, albeit mostly about headphone cables.
    It does have some calculations concerning different conductor materials though.
    Just remember that the currents and voltages in interlink cables are much lower.
    Also source resistances and load resistances in line-level interlinks are also much higher as well so Ohmic differences in cables are completely irrelevant here.
    The screen/shield should be as low Ohmic as possible).

    It's The Netherlands by the way (I should know, I live there) not the Neatherlands.

    1. Oops. Thanks. Corrected :-)

    2. I must have been thinking about some video game :-)

    3. Your friend paid ~$600 for these some years ago, so I have to ask the obvious question: what persuaded him that they were worth that much, and does he still feel the same? Have the 2 of you ever got together for a listening test of cables (or anything else)?

    4. Hi Otto.

      He was single and disposable income before the wife and kids :-).

      No, he doesn't feel the need to pay this amount anymore. We haven't actually gotten together to listen to these vs. others... I've of course listened myself (and agreed with the measurements that there's nothing I can consistently discern that "sounds" different). I don't think my friend cares about the "sound" of these anymore. He has since sold off more expensive cables he had and decided to keep these as he got a deal and still likes the high quality of the connectors and thin unobtrusive esthetics.

    5. All as expected. Difference about error of measurement.

  2. Wasn't that expected?


    If you use a decent RCA cable, that have decent good RCA connectors for the inner contact and the outer ground, does have decent shielding (coverage) that the regular audio measurements will be all the same?

    With my years of experience and different Audio Precision measurement sets over the years, I wasn't yet able to measure and difference of different RCA cables with regular audio measurements.

    Sure I can measure the resistance, the capacitance, the inductance, the wave impedance and also with a good capacitance measurement device also the loss of the insulation, but if you have a regular conditions of output and input impedance and only typical lengths for home use, I have no difference in the audio.

    But, I still hear differences between those cables, hear differences in the conductor material, in the insulation material, how the cable is constructed / build, but can't measure those differences.

    PS: My comment here belongs to the above mentioned category of RCA cables.

    But anyway. Enjoy the music.


    1. Thanks for the note Juergen... If you have any theories about how the sound can be audibly different despite sensitive testing, would appreciate the insights!

    2. Malcolm Hawksford wrote an very interesting paper back in the 90, where he takes some math with the help of the Maxwell theory that confirms my "hearing experience" in those years. You can read about diameter and materials (even the skin effect is in the mega herz region)

      And even for the risk, to get mobbed after this, right now, for already three weeks, I test and compare different internal wiring on one and the same product to tune the sound to my taste. I can't measure this, but I do rely on some sonic signature of some cables that I use.

    3. You have excellent recording equipment.
      When interlink cables make an audible difference in a way you can even tune the sound, then it stands to reason you should be able to record those differences.

      Can you make some (not too long) recordings of different cables where the only difference is the cable on a high bitrate/depth and share this ?
      There must be excerpts where it is audible and tell us what and where to listen to.

      Skin effect is not in the MHz range. It already affects 20kHz.
      The effect is an increased resistance, very small percentage wise.
      For longer speaker cables and low impedances this can cause measurable effects.
      For interlinks even a 1000% increase in resistance is not audible simply because the wire resistance (max a few Ohm) has no influence on the relatively high loads of input circuits.
      Even with input resistances as low as 1kOhm we are talking 0.05dB at 1000% increase where in reality it will be a few % at most.

      Skin effect cannot be the culprit in interlinks.

    4. A good idea to try to record the audible difference. I have tried this about 10 years ago, but didn't succeed. You are right, nowadays, I have better recording equipment and should try it again. I will keep this in mind and will try to do this when I have time for it. But right now I have to prepare to go to Asia for some event, but when I am back in two weeks, I will try to do this.

      BTW: Have you read the Malcolm Hawksford article. It is dealing also with interlinks and different diameter of the single strands. Sure with speaker cable you can measure the inductance effect within the audio range and you can also measure the vibration distortion within the speaker cable because of magnetically forces between the strands.

    5. Hawksford wrote a ton of articles.
      Haven't read them though.

      Lots of people have written articles on this subject. Not surprisingly the outcome seems to differ depending on the goal/nature of the research.

      Speaker cable and interlinks are entirely different cables indeed, where voltages and currents can differ a factor 1,000 to 50,000 easily.
      For power cables (speaker cable = power cable) resistance, inductance, diameter, length, strands/type, skineffect, isolation and capacitance do matter.
      For interlinks these parameters are not anywhere near being a limiting factor.

      Let us know the results of these experiments should you have had the time for this.
      I know how much work is involved setting up a (near) perfect measurement/comparison.
      It's an intriguing, divisive yet interesting topic though.

    6. If you type in Malcolm Hawksford and Maxwell you will find the right articles. I met Malcolm at several AES conventions and spoke with him about some of his very clever ideas. Realy clever.

    7. Read that... It's 20 year old research showing an effect previously (before 1995) not considered to be of importance.
      It is about an aspect in wire thickness (and capacitance) in longer, and thicker, speaker cables.

      It has no relevance to interconnects though.

      Mr. Hawksford, as you say, is a quite knowledgable character in the field of analog/digital and sound propagation.

  3. The discrepancy between heard/perceived differences and measurements is always interesting to me.
    Especially when tested blind.
    It would suggest a difference in output signal from different cables MUST be there.

    Here is the intersting part in all this.
    As there is no storage of signals installed it is quite easy to electrically compare the input and outout signal.
    This can be done with an opamp. Easy to build and works well within and well outside the audible range.
    Now ... audiophiles may say ... well opamps sound crap and cannot reproduce the finest stuff they like... Let's say I disagree here.

    I built such a (rather simple) device many.. many years ago as I was curious.
    It basically 'nulls' the input and output signal.
    No digital or any sampling is involved here.
    The pure analog waveform is electrically compared/nulled.

    Fun part is... you can use actual loads (speaker/headphone) and real music.
    No test tones needed.

    The interesting things that can be found is that with amplifiers (aside from gain differences and 180 degrees phase shifts) measurable differences can easily be found.
    Unfortunately this method has a drawback which is it is hard to compensate for time delay between in and out and the circuit cannot discriminate between phase shift differences and amplitude differences.
    The output signal simply shows THE difference, regardless of the cause.

    In speaker amps the differences are bigger than in pre-amps (for obvious reasons).

    The same circuit can also be used to tell the difference between the input and output signal of cables provided the 'input' signal has the same delay (read added cable length) as the measured cable.
    You thus need 2 cables which are compared.
    One is 'loaded' with a speaker or whatever load and the other one... well it doesn't have a substantial load (say 10kOhm).

    The difference signal in speaker cables seems to come down to Ohmic losses only and certainly at longer distances the FR changes.
    How slight the change is depends on the cable and connections as well as the speaker impedance variations.
    Yes, cables measure different but so slightly it isn't audible to me.
    To small differences when decent gauge cables are used.

    Of course I measured a few 'interlinks' as well. Admittedly NOT the expensive exotic ones, didn't have access to those.
    But compared silver and copper cables (from cheap to somewhat expensive) and in these cables there simply is NO difference within the audible band.

    The output signal of a silver cable compared to a copper cable of the same length made absolutely no difference.
    Comparing 2 copper cables also no difference.

    Not saying people hearing differences between cables are mad.
    Just that I cannot hear it (maybe my hearing or equipment is not good enough) and have not found a way to measure it.
    As far as I can test the electrical input and output signal/waveform (real music audio signal) for 'normal length' RCA cable is exactly the same (thus no change in phase, delay or amplitude) within the audible band.

    Can't think of a 'better' test.
    If there actually was a way to measure it I think cable manufacturers would gladly include test results to make their point.

  4. "If there actually was a way to measure it I think cable manufacturers would gladly include test results to make their point." - Indeed!!

  5. If there isn't a way to measure the differences among cables, then how do the cable companies engineer them? How do they test their products to determine the differences among their low end, mid level and high end offerings? How do they know when they have struck audio gold if they cannot do it quantitatively?

    1. That is a golden question.
      One you should ask the manufacturers. At the same time ask them if they could disclose the test results and procedures that make them value their product(s) that high.

      As for justification...
      Expect the answer to contain a lot of assumptions and BS about what 'they' discovered and how they solved that in their unique way which, of course, will not be disclosed and is always better than that of their competitors.
      YEARS of research/experience is also something they will gladly shove forward to justify exhuberant pricing.

      Of course, the usage of rare and or expensive materials and complex production methods will also affect the price as will the looks and feel.

      The concept of being able to buy more audio nirvana when you spend more is always on the audiophile's mind.
      It is a powerful suggestion not many can resist.

      In the end though there is only one justification for the price of the cable which is how many cash the consumer is willing/able to spend.
      And some do spend a LOT... so there is definitely a market.
      It is that market that determines the price.

    2. Even if we accept that there are audible differences, what are we to make of the extreme high-end stuff like Nordost? What development, material and manufacturing costs can possibly justify prices such as
      Speaker cable £26,400 pair
      Interconnect £19,800 pair
      Digital £9,700
      Tonearm £9,600
      Power £15,000
      (Odin 2 Supreme Reference, all 1 or 1.25 metre. Current exchange rates £1.00 = €1.12 = $1.25 = $1.68 CAD)
      And why do pro audio engineers seem content to use standard, well engineered cables?

    3. in about a dozen studio/control rooms I worked in, including wiring a few myself, Belden cable was the standard

      Aside from mic bays/cables, cables from console to multi-track and back x 24 channels or more, etc., even studio consoles themselves have hundreds of cables and connections

      Think of the number/length of balanced cables used when mixing a tune where all 24 or 48 tracks are used, with the tape machines located in another room, plus patching effects in the patch bay linked above. By the time the music is played on one's system, typically, it has passed through a great length of audio cable already.

  6. I would suggest a bigger sample size if you can arrange it. Maybe some entry level audiophile interconnects- e.g. Kimber, Wireworld, Audioquest, etc.

    1. I do wonder what would be the point in using more expensive cables for the null test ?
      When cheap cables produce no 'difference' then why would an extremely expensive cable would produce a difference ?
      If it would, it would actually change the signal.

      At least that's what I think you mean with bigger sample size for the test.

      Why/how would a piece of 'precious metal with fancy isolators' in SERIES with many meters of cheap thin copper on a fiberglass PCB (PCB traces) make any difference ?

    2. Hi Anthony,
      I agree with Solderdude...

      I suppose we could do a measurement shootout with all kinds of cables at different price levels. But if even generics already perform excellently to conduct the electrons with minimal interference that is unlikely to be audible, the only good this would do would be to identify some cables that might perform *worse* (ie. purposely or not cause frequency roll-off, more susceptible to interference, higher crosstalk, etc).

      At least the Crystal Cable didn't make anything worse compared to the common RCA cables I used :-)!

    3. Adding a larger sample size (including some popular contemporary brands) would squelch questions about the quality of the brand and sample that was tested.

  7. Lets not forget that inside the sound components like amplifiers and speakers etc is also, in the vast majority of cases, good old standard copper wire and in some cases ribbon cable. So surely if you believe in the idea that cables do make a difference then that quality is all destroyed by the internal cabling. Let alone all the miles of cable and thousands of electronic components that the music signal has passed through from studio to mixing etc to get to the finished product.

    1. Yup :-).

      Somehow this point never makes it into the general mindset / meme of the audiophile press...

  8. I'm minded of the times I used to visit China regularly on business, oh, around 10-12 years ago. Those that have been to China know that tea is held in very high esteem, perhaps even more than wine is among wine connoisseurs in the West.

    The price of tea ranges from the ordinary, to the eye-wateringly extraordinary. What caught my attention was the price of teapots. In a tea shop, and this was most tea shops, it was not unusual to find teapots that cost thousands of dollars. The expensive ones all held special properties, and of course the tea connoisseur simply would not drink tea out of an ordinary vessel. I asked a couple of tea shop owners what it was about tea pots that made them so coveted, and I swear the conversation could be replayed to discuss audiophile cables. There is nothing rational or logical about the price of teapots that I could work out, but that was not the point. People were passionate about them, and invested lots of time and money in finding the right one, and, I think, with the right teapot, they could enjoy their tea even more.

    For what its worth, I passed on the teapots, but I bought a lovely set of mid-price chopsticks (or finger sticks they called them) for around $100 for 6 pairs. And yes they have shops that only sell chopsticks.

    I guess one mans teapot is another man's phono plug.

    1. Nice Bob :-).

      Speaking of tea, cups and pots, I came home from Asia this weekend with a bunch... They like giving them as gifts overseas.

      Who knows if pots and cups make a difference to the taste, but whereas pots and cups may result in changes with temperature, clear tactile changes when put to the lips, I suspect there's more potential effect to the experience than a piece of metal used for electrical flow with little *physical* evidence of any difference at all.

      In any event, it's luxurious and maybe that in itself is powerful enough for some to "experience" a difference in a sighted evaluation. If that is "valuable" enough to spend a few hundred or thousand dollars on... Then so be it...

  9. My brother-in-law runs his Avantgarde horns with lamp cord.

    1. NO! SAY IT AIN'T SO!

      That guy should be banned from the Avantgarde brotherhood... :-)

  10. Well people are clearly of the hook , engineeing for a 1m long cable to transport such a low bandwith signal as audio is trivial :) A younger version of mysylf bougth in to all this BS and i had all kinds of expensive audioquest cables VdH etc. I did not even sell them used I though that was imoral once i discovered what it really was . So one set of caldera speakers cables and a whole bunch of other audiophile cables went to the recycling station... :) Thanks for the blog as usual

    1. Thanks for the note Mnyb. Your honesty has clearly secured you a place in Valhalla, Elysium, Heaven, and various other places reserved for the moral and valiant :-).


  11. Having tried a variety of relatively inexpensive value models of interconnects I have to report that I do notice subtle differences in sound quality among them as well as compared to basic cheapies.
    However I would like to expand on the idea questioning how the manufacturers actually engineer these products for sound, especially since the measurements of sound reproduction attributes show minute differences that should not be audible.
    How do they differentiate between the models in their range and know (ostensibly) how to appropriately price them based on sound quality improvements as you move up the line?
    Could it be a process of changing the scale of measurements such they zoom in and compare them at a micro level?
    Do they correlate easily measurable electrical properties to sound quality changes based on some experimentation and "golden ear" objective interpretation?
    Or- do they just make them look fancier and more exotic to justify the price- knowing that sound quality changes are difficult/ impossible to engineer consistently?
    If you truly get what you pay for in the land of interconnects, then they must be properly engineered- designed to meet quantifiable sound reproduction attributes and tested / measured to validate that the design goals were met- and to demonstrate that these attributes improve the quality of sound as you pay more for the models in their range. From the outside looking in it seems like a tall order.

  12. The manufacturers explanations/justifications to your questions can, undoubtably, be found on their websites. You can choose to trust them or just lift your eyebrows.

    When the differences between cables are worth their asking price to their buyers then who cares if others can not hear, or measure, any (worthwhile) differences ?

    Be happy such audio jewelry is available and is appreciated by many, just not by everyone.

    Others (including me)may well be content with cheap cables that seem to be technically proficiant enough to them and don't appear to degrade the sound quality in any way.