• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Belden ICONOCLAST XLR Cable Review

Rate this cable

  • 1. Poor (headless panther)

    Votes: 152 53.9%
  • 2. Not terrible (postman panther)

    Votes: 86 30.5%
  • 3. Fine (happy panther)

    Votes: 21 7.4%
  • 4. Great (golfing panther)

    Votes: 23 8.2%

  • Total voters


Staff Member
CFO (Chief Fun Officer)
Feb 13, 2016
Seattle Area
This is a review and detailed measurements of the ICONOCLAST XLR by Belden (manufactured by Blue Jeans Cable). Tested sample costs US $785 for 6 foot length:
Iconoclast XLR TPC cable interconnect review.jpg

I ordered a pair of XLR and RCA cables from the company. They nicely intercepted my order and asked if intended to review them. When I said yes, they refunded my money and sent me samples of both. You have a choice of cable (copper purity) and I was supplied with the TPC version per below:

Iconoclast XLR TPC cable interconnect product page.png

Company was very confident of the performance of the cable and suggested that I test the entry level material. I quoted the above text because this cable uses solid conductors that are very inflexible. It especially resists rotation which sadly, is something you need to do with XLR cables to align the socket/plug. Combined with their weight and large reduce, using them with desktop products is very difficult. Even with large and heave audio gear, the lack of ability to rotate them easily is a problem. Per above, they have other solutions such as right angle and more flexible versions.

Other than that, the finish on the cables is pretty and seems indestructible. Connectors were Neutrik and mated nicely with everything I plugged them into.

Each cable seems to be hand tested/measured which gives the feeling of care in manufacturing and production:
Iconoclast XLR TPC cable interconnect specs review.jpg

The cable is brainchild of the long timer Belden cable engineer, Galen Gareis. In sharp contrast to other premium cable manufactures, Galen has produced a number of super detailed papers on the design philosophy and many factors going into the production of these cables. It also debunks some of the other theories out there such as transmission lines and cryogenic treatment. In a nutshell, Galen's theory on why cables make a difference is that different frequencies in audio have different propagation delay/speed in the cable and he aims to reduces this differential as to have all the frequencies arrive closer in time.

I needed a reference comparison XLR cable so I selected the World's Best Cable (WBC) 3 foot XLR cable with Canare L-4E6S Star-quad balanced cable which I purchased for US $26 from Amazon. It too uses Neutrik connectors although they are a different model and feel cheaper. The Canare cable is ultra flexible and joy to use as I constantly plug and unplug it in my testing.

Iconoclast XLR TPC Cable Measurements
For these tests I simply used the Audio Precision APx555 as source and destination for audio. Some of the tests use its ultra-low distortion signal generator while others (like Multitone) use its internal DAC. I set the output voltage to 4 volts per nominal standard in DACs for XLR connection.

Let's start with our usual dashboard and reference WBC cable:
Worlds Best Cable Canare L-4E6S Measurements Loopback.png

There is essentially no distortion in the output of the analog generator. So the only thing we have is noise which is -122 dB down, setting SINAD to the same level. With best case hearing threshold of -115 dB, the WBC cable combined with the APx555 electronics is delivering an absolutely silent signal and distortion that is way below what we can perceive.

Switching to Iconoclast XLR TPC cable we get the same:
Iconoclast XLR TPC Measurements Loopback.png

Any tiny difference you see is run to run variation of no consequence.

Above is at one frequency (1 kHz), so let's try a range of frequencies in audible band:
Worlds Best Cable Canare L-4E6S Measurements THD vs frequency.png

Ignore the response below 30 Hz as that was variable (with both cables). Despite bandwidth now occupying 90 kHz (4.5 times the audible band), we still have identical response at or below threshold of hearing.

Instead of testing one tone at a time, let's throw 32 tones at the cable simultaneously. In "time domain," i.e. what the cable sees, this is the waveform:
Worlds Best Cable Canare L-4E6S Measurements Multitone Time Domain.png

So definitely not a clean sine wave or anything. Above is impossible to interpret so we convert it to a spectrum display using Discrete Fourier Transform:
Worlds Best Cable Canare L-4E6S Measurements Multitone.png

If the theory of differing propagation velocity causing distortion, we should see something here but we don't.

I also ran a phase difference at 1 kHz but did not see anything between the two cables:
Iconoclast XLR TPC Measurements phase.png

With all of this out of the way, the last that could cause audibility difference would be frequency response so let's test that:

Worlds Best Cable Canare L-4E6S Measurements frequency response XLR cable.png

Despite zooming way in with just ± 1 dB vertical axis, we see no difference all the way up to 90 kHz.

At this point, I could have measured a number of audio devices but I think we all can predict that they would show same results so I decided to tackle something that has been asked for a long time, null testing.

XLR Cable Null Testing
Null testing seems like an obvious strategy to sort out differences in cable. But it is not. Analog nulling will always generate a lot of noise which confuses people as to what is being seen/heard (and fiddling with gain knobs). Digital testing reduces the impact of that but creates problems of its own: synchronization. Clocks in DACs drift in time in as few as a few seconds. So even if you capture two audio pieces precisely and match them to a sample, they still drive at sub-sample level! And this is quite audible. A perfect null still shows quite audible version of the music being used even if nothing has changed.

The only solution we have here is to analyze each sample and using signal processing (cross correlation) detect actual timing and essentially resample the tracks to match each other. This process is never perfect either but can work. Our member, @pkane, has written an excellent tool called detalwave which performs this function.

To make sure the baseline process was correct, I would make two recording back and back between a DAC and ADC and compare using deltawave. While you could see that there was good matching, it was far from perfect. The closer you could get the tracks together to start with as far as time and cadence, the better the matching. I managed to develop a good process here using RME ADI-2 Pro FS as both the DAC and ADC (and hence clocks were synchronized). I then used Adobe Audition Multi-track editor to play the source track and then record it back in real line (with the XLR cable between input and outputs of the RME ADI-2 Pro).

For test track, I resorted to one of the best high-res recordings out there, produced by my friend "Professor" Keith Johnson, the first track from album, Exotic Dances from the Opera:

I used track one: Rimsky-Korsakov: The Snow Maiden (Snegurochka), suite for orchestra: Dance of the Tumblers. The album is encoded at 176 kHz sampling and 24 bits. I shortened it to about 11 seconds as to reduce the computational time and get around copyright concerns.

Once I had my workflow setup, I made two separate captures using the same WBC cable. I captured the results at 24 bits/44.1 KHz. I originally used 176 kHz but there is a ton of noise there that was interfering with proper testing (but still worked). Here is how that "null" looks:
Worlds Best Cable Canare L-4E6S Measurements Loopback.png

This is a superb null indicating that down to threshold of hearing (-115 dB), we don't have any difference. Seeing how we didn't change anything in the two runs, this is the best our analysis allows.

Now let's swap the cables for Iconoclast TPC XLR:
Iconoclast XLR TPC cable vs Worlds Best Cable Canare L-4E6S Null Test.png

So despite changing the XLR cable and capturing analog output of the DAC, there is no difference in our null. The cable is as innocent in this pipeline as it can be.

If you want to do your own testing/listening/comparing, here are the three files:

WBC cable: https://www.dropbox.com/s/2qaikozigrtdpma/Multitrack Capture_mixdown_Capture 1 WBC.wav?dl=0
Iconoclast cable: https://www.dropbox.com/s/k2z3rizs8wchqo4/Multitrack Capture_mixdown_Iconoclast TPC XLR.wav?dl=0
Null/difference file: https://www.dropbox.com/s/fp2otdat2ln5w01/WBC vs Iconoclast XLR TPC.wav?dl=0

I played the Null file by itself at max volume of the RME ADI-2 Pro headphone output and using my Dan Clark Audio Stealth headphone, I only heard silence.

Iconoclast CLR Cable Listening Tests
I used two setups for listening tests: headphone and main 2-channel system:

Headphone Listening: source was a computer as the streamer using Roon player to RME ADI-2 Pro ($2K) acting as a DAC & headphone amplifier, driving my Dan Clark Stealth headphone ($4K). I started listening with Iconoclast cable. Everything sounded the same as I was used to. I then switched to WBC cable. Immediately I "heard" more air, more detail and better fidelity. This faded in a few seconds though and the sound was just as it was with the Iconoclast.

For my main system, I used a Topping D90SE driving the Topping LA90 which in turn drove my Revel Salon 2 speakers. I picked tracks with superb spatial qualities to judge the usual "soundstage." I again started with Iconoclast XLR TPC cable. I was once again blown away how good my system sounds. :) I don't get to enjoy it often enough given how much time I spend working at my desk. Anyway, after a while I switched to WBC cable. Once again, immediate reaction was that the sound was more open, bass was a bit more tight, etc. This too passed after a few seconds and everything sounded the same again.

It is clear that our classic audio measurements don't show any defect in budget XLR cables so there is nothing there to fix. Naturally, the Iconoclast cable produced the same results as my budget WBC cable.

To get around the argument that "you can't measure everything," I set up an analog loop and captured what comes out of the XLR cable. This resulted in incredibly deep null that is inaudible by itself let alone in presence of music. And same result is achieved in two runs in sequence with the same cable with null extending deep to threshold of hearing.

My casual subjective listening tests didn't produce any reliable difference either. Yes, I could be accused of bias in not wanting to hear something (even though I did initially). If bias is an acceptable issue, then folks should never advertise other sighted tests as proof of such cables sounding different! At any rate, I have provided the two audio samples so you can do your own AB tests (blind or otherwise).

So on three fronts we see no data that backs Iconoclast cables sounding different. How could that be after tons of measurements and formulas in the company white paper? Well, the heart of it is this statement at the end of one:

Iconoclast cable sound different.png

The fact that cables sound difference was taken as a given. Theory was developed around Vp (propagation velocity) and lots of engineering went into optimizing this factor. I take company's word that they have achieved this (unlike many companies that theorize but never show a solution to said problems). But if there was no audible difference to begin with, then all this work unfortunately is for naught. A formal, double blind study to establish audibility would not have taken much time to put together prior to engaging on this effort.

I encourage the company to perform such a blind test to demonstrate that there is a real problem here. The "we already know is a problem" is not a sound engineering premise. We have tests for such things and the protocol cannot be short circuited lest you want wasted results.

One could argue that if premium cable works great and looks nice, no harm is done. Iconoclast XLR cable indeed works well as an XLR cable. The problem is that in this implementation using solid cabling, it is awful to use. It is so difficult to rotate to plug cables in and out of devices and can easily yank them backward. It can also cause damage to the sockets if you don't have mechanical support for the cable (and room for such).

In summary, while I very much appreciate Iconoclast's work and volunteering to have their cable for testing, I cannot recommend it, or verify any benefit for it. I am afraid Galen's engineering knowledge and experience is wasted in audio domain.

Email response from chief designer, Galen to my review:


We got it exactly half right. We have a little distortion added in.

ICONOCLAST is designed to get ALL the typical variables optimized that we can MEASURE and CALCULATE, and not just Vp. We do hear TIME based changes to the EM wave. Pure resistive amplitude is a passive distortion. A cable that is resistive only would be ideal. ICONOCLAST is optimized to better is better electrical. To say ICONOCLAST is “only” about Vp propagation times (still very fast in any cable as it is the speed of light in the dielectric) is not totally right. We improve Vp linearity and ALL the related variables, too. And yes, those include the foundation of R, L and C common to all cables. ICONOCLAST can’t remove physics any more than create new physics. It can show what we don’t know, though.

When you improve a variable in audio cable, it effects related variables as well. When Vp coherence is improved (and Vp coherence DOES change the cable’s properties or physics is wrong) it also impacts the OPPOSITE frequency range by lowering the open-short impedance. BOTH are tied together and BOTH need to be improved. Higher capacitance and individual wire loop DCR impacts the low frequency impedance. And yes, this changes how the amp/speaker and cable interact as we have a different reactive network or again, phsics as we know it is wrong and it isn’t. Unlike RF, where stuff is steady state, analog is an awkward frequency range in constant transition at every frequency point. The impacts of this are going to be different based on the total reactive network…it has to be. A cable’s impedance to a speaker at a frequency are NOT matched! We have simple reflections (ZOBEL networks use this property).

The application of more individually insulated and small wires splits the current into smaller signal values per wire, and this smaller wire improves skin effect, lowers current removes to reduce the proximity effects (proportional to current). Managed as a network more small wires can INCREASE the CMA area and lower DCR. IC and speaker cable use different characteristics of the technology. IC cables have no proximity effect to even consider into a high impedance load for example.

Inductance is wrapped around all of the capacitive effects. One BONDED speaker cable pair measures 0.126uH/foot inductance nominal. To lower that inductance value you need to reduce the loop area with DISTANCE and EM field CANCELLATION. Distance is already at a minimum with BONDED pairs thus we introduce EM field cancellation into the design’s. The speaker cable weave pattern used in ICONOCLAST speaker cable does exactly that. Cross weave and SEPARATE polarity paths reduce inductance to 0.08 uH/foot. It works as the physics says it should.

The speaker cable weave also limits the capacitance as the dielectric (inductance isn’t sensitive to the dielectric properties) and physical and periodic separation paths LOWER total capacitance as the average distance is increased. Every wire path is the exact same physical length, so the cable thinks it is “one” wire. Again, the physics says it will work, and it does. We ideally want to hold L and C to reasonably low values and JUST use DCR to optimize the cable if we can, that’s the end goal in a perfect world. In practice we allow higher capacitance in a speaker cable (-3 dB roll off in is the GHz, but amplifier reactive loading is a concern) to lower inductance for current delivery. In IC we like to see low capacitance as it is a voltage signal with low conductor loop DCR.

The IC cable, RCA and XLR use like physics to alter L and C. More smaller wires in a star quad reduces Inductance, but it ALSO has to raise capacitance. It does, from 12.5 pF/foot to 17.5 pF/foot nominal. This is expected as the physics says it has to be. We also increase the CMA area for longer runs. Both the RCA and XLR measure the same swept open-short impedancd by design. The RCA’s double braid improves RCA cable issue of DCR being added between devices and this can aggravate ground loops as the ground isn’t as uniform as it should be. Again, standard physics.

Belden’s task was to IMPROVE every aspect of an analog cable and we did that. We have never put to book a “sound”. A loud speaker’s specs have ZERO real meaning until you listen to ALL of the parameters at once in your room. A cable by itself isn’t ever used, but with an amplifier and speaker. Physics DEMANDS that the introduction of a reactive AC network is derived from the system’s total load to the amplifier. That you, and us too, fail to make this distinction we KNOW is true, is just the short comings of what and how we measure this dynamic interaction. We KNOW for a fact that this reactive interaction is different when we change any of the three variables, amp, cable and speakers. This fact alone suggest we can only calculate and measure certain things. I can relate as I’m restriced to this issue too, but it DOES NOT provide the final answer as to how this complex network is changing things.

Your test illustrate the limits of attribute testing. It can’t show differences in the reactive network when physics says it is definitely there. Saying I can’t hear that isn’t the same as then trying to “prove” the cable electrical don’t impact the RIGHT tests. R, L and C changes do and will alter the analog signal and they will and they have to or again, the physics is broken.

ICONOCLAST will sell what I can calculate and measure, same as we’ve provided since 2015. Our job is to provide properly made cable with KNOWNS adjusted to better suit analog. We sell the entire range of electrical cable. ALL designs will be measured and shown to work as the physics says they should. We have no magic that needs to be accepted. The same properties that make your and our “generic” cable are still at play but to a higher degree in ICONOCLAST to reach better electrical. That’s what the market wants to try and that’s what we make. The effort to make better cable is no more wasted than it has been to provide the products we buy today and improved over the last 100 years. PRICE, not performance, is the barrier to entry and with proper pricing volume there is ZERO reason to not use better R, L and C cable. None. Why would you? Analog is an addative distortion and every step matters.

Your simple testing, and mine, is what blinds us to the changes that physics is providing in each design and yet, we still can’t test them. Welcome to the club. Show me the measurements the do capture the physics in play and we’re good. This isn’t saying, “I can’t hear that”. Your tested data, although accurate to the tests resolution, is incomplete “proving” there is no difference when we know for a fact there is. A simple device is limited by what we know today and doesn’t change what’s left we can’t test. It is a tool to stay on track, it doesn’t answer all of the networks actual properties.

Sitting behind a knowledge limited test fixture won’t change things any more than making zip cord speaker cable forever. The limits need to be pushed in testing and design forcing us to ask, “as different as this really is, why can’t we test the tertiary elements that HAVE TO BE showing up in the “tested” data?”. When an analog design changes, the output has to change or the physics has stopped and it didn’t. We stopped. We make properly made cable to push those testing limits.

Galen Gareis

Any my response to him:
My response to Galen:


Hello Galen. Thanks again for the response. I included your original write-up below in the review.

As to points made within, some of your comments are about speaker cables which I did not test (I only tested XLR interconnect).

You start with the premise that: “We do hear TIME based changes to the EM wave.”

We don’t have agreement on this. Sit in a live music presentation and you hear not only the direct sound, but reflected sound from the room. What you hear then is a phase/timing “soup.” Our hearing fortunately is designed to not care or we would go crazy, listening to others in our homes with all of those reflections/timing distortions. Indeed, the brain is good at filtering that. As you listen to your loved one in your home, they do not sound different as you or they move, yet timing distortion is introduced at higher frequencies like nobody’s business.

Please see these AES papers:
“Measuring Audible Effects of Time Delays in Listening Rooms,” Clark, David, AES Convention: 74 (October 1983)
On the Audibility of Midrange Phase Distortion in Audio Systems, STANLEY P. LIPSHITZ, MARK POCOCK, AND JOHN VANDERKOOY
University of Waterloo, Waterloo, Ontario, Canada N2L 3Gl

My testing also was NOT white box as you mention below. Instead, I treat the cable as a black box and see what effect it has with respect to noise, phase, distortion and frequency response on an audio signal. I found no difference here.

Importantly, I also performed a null test. I captured the signal from both your XLR cable and a much cheaper one. The results nulled to threshold of hearing (-115 dBFS) which indicates no difference in sound with very high confidence. I further post the differential audio file which is silent. If there were changes to the waveform, this test would have detected it. But it did not.

Ultimately we, the audiophiles, care about the sound, not the cable parameters as we don’t listen to cables. In order to prove there is a difference, you either need to conduct a blind/controlled test with statistical significance, or null test above, or both. Without this, there simply is no way forward. The whole premise of needing a new cable relies on this basis being proven. If you have such data, I will happily eat my words and run with yours.

Best regards,

As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited:
I've no idea how to rate something that works perfectly well but can be replaced with something else much cheaper that also works perfectly well.

Need the piggy bank panther.

Upon reflection:
It does its job, so not terrible. It costs a bunch, that isn't fine or great. So... not terrible.
Last edited:
Results as expected. It's a good thing that company didn't charged you for this cable.

I'd say that paying more for a manufacturer tested cable is a good thing and it would be nice to see more manufacturers doing this, but still the final price for such a fancy cable is way higher than I would expect. :(
I wonder if Blue Jeans will still offer this sort of thing at the price it is @.
Wow. This almost makes me disappointed in Belden, and I just bought 100' of their cable for wiring my Subs! :rolleyes: :p

No one is immune. Hell, if I knew how easy it was, I shoulda started a cable company! :oops::D

Thank you, @amirm !!! I vote piggy bank Panther.
Top Bottom