Sharpen the pitchforks: S/PDIF cable heresy from Mad Scientist Audio


ESS PEE DIFF. S/PDIF. Sony Philips Digital Interface Format. As a derivative of the pro audio world’s AES/EBU standard and designed to transmit digital audio signals over short distances, it comes in two flavours: coaxial and optical.

Not without its own shortcomings, coaxial connections are, by a large, the audiophile’s choice when connecting a USB-less streamer or CD spinner to a DAC. Toshiba’s fibre-optic-based TOSLINK is all too often looked down upon by audiophiles despite its ability to completely isolate one’s DAC from electrical noise spilling from any upstream device.

A 75 Ohm characteristic impedance guides the design and build of modern coaxial S/PDIF cables where the centre pin usually carries the voltage and the outer ring the ‘return’. It’s helpful to see digital cables as carriers not of ones and zeroes but of analogue representations of those ones and zeroes.

According to engineers like Chris Sommovigo, a cable with a characteristic impedance of 75 Ohms optimises the connection between transmitter and receiver to give us the best sounding connection.

(The background story of how Sommovigo and his ILLUMINATI brand arrived at this number, one that would, according to Stereomatic, eventually be adopted as an industry standard. Read their take here.)

The fly in this 75 Ohm ointment are the RCA plugs demanded by the majority of hardware devices. Their characteristic impedance is typically 50 Ohms. You might own a fancy-schmancy S/PDIF cable whose conductor fully complies to the 75 Ohm standard but is terminated with RCA plugs that do not.

Of course, as is common in the world of audiophiles, and especially when it comes to cables, whether or not this 25 Ohm differential audibly impacts sound quality is a matter of (sometimes white hot) debate.

One (theoretical) workaround is to terminate one’s S/PDIF cable with 75 Ohm-compliant BNC connectors as per Sommovigo’s own Black Cat and Silverstar-branded S/PDIF cables of the late noughties and early 2010s.


But what if we thumb our nose at the 75 Ohm specification altogether? That’s the question being asked by Bob Prangnell of New Zealand’s Mad Scientist Audio.

He writes: “I am sending you one of my Heretical digital cables – the world’s first and only 37 Ohm cable”.

Except it isn’t necessarily. Prangnell continues, “The 37 Ohm I mention is not the characteristic impedance of the cable, it’s the actual resistance”. Apples and oranges then.

Here we infer Bob’s point: that the characteristic impedance could be inconsequential if the carbon fibre core of the Heretical cable “absorbs the mass of reflections that bounce up and down the cable, causing jitter” as this Mad Scientist claims.

A supporting document supplied with the cable brings this Kiwi’s train of thought back into focus: “The Heretical Digital Cable (HDC) is very different to most digital cables. It makes no attempt to have a 75 Ohm characteristic impedance.”

The carbon fiber conductor means this cable’s resting state tends towards straight. Coil it too tightly risks damage. Standard RCA terminations are brass with a rhodium plate. Keith Eichmann’s (KLE) Innovations plugs are available as an order-time option.

“However, the carbon fiber construction means that it is immune from [sic] skin effects; even up to 250 MHz (a frequency reduces [sic] silver and copper skin depth to a few micrometers.) And because of its inherent resistance, HDC absorbs any reflections caused by impedance mismatches,” continues Prangnell.

Maybe. Maybe not. I’ll leave that to readers of a more engineering persuasion to politely discuss in the comments section below. My limited engineering nouse means I’ll be noping out on the tech talk in favour of moderation.

Then comes the kicker. “The result is a digital cable that makes your DAC sound twice the price”, concludes Pragnell’s marketing spiel. Yeah naaaah.

Wild claims like this are par for the course for marketing bods but reality at DAR HQ says otherwise. Swapping out an AudioQuest Forest S/PDIF coaxial cable (US$25) between AURALiC Aries and Devialet Expert 200 didn’t bring silence or a dose of the (literal) jitters. Perhaps a slightly smoother presentation from the HDC with lower levels of cymbal splash – but I wouldn’t stake my lunch money on it.

The differences proved to be super subtle too with the multibit version of Schiit’s Bifrost. A multibit Gungnir it does not make. Completing the picture was Vinnie Rossi’s LIO and Spatial Audio M4 loudspeakers.

Wanna try one for yourself? US$99 gets you through the checkout.

Lastly – Prangnell’s letter explicitly permits re-gifting of the accompanying HDC sample. If any reader 1) has the gear required to properly measure this cable’s characteristic impedance and 2) an inclination to report on their findings, speak up and I’ll wang it your way.

Further information: Mad Scientist Audio



Written by John H. Darko

John lives in the NOW + HERE = NOWHERE. He derives an income from the ad revenues of DAR. John is also an occasional staff writer for Stereophile, 6moons and TONEAudio.

Twitter: DarkoAudio
Instagram: DarkoAudio
Facebook: DAR


Leave a Reply
  1. Ordered. The only S/PDIF cable that provides a noticeable difference for me is the Kimber Kable D-60 (Ag), between a Bryston BDP-1 and a Rega DAC. We’ll see how this cable stands-up.

    Nice write-up!

    • Let us know how you go in listening, Ken. I think it’s kinda smooth sounding but far from the revelation of the Curious in USB land.

      The offer still stands of a freebie to anyone who has the gear to *measure* it.

      • Well, I’ve been spending time with the Mad Scientist coax cable. I am impressed. It *is* smooth sounding and beats my AQ Carbon S/PDIF coax, within subtle measures, of course. I would recommend it – especially with the Eichmann plugs I got with it. The Kimber D-60 is still my favorite leash, to use your phrase, but I wonder if I could really differentiate the two in a blind test.

        And, you are correct: not the jaw-dropping experience that is the Curious USB (I have this, as well, with the second Regen link). But then again, the price of the Curious was equally jaw-dropping. Given the price difference between the Mad Scientist cable and the Kimber D-60 Ag, I think it’s a clear winner.

        • Thanks for the extra info, Ken. Your comments are of genuine value to other readers. A doff of the cap to thee.

  2. He isn’t the first to embrace carbon fibre as an audio signal connection. Steve McCormack has used carbon fibre for a while, but only in specific circuit junctions of his electronics due to the material’s high resistance. Serguei Timachev of Stealth Audio Cables has carbon cables and so do van Den Hul. There are probably more but those are at the tip of my tongue. Your man however could very well be the first to do an affordable S/PDIF cable with it. It’ll be interesting to hear whether any of your readers post something on what it sounds like -:)

    • Hey Srajan – I don’t think Prangnell’s claiming first on carbon fibre at all, only that it’s the first 37 Ohm digital S/PDIF cable (but it isn’t really that either). The man’s angle is not to fuss over 75 Ohms.

  3. The Mad Scientist Audio site says the HDC is $99 US for a 1 meter version (with the standard RCA connectors). Did your sample have the standard RCAs?

  4. The calim you quoted (‘because of its inherent resistance, HDC absorbs ANY reflections caused by impedance mismatches’) is a marketing BS.

    It cannot absorb all reflections within the cable, as otherwise the primary signal would not pass. It attenuates the primary signal to a degree – but probably just not enough to loose the connection. If the attenuation is say 3dB (just an example) then with each pass of the reflected signal, the reflection will be attenuated by another 3dB until it dies.

    Increasing the resistance of the cable creates the effect similar to increasing return loss at both transmitting and receiving end, which may help some, although the very same effect may be achived by proper design of the the both sides (receiving and transmitting) of SPDIF interface, which is much more elegant – and most of all – reliable solution from the engineering point of view. Why most manufacturers do not care to implement high return loss SPDIF inputs (and outs) and such a cables may make a positive difference – is a topic for a separate discussion.

    I believe van den Hul was first to introduce a carbon fibre cable (metal free) some 20 years ago. Since then other manufacturers followed. Their first cable – The First – even though it was designed as analog IC, was used by many as a digital IC with some success.

    But you do not need to buy a CF cable to experiment with attenuation of the digital signal within the cable (and its reflections).

    You can buy simple 75 Ohm fixed attenuator. They are widely available in various dB variants:

    I remember John Kenny (the manufacturer of JK SPDIF converters and DACs) was offering those as an accessory in the past.

  5. Hi
    Have to Agree with Adam, this makes no sense from an engineering point of view

    Coaxial transmission engineering is a mature engineering area, and high speed data transmission is a large industry….its not a matter of “voodoo” or ‘weird science’… and would agree getting a digital eye report would tell us a lot more, impedance in and of itself is only a part answer

    The company is from your article deliberately obfuscating an important electrical characteristic differences resistance versus impedance “apples and apple pie”

    There seems to be no clear point of reference in your post about what the actual impedance mismatch is in this cable? Surely if the maker makes a claim of this nature they know what the difference is?

    It is relatively easy once you know the actual measured parameters of the digital cable and the impedance mismatch to simulate what the effect on the waveform will be

    H(f)=Vout/ Vin = (1-r(f))ej(ja(f)+b(f))l (2)

    r (f) is the voltage reflection coefficient.
    1-r (f) is t (f), the voltage transmission coefficient.

    If all components in a channel are close to 75 ohm impedance, t (f) is close to 1 and has little effect on the transfer function of the channel.

    Capacitance is a big killer of digital wave forms, the lower the better

    Impedance and/or reactance is the sum total of all these characteristics in a cable.

    Reactance: varies with frequency and we are dealing with high frequencies > 6Mhz ( > CD)

    Therefore resistance of 37 ohms DC ( ac freq = 0) is not relevant as a single parameter

    Resistance on its own, is unhelpful in predicting the propagation characteristics of a high bandwidth digital transmission

    If the manufacturer would like to publish such characteristics, it is possible to see the effects in modelling

    However as a general rule…impedance mismatching is not desirable in high speed digital transmission, and from Baynesian inference: unlikely i.e. probably = 0 of being better than matched cable, but could quite clearly be “different” though if one is desiring engineering optimisation of digital transmission not desirable….A