Is Synergy In Audio Cables Real or Snake Oil?

Check any audiophile facebook group or forum and you will find dozens of enthusiasts asking the oh so common question "what are the best cables to use with my audio components?" There are never shortages of recommendations from fellow enthusiasts and sometimes even industry experts have their own recommendations too.  But is there really a "synergy" between audio cables and the components they connect between? Can this be proven with science or is it just hearsay? There are two camps with opposing views. Surely someone has to be right?

When answering this question, I can't help but reference Hitchen's razor which states as follows:

"What can be asserted without evidence can also be dismissed without evidence."

This takes ALL of the air out of the proverbial lungs of the audiophile cable manufacturer since they almost NEVER provide any objective peer reviewed proofs to their claims hence why Audioholics has been on a 20+ year quest to debunk the nonsense in audio with cables being one of our biggest exposes.

Synergy in Audio Cables YouTube Discussion

I could go all math and measurements on you to defend the objective viewpoints that "cable synergy" is complete nonsense. But, that approach often falls on deaf ears (no pun intended) by the audiophile and/or cable soothsayers that claim "measurements don't matter" or it flies over the heads of those not well versed in science or basic EE theory. Although this is something I still plan on doing in future editorials, I will instead use irrefutable logical in hopes it will definitively prove why "cable synergy" is a fallacy and actually a detriment to the hard work engineers have done designing your favorite audiophile electronics.

The Transfer Function by Matthew Poes

The transfer function of a piece of electronics fully encompasses everything coming out of that piece of electronics.  Nothing can hide.  If you look at the difference In the transfer function between a piece of equipment using two different cables and there is no difference in the transfer function, then nothing changed.  If nothing changed, there cannot be an audible difference. 

In other words, for a cable to make a difference, it must either change the signal or fail to change the signal (as compared to some base reference).  If it changes the signal, it’s flawed. If it fails to change the signal, then a difference assumes the base reference cable used in testing is itself flawed. If there is no difference, there is no difference.  Our ears are not needed to know if that is true.

While we know that cables have enough differences in their electrical properties to make very small differences in the transfer function (which we can argue about audibility all day), I have never seen evidence that audiophile cable companies are somehow providing superior measuring cables as compared to the best commercial stock used in professional settings (i.e. Belden and Mogami).

For "Cable Synergy" At Least One of Two Things Must Be True:

1. Electronics have faulty interfaces requiring a special cable to "optimize" performance.

2. Cables are faulty designs only "optimized" with certain gear.

One would hope an audio manufacturer knows how to properly design the input and output stages of their amplifiers to accommodate any 3rd party audio equipment. The only exception that comes to mind is some AV receivers not being able to drive 4-ohm difficult speaker loads, due to insufficient current drive. However, even in those outlier cases of inadequate current drive amplifier stages, no special cable is going to fix that problem. If you want to drive a difficult speaker load, you will instead need an amplifier with higher current capability. All of this is well known. No surprises.

Cable science is well understood in every discipline of engineering, including audio. Audio frequencies (20Hz to 20kHz) are in fact the easiest to design cables to since we are dealing with a very limited bandwidth where cable characteristic impedance is not an issue and long lengths of cables work just fine if resistance is kept low for speaker cables and proper shielding and low capacitance is followed for long run analog interconnects. For digital audio interfaces a cable with the proper characteristic impedance (110 ohms for AES/EBU and 75 ohms for S/PDIF) is vital for long runs to prevent impedance mismatches and associated reflections which can cause errors and loss of data that the interface error correction may not be able to deal with.  These are examples of industry standards for audio/digital interfaces and corresponding interconnects. Beyond adhering to these standards, "cable synergy" is not needed, nor desired for accurate signal transfer.

Mogami 110ohm AES (left); RG59 75ohm COAX (center); Belden 1694A 75ohm COAX (right)

An ideal analog audio cable should transmit the audio signal unadulterated.  Nobody will argue the ideal cable is NO cable at all or a superconductor with no loss tangents. "Cable synergy" would imply the cable itself is NOT ideal. It's NOT transparent, it's not lossless Instead, it's acting as a tone control to alter the signal between connecting components. Why would any audiophile want this? Most audiophiles cringe at using tone controls on their preamps so why would they use them in the form of connective cabling? The best and most effective frequency shaping in a system is done in the digital domain with room correction below the room transaction frequency (300-500Hz) in concert with multi-sub and positional EQ and with passive room treatments above the room transition frequency to control excessive reflections.  With audio interconnects and speaker cables, we are at best looking at 10ths of a dB change out to the frequency extremes between good and super-conductor good cabling which is below the threshold of audibility for anyone but Kryptonian hearing.

What is the magic in audio cables?

In 20+ years of running Audioholics, I still haven't found the proverbial pot of gold or magic dust that ONLY audio cables exhibit according to many esoteric cable manufacturers. What I always find interesting is NOBODY ever questions the usage of standard 10AWG speaker cable as amplifier test leads or shielded RG59 COAX for line level connections on the very best, most accurate audio test equipment such as the Audio Precision APx525 and 585. This test equipment is sensitive and accurate enough to measure down to -130dB or so which far exceeds the distortion threshold of even the best, most expensive audio gear on the market. Yet, we are only using inexpensive double braided, double foil 75-ohm coax to make such delicate and precise measurements. In fact, referencing Henry Ott "Noise Reduction Techniques in Electronic Systems 2nd edition states coax cable effective usage out to 1GHz, that's 50,000 times the bandwidth of human audibility! So why do we need fancy expensive cables with exotic dielectrics and battery biases ONLY in consumer audio and NO other field of discipline?

Useful Frequency Range for Various Cables per Noise Reduction Techniques in Electrical Systems, Henry Ott

Editorial Note About Audio Precision Test Cables:
Audio Precision stresses the importance of using quality cables to ensure noise free measurements. However, they define a "high quality" cable as being properly constructed per common engineering standard (ie. shielded coax with proper characteristic impedance). They mention that one of their cables is based on Mogami cable stock. There is NOTHING magical about Mogami cables. They are just well engineered cables often used in recording studios.

See: Audio Precision Cable Recommendations

PS Audio Discovers Cable Synergy with Audioquest?

I recently received a newsletter announcement from PS Audio about their partnership with Audioquest and how owner Paul McGowan teamed up with his son to "hand-curate" a select series of AQ cables that sounded best with each PS Audio product.

One of the most common questions we get at PS Audio is, what cables should I buy? For years we've recommended one of our favorites, AudioQuest. Inevitably, the next question was, well, which AudioQuest cable? And we didn't have a good answer. So, I teamed up with Darren Myers and my dad, Paul, and we spent the last six months hand-curating a select series of AQ cables that sound best with each PS Audio component. The differences we heard were remarkable. And now, we can confidently tell you which cable is the perfect match.

We were so excited about the pairings that we decided to partner with AudioQuest and offer this curated collection directly for sale from PS Audio. You can read more about our collaboration with AudioQuest, and watch a little video we made, here www.psaudio.com/connections

If you'd like to know which cables are the perfect match to your PS gear, let me know. I'd love to help.

All my best,

Scott

Intrigued, I emailed PS Audio to see if they could answer just how they were able to empirically "hand-curate" the proper AQ cable for each of their products.

Q&A On Cable Synergy with PS Audio

Audioholics: Greetings to you and your dad. I'm curious in your selection process of determining the "best sounding" Audioquest cable for each of your products. Did it involve any empirical measurements or was this solely based on listening tests? Were these listening tests conducted blind or sighted? Why wouldn't the very "best" Audioquest cable be suitable for ALL of your products?

PS Audio: Thanks for writing! Yes, all of our tests were done with one of the most important factor we can muster, which is listening tests, specifically on our Infinity IRS-V system. And great question about why wouldn’t the best cable be best for all products! Yes, a $1,030 Diamond USB cable would be amazing for the $699 Sprout100. It really would (and IS, as I have listened to). It’s profound in fact. However, the curations we offered are both for performance and value, meaning we found the best performing cables with the highest value, and if a cable passed both the listening and the value test, it was eligible for the list of curation.
Does that help? Thanks sir!
Best,
Scott

Audioholics: I appreciate your reply. So just to be clear, you base your AQ cable recommendations solely on subjective listening tests ( NOT controlled DBT) and neither Audioquest NOR PS Audio does any empirical measurements when determining which cables are "best" for each of the components in your product line?  I can't find any published test data on the Audioquest website for ANY of their products. Please advise.

PS Audio BHK 300 Monoblock Amplifiers (MSRP: $14,998)

PS Audio: Thanks for writing, and while I know AudioQuest does much high-level testing of their gear (and I cannot offer specifics there, with apologies) yes it is the case that our most profound and valuable test is our ears. Here are a few words from Paul regarding synergy, curation, and value. I hope it helps!

PS Audio’s value proposition, which remains unchanged, is based on a pretty strict ratio between parts/labor costs and retail pricing. Cables are very different and admittedly somewhat of a mystery to me. I sincerely doubt they use the same pricing formulas we do, but I am not privy to their models.

That said, we have always wanted to make available synergistic cables with our products. As Scott's note suggests, we're constantly asked what cables work best with our products (there are a combo, after all). We all hate saying "we've tried this or that and worked well, sorry if that's not something you can participate in" or send you on your way to ferret it out for yourself. Hate it. It's what happened with MG Audio cables which I so loved (and still do). People either couldn't find them or afford them. They made no sense for our more affordable products—so I had to shrug my shoulders when asked what to pair products with. That just doesn't sit well.

We can't determine AQ pricing but, what we can and did do, is match a sensible price relationship of PS gear to cable. In other words, given a product's retail price, what would we recommend spending on cables from *any* manufacturer? It makes little sense to connect a $10K power cable to a $2K Stellar phono stage (though it'd probably sound great) :)

So, given that we love AQ cables and how some match our gear in exquisite ways—and they were willing to work with us and permit us to retail their products—it seemed a perfect match. Take our recommended combos and you'll find them to make economic sense and wonderful synergy.

That is a PS combo that not only makes sense but one I can fully support.

Trusting Our Ears

This was a very cordial exchange I had with PS Audio. In fact, watching a few videos by Paul McGowan left me with the impression that he's a very likeable guy. He clearly loves this business and he's doing some great work over at PS Audio. Someone with that much experience and staying power in this industry probably  has accumulated some interesting stories worth hearing over a few beers.

I can't shake off the fact that Paul's and Scott's answers are a bit unsettling. How can a company with such a vast pool of engineering resources, the same resources paramount in making SOTA electronics, simply rely on "testing with your ears" approach for finding the "best", most "synergistic" cables to use with their products? This is NOT a sound way of approaching optimal design in anything related to audio engineering. It has no scientific rigor, no repeatability in a controlled Double Blind Test (which is designed to remove human biases), and quite frankly puts the fate of the performance of their products in the ears of a couple of middle aged men likely with impaired listening acuity. Age is a killer of listening fidelity in ALL of us and a topic reserved for another editorial.

Editorial Note: Hearing Loss vs Age

As we age, our ability to hear high frequencies diminishes even with normal usage. A rule of thumb formula for BW vs Age = 20,925 - (Agex166). An average male age 50 would under ideal circumstances only be able to hear 12.6KHz. I'm 46 years old and can no longer hear frequencies above 14kHz right in line with this equation.

The Science of Sound & Controlled Listening Tests

Dr. Floyd Toole and Dr. Sean Olive lead pioneering research at Harman regarding the science of predicting subjective listening test preferences based on how loudspeakers measured. In their research, they found the best way to control biases was to conduct Double Blind Testing as sighted tests tended to bias the outcomes based on appearances, expectations, or brand recognition by even the most disciplined listeners. I've always argued that the rigor of a true DBT is only needed when the sonic differences between two or more stimulus are small, however, Floyd and Sean have proven otherwise in their peer reviewed AES contributions and I acquiesce as my "feelings" don't meet the rigor of their peer reviewed research. In fact, because our auditory memory is so short, just 200-300ms, and because many commonly used DBT methods, like ABX, do not prime the subject to the specific differences, their sensitivity to change are low among sensory testing methods. Considering that the sonic differences between cables is at best case minuscule and at worst case, nonexistent, the need for a controlled DBT to discern differences is indisputable. A controlled DBT is the only way to eliminate the human biases and to better discern small differences in stimuli without introducing the placebo effect where a listener is expecting to hear a difference. These methods must be as rigorous and controlled as possible, eliminating any source of extraneous variance.  Fast A/B comparisons are also necessary to remove auditory memory limitations. But, most cable comparisons are almost never conducted in this manner. Instead, they are often compared sighted infused with hyperbole of what you "should hear" before the comparison is actually conducted. This is called preconditioning the listener and it's a successful sales tactic when executed by a skillful practitioner.  Further, they often argue that long listening tests are needed to fully appreciate the differences.  In fact, what this does is confuse the listener with too much novel information.  Auditory memory is too short to be able to sort these differences in a reliable manner.

One cannot declare "synergy" simply by trusting their ears as the primary test metric especially if the listening test is NOT done with any scientific rigor to eliminate biases. With that said, how can an audiophile sleep at night knowing that a company achieved the "best synergy" with their products by testing only one brand of cable? What if there was a better brand offering "better synergy"?

The Simplest Answer

Any company with a heritage of building SOTA components such as PS Audio should never sell themselves short by declaring their products require "synergy" derived from some magical properties in cables that can't be measured or quantified. In my opinion, this does a disservice to ALL of the engineers who worked hard on their products to build the most robust interfaces, and highest resolution systems to really reach the limits of audibility for high fidelity.

Following Occum's Razor, the simplest answer is usually the right one, did PS Audio truly find the "best synergy" of connecting cables for their components using biased uncontrolled listening tests as their primary metric, or did they simply find a brand they like working with which offers expensive, high margin, quality audio jewelry that audiophiles assign intrinsic value to?  Do specific Audioquest cables really offer the best "synergy" for PS Audio components or did PS Audio realize they were leaving money on the table in the shopping cart for online sales by not offering recommended cables to accompany their products? You decide.

Rebuttal from PS Audio

Ahhh, Gene, thanks. You’ve gotten to the crux of it all. How can an engineering based company like PS Audio, whose third highest expense after parts and health insurance is engineering, can, in our approach to audio design, be so seemingly cavalier with measurements? And to make matters worse, we’re a company that’s invested in some of the best audio measurement equipment in the world—and we use it all—rely upon it and couldn’t continue without it.

The answer is simple. It’s a partnership.

To untangle this I’ll use as an example our recent design work on the Stellar Phono stage. In its first iteration, engineer Darren Myers spent months in simulation designing the lowest noise, highest linearity and overload-capable circuit he could muster. On the computer it looked terrific. We committed to the design, built the prototypes, spent weeks getting the physical circuit performance to match the simulation’s predictions. None in engineering had seen a phono preamplifier perform as well on the test bench.

Because extraordinary design, like exceptional cooking, is a partnership between measurement and tasting, off to the listening room it went—where it promptly failed: flat, lifeless, stuck in the loudspeakers (if you know what measurements correlate with this observation I am all ears (not to make a pun)).

Back to the drawing board where we suggested he had too much loop feedback and in some areas, too little bias. Changes made, measurements now slightly worse, it was back to the listening room. Voila! The sound opened up and detached itself nicely from the speakers. (We still maintain in our prototype archive both circuits and they are easy for anyone to revisit if you’re ever in town).

I won’t bore you with a blow by blow playbook as there were many more trips back and forth between the two.

The same is true for cables, though we no longer design them. I gave up their measurements years ago.

And, for the record, we only do blind listening tests on cables. We have one person who switches cables and listeners who judge them. Only the switching person knows what’s what. Double blind as you know it doesn’t work, a subject I’d be happy to elaborate on at some other time. But, blind measurements are a must.

As to your kind offer of public debate, I think I’ll have to pass but thank you for the opportunity. It’d be akin to a debate between political right and left wingers: both so entrenched in what they know to be true that nothing of value would result. Life’s too short for that.

I will suggest that we would dearly love to better understand how we can use the measurement tools we have to get to the bottom of why using transfer functions and differential measurements don’t help us relate what we hear to what we measure. We know those differences exist. We know measurements don’t currently show us what those differences are. We also know how much easier our design process would be should we figure out what is happening when we make those changes. In the meantime, we’re continuing to listen (taste) the end results of our design efforts.

Be well.

Paul McGowan