CAF 2015: What Happened Here?

Herb Reichert and I, just before everything went sideways

CAF2015With this year’s Capital Audiofest, there’s good news, and there’s bad news.

The good news? This show was … very possibly … the best-sounding show I’ve been to. And I’ve been to a few. Explaining the “why” of that is going to be completely beyond me. I could venture that there was something in the venue, or the layout of the vendors, or the sheer quality of the displays, but who really cares? The average attendee won’t. What that “average attendee” heard, though, was simply astonishing. The proportion of rooms that sounded not good but great was higher than at any other in recent memory. Ska-doosh.

The bad news? You missed it. It’s over. Now, there’s nothing to do but wait till next year. The Hilton where the event was held, is getting a facelift in the next year, so it’s unclear if CAF will be back and even if it is, whether or not it’ll be “the same”. Here’s to hoping. Ticket sales seemed brisk on Saturday, but Sunday was the usual audio-show ghost town; a shame, as that’s always the best day for sonics.


Meet the Writers

I got to “do” two panel discussions/events, courtesy of my host, the inimitable Ken Furst. The first was a meet-the-writers panel, hosted by Art Dudley, the Editor of Stereophile magazine. I know Art a little, ever since I maligned his taste in beef by viciously accusing him of routinely transmogrifying his steaks into shoe leather, but sitting next to him and attempting to speak extemporaneously with any level of cogency has got to be one of the most nerve-wracking things I’ve done in audio. Did I mention he’s my hero? He is, and he scares me. Not a little, either. It’ll come as no surprise to you, those that are fans of his writing, that he’s wickedly funny and has absolutely no patience for fools. What you might not know is that he’s quite charming and mild in-person — you have the opportunity to underestimate his perceptiveness when you see him, smiling agreeably. That would be a mistake. The man is a lethal cocktail of coiled menace and devastating judgment. That night, at dinner, when his steak was served not only well-done but cold, I heard a ululating scream of outrage that I’m positive emanated from the very bowels of the earth. I’m pretty sure the waiter fainted on the spot. So did 12 bystanders. There are conflicting reports of glassware shattering, but Robin Wyatt, our gracious host for the evening, was actually speechless. Mat Weisfeld, President of VPI and a master in the dark arts of hand-to-hand combat (and also one of the many guests), was so startled by that, he promptly kung-fu’d everyone in the room into submission. All in all, it was terrifying moment and many shorts needed to be changed. We were kicked out asked to leave shortly thereafter.

Luckily, the panel went smoother than the unfortunate outburst later that evening, and Art was able to guide Alan Taffel of my erstwhile home, The Absolute Sound, “cub reporter” (snort) Herb Reichert of Stereophile, and myself through a great Q&A. There were questions of ethics, of product selection, of editorial process, and still room for some great storytelling.

Speaking of the “cub reporter” (Herb’s words), I have to offer my thanks to whomever it was that had the great good sense to bring Herb into the fold over at Stereophile. That magazine is immensely better with his contributions — I’m just going to say that out loud. Loud and proud. Herb’s a deeply kind and funny man, and given his long history on “the other side” of the editorial aisle (he sold — and survived selling — Audio Note equipment for years and years), his perspective is fresh, engaging, and insightful. His beat, taking over “The Entry Level” from departed Stephen Mejias, has remained one of the many highlights of that magazine.

I was going to insert some more praise here, but my laptop just made retching sounds at me. Fine. You call it “sucking up”, I call it Truth-Telling. Whatever.

Stupid laptop.

Meet Sandy Gross

Ha! That’s me sitting with Sandy Freakin’ Gross! Photo courtesy of Ken Furst.

My second “Major Event” at CAF this year was a Between-Two-Ferns interview with GoldenEar‘s Sandy Gross. For those of you still living under rocks, and lacking the benefits of The Google, Sandy co-founded Polk Audio and Definitive Technologies, and selling both of those companies, before launching his latest venture.

We spent some time talking a little about the in-between/all-around stuff, too — though I didn’t really get a chance to talk about Sandy’s love for fast cars and Native American art, we did explore his time as a national champion slot-car racer, and his time as a Hollywood movie producer. The guy is fascinating. Anyway, we entertained his fans with some answers around company-building, speaker-designing, and a brief exploration of some of his major contributions to audio history.

What a truly remarkable fellow.

The Cable Polygraph

Me (Left) shaking hands with David Salz of Wireworld (Right). Photo courtesy of Paul Elliott

I did manage to catch another “Major Event” at CAF this year, and one I wanted to call out, especially given all the joyful noise the Internet holds around the sonic contributions of audio cabling. As you probably know, having earned your PhD in metallurgy, acoustics, physics, and engineering, there is some debate around their efficacy. Actually, there’s a whole knot of debates around this, but put that aside for now. David Salz of Wireworld was on hand to talk us through what he calls the Cable Polygraph.

In this set of “experiments”, we were told that two Tascam AD/DA converters were directly wired together with a series of 1m interconnects pulled from a broad sweep of the cable industry. These were compared with “no wire” (directly attached). But first, a test.

Ostensibly, if the Total Cable Doubter are correct, none of these will sound different. That is, files generated from one, played across the gap, and captured by the other, would be identical. Failing that, the Lesser Doubters might maintain that lampcord and [insert an expensive brand here] would be identical to each other but perhaps not identical to the “no wire”.

David recreated the experiment for us in the room. Using Bel Canto Ref500 monoblock amplifiers right up on top of the speakers (I think they were from Neat Acoustics), David was able to use a 3″ piece of 14 gauge wire as the “no wire”. It wasn’t “no wire”, but as David discussed, most cable differences (assuming such exist!) will be fantastically suppressed with distances that short, so it was “close enough”. That was the control. Next, he played the same track (an old CD rip of “Fire and Rain”) with 3m of lampcord. Then, 3m of lampcord with the conductors separated. Then, he swapped in a progression of his own, Wireworld speaker cables, starting at $500 and moving up to $1200.

Sitting 8 rows back, I’m going to say that this demo wasn’t as convincing as David would have hoped. So, being the empiricist, I stood up and walked up to the sweet spot (which happened to be the open aisle in this room) and plonked my rather prodigious butt down in what was approximately one corner of an equilateral triangle.

To my ears, the “no cable” was best. It was open, lively, engaging with plenty of detail. Note that the recording was crap — and there was no attempt to hide this fact. But when he switched to lampcord, the image shrank, and no longer wrapped me in a soundstage as comprehensively huge. Interestingly, the sound also seemed to be quieter. I want to say it was 2-3dB quieter — and no, he didn’t fiddle with the volume control (it had a digital display, and it was constant), just the backward-key on the CD player in order to restart the track. When David switched to the lampcord with the conductors separate, the sound was even quieter — and now, it was “fuzzy” and diffuse. Still, everything was totally recognizable — the difference wasn’t enormous, per se, but it was obvious. When David swapped in his own cables at this point — again, the only change being speaker cables — the focus snapped back into place, the noise floor dropped and the volume came back up. And that was the “entry-level” cable. By the third Wireworld cable, the sound was very close to the control, which he then played last.

Back to the Polygraph. Apparently, this test with the Tascams, wired back-to-back using high-resolution audio files, was done in Germany; the results were published online — in German (here’s the link to the translated version). David played a progression of these over the system he’d just done his own tests with, and while subtle, the “sounds” of the interconnects from Audioquest, Wireworld and Cardas were all readily apparent.

I would love to check out this experiment — not that I read German — but more importantly, I’d love to try this out myself. Another day, another project.

The show coverage will roll out slowly over the next couple of weeks. There’s a lot of great stuff to show you, so, do stay tuned.

About Scot Hull 1062 Articles
Scot started all this back in 2009. He is currently the Publisher here at PTA, the Publisher at The Occasional Magazine, and the Executive Producer at The Occasional Podcast. There are way too many words about him over on the Contributors page.


  1. …Everyone stumbling on your website should also read this:

    By the way, it’s obvious that stereoplay’s results were tainted by the cable appearance, reputation and the listener state of mind or bias since the listening wasn’t done blind. In short, a wonderful piece of pseudo-scientific crap.

    • I almost hate to say it, but:

      Saying it doesn’t make it so.

      Believing it doesn’t have any impact on whether it is so.

      Believing it, because you happen to be immune to evidence, doesn’t undermine the evidence.

      Calling it “pseudoscience” really does disservice to the term. Which has a definition. And has rightful applications. But confusing “marketing” with “pseudo-science” is just as egregious as promulgating beliefs as truth.

      The fact of the matter is that there are a TREMENDOUS number of anecdotes of “failed tests”, where the testers concluded that there was no audible difference. But none of them doubt that too long of a wire, of insufficient gauge, can impact the sound. None of them doubt that wire, of peculiarly high resistance (or capacitance or inductance), can impact the sound. What they all say, however, is that wire “properly constructed” makes no difference. And when wire does make a difference, it’s not “properly constructed” — and then they promptly make changes to the parameters that define “proper construction”.

      Back in the 1970s, Polk Audio made Cobra speaker cables, which competed directly with lampcord. Even when the Cobra cables were isolated as the producer of sonic changes, they were dismissed as snake oil. “A solution to a problem no one has”. And this was at the same time that users were reporting that their amplifiers were going into oscillation when using the Cobra cables. Those users blamed the wire. The wire that could not be causing changes. Turns out it wasn’t the wire, at least, not with the oscillation — but rather that the amps had been designed to be used with wire that would massively attenuate. When that was removed, the load for the amp as designed was too much. When this was explored, amplifiers were redesigned — “properly”.

      And for the record, a degree in EE doesn’t make you a scientist (apologies to Gene).

      So, either wire matters — and design can impact sound — or physics is wrong (see the comment about wildly abusing resistance, inductance or capacitance). That’s Point One. Point Two, that small design changes can introduce small sonic changes, flows logically from Point One. Again, you don’t have to believe it for it still to be true.

      And, lastly, the article referenced shows how interested parties can do all of this themselves — should they ever want to stop bullshitting, venture past their armchairs and actually do the science.

      [mic drop]

      • Of course wire can impact the sound if their RLC properties are off the mark. Have you read Greiner’s article? For home audio purposes, with the lengths involved, a traditional lamp cord is able to do the job. The pseudoscience is to make us believe that you need to pay hundreds of dollars to improve your sound, and bring fake arguments (like skin effect, dielectric, etc.) that have no impact in audio frequencies. The burden of the proof is on the believers. The stereo play link rely almost entirely on pseudoscience, debating subjects that have been settled a long time ago.

      • “The pseudoscience is to make us believe that you need to pay hundreds of dollars to improve your sound.” <– that's marketing.

        "[Things] … like skin effect, dielectric, etc … have no impact in audio frequencies." <– that's a verifiable claim. Which means that there is the possibility that there is evidence that can be provided that would refute or falsify the claim. I'm pretty sure the link you shared around skin effect claims that the net effect (for the random cable selected) is an increase in 3% of resistance — not evenly distributed, but at the top-end of the audio band. 3%. Not a lot. But a turntable with a speed variation of 3% is pretty audibly flawed, no? .3% is too, for that matter, no? Doesn't matter. Just a case in point.

        But let me get back to your general comments. You know that stereoplay is using the same equations as Audioholics, correct? This is just math. There’s no beliefs about the results of an equation — you fill in the numbers, you get results. Whether or not those mathematical results point to an audible difference is another issue. But the math says there’s a difference. 3% at 20kHz, according to Audioholics (for the cable they were discussing). Again, can you hear that? Well, that’s a different question (and still not pseudoscience). Interestingly, the stereoplay article repeatedly says that this is something they’ve had significant trouble with, both identifying and quantifying.

        For example:

        “These four variables R, L, C and G, together, as electrical components considered a kind of lossy low-pass filter. In the beginning of the 90s stereoplay projected the concept that mapping these four parameters according to generally accepted doctrine could predict the sound of cables. However, we were unable to confirm this thesis by correlating the measurements with our listening tests.”


        “Even if the audio cable manufacturers like to describe cable differences in orders of magnitude, they’re not as critical as speakers or even acoustic modifications of the listening room.”


        “The real problem in audio cables is that from the outside, one cannot predict their fidelity. Indeed, even cables made with good materials and manufacturing processes often provide less than good results. How well the cables perform depends solely on the expertise of the developer. The fact that their function at first glance seems almost trivial and they always “somehow” make music, has seduced some inexperienced individuals to go into the audio cable business hoping for easily-earned Euros.”

        Did you miss those bits? These are not reliable indicators of absence-of-bias, but come on. These folks, a charitable interpreter might be inclined to say, don’t seem to be mucking about or actively shilling.

        An interesting aside — if cables “don’t matter”, do you believe in the sonic superiority of a given kind or brand of capacitor? Sorry — off topic.

        Having read through the entire article, all I can offer is — no one I know of has done anything like this. While I wish that the files were still available for download, and I wish someone had done some digital diffs on the resulting samples, this is still far and away the most comprehensive set of tests I’ve run across.

      • I don’t have the time to answer everything and don’t want to show disrespect, but I just wanna say that audio in general and cables in particular are a mature and well understood technical field. There never have been evidences (I don’t take the stereoplay test as evidence) that something other than RLC would affect the sound in a perceptible way (certainly not skin effect). And this is why engineers and rational audiophiles have moved on…

        (The capacitor thing is a pseudo argument by Roger Skoff. Capacitors sound different if their measured capacitance is different. There’s no enigma here).

      • As an EE, I can tell you that EE’s are not scientists. But I’m not sure what you’re saying — you’re not a rational audiophile or not an EE? Because you clearly care.

      • About the stereoplay bias… Everyone who believes that similar cables will sound different is biased in a way, as much as someone like me who believe the difference can be and are defined by typical measurements. The problem isn’t to know if someone is biased or not, he is. Perceptions are affected by appearance and reputation (on a conscious level or not), that’s a given who isn’t disputed seriously by anyone in the scientific field, hence the double blind standard.
        I am certainly not saying the stereoplay people were not in good faith.

      • So, the only evidence that you can accept — that will convince you are wrong — is one, particular and peculiar test? Only that? And how “good” does that test have to be? Will you require a large, randomized sample? And what particular statistical tool would be acceptable, assuming the pool size is greater than one?

        Seriously. That kind of requirement is not another bias? An unrealistic, foolishly narrow, bias? Especially given that DBT are designed and typically used for something other than sensory tests? And are well known to induce more artifacts than can be statistically ruled out?

        I’ve argued already that this has already been done — there was even a rather high-profile test done and passed by both Michael Fremer and John Atkinson. But we need more tests. Because … tests. And biases. And … we need methodology and tools and witnesses and pressure and screens and blood tests …. because actually doing this — borrowing a cable and trying it yourself — that is just too hard.

      • You’re dramatizing. The double blind procedure is well documented and no one has ever been able to prove their cable claims under those conditions, so you shouldn’t be the one feeling exasperation here… The point is, people can only hear great things from great cables when they know which one is playing. What seems to be “just too hard” is prove that they can hear it blind.

      • I take it you didn’t read what I wrote. Those are DBT about amplifiers that didn’t measure the same… Hardly relevant to the discussion here.
        I challenge you to find one DBT which demonstrate audibility between two cables with similar specs.

      • …The same way differences between cables suddenly vanish when you blind the listener, this kind of discussion seems to brutally stop when you ask for proofs…

      • If finding the truth will cause prejudice to the defense then yeah, I tend to agree…
        It is apparent you will exhaust every possible intellectual detour before addressing the main point, so let’s be clear. The burden of the proof lies on the people who pretend they have seen aliens, not on those who don’t believe them.

      • Bah. We’re going in circles. One last time, just to be clear what’s gone wrong with your argument:

        1. If you make a claim, you must provide evidence.

        2. If that evidence is not enough for some, it is not the job of the original claim-maker to provide more evidence just because. That’s “moving the goalpost” — a fallacy.

        3. Instead, the burden of proof shifts to the accuser at this point. There was a claim, you didn’t care for the evidence — it’s up to you to provide counter-evidence.

        Some caveats:

        A. Saying “it ain’t so” isn’t evidence. Again, you have a burden here. You also cannot assume the opposite of the antecedent — that’s another fallacy, “begging the question”. Note that this generalizes to a whole host of other logical fallacies.

        B. Having your evidence be guilty of your own original criticisms is bad form — that is, you can’t be half-assed or random. If you’re serious, be serious — take the objections seriously and build your methodology in a careful, comprehensive manner. You have to do more, not less. That is, to have your objection taken seriously, you must provide sound reasons why your objection is relevant and on-point, and then provide valid evidence that supports that objection. To wit — you’re going to need to show some relevant and properly derived data.

        About #1: Cable companies (and many, many other companies) make claims about products every day. In this case, the claim is that cable x sounds different than cable y. The evidence supplied has been testimonial. In many cases, many many testimonials. Just for the record, this is not medicine — testimonials are a valid and perfectly acceptable data and are also perfectly appropriate to the domain. The testimonials may be mistaken, they may be fraught with inaccuracies and translation issues, but so be it. This is data. Take it or leave it.

        About #2: This is the asked and answered thing. You wanted data, you got data. You didn’t like data, that’s really too bad. This is why you don’t see cable companies jumping up to spend more money on “testing”. To their mind, testing has been done. You not being satisfied doesn’t mean anything. You wanting more data will only get you more testimonials. You wanting something other than testimonials just makes you weird.

        About #3: This is where things get interesting. The prosecution at this point can either rest their case (and lose), or provide an alternative theory with the appropriate experimental data that supports their position. Note that providing more anecdotes makes this he-said/she-said and doesn’t answer the question. This sort of response is ruled out as irrelevant. To overcome that, you have to mount a different theory that is more convincing — and either demonstrate that the original claims are either logically incoherent or provide the results from an experiment testing that alternative theory. That is, if you believe a DBT is the only way to show that a cable manufacturer is actually producing a cable x that is sonically different from cable y — or not — then go do it. Just be aware that proving the negative is going to be rather problematic.

        Again, the higher burden of proof is on you, the accuser, than it is on the defendant.

        And here endeth the lesson.

  2. Another cable demonstration that hugely rely on the power of suggestion. It was probably too risky to just hide which cable (or non-cable) was playing…
    Let’s just remember that in other posts you are blaming the Ars/randi demonstration for their methodology. I believe you just don’t understand the inherent problem of sighted tests for subjective perception. Or maybe you’re just too afraid of Art Dudley.

3 Trackbacks / Pingbacks

  1. CAF 2015: Viva HiFi presents Daedalus, ModWright and VPI | Part-Time Audiophile
  2. CAF 2015: Command Performance delivers VPI, Berkeley, Wireworld, Audia Flight and Neat Acoustics | Part-Time Audiophile
  3. RMAF 2015: Coming Friday | Part-Time Audiophile

Comments are closed.