In Part One of this interview (Issue 212), Jamie Howarth, President of Plangent Processes and I discussed the fundamentals of the company’s audio speed-correction and stabilization technology, how the idea originated, and the long journey from bringing the idea of compensating for the constant variations in tape speed recording and playback – and the subsequent degradations to recording quality – to practical reality. The interview continues here.
Frank Doris: I'm sure there'll be people who will think that since you're converting analog recordings to digital, it's going to lose something sonically. But what I’ve heard in your demos of Plangent is that there are many obvious improvements – instruments become more stable, the drums swing more, you can really hear the nuances of vocalists, and everything just sounds more “locked in.” And in the real world, more people are going to be listening to digital than vinyl.
Jamie Howarth: Yes. Not to name-drop but Jim Keltner, the great session drummer commented very early on that [drummer] Hal Blaine “swung harder” in a test of Simon and Garfunkel’s “America” that unfortunately never was released. And in 2014 Bob Ludwig mastered the Bruce Springsteen catalog from our transfers and was very complimentary of the more vivid and realistic experience he felt with recordings he had worked on repeatedly since 1985. As for the analog vs. digital controversy – audiophiles seem to treat differences in analog sonics as par for the course, while digital is still belittled. I believe that high-quality digital, at least as we do it, is really, really close to the source we are listening to.
Remember, outside of shops like us or Bernie Grundman Mastering - and those who extol exclusive AAA to an extreme should recognize this – nobody else hears AAA without a generation loss. We’re the only ones who hear the actual tape. Everybody else hears the massive contraption that is vinyl, which sounds great considering how wild a process it is, or they hear digitized audio with varying degrees of precision. I can only speak to what we do – I’ve heard flat ADA vinyl test pressings from our digital sources from the best in the business. It’s close, surprisingly so, particularly Chris Bellman’s cuts. But it’s not exactly the same; there’s a loss with AAA.
And tape is also pretty crazy: the “miniature golf course” that a flimsy piece of tape has to go through in a tape machine is really a mess; the amount of chaos it has to withstand in order to be pulled across the head with some reasonable amount of stability…so when we put that stability back into it, it sounds more like the control room, what the mixing [engineer] heard – and cutting that to vinyl, it holds up. Somewhat diminished but still worthwhile. But ADD is going to beat it, in the immediacy that the stable rendition of the performance brings to the listener. We contend that any potential offset of the conversion is swamped by the improvement of the timing precision regained.
I’m not totally in the tank for digital in general – I think there's still something in the analog and digital filtration that isn't quite nailed down yet, and SACD/DSD to our surprise is never an exact duplicate of the PCM we deliver. But is it close? Yes. Digital also is still a noisy process in the analog electronics where grounding and RF control are concerned. You hear a lot of things these days about these fancy USB connectors and whether or not you “hear” a USB connector. No, you're not hearing the connector data delivery, that’s fixed. What you're hearing is deficient ground isolation and RF rejection between the two USB-linked components – a lot of the audiophile designers still don't seem to know how to absolutely reject the RF noise that's being produced by the computer or by the clock and the DAC’s innards.
Way, way back, I swear to god, in a CD player design David Smith and I did, I could move my hands over the circuit board and get different sounds because the stray capacitance of my hands was enough to shield the RF from certain parts of the circuit. But most of these problems were solved long ago, and at this stage it’s incremental.
Sample of "Born in the USA," FFT spectrum of drift, wow and flutter with sidebands on a 240 kHz bias signal, Studer A80 tape deck.
The same sample after being corrected with Plangent Processes: all speed-related issues are resolved.
FD: When I was at The Absolute Sound we were testing a CD player, and I discovered by complete accident that it sounded better when I had my hand on the top cover.
JH: If you told the pro [audio] guys that, most of them would say you’re out of your mind. Putting magic marker on the edge of the CD, most pro guys scoffed that's never going to make any difference. [Beginning in the mid-1980s, some audiophiles experimented with doing this. – Ed.] Well, yes and no. The notion from the audiophiles was that the data recovery was better, but that was a red herring – actually the light scatter reflecting from the edge of the CD was causing the electromagnetic sled the optics were mounted on to jump around all over the place trying to track the CD, and without those internal reflections, it jumped a little less. And the servo that was driving the sled and the linear motor of the sled was capable of putting spikes of noise into the audio. That was fixed fairly early on, but again and again the problem with digital turned out to be something analog, in this case the electromagnetic focusing system corrupting the analog audio.
There are so many things that can effect digital audio playback. It's like going to the f*cking moon and back. [But] it's gotten better and better and better and better.
At Plangent are working very, very hard. We do really good digital and we think we know enough about how to avoid any of the problems that could happen. With the Plangent output I as a musician can hear the differences in the tempo, the pitch, the stability, the authority of the performance. There's a realness that comes into play because there isn't a destabilization of the performance by the tape machine speed. It's more solid, it's more authoritative.
Those little tiny “pine trees” that I was telling you about turn out to be intermodulation distortion, and the azimuth stability nails the imaging. Without getting too much into the weeds about what this all means to other aspects of audio, there are a lot of sources of intermodulation distortion in the world that haven't been extensively thought about, woofers being one of 'em. The Doppler effect that exists in a high-excursion woofer is actually directly equivalent to wow and flutter. We’ve conducted some experiments that prove that out, and that the process could work there too.
FD: I never considered that.
JH: Nobody has, at least not in that model. There's a larger AM component in there, which is actually the “chuffing” and amplitude modulation (AM) distortion: the intermodulation distortion caused by the volume differences as something gets closer to you and farther away, and the inefficiencies of the voice coil as it gets farther out of the magnet, which has been known for a long time, like in the older Infinity IRS. But there’s also an FM pitch shift happening in all the other audio caused by the 6-inch woofer driver pushing air towards and away from you, just like a siren or a freight train whistle. 1 kHz has 30 Hz sidebands, for example, from the doppler of the woofer/midrange unit.
One of the reasons that Quads, electrostatics, and other panel speakers sound so good is that their excursion is very small. They have such a big surface area that they don't need the excursion in order to be able to achieve the same level. That small excursion means very, very minimal Doppler effect and very little erroneous amplitude modulation. Both are IM and in the Quads with their bipolar drive mechanism, it’s not only low physically, it’s symmetrical and damped. Hence their legendary “realness.”
You've still got a lot of problems in speakers and the vinyl record/playback chain. Mechanically, there are problems, and magnetically there are some problems. Let's imagine the worst speaker in the world that rings for three seconds, and I put 1,000 Hz into it, plus a second later I introduce 1,200 Hz tone. I just got 200 Hz as a cross-modulation product because the ringing of the original 1,000 Hz and 1,200 Hz tones are still happening. Add a 2,500 Hz tone and now I've got three signals simultaneously blurring against each other during the ring time. I’ve not seen that described that way to date – IM during the waterfall decay…
Some speakers are well damped, but without serious negative feedback a cutter head will ring until next January. Some speakers, you excite them and the tail-off is like next week. That’s why tweeters whose primary resonance is out of the audio band – ideally twice as high – are not only less tizzy and more focused, but also disappear better – it’s hard to locate them since the material isn’t telling us what it is by its resonance.
So, you really want short-attack and short-release drivers in both lathes and speakers, which is not an easy thing to do with mechanical devices. Recently that’s improving but still…
Now imagine if a tape machine has got a 60-cycle judder in it, because the motor is a 60-cycle device…and it’s unlikely the servo can go that fast to correct it. Then you're going to have IM. An A 440Hz flute is played and you’ll also get 500 Hz and 380 Hz, the sum and difference products. Not harmonics – rather it’s IM/Doppler/ring modulated hash…and it’s on everything, not just the A 440 test tone in my example. The intermodulation is on all the music provoked by spurious mechanical and electrical things besides the obvious example of a single 60 Hz motor error. And it’s directly measurable and audible.
On my soapbox for a minute: the so-called objective criteria here definitely do have a direct effect on the fidelity to the actual sound. This has gotten so politicized and linguistically manipulated in a way I find really damaging to good order and direction in audio. It’s now a disqualifier to know the tech! J. Gordon Holt and the John Atkinsons of the world [former Stereophile editors – Ed.] actually taught their readers and listeners how the tech affected what was heard. And now a trope is that if you know the tech you can’t hear the beauty, which is really smug and insulting.
Obviously good ears count and hopefully the ability to describe what one is hearing is crucial. Harry Pearson [founder of The Absolute Sound – Ed.] was a genius at this. Pearson could hear really, really, really well, although his tech knowledge was seriously limited. His tech was terrible. He didn’t know how it worked. His linguistics were superb. But he never made the inferential insulting leap that all the engineers couldn’t hear. That’s self-serving bullsh*t.
FD: Oh, I thought you meant his tech meaning the guy who worked for him. I was that guy…I worked for him for seven years! (laughter).
JH: I think Pearson was brilliant. And that gets me in dutch with the pro audio zealots, but hey, the guy could hear music and look at his original CD Baker’s Dozen best of list, it’s got Glyn Johns, Doug Sax, top audio names…but he led with his ears, not with their credit list or their ability to measure cutter accelerations. They knew both.
FD: He didn't have the engineering technical kind of thing, but man, he could sure hear. He was right on. All the heavyweights of the audio world would come to Sea Cliff to be wowed, and we had to do the dog and pony show, and here's Dave Wilson (of Wilson Audio), here's Peter Walker (founder of Quad), and all these heavyweights, and they would sit down and expect to hear the system from god. It usually did sound pretty incredible, barring when things went wrong, which happened a lot. But it was very, very stressful because I was like the guy who had to tune the race car, and if the race car was not winning, everybody would look at me and make faces.
JH: I understand that stress. It took me a long time to work that stress out on my [Plangent] system demonstrations.
FD: People don't know how stressful it is to do audio demos.
JH: They have no idea. I kind of become impervious to it over time, and that's basically just because I've gotten smacked around everywhere imaginable, and still know when I listen to the Plangent process that it works. One guy in a demo dubbed it Windex for audio, which I think is an apt description. But another top name was offended that I was implying I could improve his work. No way would I say that. We improve the chain, and that delivers their work more clearly. And that’s a good cause.
Our interview with Jamie Howarth will be continued in Issue 214.
