Current Measurement Concepts vs. Perception

[FunkyKikuchiyo]

The other elephant in the room is that gross variation in right+left hand techniques guarantee that different players sound different on the same gear. A particular guitar+pickup+amp combination is an accomodation to physiological differences as well as preferences. Gimme a wide fretboard with a compound radius, thanks.

When I was doing sales, all of the staff knew each other's "sound" like we knew each other's voices. One person could plug in and play just a couple open chords and everyone would know who it was. THAT is something that is amazing to consider, but probably best left for another thread. I think classical players grasp that better than electric players given all the time they spend working on right hand technique, nail shaping, etc.

[salvarsan]

VoVox' 11.5 foot guitar cable is $135.
Save money on the 19.7 foot one -- only $195!

If I need it that badly, I'll thread my own damn #18 twisted pair through a dacron braid.

[Steve Conner]

The question of why blue is blue is addressed (and surprisingly enough left unsolved) by a thought experiment called Mary's Room. Google for details.

There are interesting parallels between music perception and speech recognition. Namely, computers suck at both. The reason why we can't quantify the kinds of differences Frank is talking about is the same reason you can't talk to your computer. Sounds have higher-level meanings in the context of being a living, breathing human being, that are utterly lost on machines.

What this has to do with measurement, as opposed to perception, I don't know. But I do know that most electronic components don't have enough internal structure or state to produce sounds with these high-level meanings. As Salvarsan puts it, there is an elephant in the room holding the guitar.

[StarryNight]

great thread btw! two things come to mind for me:
1. if we, the pickup makers, are "artists" then its a case where the artist is making tools for the artist (i.e. musician). this complicates things greatly. The baker is an artist who usually makes bread for the person who consumes it. We make pickups for artists who decide how they will use pickups and other tools to create art for the consumer of the art. this leads me to the second point:
2. Humans will make things out of stuff that was made for some other purpose. By that I mean people are more intuitive than the machines around them. An artist by definition will create a new sound in their head before it ever exists in the visceral world. They will then use what ever tools they have on hand to produce the sound that is in their head. What I'm trying to get at is humans actually enjoy making something out of nothing. If you give a musician a guitar and say, we've done extensive modelling and testing and this guitar will make you sound exactly like Jimmy Page, here you go. Then we're taking the art out of the artist. I don't consider myself a pickup artist (except maybe with the ladies, bow bow chicka bow!) but I think of making pickups as making tools for artist to use however the hell they want. We're sort of creating wimpy (and whinny) musicians by giving them everything on a silver platter and saying this pickup is for X and this other pickup is for Y. That method of marketing has created a giant vat of wannabe's that just complain how their patch cord doesn't make them sound like Jimmi Hendrix. Spare me please!

[Rick Turner]

The "problem" I had with the VoVox cable was that an active low impedance output guitar sounded better with it as well as the expected high impedance passive instruments. That ran counter to my own built-in bias, if you will. And many musicians would gladly pay a hundred bucks for some device, any device that made that difference. I was simply not expecting to be able to hear a difference using one of my Renaissance semi-hollow, buffered piezo pickup guitars. I did. Oh, well... So much for that theory.

[defaced]

The funny thing about musicians is that they try to use a micrometer to measure a mile. Most of the time the actual measurement value doesn't make a bit of difference, so whether it's mojo or a real difference that can be measured is a moot distinction.

Remember the rules of engineering:
1) Order of magnitude is more important than accuracy
2) All you have to do is cross the threshold of "enough". Over engineering is a waste of time and money.
3) The threshold of "enough" is usually determined empirically. Hope you made friends with the old hands at work.

And about the cables... Did they allow you to compare your favorite brand, that you brought to the stand, to theirs? It the control isn't a true control, all bets are off.

[Rick Turner]

We used the cables A/B'd against several of our own (Monster, Planet Waves) in two booths with four different amplifiers, two of which I'm very familiar with.

No, it was not a controlled experiment, but nobody who heard the VoVox against any other cable was deaf to the differences, and I'd say we had a good ten people try the A/B. In every case, the VoVox was preferred. BTW, this was without most people knowing the price. If pressed for an adjective, I'd say clearer was the difference. Maybe a bit fuller, too...like wider bandwidth.

Beats me; I'm not flogging these things, I was simply amazed by them. When I find gear I like, I talk about it...like the Bose L1 system we use.

[salvarsan]

Remember the rules of engineering:

"Never leave well enough alone."

First Rule of Hobbyism: "If you can't polish that turd, gold plate it."

Sattinger's Law: "It works better if you plug it in."

[Mike Sulzer]

Sounds have higher-level meanings in the context of being a living, breathing human being, that are utterly lost on machines.

I think it is just a matter of not knowing the algorithm. You are implying that there are mental phenomena that take place outside of our body and brains, or that our brains are capable of operations that cannot be represented as algorithms. This is not the direction that the research is moving in.

[Steve Conner]

No, I am not. I concede that a computer could behave identically to a human, if it could somehow acquire the petabytes of state that gives every human action and utterance its context and makes it meaningful. Think of all the sights, sounds, smells, feelings, tastes, emotions you have ever experienced. Try to imagine every step evolution took simce the first DNA molecule crawled out of the primeval soup, and the fruits of those millions of years of trial and error. Those are "all" that's missing from the computer. It is of course also missing a framework to interpret them, but compared to the enormity of the state itself, that is trivial.

Good luck trying to get it in there, I don't envy the hard AI guys their task. If they think they're getting anywhere, it's because their own lives are so dull.

Anyway time for my new theory of musical instrument engineering. An object has mojo in proportion to the number of state variables and nonlinearities in it. Above a total of seven, optimising it stops being a science and becomes an art.

Speaker cables and resistors have none, guitar cords, tubes capacitors and semiconductors one, output transformers three, and then you start getting to things with distributed resonances, that theoretically have an infinite number. How many would you give to pickups? I'd say seven at least.

Mike defaced, how do you know when you have enough tone? I think I have too much. My guitars and amps would probably upgrade to a better player if they could.

[defaced]

Mike defaced, how do you know when you have enough tone?

When groupies and booze show up without invitation, then you have enough tone.

[Mark Hammer]

1) When you reach that part of the curve where you're at the point of diminishing returns, the odds of perceptual error and desire influencing perception go up substantially; largely because it takes tremendous investment to yield any return, and big investment has a way of shaping human perception.

2) Whenever I see discussions like this one, I am always reminded of Thomas Edison's wax cylinders that he would record for you at county fairs (as a fee-for-service). Some years ago, I read some newspaper pieces of that time, where reporters were remarking how mind-bogglingly indistinguishable the recordings were from an actual live human being. We see such comments and think "What were you guys smoking? You can't hear that?". But wax cylinder recording was such an achievement on its own, that they were simply blown away by the fact that it could happen. It took many years before people were so accustomed to it that they could now start to hear the aural differences at face value, and begin the process of trying to describe and quantify them. And, as is the case in science, the sequence goes: notice -> describe -> explain -> predict -> control. First you have to notice something. Then you have to figure out how to describe it in some fashion that lends itself to measurement and hypothesis-testing. Then, you have to figure out why its happening in the way that was observed (and measurement helps a lot there). Then you have to try and go beyond what you've witnessed and test if your explanation was right by predicting outcomes beyond those already noted and measured. Once you're able to confirm your explanation was right by predicting things, you can begin the process of controlling them and shaping them the way you'd like them to be.

So, there are undoubtedly things we haven't been able to quite clue into yet, when it comes to the assorted phenomena that concern audio and wires. It'll take some time before we're so blasé that we can then say "Well THAT doesn't sound quite right!". At which point, we begin the arduous task of trying to capture in language and measurables, just what it is that doesn't sound quite "right".

That doesn't mean we should have high tolerance for snake oil sales pitches. But the fact remains that we have reached the point we're at now precisely because someone said "Boy, does that ever sound scratchy and shrill!", and someone else looked for a way to measure what was once ineffable. Hard to imagine there was once a time before S/N ratios, THD and IMD measurements, measurement of group delay, FFTs, and all the rest of it, but there was.

[FunkyKikuchiyo]

The question of why blue is blue is addressed (and surprisingly enough left unsolved) by a thought experiment called Mary's Room. Google for details.

There are interesting parallels between music perception and speech recognition. Namely, computers suck at both. The reason why we can't quantify the kinds of differences Frank is talking about is the same reason you can't talk to your computer. Sounds have higher-level meanings in the context of being a living, breathing human being, that are utterly lost on machines.

What this has to do with measurement, as opposed to perception, I don't know. But I do know that most electronic components don't have enough internal structure or state to produce sounds with these high-level meanings. As Salvarsan puts it, there is an elephant in the room holding the guitar.

Sound likes you've been watching Jeopardy!

[FunkyKikuchiyo]

great thread btw! two things come to mind for me:
1. if we, the pickup makers, are "artists" then its a case where the artist is making tools for the artist (i.e. musician). this complicates things greatly. The baker is an artist who usually makes bread for the person who consumes it. We make pickups for artists who decide how they will use pickups and other tools to create art for the consumer of the art. this leads me to the second point:
2. Humans will make things out of stuff that was made for some other purpose. By that I mean people are more intuitive than the machines around them. An artist by definition will create a new sound in their head before it ever exists in the visceral world. They will then use what ever tools they have on hand to produce the sound that is in their head. What I'm trying to get at is humans actually enjoy making something out of nothing. If you give a musician a guitar and say, we've done extensive modelling and testing and this guitar will make you sound exactly like Jimmy Page, here you go. Then we're taking the art out of the artist. I don't consider myself a pickup artist (except maybe with the ladies, bow bow chicka bow!) but I think of making pickups as making tools for artist to use however the hell they want. We're sort of creating wimpy (and whinny) musicians by giving them everything on a silver platter and saying this pickup is for X and this other pickup is for Y. That method of marketing has created a giant vat of wannabe's that just complain how their patch cord doesn't make them sound like Jimmi Hendrix. Spare me please!

This of course opens the black hole of a debate of whether guitars can be ends in themselves. Do they exist only for the end result of making music, or can they be artful creations in themselves? Personally, I highly advocate the value of music strictly as a hobby or just for family and friends sort of thing - much folk music was that way for years - and that is the bulk of the market for our products, not for real living musicians. A person doesn't need to justify a love for playing guitar by trying to get gigs - they can play simply to unburden themselves with the normal pains of existence. So, the ultimate end in that scenario isn't the creation of the art, but the instrument being a personal tool. The music is still the ultimate end, but it is about the person spending time interacting with the instrument for its own sake.

I hear ya on spoiling players though.... I don't know how many times I've had a client's guitar in my hands listening to their complaints and just wanted to say "just learn to play the friggin' thing and your problems will go away"... but I behave.

[LtKojak]

I don't know how many times I've had a client's guitar in my hands listening to their complaints and just wanted to say "just learn to play the friggin' thing and your problems will go away"... but I behave.

I usually don't.

I see a lot of people coming to me because the instrument won't intonate... just to discover they squish the neck so hard for making a chord that the notes go sharp. At that moment, I just play a chord sequence and most times than not, I get the "wow, what did you do?"

To the ones that are genuinely interested, I show'em how to correctly finger a chord... to the other obnoxious ones, I tell'em that I'm a Supreme Being and charge'em twenty bucks for the time wasted. Works like a charm! Kid you not, BTW.