anorakDan said:
Cagey said:
Wouldn't it be nice if somebody took the time and absorbed the expense to do a computerized "plek" type of fixture with a pile of sensors on it that could quantify the response of a particular build? As a database was built, you'd know what effects the various variations would have. You'd finally be able to make objective decisions. Think of all the mythology that would get blown away!
Actually, it wouldn't be that tough. Which is probably why it hasn't been done. It would probably wreck a lot of business models.
Weird. I was thinking the very same thing just before I scrolled down to this post, Cagey! I think we would all love to see that data!
I used to do work for a company called "Gauging Systems" that specialized in measuring things that for various reasons were difficult to measure. We used lasers, ultrasonics,
hall effects sensors,
accelerometers, magnetics, etc. Most of what we measured had to do with things you couldn't touch for one reason or another, or had to be done dynamically. I wrote the software that made the data the things returned useful to humans. It was a lot of fun, and took me into places to measure some interesting things.
Actually, one of the cool things about the job was that we were often called upon to do what hadn't been done before, so you were allowed to fail and still get paid in the hope something was imminent. Think military work, or automotive R&D.
So, I have some background there. Not so much in failure - we pulled off a lot of pretty tricky stuff. But, contrary to nanny-state thinking, failure is a huge part of success. You learn what's possible or not and build on or avoid that in other applications.
There are some economies of scale that come with consumer applications that drive some devices into the dirt, cost-wise. Almost any micro-electronics is a good example of that, and for what we're talking about here, accelerometers fall into that group. They're used in smartphones for orientation sensing, cameras for anti-shake compensation, collision sensing in collision air-bag deployment decisions, and they build gajillions of those things a year. So, they're cheap. Now you're seeing them in those clamp-on headstock tuners like the Snark...
You named me "Snark"? Really?
The guitar shops want $30 for them, but you can buy them online for less than $10 just about anywhere. Surprisingly sensitive and accurate, and somebody's still making money on them at that price. Tells you what the sensor costs.
So, what would it take to place a pile of those things on a holding fixture, add some A->D circuitry, and get the data into a computer to fiddle with? Seems like it should be fairly straightforward, other than about a million details.
