Fooling around with 3 velocity layers

EDIT: Glenn-it took me a second reading to see what you meant--that the hammer attack would be linear. What follows has to do with the results of the strikes. I guess I'm trying to manipulate the velocity to register those, if they exist...

Well, my thought--mulling this over without research or testing--was that although the velocity and force would be linear, the response of the string and the hammers might not be. In other words, there might be levels of impedance. With soft strikes, assuming a softish hammer and a hard string at great tension, the string wouldn't be moved much. But there would be a specific point at which the resistance of the string was overcome--any force above this would overcome the resistance easily. Any force below it would have the hammer bouncing off the string as much as exciting it--there would be sound, but the string would win fairly fast and return to a steady state fast.

I don't know about a third level. Another point at which the resistance becomes even lower. You're an engineer, yes? There would of course be a breaking point for any object struck by a force, but would there be a level below this, a level well above the point at which the resistance is overcome, a point at which very little resistance was present--a stage of suddenly greater elasticity, say? [EDIT: But that doesn't make sense, given the sound produced, does it--a sudden increase in string elasticity would create a darker sound, but the sound instead brightens.]
Or perhaps that third stage might be instead be one at which suddenly greater excitations of transients occured?

But all of this comes out of nowhere--random ideas that occurred when I accidentally set three velocity stages, liked the sound of the accelerated movement to medium strikes, and remembered liking some triple-strike pianos from the past. That may be all I'm doing, really--recreating an old fashioned sampled sound. But I was playing a Mason-Hamlin for a few minutes yesterday and felt as though the action wasn't entirely linear--that it got a little louder and fuller, if not exactly brighter, with just a little more force, as opposed to working up to the sound linearly. (Did you try the settings?)

Actually I didn't think I said it was linear, but likely non-linear.

Many things in physics are not linear at all - the fall-off of radiation with distance follows a square law, the volume of a cube follows a cubic law (obviously).

But this doesn't mean they have discreet step values - rather they usually are continuously variable no matter the function governing input and output.  If I lift with twice the force, I can lift twice the weight and this is linear, but main point is that there are no steps in the equation.

I think we've become so attuned to the digital world utilized by most computers that we think in terms of discrete steps (however small).

Once upon a time, when I started engineering (before most people here were born), the relative merits of digital versus analogue computers were debated.  Of course digital won out for the most part, so we tend to think of digital (128 steps in MIDI) as being the norm.  Real musical instruments don't output in 128 discrete steps, but in a continuum of values.  Generally it works because we don't notice the difference.  But there are people that claim that digital music sounds cold or synthetic or some other term.  Is it possible that they can detect the steps?

Back to the hammers.  Their reaction to force is not likely linear at all;  I don't believe there is a point (or step) where the sound suddenly changes because the felt suddenly undergoes an abrupt change in physical properties.  If one could plot a curve relating hardness or tone output to velocity, the curve likely isn't linear, but curvilinear, and I also suspect that the there are regions where the curve equation changes shape and may well change rather quickly.  This may appear to be a discontinuity, and for practical purposes it may be, but one wouldn't find a gap in the curve if an infinite number of points were plotted.  Make sense?

Glenn

Yes. I didn't mean an abrupt change, but instead an accelerated rate of change. And it occurs to me that what I want to try is an expanded middle area, reducing the ratio for middle velocities there. But I'm getting confused, too, since I'm actually changing the velocity to amplitude in the Velocity window, while trying to do this.

Hey, Skip. I have to admit that I've been a coward about trying out the big libraries with 100 samples per note. I guess the question to ask is--does the big library sound good and feel good?

Can you expand on multiband compression? That means that several ranges of  amplitude levels are isolated, their highest amplitude reduced, and the overall level of each band is raised? But wouldn't the leaps in amplitude--from one band to the next-- be noticeable? Or is something done to smooth the leaps?

re: folks sometimes preferring band-limited recordings, I don't think that applies to the kids of today - apparently if they don't hear the "MP3 sizzle" they think something is wrong. ;^)  Sad. Very sad!

Greg.