The Mega voices of course are extremely velocity dependent. velocity controls of 8 or so sample volumes then the next higher velocity selects a new sample.
I would love it if Yamaha put in a patch that made use of one of the data sliders, so that moving the slider, would jump to the next sample at a consistent volume.
This would give the user precise control of what sample of a mega voice was used. It would bring a whole new realism to live playing.
I'm sure it would be simple code.
Don't think one could do it in the Kenos now. If you used a DAW, you might be able to script it, or use Logic's environment to code such a process. (don't know about other DAWS, but in general they have pretty much the same features.