I think I saw a post some time back about a program that tested the accuracy of a midi keyboard in sending midi velocity. Measures the lowest and highest velocity a keyboard can send, and may determine if the response to pressure\speed is linear. Couldn't find it by doing a search, though. Anyone know of a program that will do this?
(I can of course get a general sense of the keyboard's accuracy just by going into Kontakt's instrument editor and watching the light move as I press a key, but I want to do a more precise test.)
Thanks for any suggestions.
Program to test a keyboard's accuracy at midi velocity?
-
- KVRian
- 859 posts since 14 Sep, 2004
-
- KVRian
- 932 posts since 26 Oct, 2005 from The Void
I would think most MIDI controllers let you pick different velocity response curves. Even the ones that don't probably aren't linear but some sort of curve. So testing if it's linear seems kind of pointless.
-
- KVRAF
- 12615 posts since 7 Dec, 2004
it would be impossible to accurately test since it requires a physical interface to be present. for an accurate interface like that (solenoid servo?) i'd think you'd need to spend tens of thousands of dollars. you could just tap the keys randomly and then take the input while assuming your fingers have applied a flat distribution of pressures (which is impossible) or i suppose you could learn what the average distribution of pressures would be from a human finger and weight the results based upon the inverse of that.
taking accurate measurements probably isnt very useful from the user point of view. from the engineer's point of view, i'd think the best default responce would be one where output values map directly to levels which seem intuitive to a user. so, you'd probably have to get a few experianced keyboard players and have them hit the keys in increments of 20%, then normalize the results and take the mean as your curve.
based upon that idea, i'd think the best way for you to judge different boards is to just get a midi monitor and hit the keys yourself in those types of increments, then see what values are getting spit out. if you're hitting at 50% and you're getting a velocity of 110 (85%), that keyboard probably isnt for you.
taking accurate measurements probably isnt very useful from the user point of view. from the engineer's point of view, i'd think the best default responce would be one where output values map directly to levels which seem intuitive to a user. so, you'd probably have to get a few experianced keyboard players and have them hit the keys in increments of 20%, then normalize the results and take the mean as your curve.
based upon that idea, i'd think the best way for you to judge different boards is to just get a midi monitor and hit the keys yourself in those types of increments, then see what values are getting spit out. if you're hitting at 50% and you're getting a velocity of 110 (85%), that keyboard probably isnt for you.
-
- KVRAF
- 16814 posts since 8 Mar, 2005 from Utrecht, Holland
http://www.midiox.com
Just monitor the velocities while you hit the keys as soft and as hard as you can.
Just monitor the velocities while you hit the keys as soft and as hard as you can.
We are the KVR collective. Resistance is futile. You will be assimilated. 
My MusicCalc is served over https!!
My MusicCalc is served over https!!
-
- KVRAF
- 5851 posts since 9 Jul, 2002 from Helsinki
Another practical test would also be any midi sequencer. Record a measure of notes, and try to start quietly and slam it in the end. Practise a few times and the note velocity bars should resemble a nice curve.