|
build this midi guitar for me
StoreTags: hair brained, guitar, midi
Viewed 1570 times. 8 people liked this blog. You can rate it below if you haven't already.
-->
i don't know if this concept is being used anywhere but it occurs to me that the tracking on most midi guitars suck. the problem is that they're all pickup based which relies too heavily on acoustic movement of the strings and detection of freq's blah blah blah. i honestly don't know how they work.
why don't we run some controlled fire down all these metal parts and translate from discreet known values? is this feasible? look at the diagram i threw together.
6 carrier signals are going up the strings. when the 1 transducer at the base picks up a significant vibration on that string, the amp of that carrier increases. when contact is made at any fret or fret 0 (open string) the corresponding modulator affects the carrier on its string. the signals are summed at the top and fed into a standard audio interface. the software translator is "tuned" into 6 stations and gets it's on/off/velocity info from the carrier's amplitude. the note is determined by which modulation is applied.
it seems like this would be less prone to error in the physical realm with harmonics and crappy playing. the only latency would be on your audio interface. of course the "carrier" signals would have to be audible or at least within the range of modern interfaces....
p.s. i have absolutely no training in physics or engineering so this could be completely impossible for obvious reasons, unknown to me because i'm a bumdass. | |
Comments
03/26/08
+
PM |
QUOTE |
PERMALINK |
REPORT
mux
I don't quite get how this is any better than the current guitar controllers...
the problem with guitar controllers isn't that they don't pick up notes properly, it's that playing guitar isn't just playing clean notes; the nuances come from the bent string, the muted string, the pick harmonics, etc. Using frequency detection on the individual strings lets the MIDI kit guess at the 'base' note, then apply a MIDI pitchbend. Still, the tracking kinda sucks, mostly because it feels "mechanical" - you really have to learn to play the guitar the way the MIDI pickup works, not vice-versa.
Maybe some combination of the two technologies? Get the base note from the electrical contact, then use that info to better judge what's going on with the frequency of each string?
03/26/08
+
PM |
QUOTE |
PERMALINK |
REPORT
j_chot
there was an old voltage controlled guitar/synth interface in the 70s that worked like what you described....
can't remember the name for it though...
03/26/08
+
PM |
QUOTE |
PERMALINK |
REPORT
fakeBlooper
if i'm using a guitar as a midi controller its because i know how to play the guitar, not the keys. i'm not so much worried about translating muted chucka-chucka's and pinch harmonics. i'd rather do bends with a whammy bar cuz its the sexy. i just want a very accurate, expressive (sans the bends) midi stream into my sequencer. this system doesn't really on any freq detection... the strings don't even have to be in tune, they just need to be tight enough. the software would be smart enough to know that mod(top) is the one that counts in deciding which note to spit to handle bar chords. it also knows to ignore mod(0) since it would be on every note, unless it was fired alone.
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
pravda23
some tech heads around here huh. 
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
Squeal
I've thought alot about systems similar to this one, and the problem is that the frets will short your strings together. Keep in mind that when you fret a string, you make contact with two frets. Whenever you fret 2 or more strings on adjacent frets, you end up shorting the signals from each string together. For monophonic control, this would not be a problem. If you want to input chords, though.....
Does this make sense?
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
fakeBlooper
i've thought about that but i figure if we can have multiplex signals (analog cable tv) can't the software be smart enough to only pay attention to the mod(top)?
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
Squeal
I think that mod(top) of the higher-fretted string would end up on both strings, because they would share frets. Then the software would think that both strings are playing mod(top) when you have fretted mod(top) and mod(top-1).
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
fakeBlooper
when i say mod(top) i'm talking about the highest number mod on the string (closest to the bridge). the software knows to ignore everything except mod(top) for that channel. i think you might be more right htan me though...
03/27/08
+
PM |
QUOTE |
PERMALINK |
REPORT
Squeal
On string 1 you fret mod(top) and mod(top - 1). On string 2 you fret mod(top-1) and mod(top-2). Since both strings are touching mod(top-1) the signals from each string are transmitted to the other, so you end up with mod(top) on both strings 1 and 2.
Mull it over awhile. I can't figure out how to do it that way. I think the one that j-chot mentioned was targeted at a monosynth, so it worked great. I remember seeing a guitar that had fragmented frets on it(i.e. each fret was split into six insulated parts). That's a feat of engineering (or stupidity) in itself.
I am working on a midi guitar with no frets. Each string is part of a separate ribbon controller type circuit. Great if you're into fretless bass, but no longer much like a guitar.
03/28/08
+
PM |
QUOTE |
PERMALINK |
REPORT
tunalicker
fretless slide guitar would be fun to play with...
Register / login
|
^
EM411 is Copyright 2001-2008 EM411.com
All rights reserved. | Contact | RSS
|
