How a string bends

El_Presidente

Hey, I'm not sure if this the right forum to put this up on but I need a little help...

I'm writing a program to play guitar tab and at the moment im trying to get it to do string bends. So my question is this, How does the pitch of a note change as the string is bent up?

Seems simple but it isn't. If I were to change the pitch uniformly over time it sounds way off. I suspect it should start with a small change in pitch followed by larger changes. This would give a sort of curve (if we graphed change in pitch against time).

Going back to leaving cert physics I know that the pitch is proportional to the square root of the tension (which goes up as the string bends) but is also inversely proportional to the inverse of the length (the lenght also goes up as the string bends)

So, anyone out there have any ideas?

Toulouse

Nope...and I think I just wasted every braincell I had reading that

El_Presidente

8 braincells aint that much.

seamus

Get a book that deals with guitar tab notation. There are loads of types of bends, slow bend, fast bend, prebend,.......

Bends can increase the pitch by any number of semitones, and some GNR songs do a doubletone bend (ouch). Experiment yourself with a guitar and the tab book to see what kind of effect your program should create

Going back to leaving cert physics I know that the pitch is proportional to the square root of the tension (which goes up as the string bends) but is also inversely proportional to the inverse of the length (the lenght also goes up as the string bends)

inversely proportional to the inverse of the length

Double-negative - therefore it's proportional to the length, which is wrong. As length increases, pitch goes down, so it's either

proportional to the inverse of the length (squared)
or
inversely proportional to the length (squared)

But when you bend the string, the amount by which you increase it's length is negligible, esp compared to the increase in tension - so neither of these really apply. Assume the string is not flexible.

Gordon

Well not sure about inversely proportional square root etc but...
From one octave to the higher pitched octave the distance of the string is halved, therefore it is a logarithmical amount surely. So if you want to do a bend over a whole tone the timing would need to be logarithmical too.

Oeneus

I don't know how simple/complicated an answer you want, but when a string is bent, the tension is increased!

col_nicholson

id dont know about hte string bending, but can i have a copy of this program if it works, which im sure it will

Oeneus

You do know that this idea has already been done?

It's called Guitar Pro. Check it Out. It might give you some ideas. http://www.guitar-pro.com

El_Presidente

Yeah, I know its been done a couple of times before but I had to pick something to do for the project and it seemed as good as any.

Once its finished I'll put it up and anyone who wants it can have it.