- Joined
- 9/7/07
- Messages
- 220
- Points
- 28
Reply to jayg's "My first post"
You have solved a part of the problem, and have done a very good job of solving the Diophantine equation.
But the issue is not only where the function f is maximized, but what that maximum value is subject to f<R. How are we supposed to know in advance that max of f is 101?
Of course taking advantage of gcd(9,20)=1, we simply multiply through by 101 (see quantyst's 'partial' solution), etc. etc. So, given that we have differently signed x and y solution, how can we use this 'insight' to our advantage in order to produce a same signed x and y solution?
For example, allowing x and y to be differently signed, the Diophantine equation 9x+20y=11 has solution (x=-1, y=1). But if we restrict both x and y to being non-negative integers, then it has no solution, even though the maximum value of f(x,y)=9x+20y subject to f<11.7087 is 9 with (x=1, y=0).
You have solved a part of the problem, and have done a very good job of solving the Diophantine equation.
But the issue is not only where the function f is maximized, but what that maximum value is subject to f<R. How are we supposed to know in advance that max of f is 101?
Of course taking advantage of gcd(9,20)=1, we simply multiply through by 101 (see quantyst's 'partial' solution), etc. etc. So, given that we have differently signed x and y solution, how can we use this 'insight' to our advantage in order to produce a same signed x and y solution?
For example, allowing x and y to be differently signed, the Diophantine equation 9x+20y=11 has solution (x=-1, y=1). But if we restrict both x and y to being non-negative integers, then it has no solution, even though the maximum value of f(x,y)=9x+20y subject to f<11.7087 is 9 with (x=1, y=0).