Shop OBEX P1 Docs P2 Docs Learn Events
Optimizing Floating Point Objects? — Parallax Forums

Optimizing Floating Point Objects?

Prophead100Prophead100 Posts: 192
edited 2011-12-01 15:14 in Propeller 1
I am writing a program that uses floating point math including the most complex stuff in Float32Full but the various objects like "floatstring" are using different floating point objects of FloatMath, Float32, and Float32Full. Would the program be more compact when compiled if I go in and change them all to the same object (e.g. Float32Full)?

Comments

  • PerryPerry Posts: 253
    edited 2011-12-01 14:16
    Try "Brad's Spin tool "http://www.fnarfbargle.com/bst.html. It does a great job removing unused spin code.

    Perry
  • Mike GreenMike Green Posts: 23,101
    edited 2011-12-01 14:24
    Brad's Spin Tool won't help here. You're best off using just one floating point object, whichever has the features you need for your whole program. Also keep F32 in mind. This is a faster, smaller, more accurate floating point library that's mostly compatible with Float32Full. F32 does lack the user functions.
  • Prophead100Prophead100 Posts: 192
    edited 2011-12-01 15:14
    Thanks. I will check into the F32 and go with a single object in the final.
Sign In or Register to comment.