Mops 2.3 floating point 
Author Message
 Mops 2.3 floating point

Somebody asked me a Mops question by email yesterday and I seem to
have deleted it.  But others might be interested in this as well,
so I'll post.  Apologies to whoever-it-was.

The section of the manual on floating point mentions a class "Float".
Sorry, but this class doesn't exist any more -- the manual should
have been updated to reflect this.  There are floating point numbers,
but they are "low-level", not objects.  Previously we had that
class "Float" as well, but I felt this was redundant, and furthermore
Float objects were very hard to compile optimized code for.
Ordinary floating point numbers can be optimized quite well,
especially if they are local variables, since they can then be
kept in the coprocessor's registers.

Another consequence of the removal of Float is that floating point
quantities can't be ivars.  I realize now that this could be a
problem.  Anyway I'll reinstate Float in the next version if there's
any demand, or post the code (it's only about a page).

Cheers,  Mike.



Tue, 19 Sep 1995 05:24:06 GMT  
 [ 1 post ] 

 Relevant Pages 

1. Mops 2.3 bug fixed

2. Mops 2.3 posted

3. An alternative to floating point (was Re: Floating point non-exactness)

4. IBM 370 Floating point to IEEE floating point

5. Floating point warnings in CLISP 2.30

6. fixed point vs floating point

7. code with fixed-point or floating-point

8. fixed point vs floating point

9. fixed point vs floating point

10. Floating Point vs Fixed Point

11. Getting a floating point number from a float-object

12. vax floating pont to unix floating point


Powered by phpBB® Forum Software