floating point comparison

[Ric]

Evening all,
Could someone with a higher knowledge than mine please explain why this code snippet doesn't work?

Code: Select all

      MODE 1

      DIM Deax      3
      DIM primary   7
      DIM secondary 7
      DIM answer    3
      DIM code% 1024

      FOR I% = 0 TO 2 STEP 2
        P% = code%
        [
        OPT I%
        .calculate
        fninit
        mov eax, 0
        mov DWORD [Deax], 0
        fld QWORD [secondary]
        fld QWORD [primary]
        fcompp
        wait
        fstsw ax
        mov [Deax], eax
        sahf
        je equal
        jg greater
        jl lower
        ret
        .equal
        mov DWORD [answer], 0
        ret
        .greater
        mov DWORD [answer], 1
        ret
        .lower
        mov DWORD [answer], 2
        ret
  
        ]
      NEXT I%

      CLS
      INPUT"Number 1 ";num1
      INPUT"Number 2 ";num2
      |primary   = num1
      |secondary = num2
      CALL calculate
      PRINT "0 = Numbers are equal"
      PRINT "1 = Number 1 is greater"
      PRINT "2 = Number 2 is greater"
      PRINT
      PRINT "Answer = ";!answer
      PRINT "Value in eax ";!Deax
      END

I have taken the bulk of the floating point stuff straight from the manual.
The results is eax are always the same for the same comparison so i could use that but it would be nice to use the correct code.

Regards Ric

[David Williams]

Replace

Code: Select all

        je equal
        jg greater
        jl lower

with

Code: Select all

        jz equal
        ja greater
        jb lower

I'm so rusty with this FPU stuff that I had to consult this old web page:

https://www.website.masmforum.com/tutor ... uchap7.htm

[Ric]

Thanks David,
I did not realise that the conditional branches were different.
Ho hum
Regards Ric