Index
> Main > how do I use FPUGoto page Previous 1, 2 |
| Author |
|
|
MHajduk 05 Sep 2013, 19:25
Use sscanf to convert the obtained strings from the text format to the desired number format then store values in v1 and v2, should work.
|
|||
05 Sep 2013, 19:25 |
|
|
patchariadog 05 Sep 2013, 19:37
I tried this code but it just crashes
Code: invoke GetDlgItemTextA,[hWin],textbox1,addr v1,100 invoke GetDlgItemTextA,[hWin],textbox2,addr v2,100 cinvoke sscanf,[v8],"%s",addr v1 cinvoke sscanf,[v9],"%s",addr v2 ;rest of the code that works finit fld dword [v8] fld dword [v9] fdivp fstp qword [v3] cinvoke sprintf, addr v4, "%.16lf", dword [v3], dword [v3 + 4] invoke SetDlgItemTextA,[hWin],maximumoutputpowertext,addr v4 I made v8 and v9 dd I also tried change the %s to %f because I was not sure what to use but it crashes also |
|||
|
05 Sep 2013, 19:37 |
|
|
MHajduk 05 Sep 2013, 19:49
patchariadog wrote: I tried this code but it just crashes  v8 and v9 should be declared as buffers big enough to store strings taken from the text fields, something like this: Code: v8 rb 101 v9 rb 101 (...) invoke GetDlgItemTextA, [hWin], textbox1, v8, 100 invoke GetDlgItemTextA, [hWin], textbox2, v9, 100 cinvoke sscanf, v8, "%f", v1 cinvoke sscanf, v9, "%f", v2 finit fld dword [v1] fld dword [v2] (...) |
|||
|
05 Sep 2013, 19:49 |
|
|
patchariadog 05 Sep 2013, 20:03
I changed v8 and v9 to rb 200
then I put this code in and did textbox1 = 10.1 textbox2 = 5.5 answer should be = 1.83636363636 I get = 519010387852.254564 Code: invoke GetDlgItemTextA, [hWin], textbox1, v8, 100 invoke GetDlgItemTextA, [hWin], textbox2, v9, 100 cinvoke sscanf, v8, "%s", v1 cinvoke sscanf, v9, "%s", v2 finit fld dword [v1] fld dword [v2] fdivp fstp qword [v3] cinvoke sprintf, addr v4, "%.16lf", dword [v3], dword [v3 + 4] invoke SetDlgItemTextA,[hWin],maximumoutputpowertext,addr v4 thanks for all the help |
|||
|
05 Sep 2013, 20:03 |
|
|
patchariadog 05 Sep 2013, 20:10
sorry I somehow missed changing the %s to %f now it works thanks so much.
also on this line of the compare code you gave me Code: cinvoke sprintf, addr v4, "%.16lf", dword [v3], dword [v3 + 4] what does the dword [v3 + 4] do thanks |
|||
|
05 Sep 2013, 20:10 |
|
|
MHajduk 05 Sep 2013, 20:10
The formatting strings should be "%f" not "%s" - you want to convert input strings to single precision floating point numbers.
|
|||
|
05 Sep 2013, 20:10 |
|
|
MHajduk 05 Sep 2013, 20:13
patchariadog wrote: what does the dword [v3 + 4] do |
|||
|
05 Sep 2013, 20:13 |
|
|
patchariadog 05 Sep 2013, 20:28
okay thanks for all your help today. I have been searching for how to use the fpu in fasm on and off for about 6 months now so thank you.
|
|||
|
05 Sep 2013, 20:28 |
|
|
MHajduk 05 Sep 2013, 20:38
patchariadog wrote: okay thanks for all your help today. I have been searching for how to use the fpu in fasm on and off for about 6 months now so thank you.  |
|||
|
05 Sep 2013, 20:38 |
|
|
AsmGuru62 06 Sep 2013, 01:45
Code: fld [v1] fld [v2] fcom This is same as: Code: cmp [v2], [v1] To do the opposite (cmp [v1],[v2]) you must load values in opposite order: Code: fld [v2] fld [v1] Also, you do not have to load the 2nd argument -- can save 1 FPU opcode by direct operation with memory argument: Code: v1 dq 45.9776 v2 dq 7.0594 v3 dq 0 ... fld [v1] fmul [v2] fstp [v3] |
|||
|
06 Sep 2013, 01:45 |
|
|
MHajduk 06 Sep 2013, 07:28
AsmGuru62 wrote:
But maybe you meant something like a macro shorthand here. |
|||
|
06 Sep 2013, 07:28 |
|
|
AsmGuru62 06 Sep 2013, 13:08
Yes, I meant it like a pseudocode, because the swapping of the arguments will mean the difference in flags.
|
|||
|
06 Sep 2013, 13:08 |
|
| Goto page Previous 1, 2 < Last Thread | Next Thread > |
Forum Rules:
|