Masm32 original author?

[jj2007]

the FPU has changed considerably since the days of the 8087

Sure?  ;)

[Gunther]

Hi Jochen,

the FPU has changed considerably since the days of the 8087

Sure?  ;)

without any doubt. Since the 80387 a bunch of new instructions are added, especially for the transzendental trigonometric functions. Furthermore, the 8087, 80287, and 80387 had an own sloat on board. Since the 80486 DX, the CPU and FPU are integrated at the same chip; so the entire thing needs a shorter bus protocol and has a lot of other advantages. That are not only minor changes.

Gunther

[dedndave]

yah also - we don't have to mess with "affine" or "projective" infinity any more, either - lol
they simplified some of the initialization
it also seems like FWAIT got simpler - i could be mistaken on that one

[hutch--]

Guys,

Iczelion retired about 10 years ago, wife, life, work etc .... He was a genuinely nice guy, an excellent programmer and a good brain for architecture.

[Gunther]

Steve,

Iczelion retired about 10 years ago, wife, life, work etc ....

it seems to be the walk of life.

He was a genuinely nice guy, an excellent programmer and a good brain for architecture.

no doubt about that.

Gunther

[alanji]

Who was the Masm32 original author?

Long ago I wrote "Masm32" and put it on a floppy for use of the employer but another technician put it on the internet without my consent. I've just looked through my floppies and there are 5 (1..5) of Masm 6.11 and 1 of Masm 6.14.zip. If someone looks at the original (I can't find it at this time) my email provider is on the disk, and is something like Cantec. I'll do some more searching for the original.

I have a file on my PC "c:\Utilities\Real Asm\REALASM1.zip" which is also available at..
http://www.programmersheaven.com/download/1374/download.aspx
Submitted BY: "Unknown"

I have contacted the website (yesterday) to ask them replace "Unknown" with my name "Alan Illeman".
RealAsm was written by me for 16-bit floating point arithmetic but when we all moved to 32-bit (year?) I rewrote these functions and called it Masm32.
I cannot currently find the Masm32 floppy, but will get back to you all.

[Vortex]

I wish Iczelion would return back to programming but he has his own priorities in his life and this is natural.

[Ryan]

I wish Iczelion would return back to programming but he has his own priorities in his life and this is natural.

It's a shame he took his tutorials down.  I learned a lot from them and was still working through them when he took them down.

[Vortex]

Hi Ryan,

Here is Iczelion's Win32 Assembly Homepage :

http://win32assembly.programminghorizon.com/index.html

[Ryan]

Hi Ryan,

Here is Iczelion's Win32 Assembly Homepage :

http://win32assembly.programminghorizon.com/index.html

Thank you!

I believe the domain changed from the one I had bookmarked.

[mineiro]

I'm just curious Sir alanji, this file inside your computer, real asm, what are the date of files?

[Gunther]

Hi Vortex,

Hi Ryan,

Here is Iczelion's Win32 Assembly Homepage :

http://win32assembly.programminghorizon.com/index.html

I've bookmarked the site immediately. Thank you.

Gunther

[alanji]

Who was the Masm32 original author?

Long ago I wrote "Masm32" and put it on a floppy for use of the employer but another technician put it on the internet without my consent. I've just looked through my floppies and there are 5 (1..5) of Masm 6.11 and 1 of Masm 6.14.zip. If someone looks at the original (I can't find it at this time) my email provider is on the disk, and is something like Cantec. I'll do some more searching for the original.

I've found the 16-bit version.  Here's an example (ITOFT):

.386p
code32 segment para public use32
       assume  cs:code32, ds:code32, es:code32
       include pmath.inc
;------------------------------------
; convert signed integer to a REAL4
;
; Returns eax = REAL4
;------------------------------------
s_struc struc
        dd  ?  ; ebp
        dd  ?  ; caller
integer dw  ?
s_struc ends
retval = (size s_struc) - 8

itoft:
       push    ebp
       mov     ebp, esp
       push    bx esi edi
       mov     eax, 0
       mov     edi, REAL4BIAS + 15
       mov     bx, [ebp].integer
       or      bx, bx       ; if integer == 0
       jz      exit         ; return eax=0
       mov     esi, 0       ; assume sign = +
       jg      f1           ;
       mov     esi, 1       ; sign = -
       neg     bx
f1:
       test    bx, 8000h
       jnz     f2
       dec     edi          ; bias--
       shl     bx, 1        ;
       jmp     s f1
f2:
       movzx   eax, bx      ; integer
       shl     eax, 17      ; remove explicit 1
       shrd    eax, edi, 8  ; insert bias
       shrd    eax, esi, 1  ; insert sign
exit:
       pop     edi esi bx
       pop     ebp
       ret     retval
code32 ends
       end

The files are all dated 1994.  I found them on a floppy disk. At my age I'm a little past it. I concede that maybe someone else may have wriitten them too (smile).


[mineiro]

Hello Sir alanji
I put myself in that era and remember that we use more arj instead of zip.
Searching by realasm1.arj give some results, but dates 93, bbs times.

[dedndave]

back in the day, i wrote a 16-bit math library   :P

that was before i knew anything about the 8087
i wrote it to be compatible with MS BASIC
unfortunately, "MS float" formats were not the same as the Intel/IEEE formats
so - the library would have to be completely re-written to be good for anything - lol
and, of course, it's 16-bit - so it would need to be updated to 32-bit, as well