🚀 go-pugleaf

On my favourite subject: TopSpeed Modula-2 3.10 for Windows 3.1

Right after the previous post I got responses from a couple of people
(thanks) and that made me do some more testing. After some intense library
re-building, I can report the following:

If stack calling model is used as opposed to jpi, and thereby as default
standard_float (i.e. stack also for the x87), the bugs go away and e.g.
MATHLIB can be called in a normal fashion.

The bug in Str.RealToStr is still there but the (*%T _WIN *) version is
straight M2-code and the bug could be fixed (if you need the fix - send
me a mail).

Still, I do get occational problems when I get deep into calculations.
That may be a problem of my own, but the code is identical to the DOS
version so I think it is still a bit unstable.

I got reports on several potential problems spotted in CoreMath.A when
it comes to Windows and, if needed, it may be possible to fix some of it
without the aid of TopSpeed.

Anyway, when it comes to continued development:

 ** All systems go! **

--------------------------------
Bengt-Inge Dalenback, CATT
also at: 
Department of Applied Acoustics
Chalmers University of Technology
Gothenburg, SWEDEN
e-mail: bid@ta.chalmers.se
fax : +46 31 145154