Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!uunet!zephyr.ens.tek.com!tektronix!percy!m2xenix!puddle!f16.n15.z1.fidonet.org!Mike.Chapin
From: Mike.Chapin@f16.n15.z1.fidonet.org (Mike Chapin)
Newsgroups: comp.lang.modula2
Subject: TopSpeed 3.01 First Impressions
Message-ID: <3251.28683DBF@puddle.fidonet.org>
Date: 23 Jun 91 15:32:30 GMT
Article-I.D.: puddle.3251.28683DBF
Sender: ufgate@puddle.fidonet.org (newsout1.26)
Organization: FidoNet node 1:15/16 - Black Diamond BBS, Gillette WY
Lines: 49

I have been using system 3.01 for a week or so and have found a few
problems and a few good suprises.  I have redone several libraries that
I have on the shareware market and came up with a few suprises.  One
library corrects a deficiency in the graphics libraries of all the JPI
family; it allows the BIOS fonts to be written on pixel boundries.  This
particular library does not use any of the features of the graphics
libraries; it is completly stand-alone.  The other is a graphing library.
It uses the library for all the routines except writing text, that is
done by the preceding library.
 
I also did the translation and do M2 support for the GFX Graphics and
GFX Fonts and Menus libraries.
 
1.  The new project system looked rather daunting.  I did a few simple
things and got lost in the larger projects.  After some false starts,
I got everything working and am now a believer.  It seems that JPI did
this one right.  I would suggest a person trying to build complex systems
look at the project files for building the library.  This was a great
help to me.
 
2.  The pragma system works a more as advertised and is better documented
from both inside the environment and in the documentation.  The GFX
libraries crashed when I first tested them under Ver. 3.01.  The problem
was (*# module(implementation => off) *).  This pragma was not used in
Ver 2.x and did nothing when added.  Ver. 3.01 requires it if the module
is going to be part of a larger library.
 
3.  JPI blew it on the M2 graphics library.  The problem showed up when
I was testing the graphing library.  SetVideoMode works properly for all
systems except EGA/VGA 16 color.  In this mode the program does weird and
wonderful things.  I have told JPI about this but have not seen any
corrections yet.  I have not had time to chase down through the library
source to see where the problem is, but my graphing library is effectivly
broken until JPI fixes this problem and sends out new graphics libraries.
 
4.  As for size, the size on all executables that I used to test the
libraries was pretty close to the same as Ver. 2.x.  What I did notice was
that the BIOS font library was a little crisper, indicating a small
speedup.  This is just an impression, I didn't do any timing tests.  The
library is written entirely in M2 and does all screen writing and bit
twiddling in the high level source.
 
--Mike
 


--  
uucp: uunet!m2xenix!puddle!15!16!Mike.Chapin
Internet: Mike.Chapin@f16.n15.z1.fidonet.org