16 Nov
1999
16 Nov
'99
2:18 p.m.
I never learned to like the CBASIC and others of its pseudocoded ilk. The
MBASIC was OK, though, and when it was finally the way you wanted, "it"
could be compiled with BASCOM, which generated fairly decent code, even
assembler if you wanted. It could be linked with output from M80 as well.
You could even generate code with which to band-aid together output from
their (MS) Fortran and COBOL using either MBASIC =>BASCOM=> (object linkable
with L80) + output from compilers => thereby genrating interactive code from
stuff that was originally designed for batch. That was more common than you
might think. It was, after all, customary to scrounge useable FTN and COBOL
from the guys down the hall. My ex-partner made LOTS
of dough using CP/M
and these various compilers to generate useable output from
long-obsolete
but still functional code written for the mainframe down the hall. I guess
that's why he got the big bucks. It was not unusual to have the payroll
running on the micro while the figured out why they couldn't make it work on
the behemoth.
Even I was fairly impressed with the selection of stuff Billy and his
buddies turned out!
Dick
Dick
-----Original Message-----
From: Allison J Parent <allisonp(a)world.std.com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Tuesday, November 16, 1999 8:48 PM
Subject: RE: TI99: Is the following true?
<QBASIC was available for the M9900 from Marinchip
for $220, according to
<their ad. It was a compiled Basic, article states you didn't need line
<numbers! Sounds like Microsoft stole from Marinchip, too, since this
versio
<was available in 1981 for the Marinchip 9900
system!
Ah, kids!
Prior to the advent of the PC in 1981:
Qbasic was a MS product befor they became a OS house. They used to be
heavy in languages and big apps. DRI (AKA CPM) was know for the OSs.
MS offered languages and apps like Multiplan for APPLE (6502), 9900,
8080/z80, 16032, z8000 and I'm certain I've forgotten a few.
So anywho, Qbasic was the MS answer to C-basic which was a semi compiled
non line number dependent basic. Qbasic is not fully compiled to native
machine assembly language. I know this as I still maintain a PC app
written in Qbasic (as recently as last week)! Qbasic also tried to copy
the Borland "IDE" <Integrated Development Environment> used the
Turbo<insert
language here> they had for z80 and later PCs, in
that you could edit, run
interpretively and then compile the working result.
Oh, Qbasic also called runtime subroutines from the library for floating
point math (and other things too).
Allison
<
<> ----------
<> From: Andy Frueh[SMTP:andyfrueh@hotmail.com]
<> Reply To: ti99(a)theriver.com
<> Sent: Tuesday, November 16, 1999 4:16 PM
<> To: ti99(a)theriver.com
<> Subject: RE: TI99: Is the following true?
<>
<> Hmmm...I guess where I'm getting confused is the way it handles floating
<> point...It DOES "do it" but it does it by converting it to an integer,
<> right? I guess that's what non-coprocessor machines must do, too.
<>
<> Oh well, 3 and 4 are still true, and I really dispute this is a "real"
<> story, even if elements of it ARE true. Especially that quote! :-)
<>
<> I don't have the magazine to look this up. Can we get more info on this
<> BASIC?
<>
<>
<> >From: "Yates, Ben" <BYates(a)mobilnet.gte.com>
<> >Reply-To: ti99(a)theriver.com
<> >To: "'ti99(a)theriver.com'" <ti99(a)theriver.com>
<> >Subject: RE: TI99: Is the following true?
<> >Date: Tue, 16 Nov 1999 16:07:34 -0500
<> >
<> >Look in the July/August 81 99'er magazine. Marinchip created a 9900
base
<> >system.
<> >To answer your questions:
<> >The 9900 cannot do floating point in hardware. Neither could the 8088
or
<> >286
<> >without coprocessors. It emulates it in the ROM.
<> >QBASIC WAS a basic that sold for $220 (find it in the above magazine).
<> >
<> >
<> > > ----------
<> > > From: Andy Frueh[SMTP:andyfrueh@hotmail.com]
<> > > Reply To: ti99(a)theriver.com
<> > > Sent: Tuesday, November 16, 1999 4:02 PM
<> > > To: ti99(a)theriver.com
<> > > Subject: Re: TI99: Is the following true?
<> > >
<> > > 1. Who would name their computer PC/OT given what that means?
<> > > 2. Can't the 9900 DO floating point? Why emulate it?
<> > > 3. When has a governor EVER cared about technology...or publicly
made
<> >such
<> > >
<> > > silly comments?
<> > > 4. The R in RISC means Reduced, not rinkydink
<> > > 5. AFAIK, QBASIC is a MS program, and not something that runs on a
<> 9900
<> > >
<> > > Based on that, I'd guess the following is NOT true. :-)
<> > >
<> > >
<> > > >From: "Yates, Ben" <BYates(a)mobilnet.gte.com>
<> > > >Reply-To: ti99(a)theriver.com
<> > > >To: "'ti99(a)TheRiver.com'"
<ti99(a)TheRiver.com>
<> > > >Subject: TI99: Is the following true?
<> > > >Date: Tue, 16 Nov 1999 15:46:43 -0500
<> > > >
<> > > >from http://www.fourmilab.ch/autofile/www/section2_46_5.html
<> > > >
<> > > > <<...>> <<...>> <<...>>
<<...>> <<...>>
<> > > >Next: Dear Jim Meadlock... Up: AutoBits Previous: CAD: The Final
<> > > >
<> > > >Marinchip Defeats IBM PC/AT In Benchmark
<> > > >
<> > > >Mill Valley, California, Mayday 1986.
<> > > >John Walker, President of Marinchip Systems Ltd., announced today
<> that
<> > > the
<> > > >Marinchip 9900-based PC/OT (Personal Computer/Obsolete Technology)
<> > > >resoundingly defeated the IBM PC/AT in an intense floating point
<> > > benchmark,
<> > > >even though the PC/AT was equipped with the 80287 math
coprocessor.
<> > > >The benchmark was an optical ray tracing program involving
primaril
<> > > >floating point computations, including evaluation of trigonometric
<> > > >functions. The Marinchip 9900 PC/OT executed the program in 69.32
<> > > seconds,
<> > > >while the IBM PC/AT took 93.79 seconds to execute the same
program.
<> > > >``Our PC/OT executed this real-world engineering program 26
percent
<> > > faster
<> > > >than IBM's much vaunted PC/AT, even though our 9900 processor
was
<> > > operating
<> > > >at 2 megahertz, one third the speed of the PC/AT's 80286 CPU,
and
th
<> > > fact
<> > > >that the PC/OT was emulating floating point in software instead of
<> >using
<> > > a
<> > > >mathematics coprocessor. This benchmark vindicates our RISC
<> (Rinkydink
<> > > >Instruction Set Computer) architecture, and clearly demonstrates
th
<> > > >superiority of our proprietary QBASIC language for scientific
<> > > >applications.'', said John Walker.
<> > > >The IBM PC/AT benchmark was run in Lattice C version 2.14, using
th
<> > > ``-P''
<> > > >memory model (large code, small data). The standard Lattice 2.14
<> >library
<> > > >was
<> > > >used. The results calculated by the Marinchip PC/OT and the IBM
PC/A
<> > > >agreed
<> > > >to 15 decimal places.
<> > > >Commenting on the results, California Governor George Dookmayjeun
<> said,
<> > > >``It
<> > > >just goes to show you how a bunch of clean living Californians can
<> beat
<> > > the
<> > > >spit out of those drug-soaked greasy Florida scumbags. Look, I
don't
<> >give
<> > > a
<> > > >flying fork what you quote me as saying, but please spell my
freakin
<> > > name
<> > > >right!''. <<...>>
<> > > >
<> > > >Editor: John Walker
<> > > >
<> > > >
<> > >
<> > > ______________________________________________________
<> > > Get Your Private, Free Email at http://www.hotmail.com
<> > >
<> >
<> >
<>
<> ______________________________________________________
<> Get Your Private, Free Email at http://www.hotmail.com
<>
<