Posted-Date:	Sat, 4 Jan 1997 10:20:46 -0600 (CST)
Message-Id:	<199701041620.KAA11019@mail.texoma.net>
From:	"Mark S. Teel" <mteel AT texoma DOT net>
To:	"DJGPP" <djgpp AT delorie DOT com>
Subject:	Fw: 32-bit filenames
Date:	Sat, 4 Jan 1997 10:25:39 -0600
MIME-Version:	1.0

----------
From: Mark S. Teel <mteel AT texoma DOT com>
To: Benjamin D Chambers <chambersb AT juno DOT com>
Subject: Re: 32-bit filenames
Date: Saturday, January 04, 1997 9:52 AM



----------
From: Benjamin D Chambers <chambersb AT juno DOT com>
To: djgpp AT delorie DOT com
Subject: Re: 32-bit filenames
Date: Friday, January 03, 1997 8:22 PM


On Fri, 03 Jan 97 01:52:01 GMT tomw AT tsys DOT demon DOT co DOT uk (Tom Wheeley)
writes:
>Doh!  All of you forget that:
>  16 bit == bad old style of doing things, whatever they
> were 32 bit == incredible new invented by Microsoft way >of doing the
same old things using more memory and disk >space and run a lot faster
(but only when you upgrade >your CPU and memory).
>
>It's nothing at all to do with integer size or range of >addressable
memory or register size or anything like that >:-p

I *BEG* to differ on this point...
16-bit=good old way to do things, great system.
32-bit=incredible old way to do things, M$ had nothing to do with it,
just as small, same disk space, don't need to upgrade memory.

I admit, you do have to upgrade your CPU, unless people program wisely -
which most usually don't.  What people refer to as 16-bit or 32-bit is
_THE SIZE OF THE REGISTERS_, so I do not see how you think it does not
refer to integer size or register size.  When Intel (again, NOT M$)
introduced a 32-bit chip they offered more memory models than were
previously available - so, really, the memory addressing _IS_ a result of
the 32bit processor.  Also, with a 32bit offset, it's possible to access
65536 times as much memory as a 16bit processor.  Using more memory and
disk space is not a result of the 32bit chip - if Intel had only made
16bit processors, programs would be just as large as they are now, just
performing tasks differently.  What usually eats up disk space is the
code needed to switch the processor into protected mode (now _THAT'S_
M$'s fault - when the 386 came out, they should have made DOS PM by
default, and use virtual 86 consoles for real-mode app's.)

But the big point is, 16-bit and 32-bit are refering EXPLICITLY to the
register size, and the rest just follows that.

....Chambers

I'm not sure what we are debating here now, but, it is more than register
size.  It also involves CPU bus width (16 vs. 32 bit) and the fact that the
"old" 16 bit intel cpus required the segment-offset approach to address
memory outside of 1 meg.  This approach (and IBM's initial choice of Intel
vs. Motorola 68000 for thier first PCs) has confused/restricted/led to our
ridiculous dependance on Microsoft for sub-par software for the last 17
years or so...  Those of us who have programmed 68000s and flat memory
model cpus in general have disliked Intel/segment-offset/Microsoft/PC
programming for years!  Hail djgpp and GNU!

MST

- Raw text -