From: Kirk Wood <cpt.kirk@1tree.net>
To: speakup@braille.uwo.ca
Subject: Re: FAQ
Date: Thu, 21 Jun 2001 13:50:44 -0500 (CDT) [thread overview]
Message-ID: <Pine.LNX.4.21.0106211342580.4313-100000@ignatious> (raw)
In-Reply-To: <Pine.LNX.4.21.0106220056280.432-100000@goanna.net.au>
On Fri, 22 Jun 2001, Shaun Oliver wrote:
> I agree with that also. windows9x don't use dynamic memory
> allocation. this means that the stack is prone to becoming fragmented. and
> that is why windblows9x falles over more often than it's nt counterpart.
> all breeds of unix and windows nt/2000 all use dynamic memory allocation
> so as to give us the advantage of longer uptime. and a less fragmented
> stack, thus fewer crashes or lockups.
Actually this gets more blame then is reality. The simple fact is that the
memory fragmentation is not a large factor. What is a major factor is that
all Win9x versions (including ME) include a large sum of code from
win3x. This code is included for either of two reasons. The first was
speed and the second was backward compatibility. The later is te main
reason for 3x code appearing in ME.
The fact is that 95 had about a third of its code written in 16 bit
code. This was for compatibity and speed. When they released 98, the share
dropped to about 25%. All indications were that ME would retain a good
chunk of the 16 bit code though it would be re-written.
In all of the versions of winblows not in the NT line programs could
communicate through shared memory. This set the system up for memory
leaks. As if a program allocated memory then failed to release said memory
it was out of service until the next reboot. This was done because there
was no way to know what memory was being shared and the OS recovering the
memory lead to spectacular crashes.
WinNT took a different approach to this. In the ealy versions of NT, the
memory was duplicated to different programs. This worked fine as long as
you didn't run too many 16 bit programs. Then in version 4, they decided
to put all 16 bit programs into seperate memory space. This gave good
performance and yet protected all but other 16 bit programs from their
memory overwriting ways.
I know this is more then most want to know. But if you are going to talk
about the failings of a company bent on domination at least get it right.
=======
Kirk Wood
Cpt.Kirk@1tree.net
Nothing is hard if you know the answer or are used to doing it.
next prev parent reply other threads:[~ UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
FAQ Holmes, Steve
` FAQ Ann Parsons
` FAQ Gregory Nowak
` (OT): FAQ Buddy Brannan
` Janina Sajka
` Ann Parsons
` FAQ Shaun Oliver
` FAQ Shaun Oliver
` FAQ Raul A. Gallegos
` FAQ Gregory Nowak
` FAQ Victor Tsaran
` FAQ Shaun Oliver
` FAQ Ann Parsons
` FAQ Shaun Oliver
` FAQ Dan Murphy
` FAQ Shaun Oliver
` Screen readers for Windows... was: FAQ Victor Tsaran
` FAQ Ann Parsons
` FAQ Shaun Oliver
` Kirk Wood [this message]
` FAQ Buddy Brannan
` FAQ Shaun Oliver
-- strict thread matches above, loose matches on Subject: below --
FAQ Gregory Nowak
` FAQ Buddy Brannan
FAQ Holmes, Steve
FAQ Bruce Noblick
` FAQ Kirk Wood
` FAQ Charles Hallenbeck
` FAQ Geoff Shang
` FAQ Shaun Oliver
` FAQ Steve Holmes
` FAQ Shaun Oliver
` FAQ Steve Holmes
` FAQ Shaun Oliver
` FAQ Ann Parsons
` FAQ Stephen Dawes
` FAQ Shaun Oliver
` FAQ Stephen Dawes
` FAQ Ann Parsons
` FAQ Janina Sajka
` FAQ Gregory Nowak
` FAQ Raul A. Gallegos
` FAQ Gregory Nowak
` FAQ Stephen Dawes
` FAQ Gregory Nowak
FAQ Ann Parsons
` FAQ Gregory Nowak
` FAQ Shaun Oliver
` FAQ Ann Parsons
` FAQ Shaun Oliver
` FAQ Geoff Shang
` FAQ Shaun Oliver
` FAQ Geoff Shang
` FAQ Shaun Oliver
` FAQ Raul A. Gallegos
` FAQ Raul A. Gallegos
` FAQ Shaun Oliver
` FAQ Raul A. Gallegos
` FAQ Geoff Shang
` FAQ Ann Parsons
` FAQ Kirk Reiser
` FAQ Gregory Nowak
` FAQ Kirk Wood
` FAQ The Electronicman Scott Berry
` FAQ Shaun Oliver
` FAQ Geoff Shang
` FAQ Saqib Shaikh
` FAQ Raul A. Gallegos
` FAQ Kirk Wood
` FAQ Raul A. Gallegos
` FAQ Georgina Joyce
` FAQ Rich Caloggero
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.LNX.4.21.0106211342580.4313-100000@ignatious \
--to=cpt.kirk@1tree.net \
--cc=speakup@braille.uwo.ca \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).