Re: Wolf 3D

[nathan@visi.com (Nathan Mates)]

In article <MPG.f5e375d6f617f01989684@news.mindspring.com>,
Richard N. Cain <rcain3@mindspring.com> wrote:
>I just ran Wolf 3D with no crash upon exit.

   You're definitely in the majority. I was the only one who could
consistently reproduce this bug among the Wolfy beta team (drafted
about 10 hours before release to help squash a *much* more annoying
set of bugs, as I've got the tools to examine real weird
conditions). The other testers launched and quit the game up to 30
times on their machines to see if it was reproducible on their
machines, and never had that. Some other init/addon interactions were
already noted in the documentation, and I've never heard of any
whining that those are unreasonable.

   I was pretty sure it was an OS, init, or addon bug biting my exact
system to cause Finder code being loaded over where the window mgr
should be in ram, and many hours of examination later, I had an exact
cause nailed down. I've provided exactly where in the OS code it
appears, unlike other flaming jerks who won't read such proof. What do
I get for providing such info to a few idiots in particular? Nothing
positive but more lies and bull. But, hey, they're supposed to be
knowledgable enough to cut & paste together articles already collated
into a "FAQ", and be able to pirate a lot. Geez, good qualifications
for the truth, good qualifications for anyone to ever listen to them
again.

   Why that bug only really hit me among the beta testers isn't too
certain (other than I'm a bugmagnet able to break most things), but I
also do have my system loaded down with all sorts of custom-made
tripwires to alert me when the system's doing stupid things. I could
reproduce this crash with tripwires off, but those tools have been
darn useful in fixing other OS bugs, something the flaming illiterate
(being unwilling to read is just as bad as being unable to) idiots
don't seem to be able to do.

>I have a hierarchic CDEV, which replaced the init file, that I
>downloaded from Genie in Sep 97 (file 28883).  The CDEV is copyright
>Seven Hills 1997 and was created in July 1997.  The file info says
>it was written by Richard Bennett and is version 1.6.3.

   Yep, that update is available on GEnie and Delphi, but have not
seen it on the internet. But, as I've said many times, when the bug
resides in the OS, it's not hierarchic's fault. The aforementioned
flaming jerks won't read the proof of that, including ROM locations,
and it's only their loss in acting like they've got the real attitude
problem. Before they attack bugs in anything else, they need to get
the huge cockroaches out of their eyelids to be able to read proof.

   Wolfy's only problem here was assuming that Apple told the truth
when they said "your patch will be automatically removed." (GS
Technote #101 for *proof* to the flaming idiots) and implicitly read
in "correctly" at the end of that sentence. The patch *was* removed,
but due to idiot bugs in the OS, the toolset being patched was left in
a bad state in RAM. Apple fubar'd here, as the disassembly points I've
previously provided show, and GUPP 1.0.7 makes the implementation of
patch removal match the documentation. Simple, really.

>I also have GUPP 1.0.6f1 running, so that might have a fix for the
>problem.

   Nope. 1.0.6 was way before this problem surfaced. 1.0.7 fixes the
*OS* bug for this or any other app that follows the documentation from
Apple and is quite downloadable from Delphi.

Nathan Mates
--
<*> Nathan Mates http://www.visi.com/~nathan/      <*>
# What are the facts? Again and again and again-- what are the _facts_?
# Shun wishful thinking, avoid opinion, care not what the neighbors
# think-- what are the facts, and to how many decimal places? -R.A. Heinlein