Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!samsung!zaphod.mps.ohio-state.edu!wuarchive!uunet!apctrc!drd!mike
From: mike@DRD.Com (Mike Rovak)
Newsgroups: comp.sys.handhelds
Subject: Re: Chip question
Summary: Make lots of backups
Keywords: caution, backups, disclaimer
Message-ID: <1990Dec7.160011.22983@DRD.Com>
Date: 7 Dec 90 16:00:11 GMT
References: <1990Dec6.231427.28844@jarvis.csri.toronto.edu>
Sender: Mike Rovak
Followup-To: comp.sys.handhelds
Organization: DRD Corporation
Lines: 91

byu@csri.toronto.edu (Benjamin Yu) wrote:
} I don't know if anyone of you who used CHIP (ver 225) experienced this or it
} is just me.  When I run, say BRIX, I put BRIX on level 1 and hit CHIP to
} "interpret" it.  Now, I don't have my clock on display, but sometimes, I need
} to repeat the whole process about 6 times before I can actually play the game!
}...

It's not just you.

Background:

I have a version E rom, and have been using CHIP/225 with BRIX and SYZYGY.
I also have been using the RPL TETRIS (fast enough for *me*!).
I recently downloaded AS48, the RPL CHIP8 Assembler, posted by G. Kohl,
(which IMHO is the slickest [most elegant] and most intellectually stimulating
piece of RPL code I have ever seen) and several things occurred in the
same day:

  1. I installed my freebie EQ LIB card, and a purchased 128K card into
     the calculator and merged all the system memory.

  2. I loaded AS48 into my calculator and ran it on the supplied demo.
     I had a typo in one of the routines, and it did not run successfully
     on the first attempt.  I fixed the problem, and then it ran 
     correctly.

  3. I downloaded the HP-supplied utilities into the calculator (stop watch
     library, USAG, appointment book front end) and tried to attach the
     stop watch library to a subdirectory, not realizing that it attempts
     to automatically attach to the HOME directory on power-up.

  4. At one point, I had the stop watch library apparently attached both
     to the HOME directory and the subdirectory.  I realized my error
     and then detached the one on the subdirectory.

  5. I didn't like the fact that SW-A appears first in the LIBRARY menu,
     I preferred to see the EQ LIB menu selections fill up the first six
     slots, so I [stupidly] turned off the calculator, removed the EQ LIB
     card, turned on the calculator, verified that the only library in the
     LIBRARY menu was SW-A, then turned the calculator off, installed the
     EQ LIB card, turned the machine on, and gave up when I saw that the
     machine was going to insist that PORT 0 library objects list first
     in the menu.

I should point out here that I appreciate the value of a backup. 
I make frequent ARCHIVEs to my PC and am glad of it.  It paid off.

Imagine my surprise the next day when TETRIS ran and accumulated score
without displaying any falling pieces.  Likewise, PUZZLE24, which I thought
was an RPL program, failed to run and required a system halt (ON-C) to return
to the stack display.  I looked at the INIT subroutine, and it had some
intersting and rather large binary integers and Externals in it, which I
could have sworn were not in the original code.  I immediately and without
question nor further debate executed the three-fingered salute (ON-A-F) and
cleared memory.  Reloaded from a good backup, downloaded what was missing,
didn't do anything questionable or anomalous, and voila, everything works
and memory is intact (then I made yet another backup).

I think that the bottom line is that what we have here is a piece of
computing machinery which is complex in its hardware AND its firmware.
Disclaimers on low-level machine language code are not published in vain.
A word to the wise should be sufficient.

I have no idea which operation actually corrupted my memory.  All I am
saying is that there was ample opportunity for doing so in the course of
normal operation, and this time it happened.  Fortunately for me, it
only cost me a few extra minutes of my time.

The user should be aware that some RAM addresses control some hardware
functions in the HP48SX (e.g., display), and there are apparently no
hardware interlocks (cost constraints) to prevent a user from physically
damaging the machine, or so I gather from readings on this newsgroup.
My humble advice would be to stay out of the memory scanner and the 
ON-D stuff unless you have a reason for being there and know what you 
are doing with it, unless potentially losing your investment makes 
no impression on you.

I would also advise clearing memory at the first sign of trouble, and
frequent archiving if you modify memory contents a lot.

Just trying to help.

-- Mike

------------------------------------------------------------------------
Disclaimer: My opinions do not necessarily reflect those of my employer.
========================================================================
------------------------------------------------------------------------
     mike@DRD.Com  
     uunet!apctrc!drd!mike
========================================================================