Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!helios!tamuts!n233dk
From: n233dk@tamuts.tamu.edu (Rick Grevelle)
Newsgroups: comp.sys.handhelds
Subject: Re: Melting 48 Display
Message-ID: <9192@helios.TAMU.EDU>
Date: 17 Oct 90 01:56:32 GMT
References: <9095@helios.TAMU.EDU> <25590065@hpcvra.CV.HP.COM> <21051@orstcs.CS.ORST.EDU>
Sender: news@helios.TAMU.EDU
Organization: Texas A&M University
Lines: 40


In article <21051@orstcs.CS.ORST.EDU> maksymc@jacobs.CS.ORST.EDU (Chris Maksymiak) writes:
>
> I found a very strange "real" located at #1017Bh SYSEVAL ... 
>  ... it turns out to be: -5.72874767606E2:8 
>  yes, it has a ':' in the exponent... I found that if you divide
> it by itself, you don't get one, you get 1.1516...  any ideas?
>
> Chris Maksymiak

Sorry about this, but what you have found is not a real.  Actually this is a
very common mistake made by hackers.  You see the five nibbles, #02922h, are
part of the machine instruction located at #10179h.  More specifically, it's

#10179h: 3433920   move.p5   #02933h,c

where this instruction moves the five nibble length constant #02933h from the
instruction stream to register C beginning with the nibble pointed to by the
register P.  Which if you look at the previous instruction at #10177, should
be

#10177h: 21        move      1,p       ; P points to nibble 1


such that register C should look like


    +---+-  -  -  -+---+---+---+---+---+---+
    |   |          | 0 | 2 | 9 | 3 | 3 |   |
    +---+-  -  -  -+---+---+---+---+---+---+

Simply by flagging down all of the occurrences of a prolog is no guarantee
what you find is indeed the start of an object or the beginning of a program.
As in the case above, they are used quite regularly within the instruction
stream of the various machine routines located through out the 48's ROM.
SYSEVALing prologs such as these can sometimes trash your user memory,
including any memory held in any RAM cards that might happen to be
plugged in at the time.

Rick Grevelle