Re: ROM Editing & Prize Offer

[mjmahon@aol.com (Michael J. Mahon)]

Rich Beaudry wrote:

>A VERY interesting challenge....
>
>However, my experience disassembling (admittedly short) programs is
>that it gets difficult to separate code from data.  For example, if
>there is some numerical lookup table defined in the source code, it of
>course is stored right in the binary.  The disassembler than merrily
>chugs on it as if it were code, resulting in a lot of spurious
>output... THAT can be tough to handle....

Disassembly always _begins_ with a lot of spurious output, but as
you begin to understand what's going on, you can specify a script
to the disassembler that specifies code regions, data regions, and
labels, and things get a lot better.

Eventually, you stop re-running the disassembler and start
commenting the disassembly.

>Also, how are we to test the "new" EPROM code w/out a Vulcan?  If you
>were willing to also supply the Vulcan, that would make things MUCH
>easier.  It would also allow some analysis into the circuitry, which
>is an aid in disassembly of code as well (if you can decipher what
>output ports go where, how addresses are decoded, etc., it makes it
>much simpler).  Be aware, however, that a mistake may mean a fried
>Vulcan...  However, that risk is the same whether you supply the
>Vulcan, or you trust the "New" EPROM, and plug it into your card
>yourself...

In the case of the Vulcan, the hardware interfaces are pretty
transparent.  A good description of the IDE interface would
be more useful, since all the commands to the drive could
then be found and interpreted, and the return status bytes.

>And, any technical information on the Vulcan (especially, he says
>laughing, schematics) would help.

This can be reverse-engineered from the board, but programmed
logic certainly slows things down.  ;-)  As I mentioned, I don't
think that the schematic would be especially needed in modifying
the firmware for addressing larger drives.  That is almost certainly
in the disk address arithmetic and command formation, which is
simply passed via 8-bit latches to the drive.

My experience with the Vulcan was with the //e model, with
6502 firmware.  That's the one that I'm most interested in...

-michael

-michael

 Email:  mjmahon@aol.com
 Home page:  http://members.aol.com/MJMahon/