Friday, December 9, 2016

#39 ROM extracted

"M-Chip from Taito Extermination B06_14.1G"

Chip decapped successfully, but it was a bit dirty despite initial cleaning attempts (above).  Try again:

Better.  Using crossed polarizers, ROM looks like this:

Getting there, but there's still enough debris to make readout difficult.  One final more aggressive cleaning round:

And we have a good quality image prime for readout.

For reference, here's a sample closeup of conventional brightfield vs crossed polarizers imaging:

After imaging, a computer vision (CV) program analyzed the images to produce a ROM dump.  However, instead of getting 0x800 bytes as expected, we got 0x900!  After a closer look, we discovered the CV program had interpreted some address lines as data:

These were fixed and the ROM now looks reasonable.  After a bit of work to understand ordering:

We now have the original ROM file.

However, additional checking needs to be done before it's ready to use (ex: dust induced single bit error).

1 comment:

  1. Can the CV program spit out a list of likely candidates, or does that have to be gone over line by line?