I bought a Milton Bradley Simon off ebay to decap the chip. The patent shows it uses a TI TMS-1000, but the chip in the game is a standard 16-pin DIP. I assumed that they just put the TMS-1000 die into a smaller package and didn't bring out all the pads, but the die shots (top metal, top metal removed) show it is not a TMS-1000. It looks like they had the functionality recreated in an ASIC to cut costs, since they sold so many games.
The chip package is marked "MB4850 SCUSO640 .3 H 8235". The Milton Bradley catalog number for Simon is MB4850, and the die is labeled 640. I don't know what SCUSO stands for "semi-custom" maybe? I'm not sure who the manufacturer is.
Simon at Waiting For Friday sent me a TMS-1000 from an earlier revision of Simon. I decapped it and dumped the ROM contents. Here's the instruction decode PLA (flipped to match the diagram in the Programmer's Reference Manual) and o-output PLA. The opcodes in the instruction decode PLA are in a different order than in the manual, but the opcodes all map to the same microinstructions.
I bought a Simon PCB off ebay that had a TMS1000 labeled MP3300. It's wired the same as the REV A PCB that Simon drew up
ROM array with top metal removed
output PLA with top metal removed
instruction PLA with top metal removed
raw ROM dump
Simon Reverse Engineering
Simon code disassembly and emulation
back to chip decap page
back to Home page