Re: 6702 chip

Date: Wed, 26 Jun 2013 23:19:53 +0200
Message-ID: <21420368.27151372281593559.JavaMail.www@wwinf3710>
I have integrated Pavel Zimmer's 6702 chip simulator with my own 'C' code for the SuperPET dongle checking routines and have had a major 
success getting them to work together (i.e. my code pokes Pavel's simulator and gets the correct responses back).

I am now starting to draw the schematics out for the 6702. I have not done so much yet (only started it today) but I have already sketched 
out the chip select lines and started to look at the data lines. The chip select lines are genuinely used as active high and active low 
chip selects as identified on the schematics - but (interestingly) they can be configured for active high or active low by modifying the 
aluminium mask. As the 6702 chip select lines are directly wired to the 6502/6809 buffered address lines - this leaves the possibility of 
Commodore producing additional 6702 chips located at different addresses without having recourse to changing the hardware or PCB layout. 
It could even mean that multiple 6702's could reside in the SuperPET at the same time - but all this is (of course) supposition!

It would appear as though the diffusion and polycrystalline layers look identical across all eight data bits. As Pavel has already 
identified, the aluminium mask contains 'jumpers' whereby it would have been possible for Commodore to change the internal 'logic' of the 
6702 thus giving a different sequence of numbers. This would have been used with a different 6809 subroutine - presumably to protect other 
software for the SuperPET if it had survived long enough.

If anyone is interested in receiving a copy of my code or routine updates of the schematics as I produce them - just send me a PM and I 
will oblige.

I will be working on decoding the logic for Data Bus bit 0 next to see how one 'channel' works. Assuming the other data bits are the same 
- I should be able to work out what they do by looking at the differences in the aluminium mask for the 'jumpers'.


       Message was sent through the cbm-hackers mailing list
Received on 2013-06-26 22:00:08

Archive generated by hypermail 2.2.0.