This is good stuff!!! Thanks, Bill. Richard ---------- Forwarded message ---------- Date: Tue, 21 Mar 2000 22:51:14 -0500 From: Bil Herd <bherd@jersey.net> To: 'Richard Atkinson' <rga24@hermes.cam.ac.uk> Subject: RE: Calling Bil Herd The problem with the CPM cartridge is a design flaw with the thinking behind the cartridge where the Z80's processor gets frozen for a memory cycle and then runs for 2 clocks and then frozen again. The problem is that if you freeze it during a cycle (which will happen about 1/3 of the time) when the internal bus on the Z80 is in a "float" condition, the address multiplexorss on the CPM cart are exposed to a voltage that is neither a "1" nor a "0" (about 1.2-1.4 volts if memory servers). This has the nasty effect of causing the multiplexor (which cane be thought of as an amplifier for this purpose) to oscillate since it has an "illegal" voltage on it's input. Well if the address lines oscillate or don't stand perfectly still during certain parts of the DRAM cycle (particuarly the Row Address Setup and Hold and the Column Address Setup and Hold) ALL bets are off. Several rows (or columns) of data get selected inside the DRAM all at once causing contention and other nasty things to occur and there goes the DRAM data integrity. I don't know of anyway to fix it shy of changing brands of multiplexes on the CPM cartridge or putting some strong pull-ups on the address lines to fight the "amplified bus float" that comes out of the Z80. This design flaw was so nasty that there was no question in my mind about building a Z80 into the C128 to do CPM.... Relying on the design flaws in the cart simply wasn't an option. BTW, a year later I found an Apple schematic tucked in a drawer of the Z80 cartridge for the Apple II. Turns out an early CBM engineer had copied the design, flaw and all, from apple. Bil Herd -----Original Message----- From: Richard Atkinson [mailto:rga24@hermes.cam.ac.uk] Sent: Monday, March 20, 2000 9:55 PM To: Bil Herd Subject: Re: Calling Bil Herd Bil, Another topic that has recently cropped up on cbm-hackers is the Commodore CP/M cartridge. Can you remember any of the specifics of why the cartridge wouldn't work with later C64s, and maybe even what could be done to it to correct this? We're trying to find a hack so that CP/M cartridges can be made to run on all machines. Regards, Richard - This message was sent through the cbm-hackers mailing list. To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tcm.hut.fi.
Archive generated by hypermail 2.1.1.