On 2012-11-24, at 09:29, Gerrit Heitsch wrote: >>> No, the next step would be to design one such circuit for a 250469 board, the one where the BASIC and KERNAL are in one ROM and only the char-ROM is extra. >>> >>> Should be a bit easier since _KERNAL and _BASIC are already added together by an AND gate on the board. >> >> Ironically this seems to do more harm than good for the glue ;-) >> >>> That way one doesn't need a 74LS11, a 74LS08 should do. >> >> I save one input on the _OE/_CE gate but then have to handle three states instead of two to distinguish between BASIC and KERNAL.. which so far seems to me like I can't make it with the same EPROM content layout and chip count I had before. With shifted content I came up with something like (still have to double-check for correctness): >> >> http://dl.dropbox.com/u/58002657/cbm/c64/rom_adapter_narrow0.png >> >> Can you have a look? Maybe you have some better ideas, especially ones that would allow to keep the old EPROM content layout? > > I'm not quite awake enough yet to solve logic gate problems, :-) I think I know what you mean. > but here are some hints to think about: > > - You have A14, it is HIGH if KERNAL or CHAR are selected and LOW > if BASIC is selected. Yes, I treat it as poor man's _BASIC signal ;-) > - It might be necessary to invert A14 to get back to the original > design. A full 74LS04 would be Overkill. Use a simple inverter with > a BC547 or BS170 and 2 resistors. Yup, that's where it comes to what I wrote about doing more harm than good. I haven't yet found way to keep the content layout and not increase the chip/part count at the same time. > - Now you have A14 and _A14. That should allow you to easily > discriminate between KERNAL and CHAR by combining it with the > _CHAR signal. > > - A14 (and) CHAR = KERNAL (high) > > - _A14 (and) KERNAL = BASIC (high) I was considering this too but the preferred objectives are: - to have the same (or lower) parts count in both versions (reasons obvious) - to have the same EPROM content layout in both versions (so that only one, variant agnostic, version of the EPROM is to be built) I would change the content map if the wide board version could be redesigned so that it would handle the mapping needed for the narrow board version without parts count increase, rather than add more parts. -- SD! Message was sent through the cbm-hackers mailing listReceived on 2012-11-24 10:00:13
Archive generated by hypermail 2.2.0.
