I guess it depends what you want to have. FWIW, my dream machine (and my direction, 3 years so far!) is an S100 box with things like a 6502, Z80, 8086, 80386, 68xxx, an ARM of some type, perhaps Power PC.. with 16 MG of bus RAM, 4GB of DRAM (32 bits) connected/shared via a ribbon cable to CPUs that that can use it all in master/slave configurations and sharing XVGA, (and other video boards) , IDE CF cards, and old floppy disks etc. Operating systems like CPM, CPM86, CPM68K, MSDOS, Linux… As to speed, it should be possible to run the 32 bit CPU’s at dual speeds, one when they (rarely) use the lower 16MG and normally (> 16M) at their max speed on the overhead bus. Reasons for doing all this --- STRICTLY A HOBBY. No commercial interests, great learning experience and sense of accomplishment, but thats it. John John Monahan Ph.D e-mail: mon...@vitasoft.org Text: mon...@txt.att.net From: n8vem...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of yoda If I would have known about it I would have been there and there were some changes that definitely that could have made to make it simpler. I know the trouble of going to 32 bits and am working a couple of solutions that might pan out. The problem going beyond 68010 is there is not data bus sizing as I recall so you can't use 8 or 16 bit memory but must go to 32 bits - I think trying to jam that on the S100 bus is probably not the way to go as it is not spec'd for 32 bit. Having daughter cards does not really make sense either as the S100 bus then just becomes a slow I/O bus for the 68xxxx. An SBC is probably the way to go then - which is the strategy that Andrew and I are working on. The S100 bus just doesn't make sense for more modern processors Where were you guys when we started this 68K board. In retrospect we probably should have! The board started out as a direct copy of the Wilcox book board. That board was a bus master (only). The next step was to convert it to a master/slave combination thereby adding to its utility. This actually involved a lot of modifications and additions. I think if we were going to do anything we would next go to a 68020 or 30 much as we are doing with the 8086/80286 to 80386. However I can tell you going to true 32 bits with >16M memory access at > 20MHz clock speeds is not as easy as it sounds. Hopeful Andrew’s tracer will arrive home soon. The board will allow decent development work to get started. Remember the 68010 is a direct plugin. Personally I am worried all the traces come our right -- it’s been a long time in the oven. I hope people appreciate there is always risk with these boards coming up duds! John John Monahan Ph.D e-mail: mon...@vitasoft.org Text: mon...@txt.att.net From: n8ve...@googlegroups.com [mailto:n8ve...@googlegroups.com] On Behalf Of yoda As another thought - why not substitute a 68 pin PLCC version of the 68K (MC68000FN12) and an AT29C1024 16 bit 44pin PLCC for the eeprom. These would drastically reduce the density on the board an might make the trace routing simpler. The 16 bit eeprom would make the programming simpler as well - just a thought. I have both in hand so it would not be a problem and they are reasonably priced as well. Dave How close are we on the S-100 68K??? There is a lot of pent up demand for that board - especially me as I work on the other 68xxx system - I am trying to write platform independent code and would like to test it on the new 68K board as well. Thanks Dave Hi Well I was hoping to get the S-100 68K and S-100 80286 CPU board PCBs ordered next but they are not cooperating. The S-100 68K CPU board is nearly complete and should be done soon. It has taken much longer than I expected though. I would like it to complete trace route optimization before ordering PCBs.
Another idea is the S-100 bus terminator board. Basically there are 10 builders willing to get these boards “as is” which is the full sized PCB with some prototyping area. Nothing fancy but useful for terminating a legacy S-100 backplane. However these have *NOT* been prototyped so there is some additional risk. I think it is minor though since the S-100 backplane has this exact circuit and it works fine. Ideas? Thoughts? Comments? Questions? All welcome! Please lets discuss what the next steps should be. Thanks and have a nice day! -- -- |