You could make the circles '?'marks - for all the undocumented opcodes! I agree that standard 16 bit 64k addressing should not have kept the processor less capable processors, both with "only" 64k and both received much more acceptance than the 6809. I do think that a natural progression in the 6809 family (had it continued) would have been an integrated segmented or bank switched memory manager - this could have been worked quite nicely into the 09 architecture. I 1/2 agree (with each of you) - the original CoCo (and even the CoCo2) were somewhat limited, and more importantly packaged (CoCo1) and marketed as a cheap home console computer. Not so much a game console (although the cartridge slot does suggest this), but as a cheap "toy" computer (that happened to play a bunch of cartridge games). The CoCo3 was a marked improvement, however it was too little/too late, and as no one else had really picked up the CPU, it got the reputation as an "oddball". I do agree on the RS drive - the drive cartridge used a standard floppy controller chip (and you could use standard off-the-shelf drives) and interfaced directly to the 09 bus through the cartridge slot - much better/faster then the custom serial drive offerings by the "other guys". But, I also think that having to use a cartridge for the drive seems a bit goofy - why not have implemented a drive interface native to the machine? I designed and built my own 6809 based workstations - both the homebrew ones shown on my web page, and a couple of custom bus/racked limited production designs for a couple of "big" companies as in-house test platforms - These "serious" systems were remarkably powerful for their time, and gave a truer impression of the 6809's potential than the CoCo which hid that power beneath a flimsy and toyish looking case, an inadaquate keyboard (CoCo1) and artifically limiting system software (for example, the CoCo ROM's required DP to be "Page Zero" - and hogged most of it). Speaking of multitasking, there's a really nice feature of the 6809 that is often ignored, however I used this in a number of 6809 based products that I designed to implement very fast simple multitasking. This feature is the "Direct Page" register! - Unlike other CPU's which use "zero page", the 6809 has DP which means that you can place your "zero page" at any 256 byte boundary. This register is also saved and restored during a full interrupt, meaning that in a multitasking system, each task can have it own "zero page". This makes for compact code (many tasks never need to use extended addressing for memory references). Here's an example of a ultra simple task swapper in only 5 instruction, in which each task runs in round-robin fashion, requiring only 3 bytes in each tasks "page" - there is no system task control blocks/tables: ; Interrupt saves all registers (can be SWI for volentary swap) INTHND: STS <MYSP ; Save current task stack pointer LDA <TLINK ; Get next tasks DP TFR A,DP ; Switch to new DP :-( can't load DP directly) LDS <MYSP ; Restore new task SP RTI ; Load new task registers & launch Ah - such memories! Regards, Dave -- dave04a (at) Dave Dunfield dunfield (dot) Firmware development services & tools: www.dunfield.com com Vintage computing equipment collector. http://www.parse.com/~ddunfield/museum/index.html