HP150-II

The trivial anwaer is 'to add a maths coprocessor' :-) My inital thoguth was that the machine was designed without thinking of the comprocessor, so the 8088 was used in minimum mode. Since that can't support a coprocoessor, and since the changes to the interface logic between the 8088 and the system buses are considerable if you change it to maximum mode, the only sane thing to do is to put another 8088 on the card, run that in maximum mode, add the logic too, and disable the one on the mainboard so that the 8088 on the coprocessor card takes over. The only problem with that theory is that all 150-IIs have the coprrocessor board connector. IMHO it would have been simpler to put the logic on the mainboard, run that 8088 in maximum mode, and just have a socket for the 8087 (like on an IBM PC). Another explanationg that breaks down is that they wanted to keep the circuitry from the origianl 150 (which never officially accepted a coprocessor). Problem with that idea is that so much of the machine was redesigned going to the 150-II, it would have been easy to redesign this bit too. Who knows? Incindentally, doea anyone (else) have an HP150-II with na Elmer board? The story of that is that one of the gate arrays -- codenamed Elmer -- wasn't finished in time, so early machines have a PCB containing standard chips (real time clock, baud rate generator, TTL for the handsahke line ports, etc) in place of it. It's the same sort of size as the 8087 board, and fits alongside it (standing vertically). -tony