80-Bus News


November-December 1984, Volume 3, Issue 6

Page 7 of 55

rmstecscennass i RSMMC wen on oe


NASIO Nasties!

Thanks to Adrian Perkins’ article in Volume 3 Issue 2, I have stopped a mysterious fault that was occuring on my system. I would go into great detail about the various combinations of boards that would or would not work with the two different CP/M’s I use, but it is so complex to explain that I prefer not to go on about it in too much detail. Suffice to say that I was blaming my Belectra board when it was not to blame. On putting the old Nascom 1/0 board back into use, I found no more trace of the fault. I suspect, and that is as far as I get with my limited hardware knowledge, that the GM809 disk controller board I am using, which I thought was supplying NASIO, is not doing it in the recommended way. I don’t understand all this open collector business too well, so I may be wrong. Perhaps an expert on that board will enlighten us?


Nice to see that Adrian has also fixed the bugs in MONITOR.COM for us. I knew about the fault where the last block of anything was loaded twice, but didn’t bother to find out why it was happening. The reasons for this slapdash approach? Well, it was doing what it was supposed to do, and that was load in the Nas-Sys based games I wanted to run, and the extra block was doing no harm. My children, I caution you against spending all your time improving the operating system software, when you could be writing application programs for money, or even better, playing adventure games! And no, I didn’t do my kludges in 8080 code, Adrian. I did them on paper in Z80 assembly mnemonies, because the program was too small to bother with using assemblers, linkers and the like. I like programming in Z80, and when there were only Nascom l“s with IK, I used to do all my programming directly. Don’t be misled by the fact that the block moves are not used where the two sections of RAM get interchanged. If there was a Z80 operation that would swap two blocks of memory, I would have been delighted to use it. Incidentally, if anyone is interested, I have a labelled, commented disassembly of my code on disk. I sent it out on the Ring of Iron, but that seems to have gone into hyperspace. If anyone wants it for the purpose of making improved versions of Nas-Sys for themselves, or indeed everyone, then send me a disk (Pertec drives, DDDS).

More thoughts about monster programs.

Having written up my ideas on chaining programs in Hisoft Pascal, I was thinking about ways of writing groups of large programs, when I realised that straightforward chaining may not be enough. Suppose you wanted control to be returned by a program to the program that had activated it, no matter what that program was called? It seemed to me that what was needed was some sort of equivalent to GOSUB in BASIC, or procedures in Pascal. It would be nice, I decided, if recursion was possible. All this thinking made my brain hurt, so I did a bit of drinking instead, and soon realised that a file that behaved like a stack was needed, along with a couple of procedures called "call" and “return”.

What "call" would have to do would be as follows. Create a file called "$$$.SUB" and put the name of the program to be called in it, complete with all the extra, unexplained spaces I mentioned last time. Add the name of the currently running program to the "top" of the stack file. End the current program.

This is an OCR’d version of the scanned page and likely contains recognition errors.

Page 7 of 55