80-Bus News


March–April 1983, Volume 2, Issue 2

Page 36 of 55


has a good approach to programming, in that he recommends that you build up a library of macros, rather than re-invent the wheel each time you need a particular routine, and he supplies the code for the actual macros. He shows how to read and write disk files from machine code programs, and gives a good explanation of how the directory works, which is something most authors avoid like the plague. There are several appendices, mostly useful, and a good index too. So, what is the catch? Well, remember that the professor has been using CP/M for a long time? He started out with the 8080 processor, and all the example programs and macros are in those funny mnemonics that we modern Z80 fans can’t stand. You could use the indices to translate from one to the other, if you wanted to: if you have the assembler he recommends (the grossly, absurdly over-priced Macro 80), it will accept either sort of mnemonics. Even with this disadvantage, this is still the best book on CP/M that I have seen so far. I was preparing a series of articles about using CP/M routines in your own machine code programs, but I won’t bother.....


This one is on sale at WH Smiths, if you can get past the hordes of children fingering all the computer magazines they stock (about 500, from the look of things), reading the blurbs on all those garish cassette cases under the- S*ncl**r Sp*ctr*m (which is busy demonstrating itself to an audience of baffled shoppers, most of whom seem to think it is Channel 4 television they are watching). It is actually in the shelf with the top ten paperback books of the moment, and costs a mere £1.95, which is a nice change from computer book prices in general. It is a description of how a firm called Data General rushed out their 32 bit machine in an impossibly short time. The reviews I had read about this book made it sound as if they burned out the brains and bodies of hundreds of their staff in the process, but having read it, I think they had it pretty cushy. Only one bloke actually left, and he stayed in the business, in spite of the note he left, saying he refused to deal in any unit of time shorter than a season. The engineers in particular seem to have lived a life of comparative luxury. Did you ever hear of a logic analyser being used on a Nascom RAM A board? They don’t know they are born, as the saying goes. Mind you, I found it a good read. There are rude words in it, but only one, in various forms, since American swearing seems pretty monotonous and is rather unimaginative. Nobody says swut, turlingdrome, or (heaven forbid) Belgium! Almost forgot, it’s by Tracy Kidder, and has a p-picture of a P-Penguin on it.


I said I would do some sort of benchmark program to find out what sort of difference using virtual disks makes to the speed things run at. True to form, I have not yet done things, but I have an estimate. Loading the 32K of an Adventure game from disk takes about 8 seconds on my system, although I believe some of the more recent drives than my Pertec will do it faster. Loading the same 32K from virtual disk takes under half a second. It should be pretty obvious from this that any program that does updates of files on disk will speed up in a dramatic way, so I tried out a program called Marvin. Hey Presto! The main bottleneck, or one of them any way, disappeared, as the file update went through in about two seconds. Even the accompanying Zap program goes acceptably fast, when run from the virtual drive.


As if the facilities provided by the Pluto were not sufficiently incredible, there is an upgrade ROM available, which adds some extra commands to those already available. It costs £60.00, however, which is pretty steep for an EPROM and a revised manual. The manual is a distinct improvement on the earlier one, thank goodness. However, for people like me, it would have been a good idea if they had said where the new ROM goes. Do not put it in the nice empty socket, because that is not where it goes. Take out the old 2732 ROM, and put the new 2764 in that socket, with the new chip’s pin 1 in the hole

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

Page 36 of 55