Doctor Dark’s Diary – Episode 11

Matters arising.

I was delighted to read, in the first issue of 80-Bus News that all articles are to be paid for, although I don’t grudge the ten I have churned out for almost free, over the past however long it has been. The thought occurs to me, however, that this encouragement is bound to result in a great increase in the number of items submitted, and if they are better than mine, I shall no longer appear! So, without further ado…

In my last article, I said that my CP/M version of Nas-Sys 1 would only send to disk, and recover, files of up to 16K. This turns out to be nonsense. CP/M is much more clever than I had thought, in some ways, and just opens a new “extent” on the disk when necessary. So even the simple disk operating routines I produced for MONITOR.COM work with large files. The only time a problem could occur is if a disk read or write error occurs whilst reading or writing the first record of an extent other than the first one. If this fairly unlikely event takes place while writing to the file, it will make a mess of the file. It would be possible to write a version of my extra code that could not foul up in this way, if there was any demand for it. So let me know, folks, if you regularly use MONITOR.COM for huge files, and have been getting inexplicably strange results. In case you are wondering if MONITOR.COM is of any practical use, in the light of these horrifying revelations, I am in fact using it now, to produce this article, using my botched version of Naspen (see last article for how to convert Naspen to work to a screen at 0F800H). Of course, if this gets printed, I may be able to afford Diskpen (or is it called Gempen? The adverts don’t seem to discriminate between the two at all.) Or maybe a free copy of Gempen will just turn up magically… [Ed. – No chance! And to answer your question, Diskpen runs on a Nascom, under CP/M, using the 48x16 display; Gempen runs under RP/M or CP/M on a Nascom or Gemini with the Gemini GM812 Ivc (80x25).]

The address of Aid to Industry Systems, who made the EPROM emulator board I use as a programmable character generator, is:–

_ _______ _____, Yate, Bristol, ____ ___

Now you will be able to write to them direct, and save me a spot of postage! I have not seen their advertisement in the glossy magazines recently – I hope they are still in operation, because their board is a useful one, and is reasonably priced as well.

My thanks are due to Dave Hunt, for his answers to my question about using CB to communicate with other Nascoms. I had suspected that it might not be practical, and have been trying to come up with some sort of alternative to using CB. Perhaps it is possible to put the tape interface signal onto the telephone system by means of a small speaker, and a microphone? One thing is certain; no system anyone cares to invent will cost as much as a British Telecom modem does…

I don’t suppose any of you are at all surprised to hear that the Pilot interpreter I was writing for use with CP/M fell by the wayside. I became interested in something else for a few weeks, and when I returned to the job I found that I had forgotten how it was supposed to work, even though the source file was full of comments. In the event of there not being letters of protest about this situation, I shall possibly not write any more about Pilot interpreters, although I will certainly not say definitely that the project is abandoned. The language seems to be of no interest at all to most teachers, for whom it was designed, as they are all learning BASIC and using some peculiar machine with an owl on it. The fact that BASIC is not in the least suited to their purposes does not deter them in the least. A volume of the CP/M user library is devoted to Pilot interpreters and the like, if you are still interested, although I have not seen them. My next attempt to write the definitive version will be written in Pascal, I suspect, as this is more sensible than doing it in assembly language. (I was originally going to ignore P. J. Brown’s advice, in “Writing Interactive Compilers and Interpreters”, on the grounds (or excuse) that being in machine code, my version would run fast. See later for why this is no longer a problem.)