Scor­pio News


July–September 1987 – Volume 1. Issue 3.

Page 24 of 67

More various points.

Dear Sir,

I’m very pleasantly surprised – Scorpio News was out on time (if not before) for the second occasion. Do keep it up! To judge from your editorial, however, the revival in fortunes of 80-BUS publishing will be short-lived – which would be tragic for all concerned.

Could I thank all those readers who kindly contacted me with offers of circuit Diagrams and handbooks in response to my request? So far, 12 people offered help and I have a11 the information I need. Perhaps I’m biassed, but I can’t imagine this level of response in other computer circles.

Regarding the possible extension of the magazine to include PC look-alikes – not a bad idea if people can be persuaded to write suitable articles and for others to purchase the magazine. The main problem is going to be how to extend your circulation once the PC articles come pouring in – and another is going to be how to get articles of general (or even specific) interest to the existing or potential audience which aren’t done already by glossier publications (although most of these tend to restrict themselves to software reviews). It’s worth trying, however, even if it does offend the purists (shades of an earlier debate in INMC80?).

I have a particular interest since I would like to know how to doctor the BIOS of my Amstrad so that I can fit and use 80 track drives – one gets used to the 784k of the Gemini QDDS and the 360k currently available is pathetic. Obviously, one needs to have one or both drives software switchable to 40 track. Any volunteers?

One of the best features about this magazine and its predecessors is De Dark’s Diary. Always interesting and usually very enjoyable...... but – the most recent episode seems to have been a bit hard on one or two of us (sob!). So he doesn’t like Fortran – well, perhaps he could write a guide to Pascal – which he knows a lot about. I would welcome a decent guide to an excellent language which has so many adherents (and so would others to judge from the letters poses in the same issue). I use Fortran because most of the programs I need are written in it – and it isn’t worth the hassle of translating 30,000 or so lines of Fortran IV or Fortran 77 into Pascal. Perhaps I might be able to do it if I sampled some of his anchovy wine!

Of course the Dobbs Benchmark program is silly – no one in their right mind would seriously want to fiddle around with tangents and arctangents around the 90 degree mark – a quick look at a table of tangents would show how rapidly the value changes. The point of the program and the article was to demonstrate how well, or otherwise, various interpreters and compilers coped with such a problem and how fast or slow they ran on a range of machines which might be of interest to readers. The general point is still valid – a machine/​language combination which took a long time to produce a result or which produced a poor result could probably be relied upon to give a poor performance in other applications. I know that this is a generalisation but herd experience tends to bear it out. I did appreciate the explanation of the poor result obtained with HiSoft Pascal – which does explain why virtually no other Pascal or Fortran IV compiler has TAN as a supplied function - but it is available in some implementations of Fortran 77.

Well, how about it, Chris? I’m sure that the Editor could find a few pages (or even more) for the first instalment of your guide to Pascal.

Yours sincerely, P.D. Coker, Orpington, Kent.

Page 24 of 67