Answers: What do you do with a hard disk which won't rotate, is an acoustic coupler really useful and what became happened to Britain's PC industry. Frank Leonhardt has some answers.
It worked fine when it came out of his machine, but for me it won't even spin up to speed.
The Jumpers appear to be set correctly, the lead seems okay and other IDE drives work with the same interface.
Does this drive (a 3204) have a reputation for funnies? Will it ever work or am I wasting my time with this paperweight?
Robert Neuschul, London
Many winchester disk drives, especially those made the late '80's, have a problem with the lubricant used on the surface of the platters. After a few years this became rather sticky and this ended up in the heads getting stuck on the drive.
When the drive is running the head flies above the platter, never actually touching it (in theory). The lubricant is there to prevent damage when the head crashes. Although nearly all modern drives park automatically whenever the power is removed so that the heads should never land on the platter they may work free while the drive is being transported, or if it is jogged and end up getting stuck in the lubricant.
If you are lucky a light tap on the drive case will get it spinning again. If this fails, I have had much success by moving the head arm manually, though the manufacturers would definitely not approve. You may simply rip the heads off. As a last resort it is possible to open the drive (in a clean room environment) and prise the heads away with a screwdriver. This became necessary with an old Priam which had been left on a shelf for two years, giving the heads chance to stick properly - surprisingly it worked afterwards and I am still using it now.
The best solution would be to replace the platters - a specialist company like DataStore (0734 712431) can do this for you but it is only economic with larger drives.
Conner, who's drives were badly effected by this problem, claim to have developed a better lubricant which has been in use for some time now, though no-one can predict how any new piece of technology will survive over the years.
Frank Leonhardt
Nice computers, but they used Apricot's own version of DOS, and therefore didn't catch on. Were they called the Apricot xi range, or am I thinking of another machine?
MH
Gone but not forgotten, I have one under my desk as I write. The original machine was the Apricot PC which had twin floppy disks (3.5"); this was supplemented by the Apricot Xi which had one floppy and one hard disk.
They were not IBM hardware compatible but they did run MS-DOS. Notably features included a faster processor than IBM's, 3.5" drives while IBM used 5.25" and a much smaller and neater CPU box. Expansion was catered for with two euroconector slots (none of
this nasty edge-connector stuff) and they had integrated serial and parallel ports when most IBM clones needed an extra card. The monochrome graphics were a respectable 800x400 pixels displayed on a very crisp monitor.
Unfortunately market forces forced Apricot to go the route of IBM compatibility and the British company able to produce these exceptionally well designed and innovative machines ended up turning out bog standard clones.
Frank Leonhardt.
Antosh Gazel, Reading
The general consensus is that this problem, and any other teething troubles, have been cured and that most users are satisfied with their machines.
Michael Hewitt, Carshalton Acoustic coupler are modems which can be connected to the telephone network by placing the handset of a normal telephone into a cradle containing a loudspeaker for the mouthpiece and a microphone for the earpiece. They were necessary when British Telecom took a very dim view of anyone actually connecting wires to their network, but since the introduction of modular jacks this is no longer a problem and all modem users were glad to see the end of them.
They are still made, however, and these days they consist of a tiny speaker and microphone which care held against the handset using elasticated straps. They also work better than they did in the old days too - 1200 baud is quite reasonable but V22bis is a bit unreliable.
I have seen kits, on sale in America, designed to allow you to connect normal modem to hotel telephone systems. They consist of a screwdriver, a trailing phone jack and some crocodile clips of various sizes. I am just including this information for the sake of completeness, of course.
Frank Leonhardt
Computer Answers can be contacted on:
leo2@cix.compulink.co.uk
uab1018@dircon.uucp
Telephone 081-429 3047.
As floppy disks accumulate, I have found that a very quick way of locating the most current disk for each section or subject, from amongst a box of disks, is to use the small "POST-IT" Tape Flags. These are obtainable in four colours, so that the raised p
ortion is in red, blue, green or yellow and the lower part, which is transparent, does not obscure any writing on the disk label. As a disk fills up the tab is simply lifted and repositioned on the next one, in such a way, of course, that it sticks out well clear of the disk drive slot.
The coloured tab can be written on with a felt top pen - with a data, reference or title. The flags are also useful for marking the target disk when doing a disk-to-disk copy.
Many households have a spare television stand, or even two, supplied when a new set is bought and never actually needed. These are very sturdy and of a shape which makes them useful as a monitor or printer stand for a crowded desk, leaving a high shelf with space underneath for storing disks, the keyboard and other items.
Erica Speel, Eastcote.
The POST-IT flags mentioned here are self-adhesive tabs designed for temporarily sticking to book pages or file index cards in order to mark an item of interest. They are made by 3M.
setbkcolor(WHITE); // Set background colour
cleardevice(); // Clear screen to background colour
However, this has a rather unfortunate side-effect - any BLACK lines you may subsequently draw will come out in WHITE! This is because setbkcolor() actually fiddles with the VGA palette, mapping colour zero (normally black) to the new background colour.
Borland suggested a solution to this problem was to re-map the palette to make the entry for white (15) into black as there was no point in having two whites on the palette after all. However a more elegant solution, and one which does not mess up any palettes, is to use:
setfillstyle (SOLID_FILL, WHITE); // Set for solid fill of white
bar (0,0,getmaxx(),getmaxy()); // Draw rectangle the size of the screen.
A similar approach can be used with Watcom's graphics library, which does exactly the same thing to the palette when you change the background colour.
If you are being forced to use Microsoft 'C' and are missing Borland's Integrated Development Environment with its editor which automatically locates erroneous lines in your source after a compile, there is a way you can retain this feature.
The IDE supports transfer programs from the system menu. These work by running a dos program file and piping the output, via a converter program, to the message window. You can theoretically take the error listing from any compiler and write a converter program to massage it into Borland's format. Borland supply some example source for doing this but no actual documentation; it is not that easy to work out.
However, the converter supplied for tasm (Borland's assembler) just happens to work quite well with the output from Microsoft 'C' too, as well as MASM 5.1.
To add a transfer program select "Transfer" from under the "Options" menu. Click on the button to add a new item and enter the name of the program to run as "nmake". The command line could usefully contain the following:
$SAVE ALL $CAP MSG(tasm2msg)
This saves all the modified files which are open and runs nmake, returning all the errors in the IDE's message box. If you want to run just the compiler, the macros $SAVE CUR and $EDNAME will be useful. $SAVE CUR saves the file in the current edit window and $EDNAME expands to the name of the file so it can be included in the program's arguments.
$PROMPT causes the IDE to display a dialogue box with the expanded command line displayed for tweaking just prior to calling the transfer program. This is very useful for debugging the macro expansions!
Borland's on-line help does explain the transfer macros, though I had great trouble locating them in the printed manuals. Look under "transfer, macros" in the help index.
The final touch is to get the programs compiled under Microsoft to work with Turbo Debugger. This can be achieved by using the utility "tdconvrt" which converts an executable file with Microsoft CodeView debugging information into Turbo Debug format. Borl
and hardly mention tdconvrt except in a readme file, and it appears to have been omitted from Release 3.0 of the compiler. However, it works just as well with the latest version of the debugger so this is a little hard to understand. Unfortunately the results are not as good as with source compiled under Borland but anything is better than CodeView!
A. Hacker, Harlow.