Archive Magazine CD 1995/Archive Magazine CD 1995.iso/text/hints/volume_05/issue_01

Text File | 1995-02-16 | 22.1 KB | 465 lines

Hints and Tips
5.1
•   FileUtils − Anyone using Ben Summers’ FileUtils should note that
<ctrl-shift-F2>, <ctrl-shift-F3> and <ctrl-shift-F4> will still call up
the FileUtils routines even when, say, working in Impression. This can
give a nasty shock if you are, for example, trying to insert preset
styles using <ctrl-shift-F4> and a “change filetype” dialogue window
pops up. Mike Hobart, Cambridge.
5.1
•   Installing Desktop C − If you have recently upgraded from C to Desktop
C, you may have a little difficulty installing Desktop C onto your hard
disc. When you first try to run the ‘Install’ program, it will stop
abruptly half way through the second of the four discs, with no clues as
to what has happened. This is because the ‘Install’ program tries to
create a directory called $.User. Cmodule, which already exists from
your previous installation of C and so the program crashes. The solution
is simple, just delete the old $.User.Cmodule directory before attempt
ing to install Desktop C.   D R Kennard, London.
5.1
•   LBP-4 CaSPL − Switching between Video output (for the CC Hi-Res board)
and CaSPL output of the LBP-4 doesn’t appear to be documented in the
manual. Here is how it can be done:
5.1
From Video to CaSPL mode:
5.1
1.   Hold down “Menu” for 3 seconds until “Mode=Video” is displayed.
5.1
2.   Tap the “Menu” key once to display “Mode=CaSPL”
5.1
3.   Tap “Enter”. The LCD will read “01 Ex Exit” then “00 Ready A4”
5.1
4.   Tap the “Off-line” key to take the machine off-line (if the green
on-line indicator is lit).
5.1
5.   Tap “Feeder Select”.
5.1
6.   Tap “Menu”. The menu as described in the manual is now displayed.
5.1
From CaSPL to Video:
5.1
1.   Take the printer off-line.
5.1
2.   Hold down the “Menu” for 3 seconds − “Mode=CaSPL” is displayed.
5.1
3.   Tap “Menu” again − “Mode=Video” is displayed.
5.1
4.   Tap “Enter”.
5.1
5.   You are now back in Video mode.
5.1
Mike, Bainbridge, Derby.
5.1
Impression Hints & Tips
5.1
•   Adding a frame at beginning of a story − It is not immediately obvious
(well, it wasn’t to me!) how to add extra frames into a story that runs
through a number of frames. Suppose you have two frames, B and C (I’ll
get to A in a minute) − perhaps they are two main columns on this page −
and you want to split column B into two frames leaving a gap in the
middle. What you do is firstly shorten frame B using the frame handle at
the middle of the bottom of the frame. Then you use <ctrl-I> and create
a new frame wherever you want it under frame B (let’s call it frame B2).
Now click on frame B with <select> and then on frame B2 with <adjust>
and the text will flow into it so that it runs B−B2−C.
5.1
Now suppose you want the text to start in a frame before frame B. (Let’s
call it frame A.) The technique is to create the new frame, A, wherever
you want it using <ctrl-I>. Then click on frame B with <select> and then
on frame A with <adjust>. The text will now flow into it so that it runs
B−A−B2−C−etc. (N.B. This won’t work unless frames A and B are on the
same page because you cannot flow text backwards across a page bound
ary.) Then click on B and use <ctrl-X> to delete it. The flow is now
just A−B2−C−etc. Finally, you create a new copy of frame B by using
<ctrl-V>, click on frame A with <select> and the new B with <adjust> and
the flow will be A−B−B2−C−etc.
5.1
•   Dual window problems solved? − A few issues ago, I mentioned the
problem that if you open a second window on a document and then expand
and contract that window (so that you can alternately use two different
scale views) each time you shrink view 2 using the icon at the top right
of the window, it doesn’t stay at the place in the document that you
have just been working on. It always seemed (in my case) to go back to
somewhere earlier in the document. No one has offered any solution but
it seems to me that it goes back to wherever you last added a frame into
that window. So if you want to “keep your place” in the second window,
use <ctrl-I> to create a new frame at that place in the document and
then use <ctrl-X> to delete it again. Crude, but it seems to work. (No,
it doesn’t, not always. Drat, I thought I had solved it! Come on,
someone must be able to solve this one − I’ll give a free Archive mug to
the first person to solve it. Ed.)
5.1
• Equasor and matrices − The biggest drawback Equasor has for me as a
mathematics teacher is its inability − at least in version 1.00 − to
handle vectors and matrices.
5.1
There is, however, a way in which a very credible matrix can be created
− with a small amount of extra work:
5.1
I build the columns as multiple fractions, and then enclose everything
in brackets; I drop the file into Draw, remove the fraction bars and
then align the brackets with the numbers.
5.1
Jochen Konietzko, Köln, Germany
5.1
•   Special characters − Now I’m sure you ALL knew this, but I didn’t
until I read this month’s PipeLine article. The Alt key is extremely
useful for inserting special characters in the text. I had discovered
that <Alt-space> (that’s holding the Alt key down and tapping on the
space bar) gave a ‘hard’ space where you don’t what the words (e.g. the
halves of a postcode) to be split across the end of one line and the
beginning of the next, and I had discovered the idea of using Alt
together with the numbers on the numeric keypad to produce special
characters, but how about this...
5.1
     a² + b² = c³
5.1
I produced that in seconds without using superscript. The ² is produced
with <Alt-2> and the ³ with (yes, you’ve guessed) <Alt-3>. In other
words, you hold down the Alt key, as if it were the shift key, and type
2 or 3. Here is a table of the ones I have found but note that this only
applies to Trinity font which I am using. Other fonts may be different
but, generally, they tend to stick to using the same symbols for the
same ASCII numbers which the Alt key produces. The third column is the
ASCII number produced.
5.1
<Alt-1>   ¹   185
5.1
<Alt-2>   ²   178
5.1
<Alt-3>   ³   179
5.1
<Alt-9>   ±   177
5.1
<Alt-0>   °    176
5.1
<Alt-C>   ¢   162
5.1
<Shift-Alt-C>   ©   169
5.1
<Alt-M>   µ   181
5.1
<Alt-R>   ®   174
5.1
<Alt-S>   §   167
5.1
<Alt-X>   »   187
5.1
<Alt-Y>   ¥   165
5.1
<Alt-Z>   «   171
5.1
<Alt-`>   ¬   172
5.1
(top left key under <esc>)
5.1
<Alt-,>   ×   215
5.1
<Alt-.>   ÷   247
5.1
<Alt-space>      160 (hard space)
5.1
If you have found any that I have missed, perhaps in other fonts, let us
know. Ed.
5.1
• Spell checking Edit documents − Whenever I typed a longish Edit text
file, I used to regret the absence of a spell checker. For Impression
users, this poses no problem:
5.1
Simply drop the Edit file in question into an empty Impression frame,
have Impression check the spelling, then export the text back into Edit
by saving just the text story without styles. You can drag the text
story onto the Edit icon to have another look at it or save it in the
usual way directly to a disk. The import/export takes just a few
seconds, so this is nearly as good as a spell checker built into Edit
itself. Jochen Konietzko, Köln, Germany  A
5.1

5.1
Silicon Vision
5.1
New
5.1

5.1
Colton
5.1

5.1
4.12 p21
5.1

5.1
Matters Arising
5.1
•   Archimedes’ Tutor − Simon Anthony has produced a Beginner’s D.I.Y.
Introduction to the Archimedes using the !Hyper application on Shareware
Disc 19. If anyone is interested they should contact him directly at 62
Rutland Road, West Bridgford, Nottingham, NG2 5DG.
5.1
•   ArmSort − Fast Array Sorts − The author replies... The reviewer
(Ashley Bowden in Archive 4.12 p53) seems to have created a problem with
the flag values. The ArmSort documentation DID clearly state that for
the CALL sort “The parameters must all be variables, but may be of any
name”. If CALL sort,test%() ,128 is used, then it is Basic that objects
with “Syntax Error”, ERR=16, because it is a Basic requirement that all
CALL parameters are variables not constants. Thus it is rather unfair
for the reviewer to claim that to have to use variables is “.. very
unorthodox, and mention should be made in the documentation.” ArmSort
errors are comprehensive and are very clearly documented.
5.1
This ‘problem’ wasted about a third of the review, which could have been
better used to mention the PROCsort and SWI interfaces or the ability to
sort on absolute numeric values, or with null strings low or high.
5.1
May I also suggest that all software reviews should clearly state what
version is being reviewed: the ArmSort review copy was Version 2.12,
sent to Archive in February. The current Version is 2.51, which has the
following improvements: sort time reductions of between 30% and nearly
50%; the ability to sort strings from any position, or by length; the
ability to specify a key sequence; new procedures for a fast binary
search of sorted arrays; and all programs now do run easily from the
desktop. All users have had free upgrades and seem happy with the
product. Martin Avison, Avisoft.
5.1
(Sorry if we seem to have reviewed an out-of-date product but when
someone sends software to us for review, we have to assume it is a
“stable” version. If the software has gone from version 2.12 to version
2.51 since February, this indicates continuous development. This is fine
but, if significant changes are made, the onus is on the software
supplier to send us the latest version. This product was still on the
“software to be reviewed” list in the May edition of Archive. If Martin
had noticed this and sent us the latest version at that stage, the
difficulty might have been avoided. Ed.)
5.1
•   MidiVision Update − Koen Lefever has sent us an updated version the
MidiVision program on Shareware Disc 33. It is now compatible with the
new Acorn MIDI Podule ROM and capable of detecting real time messages.
The new version has been put on this month’s program disc.
5.1
•   PRES Disc Interface & DFS − In the review in Archive 4.11 p27 we
referred to difficulties in reading 40 track BBC discs. PRES suggested
that the problem may be due to “residual data” on the discs. In other
words, they are suggesting that the problem may lie with discs that have
been written onto by different 40 track drives, some with narrow heads,
as in 40/80 drives, and some with wider tracks as in true 40 track
drives. It would be interesting to hear other people’s experiences on
this. We have also found that it is sometimes necessary to slow the head
stepping time right down as low as it will go in order to read 40 track
discs. (We also used a rather old address for PRES. It is currently P.O.
Box 319, Lightwater, Surrey, GU18 5PW. 0276−72046, fax 51427.)
5.1
•   RISC-OS Dreaming − I would like to reply to Oliver Gunasekara’s
letter, RISC-OS Dreaming, published in Archive 4.12 p15. Many of
Oliver’s suggestions have been extensively discussed in the USENET news-
group dedicated to Acorn machines. Well-informed contributions from
Acorn staff (in a private capacity) and experienced programmers (RISC-OS
and otherwise) allow fairly definite predictions to be made as to the
likelihood of some of these suggestions being realised in later version
of RISC-OS.
5.1
Firstly, there is almost no chance of proper demand paged virtual memory
being implemented under RISC-OS on current generation ARM processors.
The problem is that much (if not all) of RISC-OS runs with the ARM
processor in its supervisor mode. In this mode, the ARM-2 and ARM-3 are
unable to implement virtual memory. Not enough information is saved to
enable execution to be resumed after an attempt to access memory not
physically present causes the MEMC memory controller to interrupt
execution. Thus, virtual memory would require a drastic rewrite of RISC-
OS so that any code that might access virtual memory runs in user mode
or avoids memory not physically present. Acorn almost certainly do not
have the resources to do the necessary rewrite. Simpler, less effective
forms of virtual memory such as swapping out non-running programs are
possible, but would still be very complicated.
5.1
Pre-emptive multi-tasking is, on the other hand, quite possible. It is
just that doing it without introducing subtle bugs into software written
under the assumption that multi-tasking is co-operative is fiddly. As
with virtual memory, the fact that RISC-OS did not allow for pre-emptive
multi-tasking, when originally written, makes it messy to implement
later on. Even running other tasks when the current task is blocked,
waiting for i/o, is rather tricky to implement though, according to the
grape-vine, this does indeed happen to some extent in RISC-OS 3.
5.1
As to nicer system fonts, better design, TCP/IP file-sharing and hot-
linking − let’s hope it happens. There’s no technical obstacle and these
are definitely things necessary for Acorn to keep up with the state of
the art. A new file-system allowing more files in a directory and longer
filenames would also go down well. It would be a real boon to people
like myself who have to use RISC-OS machines in conjunction with UNIX or
MesS-DOS boxes.  Andrew Stevens, Overton.
5.1
•   RISC-OS Dreaming − I would agree with many of the points about RISC-OS
2.0 put forward (in Archive 4.12 p15), by Oliver Gunasekara. However,
I’m not sure that making RISC-OS look more like other graphical
interfaces will necessarily bring us much benefit.
5.1
I am responsible for several different computer systems (micros and
minis) each of which has its own user interface from the quirkiness of
the Apple and Atari systems, to the smooth, powerful X-Windows systems.
Some of the staff prefer one system, some prefer another. It’s a bit
like choosing your favourite car, you get the one that suits you.
5.1
Strangely enough, considering its sales, the least popular GUI in our
department is Windows 3.0. Even allowing for its deficiencies, RISC-OS
seems to have become the preferred choice for more than half our
department. Whether this is due to RISC-OS itself, applications like
Impression or the speed of the ARM3 powered machines, I’m not sure. What
I am sure of is that RISC-OS is quite capable of winning converts from
other systems.
5.1
Even so, let’s hope that RISC-OS 3 contains a lot of those improvements
mentioned by Oliver. D Hutchinson, Sheffield.  A
5.1



The Engineer Speaks...
5.1
Ray Maidstone
5.1
For the last few months, Ray Maidstone has been, effectively, the N.C.S.
Service Centre. He has done a great job fixing dead or dying Archimedes
computers for us and upgrading the memory of A310s. In doing so, he has
come across various recurring problems which could be avoided if a few
precautions were followed.
5.1
Here are some hints & tips which should help to keep your Archimedes in
tip-top condition. A word of warning though, some of the things I am
suggesting should not be attempted unless you really feel competent to
do so. The general rule is, if in doubt, ask for help!
5.1
•   300 and early 400 series keyboards − (See also the article on page 37)
These keyboards do not have sprung contacts coming together, but foil
closing over multiple contact points on sprung pads. Various people have
informed me that they have sprayed their keyboards with switch cleaner
to improve things when they have had intermittent contacts on one or
more keys. They may have got away with it, but it is my view that these
keyboards should never be sprayed with switch cleaner. The only
recommended method of servicing this item is to dismantle (but ONLY if
you feel competent to do so!) and remove foreign particles with a soft
brush and a vacuum cleaner. Anything sticky or difficult to remove is
best left to your Service Centre. WARNING: Using switch cleaner which
contains an organic solvent can destroy components within the keyboard
and slowly but surely render it useless. Be warned! (We have a dead
keyboard available for inspection if you are in any doubt! It was killed
by using Tandy’s Tix Electronic Switch Cleaner. Ed)
5.1
•   Mouse − If you unplug your mouse, the recommended procedure for
reconnection is to first disconnect the keyboard from the computer, plug
the mouse into the keyboard then reconnect to the computer. The reason
for this is that the mouse contains one solitary chip, which has no
protection from spikes at switch-on from the outside world, whereas the
keyboard has circuitry that is slightly better equipped to cope with any
misadventures. Plugging the mouse into the keyboard and then plugging
the keyboard into the computer is the safest method and could well save
you money.
5.1
•   Monitor smear − On the 400 series computers, it is possible to get a
composite video signal out of the sync socket which you can use for a
mono monitor. This is done by making links LK5 and LK6. However, if
these links are made, it can cause problems with certain high resolution
monitors. You get a certain amount of smearing on the ends of horizontal
lines. This effect is caused by the electronics of the Archimedes and
can only be cured by removing the links. (I have a confession to make.
Until we discovered this problem, we always tested our 400 series
computers using a mono monitor. We added links 5 and 6 but did not
remove them before sending out the computers. If you have had a 400
series computer from us, I suggest you check for this effect and remove
the links if necessary. Ed)
5.1
•   Video RGB outlet − On the 310 and early 400 series, it is possible to
destroy the VIDC chip by plugging in the video lead to either the
computer or the monitor if one or both of these items are switched on.
The electronics of this socket do not accept the slightest static
variation and any deviation will damage the VIDC chip. Whoops! − there
goes another £30! I have designed a multiple diode module as a service
modification which upgrades these old machines to the new static-
protected specification. Whilst Acorn confirm that my modification is
“Archimedes-legal”, they will not put their name to it as yet. This
modification has been particularly well received by schools, as it stops
pupils being able to blow up their computers!
5.1
•   Fan filters − With the fan driving air into the filter (lower half of
the diagram below) it is possible for the filter, in its blocked state,
to completely eliminate all airflow within the computer. However, with
the fan sucking air through the filter (upper half of the diagram
below), even if the filter becomes totally blocked, residual air
currents will still flow within the computer, providing some “last
resort” cooling. Also, with the filter on the outside of the fan,
cleaning couldn’t be easier. All that is needed is to slide the lid back
10 cm (or 4 inches for those of us still in Imperial!) and suck the dust
off the filter with a vacuum cleaner.
5.1
•   Hard drive parking − After much debate and the dismantling of two
unserviceable hard drives, I have found that, in order to give your data
maximum protection at closedown, *BYE must be followed by *SHUTDOWN.
These two commands, although slightly similar, do not seem to perform
the same task and, to my satisfaction, have been proven to be necessary
in the order stated. The effects of closing the drive down can be
verified acoustically and visually, e.g. *BYE will produce one staccato
tick noise and a short flash on the hard drive LED, but this does not
mean the drive is fully parked. Typing *SHUTDOWN will produce a multiple
wink of the hard drive LED and a triple tick noise from the drive. Take
my word that this guarantees total shutdown of the drive. If *SHUTDOWN
is typed without having typed *BYE, a single tick will be heard and the
drive will simply have moved the head to a different track, but it will
not be fully parked. (Different effects happen on different drives under
this condition.)
5.1
(Some may call this a “belt & braces” method, but when it comes to
protecting several hundred pounds worth of hard drive, I prefer to err
on the side of caution. Ed)
5.1
•   Removable hard drive parking − The command DISMOUNT brought up by
clicking <menu> on the SCSI icon, does not mean your drive is parked. It
simply means that the drive has shut its motor down and that the
software has “forgotten” about the drive in question. At this point, if
you switch the power off, the heads have been left out over the disc at
the last track you were using and will be in the same position next time
you switch on the power. In order to eliminate the possibility of power-
on destroying any data bits, the centre button on the removable drive
must be pressed and the drive fully disengaged with the lever. This
operation mechanically parks the heads off the disc and is the only way
to guarantee this.
5.1
•   Stray metalware − When removing the motherboard on the 300/400 series,
it is very possible to dislodge one or more of the spring speed nuts
(for the two rear self tapping screws that “get in the way” when
removing the board). Make sure these are firmly in place when reassem
bling − better still, remove them, clamp them slightly more closed and
refit. This will prevent any further mishap which could cause shorting
under the main PCB.
5.1
•   Stuck floppy discs − It is unfortunately possible, now and again, for
a floppy disc to become stuck in the drive. This will be due to one of
the following causes, and if you wish to keep costs to a minimum, NEVER
tug at, or forcibly remove, the stuck item.
5.1
(Again, we have u/s floppy drives available for inspection if you need
proof. Ed.)
5.1
1. The metal sliding cover on the disc can become widened, perhaps due
to being bent in the post, or being sat on or whatever. As the metalwork
of the slider opens, it will act like a ratchet allowing the disc to go
in but not to come out without difficulty.
5.1
2. The spring that returns the metal cover can become unlaced and behave
like a fish hook, snagging the unlocking post within the drive.
5.1
3. The metal screening plate on the upper disc reading head within the
drive can become dislodged and catching the plastic work of the disc
(particularly on the older Sony drives).
5.1
In ALL cases, if the drive is removed by somebody competent to do so,
you should find that delicate easing will allow the disc to come out
again without damaging the drive and, 9 times out of 10, the disc can
also be saved.
5.1
In the case of the metal head shield of the Sony drive, simply throw it
away as it was later found to be unnecessary. Where the metal gate of
the drive is distorted, it is quite acceptable to remove this item by
unfolding it and removing the spring. The disc will operate without the
metal cover but keep little fingers out − remember the 5¼“ floppies?  A
5.1