Things you did with ZX-81's
Artemis
"Steve Wood" <scw...@home.com> wrote:
> I remember reading an article on the resonant frequencies of certain
> ZX81 internal components. The article included theories about how
> to broadcast modulated AM tones using these resonances.
>
> I got my ZX81 to play a flatulent, dirge-like version of "Daisy" over
> a nearby AM radio. I remember being pretty impressed at the time.
It would have been a better trick if you could have made the
16K memory pack work worth a darn. Uncle Clive liked to
boast he could do for a nickel waht any fool could do for
a dollar. I think he use a nickel plated connector (sheesh!)
Still, there was a lot you could eke out of the ZX81 (aka Timex 1000).
I heard that foreign governments of the eastern block were buying
them up like popcorn. But given the poor performance of the
memory pack I can see why the company went under. Too bad.
Artemis
>
> --
> Steve "But then I got an Apple IIc and everything changed" Wood
> http://members.home.net/scwood
> Vi Veri Vniversum Vivus Vici
>
> David Perrin <da...@reggie.demon.co.uk> wrote in message
> news:G4SqSAAL...@reggie.demon.co.uk...
> > >One of the most impressive hacks I ever saw was chess on a 1k ZX81 -
> > >no castling, only two openings and not a strong player, but never mind
> > >the quality, feel the narrowth.
> >
> > After playing with my ZX81 for a bit I managed to write a 10x10 life
> > program in 47 bytes.... although you were limited to a setups using a
> > random seed generator. Gosh, the winter evenings just flew by in the
> > early 1980's......
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
Don Stokes
>> I got my ZX81 to play a flatulent, dirge-like version of "Daisy" over
>> a nearby AM radio. I remember being pretty impressed at the time.
BTDT. I found that tweaking the FAST/SLOW modem fast enough (read: from
machine code) could make it play tones in rather the same way as one
made sounds by toggling the speaker on an apple ][. 'course it sounded
pretty awful and had the distinct disadvantage of losing the video
display, but it played a sufficiently recognisable tune to astonish
friends.
Artemis <artemis...@mailcity.com> wrote:
>It would have been a better trick if you could have made the
>16K memory pack work worth a darn. Uncle Clive liked to
I found that if you went over the rampack's connector with a flat blade
screwdriver, bending the edge connector "pins" to have more grip, the
rampack would stay put. You had to leave the thing there -- if you took
the pack off, you needed to fix the cheaparse connector again.
>a dollar. I think he use a nickel plated connector (sheesh!)
Not that I recall. Just copper.
>them up like popcorn. But given the poor performance of the
>memory pack I can see why the company went under. Too bad.
The attitude that cheaper was better did kill the company. But the
ZX81 and Speccy were resounding successes, despite the cheapness.
It wasn't until the QL that things started to go wrong, but it was
the with C5 electric car where wheels finally came off (the company,
not the car).
Clive sold everything to Alan Sugar (Amstrad), including the right to
use the name Sinclair. Hence the (mildly successful) Z88 coming out
under the name Cambridge Computer.
Where is Uncle Clive now?
-- don
BKoons2489
there after! What really made this little guy so unreliable was the power
supply. If I ever get motivated to use it again I`ll wire in a new more stable
supply source.
ACME
euph...@freenet.co.uk
On 1999-06-05 d...@news.daedalus.co.nz(DonStokes) said:
:I found that if you went over the rampack's connector with a flat
:blade screwdriver, bending the edge connector "pins" to have more
:grip, the rampack would stay put. You had to leave the thing there
:-- if you took the pack off, you needed to fix the cheaparse
:connector again.
I had a Memotech rampack for my ZX81. Not only was its case made of
sheet aluminium rather than the usual moulded plastic, but it was the
same width as the ZX81 (and sculpted to fit neatly onto its back, and
not much higher) and came with a couple of bits of extremely sticky
Velcro to attach the thing in place. The Velcro was necessary, too.
Memotech brought out their own computer, eventually. It was a very neat
looking affair, and wouldn't have looked out of place as the keyboard of
a nice big CP/M box. Evidently Memotech agreed, as they sold as an
add-on a huge box called the FDX, as wide as the MTX512, with space for
two disk drives and an 80 column card; that turned the MTX into a CP/M
box. I desperately wanted one. (In fact I still want one - anyone got
one lying around?)
:Where is Uncle Clive now?
Apparently he's suddenly woken up to Linux, and wants to make a
RISC-based portable (maybe Z88-scale?) using it. So that'll be along in
about 2 years, then...
--
Communa -- you know soft spoken changes nothing
barnacle
I came across (mid 80s ??) a thing remarkably similar to the zx81/timex called
a 'jupiter' (IIRC) which ran Forth as an OS. It fitted into the same
size/shape box, and was apparently put together by an ex-Sinclair employee.
There was a time when you could buy the PCBs and cases with no chips on, bt
not the circuit diagram.
Not having access to the original ASIC chip he'd redesigned the whole thing,
with a much nicer video interface using a minium of chips (why use a
multiplexer when a resistor pack will do...) and I'm not even sure that he
still used a Z80. I think it might have been a 6502 or 6800.
This was such a nice little box that I (ahem) reverse engineered it and grew a
few of my own with bigger memory to use as various system controllers.
--
barnacle
Phil Guerney
news:7jcvru$vt8$1...@quince.news.easynet.net...
>I'm not even sure that he
> still used a Z80. I think it might have been a 6502 or 6800.
No - it was a 3.25MHz Z-80A.
If anyone in the UK (or elsewhere!) can sell me a Jupiter Ace I would be
thrilled. It was one of the micro's I had my heart set on when I started to
collect these things, but the Ace has proved elusive in Australia. Postage
should not be too great - it sure was not heavy!
Just slightly hoping for a reply.
Phil
in Brisbane, Australia.
Roberto Waltman
>a 'jupiter' (IIRC) which ran Forth as an OS.
It was a "JupiterAce" and it did use a Z80.
You can find the schematics, ROM dumps and an emulator for
it at the following URL's:
http://www.home-micros.freeserve.co.uk/JupiterAce/JupiterAce.html
http://users.aol.com/autismuk/ace/index.htm
RW.
Alexios Chouchoulas
"Phil Guerney" <gue...@net.uq.edu.au> writes:
> If anyone in the UK (or elsewhere!) can sell me a Jupiter Ace I would be
> thrilled. It was one of the micro's I had my heart set on when I started to
> collect these things, but the Ace has proved elusive in Australia.
It's equally elusive in the UK, unfortunately. I've been looking for
one for the past four years, but to no avail. It's such a cute little
micro, very anti-Sinclair in its white case.
Alexios
--
o88 o888o o888o --------------------------------------------------------------
88o8' `88' `88 Alexios Chouchoulas ale...@dai.ed.ac.uk
88' ,88' ,88' (aka The Unpronounceable One) University of Edinburgh
o8888888888888888 "Eimai mia micri soupiera" Artificial Intelligence
barbara_n
<artemis...@mailcity.com> wrote:
>It would have been a better trick if you could have made the
>16K memory pack work worth a darn. Uncle Clive liked to
>boast he could do for a nickel waht any fool could do for
>a dollar. I think he use a nickel plated connector (sheesh!)
>
The 64K memory pack (bought as a kit and had someone build
it for me) worked ok, as long as it didn't come loose. Then
transferred the operating system to upper memory and could actually
write a decent program. Sort of.
Barbara N.
Reply to need...@cvconline.com
Alan Thompson
it when it came out. 3Kb RAM originally, but I got a 16KB add on pack
when they were going out of business and flogging off as much as they
could. Sorry, I'm not selling. I have a long term plan to see if I can
make my PC talk to the thing. Mind you, maybe I should check and see
if it's still working...
In reference to Communa's earlier post, I too had the Memotech rampack
and also lusted after the 'puter they brought out ... anodized
aluminium case, mmmmmm ... I think they stopped making it by the time
I had some cash together and I had to settle for a C64. Good thing I
didn't get an Oric;-)
--
The letter 'e' doesn't really occur in my e-mail address.
A.
On 6 Jun 1999 17:14:36 GMT, ale...@hestia.dai.ed.ac.uk (Alexios
Chris Hedley
esla...@mwahahahaha.com (Alan Thompson) writes:
> I had some cash together and I had to settle for a C64. Good thing I
> didn't get an Oric;-)
I always had the impression that the Orics (in particular the later
models; the Atmos?) were supposed to be Quite Good tm, but, in common
with many of the other companies that sprung up at the time to vend
their new and exciting hardware, they didn't have a clue about how to
run a business successfully. It's somehow sad reading all these posts
about the micros of the '80s, and I can only add my name to the long
list of people who say that things aren't as interesting as they used
to be hardware-wise; there was an incredible diversity of totally
incompatible systems, many inventive in their own way and with their
strengths and weaknesses and inevitably their own band of diehard
supporters. Anybody have a complete(ish) list of what was on the
market between, say, '80 and '84? I have vague recollections of
several machines that are never talked about anymore, such as the
Lynx (nice graphics, shame about the speed), the Elan/Enterprise/
whatever it was called in any particular week (by the time they'd got
the name together enough to release the bloody thing just about
everyone'd lost interest, and ISTR it was *far* too ambitious a
project for the then already aged Z80-whatever CPU), various Commodore
things like the Plus-4 which looked interesting on paper but were
again a bit too late; er, suddenly I realise I can't remember half
as many micros as I thought I could...
Chris.
Adam Atkinson
>Anybody have a complete(ish) list of what was on the
>market between, say, '80 and '84?
No. But I haven't seen anyone mention the UK101 or the Nascom 2 in
this thread yet. (I had a Dragon 32 at the time, though what I really
_wanted_ was obviously a BBC Model B.)
A little earlier than this, people I knew mostly had Apple II,
Commodore PET, TRS 80, and Acorn System 1.
--
Adam Atkinson (gh...@mistral.co.uk)
You mean, you'll put down your rock, I'll put down my sword, and
we'll try to kill each other like civilized people?
Chris Hedley
"Adam Atkinson" <gh...@mistral.co.uk> writes:
> No. But I haven't seen anyone mention the UK101 or the Nascom 2 in
> this thread yet. (I had a Dragon 32 at the time, though what I really
> _wanted_ was obviously a BBC Model B.)
The UK101 and Nascom 2 were slightly before my time; they always brought
images to mind of guys in lab-coats wielding soldering irons and
oscilloscopes... I also had a Dragon 32 at the time (I should really
get around to rescuing it from my parents' loft) and could never really
make up my mind about the BBC; being a natural born cynic, I wondered if
it really lived up to its hugely overhyped reputation, and when I finally
got an Electron decided it probably didn't, but it was still a groundbreaking
machine of its time.
> A little earlier than this, people I knew mostly had Apple II,
> Commodore PET, TRS 80, and Acorn System 1.
The only one of those I've actually encountered was the PET. I quite liked
it, in spite of its discs not seeming to offer any performance gain over
a cassette unit! Weighed a bloody ton as well, as I recall.
Chris.
Tony Duell
: In article <375be471....@client.ne.news.psi.net>,
: esla...@mwahahahaha.com (Alan Thompson) writes:
: > I had some cash together and I had to settle for a C64. Good thing I
: > didn't get an Oric;-)
: I always had the impression that the Orics (in particular the later
: models; the Atmos?) were supposed to be Quite Good tm, but, in common
AFAIK (and the pair of machine I have confirm this), the Oric 1 and Atmos
were electrically the same. The keyboard was different (the Atmos had one
you could actually type on ;-)), and there may have been differences (bug
fixes?) in the ROM.
It was almost a 6502-based spectrum. The main chips (from memory) were :
6502 CPU
AY-3-8910 (?) sound chip + some I/O ports
ULA (video, system control, etc)
ROM
either 16K or 64K RAM (4416 or 41464 DRAMs). IIRC only 48K of the latter
was useable
The display was a strange serial-attribute thing IIRC.
_The_ machine to look for was the Tiger (well, it would have been if they'd
sold many of them). It was also a Tangerine design, but it was somewhat
over-specced :
Z80(B, I think) + 64K RAM
68B09 + 8K RAM
NEC 7220 Graphics copro + 96K video RAM (3 32K bitplanes)
Prestel modem
Centronics, RS232, Econet-like network, cassette
Full keyboard
All in a case about the size of a Beeb... Oh well. Never hear of them now...
: with many of the other companies that sprung up at the time to vend
: their new and exciting hardware, they didn't have a clue about how to
: run a business successfully. It's somehow sad reading all these posts
: about the micros of the '80s, and I can only add my name to the long
: list of people who say that things aren't as interesting as they used
: to be hardware-wise; there was an incredible diversity of totally
Fortunately there are some eccentrics around the world who are preserving/
restoring these old machines...
: incompatible systems, many inventive in their own way and with their
: strengths and weaknesses and inevitably their own band of diehard
: supporters. Anybody have a complete(ish) list of what was on the
: market between, say, '80 and '84? I have vague recollections of
: several machines that are never talked about anymore, such as the
: Lynx (nice graphics, shame about the speed), the Elan/Enterprise/
: whatever it was called in any particular week (by the time they'd got
: the name together enough to release the bloody thing just about
: everyone'd lost interest, and ISTR it was *far* too ambitious a
: project for the then already aged Z80-whatever CPU), various Commodore
: things like the Plus-4 which looked interesting on paper but were
: again a bit too late; er, suddenly I realise I can't remember half
: as many micros as I thought I could...
Don't forget the Tandy CoCo3 (not common in the UK). It's a sort-of
Dragon with up to 512K RAM, 2MHz 68B09, 80 column text display, 640*192*4
colour graphics, etc. Runs OS-9, and can just about support a couple of
users.
Or the Torch XXX, a Unix workstation (68010-based) with a macintosh-like
desktop frontend (but in colour), and a BBC micro-like 1MHz bus for
expansion. No I am not joking.
Or the Gemini Galaxy, which was based on the NASCOM bus. It supported a hard
disk, and could be a network server. Most of them had 2 Z80s inside, one
for user programs (it ran CP/M) and one on the (text-only) video card.
You're making me feel old...
: Chris.
-tony
Adam Atkinson
>could never really
>make up my mind about the BBC; being a natural born cynic, I wondered if
>it really lived up to its hugely overhyped reputation, and when I finally
>got an Electron decided it probably didn't
The Electron's not really the same as a BBC B. I had access to plenty
of BBC Bs, and I promise you they were nicer than Dragon 32s. Though
6809 machine code was nicer than 6502. (I obviously had to use machine
code on the Acorn System 1 since, well, what else was I going to do
with it?)
>> A little earlier than this, people I knew mostly had Apple II,
>> Commodore PET, TRS 80, and Acorn System 1.
>The only one of those I've actually encountered was the PET. I quite liked
>it, in spite of its discs not seeming to offer any performance gain over
>a cassette unit! Weighed a bloody ton as well, as I recall.
Indeed it did. I don't think the 4k "chiclet keyboard" PET had disks.
I remember a cassette player being part of the keyboard, or something.
I suppose it was an improvement on TTYs with paper tape, which we used
at school at the time. I wonder if I still have "Cells and Serpents"
on paper tape somewhere?
--
Adam Atkinson (gh...@mistral.co.uk)
Viva la pappa col pomodoro! (G. Stoppani)
Pete Fenelon
> I always had the impression that the Orics (in particular the later
> models; the Atmos?) were supposed to be Quite Good tm, but, in common
> their new and exciting hardware, they didn't have a clue about how to
> run a business successfully.
The firmware was bugged to hell in the Oric-1 and the calculator-style
keyboard was bloody awful - worse than some of the rubber monstrosities
about. The serial-attribute system made graphics even more idiosyncratic
than the Spectrum, and the software was... hmmm... well, there weren't
many true classics.
The Atmos was a fair bit cuter - most of the more gruesome bugs had been
ironed out and the keyboard was tolerable.
Tangerine paid the price of coming into the market a bit too late with a
product that wasn't quite good enough. Sure, when the home micro boom was
on it was possible to make a good living selling the sixth or seventh most
popular platform, but when the home micro market shook out in the UK
(Spectrum, 64, niche market for the BBC and Amstrad) there was no room for
"weird" machines. Economies of scale reduce diversity in "mature" markets...
It's a pity, because Tangerine used to build lovely machines in the
elder days -- the Microtan 65 could build up into one of the more
impressive 6502-based hobbyist/kit systems... I expected far more from
their "home micro"...
pete
--
pe...@fenelon.com `there's no room for enigmas in built-up areas' HMHB
Pete Fenelon
> _The_ machine to look for was the Tiger (well, it would have been if they'd
> sold many of them). It was also a Tangerine design, but it was somewhat
> over-specced :
> Z80(B, I think) + 64K RAM
> 68B09 + 8K RAM
> NEC 7220 Graphics copro + 96K video RAM (3 32K bitplanes)
> Prestel modem
> Centronics, RS232, Econet-like network, cassette
> Full keyboard
Haven't seen one of them for years -- they used to be not uncommon
in travel agents' (probably cheaper than getting a CP/M crate and a
Prestel terminal?) Loved the multi-coloured function keys. They were
certainly classier than the tacky mono Z80 CP/M boxes people tended to
have in business at the time!
> Or the Torch XXX, a Unix workstation (68010-based) with a macintosh-like
> desktop frontend (but in colour), and a BBC micro-like 1MHz bus for
> expansion. No I am not joking.
Lovely machine. I can't remember what the third X was for (X11, X.25 and
something else...). For a while a TripleX was the X.25 gateway for my
then-employers...
> Or the Gemini Galaxy, which was based on the NASCOM bus. It supported a hard
> disk, and could be a network server. Most of them had 2 Z80s inside, one
> for user programs (it ran CP/M) and one on the (text-only) video card.
I acquired one of these a few months back. Real nostalgia trip... had
a fun weekend playing with WordStar and doing some assembly-language
hacking :) Alas, it was taking up far too much space so I swapped it
(and a TRS-80 model III) for a SCSI box...
Tobias Goeller
> I found that if you went over the rampack's connector with a flat blade
> screwdriver, bending the edge connector "pins" to have more grip, the
> rampack would stay put. You had to leave the thing there -- if you took
> the pack off, you needed to fix the cheaparse connector again.
I fixed the "crash because of loose rampack"-problem with some
two-component-glue. It still lasts :-))
CU
Tobias
--
COM.BOX WINET \|/
Tobias Goeller ({o.o})
\./
t.go...@combox.de U
bme...@bruce.cs.monash.edu.au
>In reference to Communa's earlier post, I too had the Memotech rampack
>and also lusted after the 'puter they brought out ... anodized
>aluminium case, mmmmmm ... I think they stopped making it by the time
>I had some cash together and I had to settle for a C64.
And it was such a lovely machine --- a built in assembler and machine code
debugger in ROM. Solid as a rock, too. I had a few short circuits on that
thing without anything breaking. And once we forgot the power supply for
our 16 processor Z80 system when going to an show, and so the poor MTX-512
had to support it as well as itself. It couldn't quite handle 16 processors,
but I think we got 12 going. Each processor had an LED connected to it, and
if too many of those were on, the MTX lost its video output, but when the
LED's switched off, everything was back to normal. I loved that machine.
I still have 1.5 of them in storage (well, my original was dumped on the
ground by an IBM employee, and as it had been wire-wrap-extended to use a
HD64180, static memory, several serial ports and a shared memory interface
to the 16 processor system, it didn't like that. So IBM's insurance paid for
me to get a used one from someone else --- this was at the time when
they were already very hard to get).
Anyone remember "NODDY"?
Bernie
--
============================================================================
"It's a magical world, Hobbes ol' buddy...
...let's go exploring"
Calvin's final words, on December 31st, 1995
Chris Stratford
"Adam Atkinson" <gh...@mistral.co.uk> writes:
> The Electron's not really the same as a BBC B. I had access to plenty
> of BBC Bs, and I promise you they were nicer than Dragon 32s.
They were also twice the price (399 for a BBC Model B, 199 for a
Dragon 32), so they should be a lot nicer.
Chris.
--
Chris Stratford Email: Chris.S...@uk.uu.net
UK Postmaster Voice: +44(0)1223 250690
UUNET Fax: +44(0)1223 250650
An MCI WorldCom Company WWW: http://www.uk.uu.net
Simon Craythorn
> this thread yet. (I had a Dragon 32 at the time, though what I really
> _wanted_ was obviously a BBC Model B.)
I had a Nascom 2 for a while around '82 ish. Still have a Nascom 1 in
the loft somewhere I think.
Simon.
David Given
Tobias Goeller <t.go...@combox.de> writes:
> Don Stokes wrote:
>
>> I found that if you went over the rampack's connector with a flat blade
>> screwdriver, bending the edge connector "pins" to have more grip, the
>> rampack would stay put. You had to leave the thing there -- if you took
>> the pack off, you needed to fix the cheaparse connector again.
>
> I fixed the "crash because of loose rampack"-problem with some
> two-component-glue. It still lasts :-))
I saw adverts in some ancient copies of _Your Computer_ for little strips
of ribbon cable with the appropriate plugs at each end, designed so you
could put your rampack on the table and connect it to the machine with the
cable. These were known as `dongle danglers'.
BTW, anyone remember the Aquarius, a machine I saw reviewed in said
magazine? They thought quite highly of the Lynx, too. And the TI99 (which
had 256 bytes of CPU addressable memory)...
--
+- David Given ---------------McQ-+
| Work: d...@tao-group.com | I think, therefore you are.
| Play: dgi...@iname.com |
+- http://wired.st-and.ac.uk/~dg -+
C Lamb
: In article <375D131F...@combox.de>,
: Tobias Goeller <t.go...@combox.de> writes:
: > Don Stokes wrote:
: >
: >> I found that if you went over the rampack's connector with a flat blade
: >> screwdriver, bending the edge connector "pins" to have more grip, the
: >> rampack would stay put. You had to leave the thing there -- if you took
: >> the pack off, you needed to fix the cheaparse connector again.
: >
: > I fixed the "crash because of loose rampack"-problem with some
: > two-component-glue. It still lasts :-))
: I saw adverts in some ancient copies of _Your Computer_ for little strips
: of ribbon cable with the appropriate plugs at each end, designed so you
: could put your rampack on the table and connect it to the machine with the
: cable. These were known as `dongle danglers'.
I think most people used the much more mundane solution of a sausage of
Blu-Tak between rampack and ZX-81 on the upper edge. This only had one
disadvantage - after 6 hours of playing Donkey Kong (remember the
little inverse video X that was you?) the Blu-Tak turned into something
resembling used chewing gum in a sauna.
regards
C
Adam Atkinson
>> The Electron's not really the same as a BBC B. I had access to plenty
>> of BBC Bs, and I promise you they were nicer than Dragon 32s.
>They were also twice the price (399 for a BBC Model B, 199 for a
>Dragon 32), so they should be a lot nicer.
And this is why I had to settle for a Dragon 32 even though I really
wanted a BBC B :-)
--
Adam Atkinson (gh...@mistral.co.uk)
I have a spelling chequer. It came with my pea sea. It plane lee
marques for my revue miss steaks eye can knot sea.
Chris Hedley
a...@p850ug1.demon.co.uk (Tony Duell) writes:
> AFAIK (and the pair of machine I have confirm this), the Oric 1 and Atmos
> were electrically the same. The keyboard was different (the Atmos had one
> you could actually type on ;-)), and there may have been differences (bug
> fixes?) in the ROM.
The Atmos certainly looked more impressive, if nothing else. :)
> either 16K or 64K RAM (4416 or 41464 DRAMs). IIRC only 48K of the latter
> was useable
With RAM at the huge premium it was in those days, I'm surprised that
there was 16K of redundant memory in it, unless it was used to shadow the
ROM to speed things up (I don't doubt you, though)
> _The_ machine to look for was the Tiger (well, it would have been if they'd
> sold many of them). It was also a Tangerine design, but it was somewhat
> over-specced :
I don't think I've even heard of that one, and I thought I pretty much had
my finger on the pulse in those days... oh well! Sounds a nice machine,
shame it never really made the "big time."
> Fortunately there are some eccentrics around the world who are preserving/
> restoring these old machines...
This month's PCW reports that Clive Sinclair is thinking about dusting off
his soldering iron and reentering the microcomputer market; he has ideas
about a custom-made ARM-based machine for Linux. It certainly has potential.
> Don't forget the Tandy CoCo3 (not common in the UK). It's a sort-of
> Dragon with up to 512K RAM, 2MHz 68B09, 80 column text display, 640*192*4
> colour graphics, etc. Runs OS-9, and can just about support a couple of
> users.
I only have vague memories of this one. Sounds like a really neat machine.
> Or the Torch XXX, a Unix workstation (68010-based) with a macintosh-like
> desktop frontend (but in colour), and a BBC micro-like 1MHz bus for
> expansion. No I am not joking.
Wasn't that the one that actually used the BBC/B motherboard for all the
I/O stuff, and was in essence a BBC with the 68000 Unix "tower" attached
to the 1MHz bus all in one case (complete with monitor, floppies and
stuff)? I wish I still had some of my old mags from those days. Perhaps
PCW will cover it again one day in their "retro" section.
> Or the Gemini Galaxy, which was based on the NASCOM bus. It supported a hard
> disk, and could be a network server. Most of them had 2 Z80s inside, one
> for user programs (it ran CP/M) and one on the (text-only) video card.
All this talk of the Nascom is starting to make me feel embarrassed. Time
to read up about it, even if I am 20 years late. :)
> You're making me feel old...
I'm making *me* feel old...
Chris.
Chris Hedley
d...@tao.co.uk (David Given) writes:
> BTW, anyone remember the Aquarius, a machine I saw reviewed in said
> magazine?
I have vague memories of it; ISTR it was regarded pretty much as a
"games console" of the time, but I stand to be corrected.
> They thought quite highly of the Lynx, too. And the TI99 (which
> had 256 bytes of CPU addressable memory)...
I only encountered a Lynx once, but quite liked it. Very nice
graphics, as I recall. As for the TI99, didn't that hold the rather
dubious accolade of being one of the slowest machines ever, in spite
of its then almost unprecedented 16-bit CPU? The weird memory model
might explain this if it's true...
Chris.
Chris Hedley
> 6809 machine code was nicer than 6502. (I obviously had to use machine
> code on the Acorn System 1 since, well, what else was I going to do
> with it?)
I'd agree that the BBC was definitely the better machine in practically
all respects. Its downfall was its price, and, in spite of its
advantages, it was massively overhyped; I think that this was most
evident by sales of the model A, which was so crippled (even moreso
than the Electron IMHO) that it just wouldn't have sold at that price
if it wasn't for the name.
When it came to "hands-on" stuff, after any novelty value had passed
its course, I probably made equally productive use of both systems.
> Indeed it did. I don't think the 4k "chiclet keyboard" PET had disks.
> I remember a cassette player being part of the keyboard, or something.
> I suppose it was an improvement on TTYs with paper tape, which we used
> at school at the time. I wonder if I still have "Cells and Serpents"
> on paper tape somewhere?
Perversely, I actually preferred the likes of the VT100; needless to say,
my career has consisted largely of doing server rather than client related
tasks. :)
Chris.
Rob Nicholson
> models; the Atmos?) were supposed to be Quite Good tm, but, in common
They were okay - I had lots of fun with them. The teletext/attribute based
hires screen was a pain.
> their new and exciting hardware, they didn't have a clue about how to
> run a business successfully. It's somehow sad reading all these posts
Backed by British Car Auctions I believe...
> supporters. Anybody have a complete(ish) list of what was on the
> market between, say, '80 and '84? I have vague recollections of
There's loads of web sites on old computers.
> several machines that are never talked about anymore, such as the
> Lynx (nice graphics, shame about the speed), the Elan/Enterprise/
> whatever it was called in any particular week (by the time they'd got
> the name together enough to release the bloody thing just about
> everyone'd lost interest, and ISTR it was *far* too ambitious a
The Flan was doomed from the start. Great idea, poor marketing and followup.
Rob.
euph...@freenet.co.uk
On 1999-06-07 esla...@mwahahahaha.com said:
:In reference to Communa's earlier post, I too had the Memotech
:rampack and also lusted after the 'puter they brought out ...
:anodized aluminium case, mmmmmm ... I think they stopped making it
:by the time I had some cash together and I had to settle for a C64.
:Good thing I didn't get an Oric;-)
tee hee :> I know people who swear by Orics. For some reason they were
very popular in France, and I believe a successor to the Atmos found its
way to market over there.
I got my Memotech long after I believed that the company had ceased to
exist, from a little place in Swindon that appeared to sell bankrupt
stock. I also got my first PC (an XT that is still alive today, bar the
serial port) from there, on the strength of it.
euph...@freenet.co.uk
On 1999-06-07 gh...@mistral.co.uk said:
:On 07-Jun-99 17:45:54, Chris Hedley said:
:>Anybody have a complete(ish) list of what was on the
:>market between, say, '80 and '84?
:No. But I haven't seen anyone mention the UK101 or the Nascom 2 in
:this thread yet. (I had a Dragon 32 at the time, though what I
:really _wanted_ was obviously a BBC Model B.)
I'll see you a UK101 (I saw a schematic of that, but no more) and raise
you a Laser 200 (sub-Spectrum computer with colour blocky graphics and
rubber keys, for about 69 quid) and a Comx 35 (did anyone ever have one
of these? 1802-based, with "the slowest version of Basic ever
encountered", according to one site).
:A little earlier than this, people I knew mostly had Apple II,
:Commodore PET, TRS 80, and Acorn System 1.
What was the first British computer to really make it big? I get the
impression that the BBC is very much Britain's Apple II, but other than
that... I'm only thinking of UK sales here.
Pete Fenelon
> I only encountered a Lynx once, but quite liked it. Very nice
> graphics, as I recall. As for the TI99, didn't that hold the rather
> dubious accolade of being one of the slowest machines ever, in spite
> of its then almost unprecedented 16-bit CPU? The weird memory model
> might explain this if it's true...
's'right -- the BASIC was also, if I recall rightly, double-interpreted!
(yep, the interpreter was itself interpreted!).
The weird memory architecture of the TI reminds me rather of the Siemens
(Infineon these days I guess) C16x family of microcontrollers...
Peter Maydell
>On 1999-06-07 gh...@mistral.co.uk said:
> :No. But I haven't seen anyone mention the UK101 or the Nascom 2 in
> :this thread yet. (I had a Dragon 32 at the time, though what I
> :really _wanted_ was obviously a BBC Model B.)
>
>I'll see you a UK101 (I saw a schematic of that, but no more) and raise
>you a Laser 200 (sub-Spectrum computer with colour blocky graphics and
>rubber keys, for about 69 quid) and a Comx 35
How about a CGL M5?
http://www.chiark.greenend.org.uk/~pmaydell/hardware/hallath/
Not very nice, not very common, but the Guttang Guttong puzzle/arcade
game is fun :->
Somebody's mentioned the TI99/4a already, which was my first computer.
I learnt to program on that primarily because virtually the only
source of games for it was BASIC listings you typed in yourself...
[and I learnt to curse the choice of FCN-= as a reset key. Reach for
shift-= (+) in the middle of a line of BASIC and find yourself staring
at the testcard 'press any key to begin' screen and cursing :->
http://www.chiark.greenend.org.uk/~pmaydell/hardware/tapeth/
You could actually expand the TI into quite a powerful machine.
I used to drool over descriptions in the user group magazines of
systems with the peripheral expansion box, twin disk drives,
32K of RAM and the ability to do machine code, RS232 and more.
It's just a shame the basic machine was so crippled.
Peter Maydell
Ariel Scolnicov
> On 1999-06-05 d...@news.daedalus.co.nz(DonStokes) said:
>
> :Where is Uncle Clive now?
>
> Apparently he's suddenly woken up to Linux, and wants to make a
> RISC-based portable (maybe Z88-scale?) using it. So that'll be along in
> about 2 years, then...
Do we get a power switch this time round?
--
Ariel Scolnicov |"GCAAGAATTGAACTGTAG" |ari...@compugen.co.il
Compugen Ltd. |Tel: +972-2-6795059 (Jerusalem) \ NEW IMPROVED URL!
72 Pinhas Rosen St. |Tel: +972-3-7658520 (Main office)`--------------------
Tel-Aviv 69512, ISRAEL |Fax: +972-3-7658555 http://3w.compugen.co.il/~ariels
Rob Nicholson
> of these? 1802-based, with "the slowest version of Basic ever
> encountered", according to one site).
Goodness - you mean they actually made the Comx!! I remember seeing it at a
show in London. Wasn't it advertised as "using the chip used in the space
shuttle"?
Rob
Tony Duell
: > either 16K or 64K RAM (4416 or 41464 DRAMs). IIRC only 48K of the latter
: > was useable
: With RAM at the huge premium it was in those days, I'm surprised that
: there was 16K of redundant memory in it, unless it was used to shadow the
: ROM to speed things up (I don't doubt you, though)
I'm almost certain it didn't shadow the ROM (not much point on a 6502
machine...). I remember there being 2 RAM sockets on the mainboard, and
that they could take 64K*4 DRAMs. I've never heard of a way to map RAM in
place of the ROM, so I think you really were limited to 48K.
I must admit that I've not seriously hacked inside an Oric for many
years. Maybe I should dig mine out again..
: > Or the Torch XXX, a Unix workstation (68010-based) with a macintosh-like
: > desktop frontend (but in colour), and a BBC micro-like 1MHz bus for
: > expansion. No I am not joking.
: Wasn't that the one that actually used the BBC/B motherboard for all the
: I/O stuff, and was in essence a BBC with the 68000 Unix "tower" attached
No, you're thinking of the Unicorn (which was a 68000 add-on for the
Beeb). There was also the Graduate which was an 8088 (PC compatible-ish)
BBC add-on
The XXX was a stand-alone workstation. The mainboard (called a Stickleback)
had the following main chips on it :
68010 CPU
68451 MMU
68450 DMA controller
1Meg DRAM
5380 SCSI interface
Z8530 serial chip (it could do X25 on one of the serial ports I think)
(Optional) 7990 + 7992 ethernet
6303 'service processor' (basically low-speed I/O and bootstrap)
16K byte boot ROM (aka 'caretaker' or 'simon')
146818 real time clock
6840 counter/timer
6850 ACIA
6845 CRTC
64K video memory (shared between the 68010 and the 6803)
Assorted TTL glue and PALs
It had _3_ expansion buses : SCSI, VME (not quite a full VME, but not
bad) and the BBC 1MHz bus (called X-bus on this machine).
The boot sequence was strange. The service processor halted the 68010 and
copied the 8-bit-wide boot ROM into the 32 bit wide video memory. Then
the 68100 started running that program and pulled the rest of the system
in from (SCSI) disk.
The machine's case was built in layers. THe bottom tray contained the PSU
and motherboard. On top of that you put one or more (you had to have at
least one for the PSU to fit) 'storage layers'. The standard one
contained a floppy drive (80 track), a hard disk (ST506 interface, 20 or
40mbytes, I think) and an OMTI 5000 controller to link them to the SCSI bus.
On top of that you could put a VME cage if you wanted one. There was a
'slimring' which had one spare slot for a single VME card and a
'quinring' which had 4 slots (I think - one of them was used for the
cable going back to the motherboard).
If you didn't have the VME system, you could add a cheaper RAM expansion
board called a 'Limpet' This plugged into 2 IC sockets on the motherboard
and cabled to the VME connector. It gave you another megabyte of RAM. If
you had the VME cage, you could put standard VME RAM cards in it.
The other strange feature was the power switch. It was a touch contact on
the front of the case. Touching it turned on a relay in the PSU (the
control circuit was powered from a NiCd battery that also backed up the
real time clock) and powered the system up. To turn it off, you touched
the contact again. This sent an interrupt to the service processor which
then told the main CPU to sync filesystems, etc. The service processor
then told the PSU to turn off. Strange.
-tony
Scott
<cbh@REMOVE_THIS.teabag.demon.co.uk> wrote:
>I only encountered a Lynx once, but quite liked it. Very nice
>graphics, as I recall.
>
I still have my Lynx. I love the thing! Some of the guys who worked on
the Amiga designed it. I always wanted to try developing for it.
Thought it might be fun. Just to see how the thing worked even.
>As for the TI99, didn't that hold the rather dubious accolade of being
>one of the slowest machines ever, in spite of its then almost
>unprecedented 16-bit CPU? The weird memory model might explain this
>if it's true...
>
I have a TI in my closet. Well, two of them actually. The memory was
accessible throught the video chip or something weird like that. I
never understood how it worked. All I know is that you could add this
honking big 32k memory expansion unit and not use it for most things.
--
--
Scott Maxwell - scottm25 (at) bigfoot (dot) com
My shrink? I just go to him for refills!
bme...@bruce.cs.monash.edu.au
>I'll see you a UK101 (I saw a schematic of that, but no more) and raise
>you a Laser 200 (sub-Spectrum computer with colour blocky graphics and
>rubber keys, for about 69 quid)
The really sad part about the Laser (aka VZ-200) series is that the gfx
chip inside could do a whole lot more than they used. IIRC, it could do
256x192 in <mumble> colours. But they only gave it 2k of memory (at 0x7000,
or 28672 --- scary, it's been 15 years or so since I last needed that
knowledge...), and didn't connect a few traces, so actually using any of
the higher modes wasn't possible.
Bernie
P.S.: My usual question: ANYONE out there who happens to have a copy of
"Wolf3" (run-through-maze-and-shoot-the-guards-but-be-careful-with-
those-grenades-they-can-trigger-chain-reactions type of game) lying
around? I wrote that a long time ago, and I'd love to play it again,
but it seems I have lost all my tapes in one of the many moves since.
Josef Möllers
[ ... ]
> graphics, as I recall. As for the TI99, didn't that hold the rather
> dubious accolade of being one of the slowest machines ever, in spite
> of its then almost unprecedented 16-bit CPU? The weird memory model
> might explain this if it's true...
"Memory model" might be the wrong term: "CPU architecture" might be
better: the 9900 series CPUs had their registers in memory. A base
pointer addressed the first register and context switch was simple: just
reload the base register ... Later models had some kind of cache, iirc.
I still have a manual of these chips buried somewhere.
Josef
--
PS Die hier dargestellte Meinung ist die persoenliche Meinung des
Autors!
PS This article reflects the autor´s personal views only!
euph...@freenet.co.uk
On 1999-06-08 pe...@fenelon.com said:
:> Or the Torch XXX, a Unix workstation (68010-based) with a
:>macintosh-like desktop frontend (but in colour), and a BBC
:>micro-like 1MHz bus for expansion. No I am not joking.
:Lovely machine. I can't remember what the third X was for (X11, X.
:25 and something else...). For a while a TripleX was the X.25
:gateway for my then-employers...
uniX?
David Given
pm...@watchdragon.demon.co.uk (Peter Maydell) writes:
[...]
_Your Computer_ reviewed this, but called it a Sord M5. But it's
definately the same machine, with the funky corner-missing keys.
[...]
> You could actually expand the TI into quite a powerful machine.
> I used to drool over descriptions in the user group magazines of
> systems with the peripheral expansion box, twin disk drives,
> 32K of RAM and the ability to do machine code, RS232 and more.
> It's just a shame the basic machine was so crippled.
When we say RAM, are we talking real (on-chip) RAM or fake (interpreted)
RAM?
--
+- David Given ---------------McQ-+ "...it's not that well-designed GUI's are
| Work: d...@tao-group.com | rare, it's just that the three-armed users
| Play: dgi...@iname.com | GUI's are designed for are rare." --- Mike
+- http://wired.st-and.ac.uk/~dg -+ Uhl on a.f.c
Ben Hutchings
: In article <7jjjiv$2og$3...@teabag.demon.co.uk>, Chris Hedley
: <cbh@REMOVE_THIS.teabag.demon.co.uk> wrote:
: >I only encountered a Lynx once, but quite liked it. Very nice
: >graphics, as I recall.
: >
: I still have my Lynx. I love the thing! Some of the guys who worked on
: the Amiga designed it. I always wanted to try developing for it.
: Thought it might be fun. Just to see how the thing worked even.
<snip>
Aren't you thinking of the Atari Lynx game system?
--
Any opinions expressed are my own and not necessarily those of Laser-Scan.
Ben Hutchings
<snip>
: : > Or the Torch XXX, a Unix workstation (68010-based) with a macintosh-like
: : > expansion. No I am not joking.
: : Wasn't that the one that actually used the BBC/B motherboard for all the
: : I/O stuff, and was in essence a BBC with the 68000 Unix "tower" attached
: No, you're thinking of the Unicorn (which was a 68000 add-on for the
: Beeb). There was also the Graduate which was an 8088 (PC compatible-ish)
: BBC add-on
<snip>
I remember using a computer based on the BBC, labelled "Torch" with a
couple of disk drives, which AFAICR was running CP/M. I think it had
extra memory (64K total?) as well as an extra processor. At the time
I knew nothing of operating systems, so I couldn't really be sure what
it was running. Is it possible that the 8088 add-on ran CP/M?
Mike Swaim
: In article <7jk5uk$j8r$1...@watchdragon.demon.co.uk>,
: pm...@watchdragon.demon.co.uk (Peter Maydell) writes:
:> You could actually expand the TI into quite a powerful machine.
:> I used to drool over descriptions in the user group magazines of
:> systems with the peripheral expansion box, twin disk drives,
:> 32K of RAM and the ability to do machine code, RS232 and more.
:> It's just a shame the basic machine was so crippled.
: When we say RAM, are we talking real (on-chip) RAM or fake (interpreted)
: RAM?
Real. The assembler cartridge came with 4K nonvideo RAM. (And battery
backup, too, I believe.)
--
Mike Swaim, Avatar of Chaos: Disclaimer:I sometimes lie.
Home: sw...@c-com.net
Alum: sw...@alumni.rice.edu Quote: "Boingie"^4 Y,W&D
Danny Lingman
a large rubber band and plasticine to hold the 16K pack on
the back of the TS-1000.
It was nice having a basic aware keyboard. I wonder how hard
it would be to hack something up that would do the same thing
for C++...
Dan.
J. Chris Hausler
Shack MC-10, a stripped down low cost version of the COCO and
thus sometimes called the POCO. It used a 6803 (a 6801 with its
internal rom disabled) instead of the 6809 in the COCO's. I
believe it was an attempt to attract the same market as the ZX-81
(ie low cost). Although color, it was a little too little a
little too late and didn't last long on the market. I had a lot
of fun with mine the year or so I played with it. Anyone else
remember it?
Chris
Adam Atkinson
>Curiously, in spite of the Apple 2's reputation, I've never encountered
>anyone who owned one and only saw one "in the flesh" on one occasion. And
>that was in a shop display... ISTR the most likely reason was (at risk of
>being repetitive) its price; nearly double that of a BBC/B IIRC.
The Apple II predates the BBC B by quite a bit, though. I knew several
people who had (or whose parents had...) Apple II machines in, say,
1979 or 1980. Other people (or their parents) had TRS80s or PETs.
By the time I was in a position to buy my own computer (82? 83?), none of
these machines were candidates.
--
Adam Atkinson (gh...@mistral.co.uk)
Ordinary decent people in this country are sick and tired of being told
that ordinary decent people in this country are fed up with being sick
and tired. I am certainly not, and I'm sick and tired of being told that
I am. (M. Python)
st...@blighty.com
says...
>
>In article <7jk09e$mmr$4...@news1.cableinet.co.uk>,
> euph...@freenet.co.uk writes:
>> What was the first British computer to really make it big? I get the
>> impression that the BBC is very much Britain's Apple II, but other than
>> that... I'm only thinking of UK sales here.
I'd have thought the Speccy wins over the Beeb.
>Curiously, in spite of the Apple 2's reputation, I've never encountered
>anyone who owned one and only saw one "in the flesh" on one occasion. And
>that was in a shop display... ISTR the most likely reason was (at risk of
>being repetitive) its price; nearly double that of a BBC/B IIRC.
Significantly earlier, too.
I used an Apple ][ at school (in the UKoGBaNI) well before I got
my ZX80... '78? '79?
I went through the ZX80 and an Atom before the Beeb arrived, in '82?
So there's maybe four years difference between the two.
Cheers,
Steve
--
-- Steve Atkins -- st...@blighty.com
Tony Duell
: I remember using a computer based on the BBC, labelled "Torch" with a
: couple of disk drives, which AFAICR was running CP/M. I think it had
: extra memory (64K total?) as well as an extra processor. At the time
: I knew nothing of operating systems, so I couldn't really be sure what
: it was running. Is it possible that the 8088 add-on ran CP/M?
No, that's yet another Torch machine. It started off life as a little
Z-80 card (called IIRC the Communicator) that plugged into an normal
Beeb's 1MHz bus. There was a separate box that contained a pair of disk
drives and a PSU - you were supposed to run the Beeb of this PSU as well
as some versions of the Beeb PSU couldn't handle the extra load of the
Z80 card.
IIRC the card contained a Z80, 64K or RAM, a couple of interface chips
and an EPROM containign a CP/M clone (CPN?). There was also a host ROM
that went into one of the sideways ROM sockets on the BBC.
Later on, Torch sold all-in-one machines with a BBC motherboard + Z80
card + monitor + drives in one box. Maybe that's what you used.
-tony
Tony Duell
: IIRC, that was a weird coprocessor thing with a Z80 in it. The Acorn
: Master Plan was that the BBC was never intended to be a stand-alone
: computer, it was really supposed to be the I/O processor for another
: computer entirely; hence the Tube high-speed interface. Acorn only got
: round to producing the 65C02 second processor in quantities. This had
: 50-odd kilobytes of memory, ran at 4MHz (I think), and was really rather
: nice. As the second processor didn't have to handle any I/O, it didn't get
: interrupts very often, and so was good for number-crunching.
The Acorn Z80 second processor is quite common. The other ones (32016 (or
16032) and ARM 1) are much rarer. There was also an 80186 internal
coprocessor for the Master IIRC.
: The Torch was a third-party second processor, designed to run CP/M, that
: didn't really comply with Acorn's standard. I believe the floppy disk
Correct. IIRC Tube copros connected to the 1MHz bus not to the Tube. The
Z80 and Graduate (8088) both did (I have them). The Z80 one had a
Sideways ROM that contained the interface routines for the BBC side. The
Graduate used an interesting feature of the BBC MOS that means you can
map a ROM on the 1MHz bus (256 _bytes_ at a time) and execute it on
reset. IIRC this happens if NMI is asserted on reset.
: interface was connected to the Z80 rather than the 6502? People kept
No, the Z80 board used the normal BBC disk controller (on the 6502) and
drives. The graduate, on the other hand, had an internal 8272 disk
controller and 2 40 track drives, just like a PC. In fact it was normal
to have drives on both the BBC and the Graduate...
-tony
Scott
wrote:
>Scott (scot...@bigfoot.com) wrote:
>: In article <7jjjiv$2og$3...@teabag.demon.co.uk>, Chris Hedley
>: <cbh@REMOVE_THIS.teabag.demon.co.uk> wrote:
>
>: >I only encountered a Lynx once, but quite liked it. Very nice
>: >graphics, as I recall.
>: >
>: I still have my Lynx. I love the thing! Some of the guys who worked on
>: the Amiga designed it. I always wanted to try developing for it.
>: Thought it might be fun. Just to see how the thing worked even.
><snip>
>
>Aren't you thinking of the Atari Lynx game system?
>
Oops. Yep.
jrh68
> IIRC, that was a weird coprocessor thing with a Z80 in it. The Acorn
> Master Plan was that the BBC was never intended to be a stand-alone
> computer, it was really supposed to be the I/O processor for another
> computer entirely; hence the Tube high-speed interface. Acorn only got
> round to producing the 65C02 second processor in quantities. This had
> 50-odd kilobytes of memory, ran at 4MHz (I think), and was really rather
> nice. As the second processor didn't have to handle any I/O, it didn't get
> interrupts very often, and so was good for number-crunching.
>
> obscure with 80 in the name, and a couple of risc things. Prototypes were
> made, and are now quite valuable.
>
> The Torch was a third-party second processor, designed to run CP/M, that
> didn't really comply with Acorn's standard. I believe the floppy disk
While at high school ( 6 years ago ) we stumbled across an old machine
in
an old room. It was made by cambridge computers ?? It was an all in one,
sort of like a Lisa, and I think used a BBC to do I/O, and it had an
NS 32016 ??? as its main processor. It had a tube interface to the 6502,
and had some funky OS, which was like a BBC too.
Jon.
Justin Pearson
> On 1999-06-08 pe...@fenelon.com said:
> :> Or the Torch XXX, a Unix workstation (68010-based) with a
> :>macintosh-like desktop frontend (but in colour), and a BBC
> :>micro-like 1MHz bus for expansion. No I am not joking.
>
> :Lovely machine. I can't remember what the third X was for (X11, X.
> :25 and something else...). For a while a TripleX was the X.25
> :gateway for my then-employers...
>
> uniX?
> --
> Communa -- you know soft spoken changes nothing
I seem to remember that it was CP/M. It was basically a BBC Micro
with a Z80 Co-processor in different casing. Co-processors were quite
interesting on the BBC. You used the BBC for I/O and graphics and the
co-processor for the rest. Communication was over a 1Mhz bus (I
think). There was a 6502 co-processor, a Z80 box running CPM, and a
16032.
Maybe the later Torchs where unix based.
Justin
sup...@cs.york.ac.uk
> David Given (d...@tao.co.uk) wrote:
> : IIRC, that was a weird coprocessor thing with a Z80 in it. The Acorn
> : Master Plan was that the BBC was never intended to be a stand-alone
> : computer, it was really supposed to be the I/O processor for another
> : computer entirely; hence the Tube high-speed interface. Acorn only got
> : round to producing the 65C02 second processor in quantities. This had
> : 50-odd kilobytes of memory, ran at 4MHz
3MHz on the original; the Master Turbo (an internal version for the Master
128) ran at 4MHz.
> The Acorn Z80 second processor is quite common. The other ones (32016 (or
> 16032) and ARM 1) are much rarer. There was also an 80186 internal
> coprocessor for the Master IIRC.
Yes, I have two Acorn Z80s and two Torch Z80s. The Acorn one runs real
CP/M, the Torch one uses a not-overly-compatible system called CPN. The
80186 internal copro was called the Master 512.
> : The Torch was a third-party second processor, designed to run CP/M, that
> : didn't really comply with Acorn's standard. I believe the floppy disk
> Correct. IIRC Tube copros connected to the 1MHz bus not to the Tube. The
> Z80 and Graduate (8088) both did (I have them).
No, The Torch (I think that's what Tony means :-)) connects to the Tube,
not to the 1MHz Bus. And there were indeed problems caused by the way it
nearly-but-not-quite adhered to Acorn's normal Tube protocols.
--
Pete Peter Turnbull
Dept. of Computer Science
University of York
Chris Stratford
> I'll see you a UK101 (I saw a schematic of that, but no more) and raise
> you a Laser 200 (sub-Spectrum computer with colour blocky graphics and
> rubber keys, for about 69 quid) and a Comx 35 (did anyone ever have one
> of these? 1802-based, with "the slowest version of Basic ever
> encountered", according to one site).
That comment sounds familiar. There's a copy of the Your Computer
review of the Comx 35 at:
http://www.gondolin.org.uk/hchof/reviews/text/yc-comx35.html
If anyone's interested (sorry, couldn't resists a quick plug).
I've yet to find one of these machines for sale anywhere yet, though.
One day...
Chris.
C Lamb
: While at high school ( 6 years ago ) we stumbled across an old machine
: in
: an old room. It was made by cambridge computers ?? It was an all in one,
: sort of like a Lisa, and I think used a BBC to do I/O, and it had an
: NS 32016 ??? as its main processor. It had a tube interface to the 6502,
: and had some funky OS, which was like a BBC too.
Yep, I've seen one of those too, at Birmingham University (UK) in
(then) Dr Borchards office. It was (_very_ loosely_ the precursor to
the Archimedes. I have a feeling it was the first National Semi. 32bit
processor. There is a small writeup on it at the back of the Watford
Electronics BBC Master Reference manual.
This was all back in 1988/89.
There was a lab of BBC Masters outside his office and I used them
mostly as a terminal to the Gandalf and then on to the IBM3090-200S
that Birmingham had at the time (probably the worst machine I ever
worked on).
regards
Chris
: Jon.
barnacle
>euph...@freenet.co.uk writes:
>
>> On 1999-06-05 d...@news.daedalus.co.nz(DonStokes) said:
>>
>> :Where is Uncle Clive now?
>>
>> Apparently he's suddenly woken up to Linux, and wants to make a
>> RISC-based portable (maybe Z88-scale?) using it. So that'll be along in
>> about 2 years, then...
>
>Do we get a power switch this time round?
>
legal action, you'll get the computer only three months after that. Happened
to me twice. Didn't try a third time.
--
barnacle
Rob Nicholson
> the Amiga designed it. I always wanted to try developing for it.
> Thought it might be fun. Just to see how the thing worked even.
Ah - we're talking about the Atari Lynx. I was thinking of the Camputers
Lynx, a crap Z80 system.
Yes, the Atari Lynx was wonderful - I wrote a handful of games for it.
Rob.
Dave Daniels
David Given <d...@tao.co.uk> wrote:
> BTW, anyone remember the Aquarius, a machine I saw reviewed in said
Yes, I remember the adverts for the Aquarius. A 6502-based machine
made by Mattel, IIRC. ISTR thinking that it looked somewhat
underpowered even then. The term 'sunk without trace' comes to
mind to describe how successful it was (IIRC).
Dave
--
ANTISPAM: Please note that the email address above is false. My
correct address is:
dave_daniels<at>argonet<dot>co<dot>uk
Please replace the <at> and <dot>s with @ and . respectively when
replying - Thanks!
Dave Daniels
<euph...@freenet.co.uk> wrote:
> What was the first British computer to really make it big? I get the
I would say that the ZX81 was the first really successful British
machine. Ready built it was 79 UKP, IIRC. The kit version was 50
UKP. I would say that it was the first cheap non-trivial machine.
At that kind of price you had the Science of Cambridge MK14, which
had a hex keypad and a eight digit LED display, and the Microtan
(?) which, although expandable into a 'real' computer, was not
much more sophisticated in its most basic form. Moving slightly up
market you had machines like the Acorn Atom which cost about 120
UKP in its most basic form but the ZX81 was cheap and more than
usable. It had a proper (if minute) keyboard, a nice Basic
interpreter, worked with the TV and had a very good manual. The
manual had lots of information on the operating system variables
and a list of Z80 opcodes, just what was needed to send thousands
of spotty British youths down the path of enlightenment to
geekdom. The microcomputer scene was developing quite nicely when
the ZX81 came along, but it made computers affordable.
Dave Daniels
Rob Nicholson <rob.ni...@unforgettable.com> wrote:
> show in London. Wasn't it advertised as "using the chip used in the space
> shuttle"?
That rings a bell - I think it used a 1802 microprocessor.
Dave Daniels
David Given <d...@tao.co.uk> wrote:
> There were plans for a 68k-based second processor, Z80/8080, something
> obscure with 80 in the name, and a couple of risc things. Prototypes were
> made, and are now quite valuable.
The Z80 second processor was released. It used a 6MHz Z80 and ran
CP/M. The 'Cambridge Workstation' was a Beeb with 32016 (or was it
a 32032?) second processor. There was also the ARM second
processor, which cost three or four grand and used the ARM 1. The
BBC Master had a 80186 second processor that could be fitted
internally. It ran DOS Plus and GEM and turned the Master into a
reasonably compatible PC. There was also a second version of the
6502 coprocessor that could be mounted internally in the BBC
Master. That was the one that ran at 4Mhz. The original one, for
the BBC micro, ran at 3Mhz.
I am not aware of any other second processors that were released.
According to something I read in one of the Acorn magazines, the
second processor idea allowed Acorn to experiment with different
types of processor to see which one offered the best prospects for
use in future products. They came to the conclusion that the only
way they could obtain a large jump in performance was if they
designed their own. Thus, the ARM was born.
Liam Busey
>
> In article <u2djj7...@pearl.tao.co.uk>,
> David Given <d...@tao.co.uk> wrote:
> > BTW, anyone remember the Aquarius, a machine I saw reviewed in said
>
> Yes, I remember the adverts for the Aquarius. A 6502-based machine
> made by Mattel, IIRC. ISTR thinking that it looked somewhat
> underpowered even then. The term 'sunk without trace' comes to
> mind to describe how successful it was (IIRC).
The Aquarius was Z80 based.
Liam Busey
Tony Duell
: > Correct. IIRC Tube copros connected to the 1MHz bus not to the Tube. The
: No, The Torch (I think that's what Tony means :-)) connects to the Tube,
Oops... Yes, just looked at the card. The Torch Z80 (and that's what I
meant) connects to the Tube. The Graduate connects to the 1MHz bus, though.
Never tried to connect both to the same Beeb at the same time. Nor have I
tried to link a Graduate to the X-bus of a XXX (:-)).
: not to the 1MHz Bus. And there were indeed problems caused by the way it
: nearly-but-not-quite adhered to Acorn's normal Tube protocols.
Well, the Torch Z80 didn't use a Tube ULA. It used a 6522 and an 8255
communicating through the ports. Oh well...
-tony
Juergen Nickelsen
> Significantly earlier, too.
>
> I used an Apple ][ at school (in the UKoGBaNI) well before I got
> my ZX80... '78? '79?
>
> I went through the ZX80 and an Atom before the Beeb arrived, in '82?
>
> So there's maybe four years difference between the two.
The Apple ][ came out 1977 or 1978, but was continued as II+ (sp?) and
then //e, IIc (sp?), and IIGS well into the mid 80s. (And let's
conveniently ignore the Apple /// for now.) I think the IIGS came out
1986, but then the Apple II era was over and it didn't have much of a
success. I always thought of it (without being hindered by knowledge
of the facts) as a pseudo-Mac for those Apple II users who could not
afford a real one.
To get back to my original point, there may well have been four years
difference between the Apple II and the Beeb, but there were also four
years overlap -- if the Beeb lasted that long.
--
Juergen Nickelsen
Charles Richmond
>
> jrh68 (jr...@student.canterbury.ac.nz) wrote:
>
> : While at high school ( 6 years ago ) we stumbled across an old machine in
> : an old room. It was made by cambridge computers ?? It was an all in one,
> : sort of like a Lisa, and I think used a BBC to do I/O, and it had an
> : NS 32016 ??? as its main processor. It had a tube interface to the 6502,
> : and had some funky OS, which was like a BBC too.
>
> Yep, I've seen one of those too, at Birmingham University (UK) in
> (then) Dr Borchards office. It was (_very_ loosely_ the precursor to
> the Archimedes. I have a feeling it was the first National Semi. 32bit
> processor. There is a small writeup on it at the back of the Watford
> Electronics BBC Master Reference manual.
> This was all back in 1988/89.
>
data bus. In this respect, it was kind of like the MC68000. The NS 32032
version was essentially the same processor built on a 32-bit external data
bus. I saw an instruction set (order code) listing for the NS 32016. I
found it interesting that it had a CASE assembly language instruction. As
I remember, it used a table of jump addresses and picked the address to
branch to based on a small integer from a register. (The details may be
in error...it's been 20 years since I looked at that listing.)
I heard rumors that there were problems with the NS 32032 instruction
execution...some bug in the hardware implementation. Does any one have
any details about this?
--
+-------------------------------------------------------------+
| Charles and Francis Richmond <rich...@plano.net> |
+-------------------------------------------------------------+
euph...@freenet.co.uk
On 1999-06-09 rob.ni...@unforgettable.com said:
:Newsgroups: alt.folklore.computers
:> you a Laser 200 (sub-Spectrum computer with colour blocky
:>graphics and rubber keys, for about 69 quid) and a Comx 35 (did
:>anyone ever have one of these? 1802-based, with "the slowest
:>version of Basic ever encountered", according to one site).
:Goodness - you mean they actually made the Comx!! I remember seeing
:it at a show in London. Wasn't it advertised as "using the chip
:used in the space shuttle"?
God knows. By the time I got into computers, the company had already
come and gone. I seem to recall that they managed to never actually get
the thing into production. The 1802 was a slow chip anyway, which really
didn't help matters.
euph...@freenet.co.uk
On 1999-06-09 josef.m...@pdb.siemens.de said:
:> graphics, as I recall. As for the TI99, didn't that hold the
:>rather dubious accolade of being one of the slowest machines ever,
:>in spite of its then almost unprecedented 16-bit CPU? The weird
:>memory model might explain this if it's true...
:"Memory model" might be the wrong term: "CPU architecture" might be
:better: the 9900 series CPUs had their registers in memory. A base
:pointer addressed the first register and context switch was simple:
:just reload the base register ... Later models had some kind of
:cache, iirc.
The 9995 was the one used in the TI99, IIRC; it had an 8-bit bus, and I
think it had 256 bytes on chip (a handy fast place to put the registers;
you could get 8 sets in there, which seems to be a kind of precursor to
the SPARC's register windows). Unfortunately, if you have to access RAM
through a 9918 (or 28, 29, etc), it's slow as flowing treacle...
Memotech users tended to find that it also produced a display just a bit
too wide for standard TVs. (At least, I did.)
Lennart Benschop
<euph...@freenet.co.uk> wrote:
>
>
>On 1999-06-09 rob.ni...@unforgettable.com said:
> :Goodness - you mean they actually made the Comx!! I remember seeing
> :it at a show in London. Wasn't it advertised as "using the chip
> :used in the space shuttle"?
>
>God knows. By the time I got into computers, the company had already
>come and gone. I seem to recall that they managed to never actually get
>the thing into production. The 1802 was a slow chip anyway, which really
>didn't help matters.
think in 1984). I even played with one for a while at an electronics
exhibit. The importer provided 'free' software for it in the form of BASIC
listings that you could type. There were also some cassettes. For the
Commodore 64 you could get much more software, the C64 had more processing
power and the price was about the same . No wonder the Comx35 didn't get a
huge market share.
--
Lennart
euph...@freenet.co.uk
On 1999-06-10 a__fake_...@127.0.0.1 said:
:Newsgroups: alt.folklore.computers
:In article <u2djj7...@pearl.tao.co.uk>,
:David Given <d...@tao.co.uk> wrote:
:> BTW, anyone remember the Aquarius, a machine I saw reviewed in
:said
:Yes, I remember the adverts for the Aquarius. A 6502-based machine
:made by Mattel, IIRC. ISTR thinking that it looked somewhat
:underpowered even then. The term 'sunk without trace' comes to
:mind to describe how successful it was (IIRC).
Was it 6502...? For some reason I believed it to be Z80, but I could
easily be wrong. It sank without trace twice, in fact; they produced an
Aquarius II, which had a real keyboard (although Mattel had dropped it
by then, so it was produced under the name of Radofin, its designer
company).
euph...@freenet.co.uk
On 1999-06-09 lin...@wcars05r.ca.nortel.com(DannyLingman) said:
:It was nice having a basic aware keyboard. I wonder how hard
:it would be to hack something up that would do the same thing
:for C++...
Ugh. It was bad enough for BASIC. All those icky modes and things...
Besides, I thought that's what the modes in Emacs were for?
euph...@freenet.co.uk
On 1999-06-09 jcha...@delphi.com said:
:Following this thread, I haven't seen anyone mention the Radio
:Shack MC-10, a stripped down low cost version of the COCO and
:thus sometimes called the POCO. It used a 6803 (a 6801 with its
:internal rom disabled) instead of the 6809 in the COCO's. I
:believe it was an attempt to attract the same market as the ZX-81
:(ie low cost). Although color, it was a little too little a
:little too late and didn't last long on the market. I had a lot
:of fun with mine the year or so I played with it. Anyone else
:remember it?
I remember it. I quite fancied one, except that the 6803 was (iirc) a
cut-down 6800. If there were ever a CPU that shouldn't have been cut
down... A 6809-based machine would have been much more fun, and I can't
help wishing that Sinclair had discovered this processor first. Instead
there was only ever the Dragon 32 (given that over here, the CoCo was
hopelessly overpriced).
euph...@freenet.co.uk
On 1999-06-10 a__fake_...@127.0.0.1 said:
:<euph...@freenet.co.uk> wrote:
:> What was the first British computer to really make it big?
:I would say that the ZX81 was the first really successful British
:machine. Ready built it was 79 UKP, IIRC. The kit version was 50
:UKP. I would say that it was the first cheap non-trivial machine.
That's what I thought; it's certainly where I started. The price dropped
in later years; I got my first new one for 45 quid, including a RAM pack
(the beloved Memotech one). My next computer was a Commodore 16, for 5
quid more; same memory size, but colour and highish res graphics. I
loved that machine; I still have it, although I truly doubt that it
still works.
euph...@freenet.co.uk
On 1999-06-09 sw...@gemini.c-com.net said:
:David Given <d...@tao.co.uk> wrote:
:: In article <7jk5uk$j8r$1...@watchdragon.demon.co.uk>,
:: pm...@watchdragon.demon.co.uk (Peter Maydell) writes:
::> You could actually expand the TI into quite a powerful machine.
::> I used to drool over descriptions in the user group magazines of
::> systems with the peripheral expansion box, twin disk drives,
::> 32K of RAM and the ability to do machine code, RS232 and more.
::> It's just a shame the basic machine was so crippled.
:: When we say RAM, are we talking real (on-chip) RAM or fake
:(interpreted) : RAM?
:Real. The assembler cartridge came with 4K nonvideo RAM. (And
:battery backup, too, I believe.)
The graphics chip maxed out at 16k, in any case, so expanding that RAM
would have been quite out of the question.
euph...@freenet.co.uk
On 1999-06-09 cbh@REMOVE_THIS.teabag.demon.co.uk(ChrisHedley) said:
:euph...@freenet.co.uk writes:
:> What was the first British computer to really make it big? I get
:>the impression that the BBC is very much Britain's Apple II, but
:>other than that... I'm only thinking of UK sales here.
:Curiously, in spite of the Apple 2's reputation, I've never
:encountered anyone who owned one and only saw one "in the flesh" on
:one occasion. And that was in a shop display... ISTR the most
Ditto. And that was just four years ago.
:likely reason was (at risk of being repetitive) its price; nearly
:double that of a BBC/B IIRC.
Yep. Meanwhile one of the guys next door has (I think) promised me one
of his two BBC Masters - hence my (er) accusation that the BBC occupied
the position in the UK which the Apple II took in the US. (For sheer
popularity, the Commodore 64 is probably the US equivalent to the
Spectrum in the UK. But for aspirational quality...)
euph...@freenet.co.uk
On 1999-06-08 pmay...@chiark.greenend.org.uk said:
:>I'll see you a UK101 (I saw a schematic of that, but no more) and
:>raise you a Laser 200 (sub-Spectrum computer with colour blocky
:>graphics and rubber keys, for about 69 quid) and a Comx 35
:How about a CGL M5?
:http://www.chiark.greenend.org.uk/~pmaydell/hardware/hallath/
:Not very nice, not very common, but the Guttang Guttong
:puzzle/arcade game is fun :->
I think I actually saw one of these once... Just fleetingly, walking
round a showroom, but enough to realise that it existed.
Who remembers the portable that (was it...?) Prism produced? A CP/M Plus
machine, with 128k memory(???), keyboard that slid out from the main box
and a built-in modem (may have been an option) for a grand? And there
was another TMS9995-based computer too, available as a kit for 300 quid.
That one had 64k main memory and was rather more use than the TI99/4a; I
think it may have been sold as a control computer. And what about the
TI99/2?
Also, how many Spectrum clones got produced in Russia? I remember
reading about one called the Hobbit, which had a disk interface. On the
same line, can you still get SAM Coupes?
Last question: can anyone flog me a disk interface for the Spectrum?
euph...@freenet.co.uk
On 1999-06-10 Chris.S...@uk.uu.net said:
:euph...@freenet.co.uk writes:
:> I'll see you a UK101 (I saw a schematic of that, but no more) and
:>raise you a Laser 200 (sub-Spectrum computer with colour blocky
:>graphics and rubber keys, for about 69 quid) and a Comx 35 (did
:>anyone ever have one of these? 1802-based, with "the slowest
:>version of Basic ever encountered", according to one site).
:That comment sounds familiar.
Well, I was hoping someone would say that, so that I could turn round
and say:
Then it'll be your site I cribbed it from. :> Thanks.
euph...@freenet.co.uk
On 1999-06-10 nailed_barn...@hotmail.com(barnacle) said:
:>> Apparently he's suddenly woken up to Linux, and wants to make a
:>> RISC-based portable (maybe Z88-scale?) using it. So that'll be
:>>along in about 2 years, then...
:>Do we get a power switch this time round?
:It'll be *advertised* in a couple of years, and then once you've
:threatened legal action, you'll get the computer only three months
:after that. Happened to me twice. Didn't try a third time.
Hmm. Seems to me that although pretty much everyone in the UK started on
one or the other variant of a Sinclair computer, the man himself
employed some highly dubious business practices...
William Hamblen
>I remember it. I quite fancied one, except that the 6803 was (iirc) a
>cut-down 6800. If there were ever a CPU that shouldn't have been cut
>down... A 6809-based machine would have been much more fun, and I can't
>help wishing that Sinclair had discovered this processor first. Instead
>there was only ever the Dragon 32 (given that over here, the CoCo was
>hopelessly overpriced).
The 6801 and 6803 are supersets of the 6800, with the addition of a Y
index register (much needed) and a few more instructions.
The Dragon 64 was built in the USA for a short while by TANO Corp of
New Orleans from parts supplied by Dragon Data. After production
ended California Digital sold the remaining stock for something like
$15 each. They still might have some left. California Digital also
had CP/M 2.2 for $5 a copy.
J. Chris Hausler
>I remember it. I quite fancied one, except that the 6803 was (iirc) a
>cut-down 6800. If there were ever a CPU that shouldn't have been cut
>down... A 6809-based machine would have been much more fun, and I can't
>help wishing that Sinclair had discovered this processor first. Instead
>there was only ever the Dragon 32 (given that over here, the CoCo was
>hopelessly overpriced).
instructions to allow the A and B registers to be operated
on as one 16 bit 'D' register as well asmproved data
movement instructions, especially the ABX instruction. It
also was a "single-chip" machine and there was a version
where the on board ROM was an EPROM instead (68701). One
popular version of the 6801 was the 6801L1 which had a
debugging program LILBUG in the ROM. I have a Motorola
single board 6801EVM which came with this chip. The EVM
would also support the 68701 with the PROBUG debugger (it
ahd the routines to program the EPROM) on a separate 2716.
Chris
John F. Eldredge
of the ZX-81. The primary difference between the two computers was
the lack of Slow mode, so you had to put up with the display losing
synchronization every time you pressed a key. My first computer was a
MicroAce, a clone of the ZX-80. Apparently, it was an illegal clone,
as the company vanished without a trace a few months after I bought my
computer. The MicroAce was sold only as a kit; I remember that I
spent $200 on it, took two days to put it together, then had to ship
it back to the factory and pay another $50 or so to have them figure
out why it didn't work. It had only 2K of RAM, but I never succeeded
in typing in a long enough program to run out of memory. This was
because the heat sink on the power regulator was mounted on the
motherboard, right next to the RAM and CPU. After about 10 minutes of
operation, the computer would overheat, the display would black out,
and the computer wouldn't work until you turned it off for a few
minutes to cool down.
--
John F. Eldredge -- eldr...@poboxes.com
PGP key available from http://www.netforward.com/poboxes/?eldredge/
--
"There must be, not a balance of power, but a community of power;
not organized rivalries, but an organized common peace." - Woodrow Wilson
barnacle
The view of myself and several people in similar circumstances was generally
that the chap had some damn good ideas but he seemed to used the punters as
venture capital to get them off the ground. Maybe not true, but that's how it
felt. Remember 'Black Watch' and the 'smaller than a matchbox' radio?
btw...my zx80 didn't go first time and I was immensly pleased with myself for
diagnosising a faulty diode in the kb scan...
--
barnacle
Ariel Scolnicov
[...]
> It was nice having a basic aware keyboard. I wonder how hard
> it would be to hack something up that would do the same thing
> for C++...
Indeed, who can forget the experience of walking up to a Spectrum and
typing in:
10 PRINT RINT
--
Ariel Scolnicov
Peter Stephenson
in to the temptation to write this before. Please forgive me, but:
Me too! Me too! Me too!
Bruce Lin
> Indeed, who can forget the experience of walking up to a Spectrum and
> typing in:
>
> 10 PRINT RINT
Why, what did this do?
Bruce
Ariel Scolnicov
> Oh God, I'm really sorry about this. Really. I've never ever given
> in to the temptation to write this before. Please forgive me, but:
>
> Me too! Me too! Me too!
>
> >lin...@wcars05r.ca.nortel.com (Danny Lingman) writes:
I've been called a lot of names before, but this is the first time
someone's called me a Lingman...
[...]
--
Ariel Scolnicov |"GCAAGAATTGAACTGTAG" |ari...@compugen.co.il
Compugen Ltd. |Tel: +972-2-6795059 (Jerusalem) \ NEW IMPROVED URL!
72 Pinhas Rosen St. |Tel: +972-3-7658520 (Main office)`--------------------
Tel-Aviv 69512, ISRAEL |Fax: +972-3-7658555 http://3w.compugen.co.il/~ariels
Don Stokes
It would attempt to print the variable RINT. What did you think it would
do?
But if you went up to a Speccy or a ZX80 or ZX81 and typed '10 PRINT',
the above is what you'd get. Y'see, Sinclair's BASIC interpreter had
a somewhat modeful program entry mode. You start in K (keyword) mode,
with an inverse-K cursor, which would let you enter a line number and/or
a keyword -- the keywords being marked on the keyboard. The keyword
associated with 'P' was PRINT, 'N' NEXT, F, FOR and so-on. (A lot of
ZX programmers used N as a FOR loop control variable, because to
terminate the loop, you just hit <line-no>NN to get NEXT N.) Once you
hit a keyword, you went into L mode and could type normal expressions.
Function keywords were entered by hitting the FUNCTION key (shift-NEWLINE,
IIRC) which put you into F mode, and then hitting the appropriate key to
get the keyword. Then there was G mode for graphics characters.
Note that keywords entered letter by letter were *never* understood.
Each key was labelled something like:
GOSUB
.-------.
| ** |
| H # |
`-------'
SQR
The '**' is the "raise to power" operator -- two '*'s didn't cut it -- and
could be got via the SIFT key. GOSUB is the keyword (ie what you get if
you hit H unshifted in K mode), SQR is the function (from F mode), and the
'#' is supposed to represent the "grey cell" graphic character that you
got in G mode.
This is for the ZX81 -- I never had a speccy meself, but I recall it being
very similar.
-- don
Ben Hutchings
: On 1999-06-09 lin...@wcars05r.ca.nortel.com(DannyLingman) said:
: :It was nice having a basic aware keyboard. I wonder how hard
: :it would be to hack something up that would do the same thing
: :for C++...
: Ugh. It was bad enough for BASIC. All those icky modes and things...
: Besides, I thought that's what the modes in Emacs were for?
Now we have Microsoft(R) Intellisense(R) which helps you to complete
words in your program, and appears to be the main cause of crashes in
the VC++6 IDE. That's progress in the software industry for you! I
think the same feature is used in VB and VJ++ as well.
--
Any opinions expressed are my own and not necessarily those of Laser-Scan.
Simon Craythorn
> (A lot of
> ZX programmers used N as a FOR loop control variable, because to
> terminate the loop, you just hit <line-no>NN to get NEXT N.)
Old habits die hard. I still use n for my loop variable in C!
Simon.
Danny Lingman
|> p...@gol.com (Peter Stephenson) writes:
|>
|> > Oh God, I'm really sorry about this. Really. I've never ever given
|> > in to the temptation to write this before. Please forgive me, but:
|> >
|> > Me too! Me too! Me too!
|> >
|> > >lin...@wcars05r.ca.nortel.com (Danny Lingman) writes:
|>
|> I've been called a lot of names before, but this is the first time
|> someone's called me a Lingman...
|> [...]
|>
Ariel,
I think he was referring to my use of rubberbands and plasticine to hold
the memory pack on to the back end of the TS-1000/ZX-81
Dan.
bma...@iglou.com
On 1999-06-13 bruc...@tucson.Princeton.EDU said:
>misc:51759 In alt.folklore.computers Ariel Scolnicov
>><ari...@compugen.co.il> wrote: Indeed, who can forget the
>>experience of walking up to a Spectrum and typing in:
>> 10 PRINT RINT
>Why, what did this do?
On the Spectrum or ZX-81, typing the letter P by itself would produce the
word PRINT. If you were accustomed to Basic on other machines, your natural
impulse would be to type PRINT, which would result in PRINT RINT.
Net-Tamer V 1.08X - Test Drive
Martin Ibert
wrote:
>> 10 PRINT RINT
>
>Why, what did this do?
You didn't get the joke. When typing programs into a Spectrum, the
thing tried to be helpful. It knew that after a line number there must
be a command, so as soon as you hit "P", it wrote "PRINT " for you. If
your fingers just kept typing the remaining letters of "PRINT", you
end up with "PRINT RINT", so that you have to go back and delete the
"RINT" part and put in what you meant to print in the first place.
Or so I seem to remember. I've never actually owned such a beast.
--
>> Please visit http://www.ibert.com/ for further information. <<
-----------------------------------------------------------------
Blind faith in your leaders, or in anything, will get you killed!
-----------------------------------------------------------------
Remember the heroes of Tiananmen Square, Beijing, P. R. of China!
Bruce Lin
Oh no, it's all coming back to me in a horrible flashback. In Grade 3 at
my public school, our teacher had a little ZX-81 in the back corner of the
class hooked up to a little black and white television. My friend and I
were pretty bright so we'd race through our homework so that we could play
around on the computer. We'd type in programs from that Usborne series of
books ...
Although I've forgotten about all those strange keywords, now I realize
why I use "n" for my program loops. One other vivid memory I have of that
computer:
I'd seen my teacher type in programs. He'd enter a line of code, and then
it would appear at the top of the screen. My friend and I, having not
quite caught what happened in between, started a program with a line like
"10 PRINT". Then, we proceeded to push the space key several hundred times
to get the line to move up to the top of the screen like it was supposed
to.
Fortunately, my teacher taught us what the carriage return key was for
before we became the first poster children (literally; we were about eight
at the time) for carpal tunnel syndrome.
Bruce
Aaron Crane
mar...@ibert.com (Martin Ibert) writes:
> When typing programs into a Spectrum, the thing tried to be helpful. It
> knew that after a line number there must be a command, so as soon as you
> hit "P", it wrote "PRINT " for you.
It wasn't just to be "helpful" -- it was a clever way of reducing both
storage costs and parsing costs for programs. The crucial point (on the
ZX81 at least) was that characters had variable-width representations
(though the width was always an integer number of glyph cells). For
example, one character (sorry, I don't remember the code) printed as the
five glyph cells `PRINT'[1] and this single character was stored in memory
as part of the program. Indeed, every BASIC keyword was a single character.
Considering that the ZX81 had only 1k of RAM as standard, this greatly
increased the maximum length of program that could be stored. In addition,
the interpreter only had to dispatch on a single byte in each line of code,
without parsing a keyword.
[1] Actually, I think there was a space after the letters, which would make
the PRINT character six glyphs wide.
--
Aaron Crane <aaron...@pobox.com> <URL:http://pobox.com/~aaronc/>
** Please send on-topic followups by Usenet, not email **
Michael Davis
Aaron Crane <aaron...@pobox.com> wrote:
>In article <37651c00.217116@news>,
>mar...@ibert.com (Martin Ibert) writes:
>> When typing programs into a Spectrum, the thing tried to be helpful. It
>> knew that after a line number there must be a command, so as soon as you
>> hit "P", it wrote "PRINT " for you.
>
>It wasn't just to be "helpful" -- it was a clever way of reducing both
>storage costs and parsing costs for programs. The crucial point (on the
>ZX81 at least) was that characters had variable-width representations
>(though the width was always an integer number of glyph cells). For
>example, one character (sorry, I don't remember the code) printed as the
>five glyph cells `PRINT'[1] and this single character was stored in memory
>as part of the program. Indeed, every BASIC keyword was a single character.
>Considering that the ZX81 had only 1k of RAM as standard, this greatly
>increased the maximum length of program that could be stored. In addition,
>the interpreter only had to dispatch on a single byte in each line of code,
>without parsing a keyword.
Actually, the Apple ][ also stored keywords as single bytes, and printed them
as full words, for the same reasons of saving space and run time. There was also
a neat design thing where there was a vector of addresses of the functions
which implemented the BASIC keywords. And the tokenized keyword n pointed
to the n'th address which implemented that keyword (actually, one address
before it, because it would push the two bytes of the address onto the
stack, and do a rts (return from subroutine). Quite efficient I thought.
But you still had to type in the whole keyword.
--
// Michael Davis -- Solaris code slave and happy Linux User.
//
// From sunny Toronto...
Scott
<mbd...@DELinterlog.com> wrote:
>Actually, the Apple ][ also stored keywords as single bytes, and printed them
>as full words, for the same reasons of saving space and run time.
>
The Commodore machines did the same thing too. Actually the Atari's did
too. All the old 8-bits tokenized BASIC keywords as far as I know.
The CBM and Atari machines also had the shorthand keyword entry.
--
--
Scott Maxwell - scottm25 (at) bigfoot (dot) com
My shrink? I just go to him for refills!
Philip Blundell
>You didn't get the joke. When typing programs into a Spectrum, the
>thing tried to be helpful. It knew that after a line number there must
>your fingers just kept typing the remaining letters of "PRINT", you
>end up with "PRINT RINT", so that you have to go back and delete the
Actually, I find it hard to believe that anybody's fingers would start
running away from them on a Spectrum keyboard. From what I remember,
typing on one was hard enough work, and slow enough, that you really
had to mean every keystroke. :-)
p.
Ariel Scolnicov
> In article <djemjer...@planet.rmsq.com>,
> Aaron Crane <aaron...@pobox.com> wrote:
> >In article <37651c00.217116@news>,
> >mar...@ibert.com (Martin Ibert) writes:
> >> When typing programs into a Spectrum, the thing tried to be helpful. It
> >> knew that after a line number there must be a command, so as soon as you
> >> hit "P", it wrote "PRINT " for you.
> >
> >It wasn't just to be "helpful" -- it was a clever way of reducing both
> >storage costs and parsing costs for programs. The crucial point (on the
> >ZX81 at least) was that characters had variable-width representations
> >(though the width was always an integer number of glyph cells). For
> >example, one character (sorry, I don't remember the code) printed as the
> >five glyph cells `PRINT'[1] and this single character was stored in memory
> >as part of the program. Indeed, every BASIC keyword was a single character.
> >Considering that the ZX81 had only 1k of RAM as standard, this greatly
> >increased the maximum length of program that could be stored. In addition,
> >the interpreter only had to dispatch on a single byte in each line of code,
> >without parsing a keyword.
>
> Actually, the Apple ][ also stored keywords as single bytes, and printed them
> as full words, for the same reasons of saving space and run time. There was also
> a neat design thing where there was a vector of addresses of the functions
> which implemented the BASIC keywords. And the tokenized keyword n pointed
> to the n'th address which implemented that keyword (actually, one address
> before it, because it would push the two bytes of the address onto the
> stack, and do a rts (return from subroutine). Quite efficient I thought.
>
The beeb also tokenised BASIC keywords. You could type in "P." and it
would expand to "PRINT". AFAIK, the tokens didn't index any vector.
The command line was limited to 255 characters, so sometimes utilising
token expansion was the only way to get a long line into the
interpretter.
There was also an amusing bug: if you typed in
10 G.10:G.10:...
(where ... means repeat until all 255 characters are full) and
pressed Enter, something would get overwritten, and the machine would
crash, accompanied by strange sounds from the keyboard.
> But you still had to type in the whole keyword.
--
Kalle Olavi Niemitalo
> example, one character (sorry, I don't remember the code) printed as the
> five glyph cells `PRINT'[1]
I have the manual here :) The code for PRINT is 245. The codes are
sorted according to the keys which must be pressed: RUN comes from R
and has code 247, SAVE comes from S and is 248. Yet another way to
save a lookup table :)
The manual (ZX81 BASIC PROGRAMMING) is very good; it lists such things
as system variables (and which of them produce interesting effects
when POKEd), machine language instructions and variable storage
formats. It and the COMMODORE 64 PROGRAMMER'S REFERENCE GUIDE are
still the greatest manuals I know of -- they start from the basics but
don't stop there.
I can remember how I was a kid who looked at the mysterious Z80
assembler instruction table and wondered what those strange codes
could mean. So I tried it -- I made a BASIC program which POKEd a
dozen of interesting-looking codes such as "prefixes instructions
using iy" and "jp nz,NN" in memory and ran them with USR.
Surprisingly, the machine didn't crash. :) But when I LISTed the
program again, it had all kinds of blinking cursors in it. I was
never able to reproduce that effect.
> [1] Actually, I think there was a space after the letters, which would make
> the PRINT character six glyphs wide.
It wasn't that simple. Keywords always had spaces around them; FOR
N=1 TO 10 printed with a space at both sides of TO. But when keywords
were adjacent, as in PRINT SIN RND, there was only one space between
them. So the spaces weren't really part of the keywords.
I can't remember how it changed when the cursors came in the play.
Don Stokes
Aaron Crane <aaron...@pobox.com> wrote:
>It wasn't just to be "helpful" -- it was a clever way of reducing both
>storage costs and parsing costs for programs. The crucial point (on the
Actually, almost every BASIC ever built did this, but most interpreted
multi-character strings and "tokenised" them down to single byte/word
codes taht could be interpreted faster. The only exception I can think
of off the top of my head (I'm sure there were others) is TRS-80 Level 1
BASIC, which kept the whole line in memory, and you could use
abbreviations like "P." for "PRINT" to save space and improve execution
speed. ISTR BBC BASIC had a similar "shorthand", but the shorthand
words were tokenised, so when you LISTed the program you saw the full
keyword. With L1 BASIC, you still saw the abbreviated keyword, complete
with '.'.
The ZX single key entry did significantly speed up program entry on the
'orrible membrane (rubber for the Speccy) keyboard. But once entered,
things looked a lot like they did in other BASICs.
-- don
Peter Stephenson
because any formatting you entered was lost when it 'un-parsed' it to
display it.
On Mon, 14 Jun 1999 20:52:03 -0400, Scott <scot...@bigfoot.com>
wrote:
>In article <7k3ttt$2...@shell1.interlog.com>, Michael Davis
><mbd...@DELinterlog.com> wrote:
>
>>Actually, the Apple ][ also stored keywords as single bytes, and printed them
>>as full words, for the same reasons of saving space and run time.
>>