Keep in mind these are PLA's, *not* ASICs
(ULA is specifically Ferranti) The sort of thing you laid out by marking connections between the transistors in the array.
Almost impossible to imagine now.
In May 1983, Sinclair Research Managing Director Nigel Searle began briefing the press about the successful British micro maker’s next big release. It was 13 months after the company had launched the Spectrum and although that machine had become a huge success, punters and market-watchers were keen to hear about what Sinclair …
(ULA is specifically Ferranti) The sort of thing you laid out by marking connections between the transistors in the array.
Almost impossible to imagine now.
Thankfully I was put off by lack of availability before the bad quality QL stories started appearing in the technical press. Ended up buying a 64K Einstein with single floppy and 40 column display, later expanded with add ons to double floppy and 80 col display. Worked well for a few years until I could afford a 386 PC compatible and modem.
Just before Xmas 84? I remember Your Sinclair had a pullout section devoted to the QL. I think it's on the internet somewhere on a QL fan site. Anyway I can remember thinking it looked cool and being 13 wanting new computers with 'moar ports' was compelling.
My ambition was to have a computer setup that would allow me to 'hack into and rule the world'. Isn't it great when your imagination outstrips your actual talent! Not to mention your piggy bank.
I think I mentioned to my Dad at the time that the credit card companies were letting customers pre-order the QL without adding it to their monthly credit limits (I guess back then £500 was a high limit). Suffice to say he just shrugged like Dads do, said "Oh really!" and the matter was never taken further.
Looking back I think I dodged a bullet. I wasn't even tempted when two or three years later Dixons were selling them off for next to nothing.
The next computer was a 128K Spectrum+2 (Amstrad Style) and then about a year or two later a 512K Apple Mac (with external floppy too for dual drive). Now I had a computer suitable for taking on the world but aged 17, the finances still didn't allow it. Macs were still damn expensive to run back then.
Suffice to say it was my last Apple PC.
If my 13 year old self could see the gear I have around my flat now, he'd explode.
I can't be the only person to have had one of those?
Also, as a niggle: I'd argue that the 68008 is either 8 bit or 32 bit as it's an 8-bit bus with a 32-bit instruction set architecture. Which I guess means it's an 8 bit machine in context, given that the hardware engineering seems to have been the primary goal of the project, Sinclair being a company that made money through selling hardware.
Just you and 11,999 others. They fetch a pretty penny today if you still have yours due to the rareity.
To put that in perspective, even the QL is said to have shifted 150,000 units.
And to further put that in perspective, 1.5 million BBC Micro's, 3 million Amstrad CPC's, 7 million Sinclair Spectrum's and 15 million Commodore 64's.
Heh. I worked at Sandy UK while Alan Miles and Bruce Gordon of MGT designed the Sam Coupe on the workbench right next to mine. I have fond memories of that time. The first ever Spectrum software to run on the prototype was a helicopter gunship game I loaned them.
Meanwhile, I was busy building SuperQBoards and expanderams and dual floppy drives.
The main problem with those was the video memory was so big compared to the woefully underpowered near-Spectrum speed CPU that anything that wasn't in Mode 4 (Spectrum compatible mode) scrolled up the screen like treacle and thus made it useless for games compared to the 16-bit machines around at the time. On a good day it could probably take on a C64, released several years earlier.
The Sam also had a hastily finished Basic/'OS' that was later improved upon by Roms or booting from disc.
How the hell do I manage to remember stuff like this?
How come each time you read about companies in the west producing computers or cell phones or similar. There is always something about "for Christmas" or "too late for Christmas". Are we not like the pig who asks another pig if he believes in life after Christmas. I suppose the pig is more European than American. Anyway, there we have this company employing talented people but forced by Santa to deliver, and fail in every possible way, in fear there is no life after Christmas. I have a feeling the US car industry failed for similar reasons, there had to be a new look for every Santa every year.
Christmas is THE retail market of the year and nothing in the rest of the year comes close.
68000 = great processor; one of the best ever.
68008 = total dog of a crippleware part.
That was the turning point for the design and it never had a chance after that decision.
I don't know about the reality of it, but when I found out it had an 8bit bus, it rather negated all the nice "32bit" promotion that was going on at the time.
Almost bought one... went with an second hand Apple ][+ instead.
Had a QL back in the day. Did lots of code - bought from Tony Tebby all the system level docs (I remember the QJump name, but never did know till now how he was connected - thought he'd one the OS as a contract). I did the Binkleyterm port to get QL onto Fidonet. Did an HP terminal emulator. Microdrives were AWFUL but with a floppy interface it was not a bad machine. You could see the cultture clash between Tony's great software and the awful Sinclair hardware!
Well, the QL may be 30 years old and have a somewhat chequered history.
I recently resurrected one of the earliest models and reverted it to using the Dongle version of the ROM. It's microdrive units were actually amongst the best I have seen - formatting cartridges at around 230 sectors (close to the physical limit of 255 sectors).
The QL is still strongly supported with new hardware, such as the QL-SD Interface - the first production model of which has been released to coincide with the 30th Anniversary and can be found on www.sellmyretro.com.
There were some classic games converted to the QL, and highly innovative hardware, such as the Super Gold Card, which provided the original QL with 4MB RAM, a 68020 processor and ability to use 4MB ED floppy disk drives (formatted 3.2MB capacity). That, coupled with Tony Tebby's last version of the QL operating system (SMSQ/e) has kept the QL as a fast multi-tasking enthusiasts machine capable of so much more than the original design and early reviewers would ever have dreamt about.
Congratulations on a thoroughly-researched article. Whatever mistakes management at Sinclair made, the technical talent that the company recruited was world class.
If there's anyone out there who can print me out a bunch of Archive and Quill files from 4MB ED floppies.
d love to hear from them.
Yes sure - contact me via www.rwapsoftware.co.uk - I have a Super Gold Card and ED disk drives - so can convert the files and print out or email them back to you.
By the way - I am trying to collect as much QL commercial software as possible to preserve it - there are quite a few missing titles - see the QL Wiki on www.rwapadventures.com/ql_wiki
£7.95 for delivery?
Would cost about the same today.
Possibly, but as Psion showed with their hardware and its software, they had a lot to be arrogant about.
The Series3a, 3c and 3mx are three of the finest bits of hardware and software I've ever owned.
I chortled to myself when I saw that quote attributed to Alan Sugar. Sweet irony.
I loved my QL... one of the first things I wrote for it was a cross assembler to facilitate programming on my Speccy :)
I have fond memories of the QL, I went through a few of them having been introduced to them by my grandfather. He had a stack of 5 or 6 of them and was forever getting upgrades to them Goldcards etc.
This is what got me interested in computing when I was a child in the 80's
I remember the little microdrives, loading Psion's suite, and getting the game programming book and typing out lines of code to run tiny little games.
I probably still have a stack of microdrives and the odd QL sitting in the attic (possibly with the old Commadore 16 + 4 which we also had)
Excellent article, thank you.
I look forward the the piece about the OPD. ICL had some excellent technology - CAFS, a superb SSD storage system back in 1990, VME (still the best OS for its purpose I ever used) - but OPD was one the bravest, wildest bits of What-The-Fuckery I ever saw.
Another side note: whenever I have an anxiety dream I always seem to walk into a data hall with walls painted ICL orange. Some things run deep.
"Another side note: whenever I have an anxiety dream I always seem to walk into a data hall with walls painted ICL orange. Some things run deep."
ITYM "Hot Tango"
Personaly I was always an "Aztec Blue" fan.
This seems like only half the story for me. The QL failed as a business machine but it found a much loved home as a hobbyist and enthusiasts computer, spawning a large cottage industry.
QUANTA the QL user group is still running after 30 years not to mention all the individuals who still make hardware and software for it today.
Familiar with Sinclair's earlier audio products -- optimistic power output claims and some reliability questions -- I wasn't about to risk my money on the QL, however cheap.
As a writer, the clincher was the flattish keyboard which was a non-starter for any real work of my sort.
The Apricot, with a reasonable keyboard, MS DOS and 3.5 inch disks was clearly a more serious product and ran reliably for six or seven years.
> Flattish keyboard
Yet, ironically, that's what we're all typing on these days, whether on laptops or (glances around the office) desktops.
In this sense the QL really was ahead of its time.
Sadly, I have to agree with you. At least my desktop computer can still use an IBM classic (clicky) keyboard. But, yes, most laptops are woeful by comparison. As for trying to write on a tablet or smartphone -- er !!!
It has to be said though that Sinclair's efforts at keyboards were particularly abysmal -- even the Commodore PET and BBC/Acorn models had something you could type on.
It seems to me a lot of the tech industry has the same story. Interesting tech, bodged, buggy and released too early, doesn't quite work properly. The difference is they either 'catch' i.e. a compelling bit of software forces sales and a market is created or they don't. Apple had the spreadsheet, Microsoft had the IBM pc, There is always something. All these things need a bit of luck, a bit of tech co-incidence, a bit of serendipity. Sinclair was inventive and got lucky quite a few times but it was bound to run out in the end. He was his own worst enemy.
Well researched, well written and fascinating article - Thank You.
I lusted impecuniously after a QL but was under no illusions about the microdrives as I had plenty of opportunity to play with OPDs at work. The moment there was a QL overnight price drop from 400 to 200 pounds that was it and I rushed out to Dixons to get one.
Not-so-fond memories of trying to load a huge planetary orbit calculator from microdrive though :( I agree those devices were what finally pushed the perception of the QL onto the wrong side of the usability line.
The biggest problem with the Microdrives was that they created not so much a walled garden, but a bricked-over garden, isolating any data produced on a Sinclair machine from the rest of the entire universe. With any form of disk system you can transfer data to other systems by writing code to suck the data off the disks. The first thing I did when I got a BBC B was connect my Spectrum to it and dump over all the microdrives onto disk.
Great article, some interesting new stuff there, but a couple of minor errors: all QLs had the BS6312 (BT phone-plug) type sockets for the RS232 ports, except for the ones built by Samsung, which were only for export markets; and the ultimate CST Thor model was the Thor XVI which was a complete hardware redesign using an MC68000 (the Thor 20 had some kind of daughterboard on the QL motherboard for its MC68020 CPU).
Similar to the BT connectors but with the latch on the opposite edge and possibly a different keying arrangement IIRC?
I knew of at least half a dozen ICL OPD's that had an office life of around a decade. Established users didn't want to get rid of them.
You should watch the drama 'Micro men', with Alex Armstrong simply great as Sir Clive (and Martin Freeman as Chris Curry).
It basically puts the demise of both (Sinclair and Acorn) once great companies down to a simple thing:
Acorn were very successful with a mid-market product in the BBC Micro; but went down market with the Electron and boom, was their undoing
Sinclair were very successful with an entry level product 80/81/Spectrum; but went up market with the QL and boom, was their undoing
IIRC they were both designed during the boom and in enough numbers to mortally wound them during the bust.
Micro Men is somewhere on YouTube in 10 or so parts. Fond memories.
All in one byte:
Duur. It has taken me 33 years to realize "ZX80" refers to "1980", ZX81 -> 1981. I assumed it was something to do with the microprocessor.
Microdrives were ok if you formatted them several times before you used them in anger - this stretched the tape until it settled down, and from then on they were fine.
My wife used to work for a public utility, and they used the Psion database/Wp to store and produce stock letters for customers. To help her out I asked her to borrow the manual and she brought home a strange plastic box with a wierd spring loaded catch built in - this turned out to house the install floppies, which nobody in the offvice even knew existed. I was able to install on my early PC (Amstrad) and rewrite many of the routines and templates, gaining her many brownie points. I liked the software a lot.
The QL was released when I was in my first year at Uni - doing a course which had a large element of microprocessor design. Our lecturer commented that it was a bit of sharp practice described the 68000 (lovely processor btw) as 16 bit. I still recall his comment ...
"It's a bit like sending of five quid for a coat hanger and cigarette lighter, to receive a bent nail and a red headed match"
He *may* have once worked for Sir Clive ....
Sinclair was just doing what IBM were doing in describing their 8088-based computer as '16-bit'.
The Spanish government, on seeing the home computer market take off, came up with a tax on machines with less than 64K of RAM which weren't localised in Spanish - or in other words a Spectrum tax.
A local manufacturer helped with localising the 48K+ to get round it and later on contributed to much of the design for the 128K+ as well as manufacturing because Sinclair himself was more interested with the QL.
I am the PM that firmware version was named after. Yay me. :)
I recall we named many of the ROM versions using the initials of various female secretaries and PA's who worked there. 'FH' being a particular favourite of one junior engineer named in the article.
So a self declared very clever man, so clever he couldn't take advice from anybody, drove his company over the cliff eh?
Thanks God that would never happen today.
Arrogance created incompetence and together they create disaster.
Nonsense. Sinclair, with Nigel Searle, wrote the trig algorithms for the world's first single-chip scientific calculatror, the Sinclair scientific. Even the chip manufacturer (TI) said it was impossible. 6 years later he virtually created the UK home computer market with the ZX80 and 81. To say nothing of the Spectrum, miniature TVs and digital watches.
As for "arrogant" and "not taking advice", these are both basic requirements for starting a business.
A properly clever bloke, brilliant inventor, tenacious entrepreneur and deserves his knighthood (unlike most "Sirs").
One of Clive's many strengths was his ability to motivate people to do the impossible.
For 2 years in mid 80s I tried my hand at working for myself doing computer repairs. I specialised in the QL and made a sort of living. The most common issue was the keyboard membrane; I got through hundreds. One thing I did try was to produce a battery back up add-in for the RTC.
From page 1:
"Alun (sic) Sugar, in his autobiography, calls Potter and co. “an arrogant bunch of tossers”."
With respect, Sir Alan, Lord of Sugar, you are also an arrogant bunch of tossers.*
*Opinion. Partly based on the unreliability of the 3" disk drive in my Spectrum +3.
Iwas loaned a QL a few years after it had lost most of its value and about a year before the Atari ST as widely available, so I did at least get to enjoy using it. Superbasic really was quite nice compared to what I had on my Spectrum. But those darn Microdrives, awful from beginning to end. Could write a chapter on my Spectrum Microdrive odyssey, finally ending with the Opus Discovery One 3.5" disk drive expansion box, with £6 blank floppies which were just so much more reliable.
As a keen QL owner, I suggested to the company i worked for that they would make good cheap programmable serial terminals for testing our X.25 PADs.
So they got one in. Unfortunately every time we connected a QL to a PAD terminal port, the PAD crashed.
We quickly worked out that the multiplexed serial ports worked by asserting hardware flow control off each time the other port was using the hardware. Unfortunately this happened at 50Hz, and our poor PADs couldn't cope with that many interrupts a second!
Still, there's a silver lining to every cloud. When the company was taken over I was in charge of a project they canned, and so was I. However as an informal consolation they gave me the QL so now I have two :-)
DTraceunder the GPL
Biting the hand that feeds IT © 1998–2018