In many ways, the tuple (BBC Micro, Acorn Computers, arm) is analogous to (IBM PC, Intel, x86).
There was a radical difference in the relationship between the two corporations in each tuple. In the BBC-Acorn relationship, Acorn designed and manufactured the computer; BBC just offered their brand, did marketing, and supplied some high-level requirements. In the IBM-Intel relationship, IBM designed and manufactured the computer, and Intel was the CPU vendor, with many other customers. The 6502s used in the pre-ARM BBC systems were from MOS Technology–or one of their licensees, such as GTE/CMD–so those companies were really the Intel equivalent here
CMD: Commodore Micro-Devices
There is another CMD, Creative Micro Designs, who sold aftermarket peripherals for Commodore 64/etc
Remarkably, the movie Micro Men also overlooked much of Sophie Wilson’s role, despite her work on ARM becoming one of the most significant technological advances in computing history, accept for a token cameo role as the pub landlady.
We should not make assumptions about someone’s gender identity before they “transitioned”. She was living as a man, but we don’t know how she identified at the time.
But as I said in an edit, it’s bad form to deadname someone even if you are referring to a period when they went by the deadname.
I think the OP was just saying that her role was underplayed more than they were complaining about the title. If you check their comment it says nothing at all about gender.
We talk about the past events all the time using names that weren’t applicable during the relevant time period. The Aztecs didn’t call themselves Aztecs. This shouldn’t be a difficult concept in general.
I try really hard to avoid getting anywhere near these contentious things ... but I think Wikipedia's handling of this seems reasonable, at least for some value of reasonable.
If there was a docu-drama about my early days, I would expect them to use my birth name, rather than my married name. Unless the events happened after my marriage.
Quoting the OP:
"Notably, no mention of Sophie Wilson"
The OP's question was literally asking about why the name Sophie Wilson was not mentioned or given proper credit for their contribution. Please stop twisting it to make it seem like there has been some transgression or slight, that simply does not exist.
I am not accusing anyone of any transgressions. I think you’ve just misinterpreted the OP’s comment as being about gender (as they’ve now confirmed here: https://news.ycombinator.com/item?id=44939643)
> To make a short reference to something.
Sense 2 (the sense you were thinking about) is a specialised sense used in "philosophy, linguistics", and even then the context makes it clear when this sense is meant. 'No mention of "Sophie Wilson"' might conceivably be referring to the name, but 'No mention of Sophie Wilson' refers to the person.
Historical retrospectives show systematic erasure of trans women's contributions to STEM. (Certainly this happens in other fields, too, but I haven't studied them enough to notice the pattern.) This is worth talking about, if it has happened here, and does not need to be derailed by a pointless semantics argument.
That's the general rule, but some people make exceptions. Sophie Wilson was involved in the production of Micro Men, so presumably signed off on however she was depicted in it. (Then again, Clive Sinclair was also involved, and per https://web.archive.org/web/20250711183307/https://www.indep... objected to his portrayal, so…)
The irony here is that concepts like "living as a man" and "living as a woman" are inherently sexist.
Except the BBC micro didn't use an ARM processor - it used a 6502. Whereas the IBM PC did use the Intel processor.
https://www.retro-kit.co.uk/page.cfm/content/Acorn-BBC-Archi...
And the IBM PC used an 8088.
To that end the ARM instruction set was heavily inspired by the 6502 in the Beeb and cruicially the BBC Micro was used to simulate the ARM before it went into production.
Latter the original ARM development kits were connected as second processors to Beebs courtesy of the Tube connector.
I think it's fare to say that without the BBC Micro there would be no ARM processors.
https://arstechnica.com/gadgets/2022/09/a-history-of-arm-par...
That seems an unlikely goal given BBC BASIC was interpreted! Happy to be proven wrong but I've never heard one of the original team say that.
> The second thing they didn’t do was that they weren’t fast, they weren’t easy to use. We were used to programming the 6502 in the machine code and we rather hoped that we could get to a power level such that if you wrote in a higher level language you could achieve the same types of results. So you could write 3D graphics games. You could do whatever you wanted to do without having to go all the way down to assembly language and for these processors that were on sale at the time that wasn’t true. They were too slow. So between the two things we felt we needed a better processor.
https://archive.computerhistory.org/resources/access/text/20...
I think it’s possible that some applications would be faster - floating point for example due to the ARM’s 32 bit registers and the 6502’s lack of even 16 bit arithmetic - but probably not in general.
The details of "Does that mean the same speed, or faster?" weren't really relevant, because the speed difference was anywhere between 10X and 100X - which was an astounding, if rather fuzzy, target.
ARM 1 got surprisingly close to that out of the gate, with the added benefit of being unexpectedly power efficient.
It also had more (16 user registers vs, counting optimistically, 3) and larger (32 bit vs 8bit) registers, compared to the 8 bit registers of the 6502, and a 3-stage pipeline.
I expect that means the BASIC interpreter kept the current program position in a register, where the 6502 one used memory (likely in self modifying code), and could fetch the next token in a single cycle vs at least 5 or so for the 6502 version.
Having a faster CPU and more memory also may have meant they could be smarter in the way programs get stored.
I guess all that combined means there are programs were that goal can be met, for example programs computing a Mandelbrot image.
ARM2 added a hardware multiplier.
Whereas I suspect that I am nowhere near the only person on this page who once disassembled ROMs on a BBC Micro. We can state, in contrast, that there was no such self-modifying code. Again, BBC BASIC was in ROM.
Those lucky enough to have a copy of Jeremy Ruston's book after all of these years, or the retrocomputing enthusiasts who still have working Beebs, could even tell you exactly where in ROM the code was that fetched the next token for execution.
I never actually owned a copy of the book, and somewhat envy anyone who still has a copy; although to compensate I do have part of one of my own disassembly listings still, buried somewhere. (-:
I know nothing about the first ARM, but ARM2 of Archimedes (anno 1987) was significantly faster than a MC68k (both at 8MHz), both much faster than a 6502 at one (typical) or two (in the beeb) MHz.
A BASIC interpreter using the ARM 1 or 2 might not have been literally faster than machine code on a 6502 (certainly not for some silly micro benchmarks), but, the stated goal, allowing high level programming where earlier assembly was required, certainly was met.
One of the optimizations Google introduced on Android 7, when they backtracked for AOT at installation time introduced in Android 5, was that the DEX interpreter was rewriten from scratch in cleverly manually written Assembly, before handing it over to the JIT/AOT infrastructure.
BBC Basic was also one of the few that allowed direct inline Assembly, instead of having to go through DATA blocks.
Just to make the comparison more concrete the ARM1 ran at 8MHz and was 32-bit and with a lot more registers compared to the BBC Micro's 6502 running at 2MHz. It was a lot faster but not fast enough to make BBC BASIC run at BBC Micro Assembly language speed.
Most folks nowadays wouldn't even think that is an option, hence the comparison with a product several decades later doing that instead of C kind of approach.
Back to ARM v1, maybe the design did take into account how to improve the developer's life of those writing in Assembly, which was critical for implementation of the whole userspace, meaning BBC Basic.
It really is.
Both Acorn (BBC Micro designers/manufacturers) and Apple Computer (Apple II ditto) bought in the MOS 6502 chip.
When it came to successor models, both tried the 65C816.
Acorn made the Communicator:
https://chrisacorns.computinghistory.org.uk/Computers/Commun...
Apple made the Apple IIgs:
https://americanhistory.si.edu/collections/object/nmah_55324...
Both were not delighted with its performance and looked elsewhere.
Apple went for the 68000 for its next mass-market model, the Mac. (I am aware the timeline is more complex; this is a simplification.)
Acorn evaluated the 68000, the 80186, the NatSemi 16032 and others.
It designed its own chip instead: the ARM, Acorn RISC Machine.
This was first launched as an add-on accelerator for the BBC Micro.
https://www.computinghistory.org.uk/det/48862/Acorn-1MB-A-Se...
It was later launched as an Acorn computer, running a 32-bit port of Acorn's MOS and BASIC. It included a 6502 emulators and so could run some BBC Micro software.
https://www.onirom.fr/wiki/blog/21-04-2022_Acorn-Archimedes/
That ARM chip is the direct ancestor of the chip in all Android devices, all iOS devices, and modern Macs.
Besides, it did say 'in many ways' so I think that this is really needless nitpicking.
I owned a BBC Micro Model B (was given one when a friend was given a 486 PC) and I used the Archimedes at school. The BBC Micro was archaic compared to the Archimedes and PCs of the time.
It was six years older and this was the era when CPUs evolved massively from one generation to the next. Just like the Acorn 'Atom' was 'nothing like the BBC Micro' in spite of using the same processor.
For some contrast: when the BBC Model A came out in 1981 originally it had cassette tape as mass storage, an 8 bit cpu clocked at 2 MHz, 16K RAM and if you were very lucky yours came with the optional floppy drive which cost nearly as much as the machine itself. When the first ARM was sold to the public, six years later it came with a 32 bit RISC CPU clocked at 4 MHz, 512K or 1M of RAM, an ST 506 based harddrive option.
That's just six years of progress, and we're skipping over many steps in the lineage, the BBC Master series, the tube expansions and the Olivetti saga. You could pick 1980 to 1990 and write a pretty large book about personal computing progress during those years and you likely would still miss important events.
But the lineage was - for those that owned all of the intermediary machines as well - pretty clear, and that is before we get into the lineage of the software that the Archimedes shipped with, MOS and BBC Basic, which both worked more or less as you would expect given the new machines capabilities.
My brother had one. Really cool machine, and as far as I remember, on a completely different level than anything that had existed before it. Soon succeeded by the Risc PC, which I mostly remember for being able to accept various configurations of additional processors (it could get either an x86 as co-processor, or several additional ARMs).
Pictures:
https://chrisacorns.computinghistory.org.uk/Computers/A3000....
Note the logo at top right of the keyboard.
Also note the bright red function keys.
As I wrote in another comment I was pretty close to the fire and had very early access to the ARM architecture based beebs courtesy of a friendly contact but I realized soon enough that the future for home computing and SMB business computing did not lie with either Acorn, Atari, Commodore or any of the other contenders. In '88 or so (my memory is a bit hazy about the dates, there was a lot going on in my life back then) I moved to x86 professionally with a side of Atari ST (using the fantastic Mark Williams C compiler + documentation) for more fun stuff and with the advent of the availability of the internet for the masses I ran SGI Irix for a couple of years until I settled on Linux which has been my daily driver for decades now. Hardware architecture used to be a super important factor for me, now the only thing that matters is whether or not I can run Ubuntu and whether the hardware is sufficiently powerful to get me through my working day. The fact that my daily driver is a 12 year old laptop is a nice indication of how far we've come, it is pretty rare that I put together machines where performance really matters.
But regardless of all of that I have a fond spot in my heart for the BBC, regardless of processor used, it was the machine that allowed me to finally do some more structured programming and explore other languages without breaking the bank.
This is not true. It had lots to do with it.
* The Archimedes used a CPU whose designers (Sophie Wilson and Steve Furber) have specifically said they built it to be conceptually similar to the 6502. Source: I have met both and seen them say this in person.
* The Archimedes ran RISC OS which is a rewrite of the BBC MOS. Source: I have interviewed the project leader, Paul Fellows.
https://www.theregister.com/2022/06/23/how_risc_os_happened/
The Archie came with !6052tube and could run some BBC apps.
Source code: http://bbc.nvg.org/rom/Acorn/os/?C=N;O=A
(detail: https://www.retro-kit.co.uk/user/custom/Acorn/32bit/A310/310... )
(full article for reference: https://www.retro-kit.co.uk/page.cfm/content/Acorn-BBC-Archi... )
... and besides, it runs BBC BASIC!
But in the 65XX family there is the 65816, a chip that tried really hard to maintain as much backward compatibility as possible. It saw some commercial deployment (Apple, Nintendo). At that point in time backwards compatibility began to have real value and intel really made some lucky calls: the weird addressing modes resulting from the lack of register width eventually culminated in a setup that worked very well for CPUs that were running multi-tasking OS's. The 386 was a very nice match for such code and this model was a major factor in the success of the line (which really was creaking badly with the 80286 out vs the 68K, which effectively had a 32 bit flat model built in because of its ability to run position independent code).
But in 1987, when the 80386 hit GA it was pretty much game over for the rest even if it took a while for the other empires to crumble, only ARM survived and that is mostly because Acorn had a completely different idea about power consumption and use of silicon than Intel did. The current crop of x86 hardware is insane in terms of power consumption and transistor count, ARM is so much more elegant (in spite of its warts).
>Notable exception: the 6800 is in many ways simply an improved 6502 but by a different manufacturer.
It is directly binary compatible. A Core i7 can boot and run MS-DOS from the 8088.
I offer a freebie DOS distro for modern PCs:
https://github.com/lproven/usb-dos
What more do you want than "executes the same binaries"?
As a schoolboy I was one of a handful who were in the computer club. We had a CBM (PET) 3016, a few Acorn System Ones and a UK101 that was built by our physics teacher.
One day this big grey prototyping keyboard case turned up. There was a microcassette unit fitted for loading and saving programs, and the whole thing was connected to a colour TV via an umbilical cord that looked like a vacuum cleaner hose.
We were given task sheets with projects to complete on this unit, and we could control the TV from the keyboard, read Teletext pages AND download programs.
It was a fun piece of kit that stayed with us for a couple of months.
In hindsight, I realised that the unit was a pre-production BBC Micro and we'd been part of a pre-launch test programme thanks to that same physics teacher.
(Looking at Wikipedia, I think they might have been R260s running RISC iX on the other partition).
https://chrisacorns.computinghistory.org.uk/Computers/A500.h...
I recall a few years ago writing a BBC basic program under Riscos that used the Raspberry Pi's BCM2835 undocumented random number generator - no problems. You can see how simple the source code is here:
Someone uploaded a video of it in action
Does this 60% number include laptops? If not, I'd wager that number is greater than 95%.
I think the right adjective for the reaction of those present was 'incredulous'. A small team with no previous experience had created a powerful 32-bit design from scratch when 8-bit architectures were still commonly used.
Had anyone told us that 40 years later we'd all be carrying around the 'descendants' of that first Acorn RISC Machine in our pockets then we'd have been utterly astonished.
Turns out the design was such a low power design that just the voltage from the data lines was enough to run the chip.
I always get the impression though that while the UK and European home computer era continued from a diverse eight-bit era of C64s, Spectrums, Amstrads and BBCs to the sixteen-bit era of Amigas and Atari STs, before the PC became dominant, in the US the early eight-bit home machines gave way much earlier to consoles - the NES at first, then the SNES and Megadrive.
At the time the Archimedes blew the nascent PC and every other machine out of the water, and yet couldn't get a toe-hold in the US market for reasons I've never quite understood. At the same point MS Windows looked shoddy at best in comparison to RiscOS.
RiscOS wasn't even on the table for the likes of IBM and that is what it would have taken to succeed in the business market. But for many years the preferred machine to create Videotext or ATEX (automatic typesetting system) bitstreams was to have a BBC micro and there were quite a few other such interesting niches. I still know of a few BBCs running art installations that have been going non-stop for close to 45 years now. Power supplies are the biggest problem but there are people that specialize in repairing them, and there are various DIY resources as well (videos, articles).
But I just don't think Oracle were able to sell it – and Oracle's sales people are really good, if they can't sell your product, the problem is likely the product or market fit not their sales ability
RiscOS simply had to start the whole OS cycle from scratch, it wasn't as good as what was already available on the Amiga and it wasn't Unix. It was fun to work with if you came from the BBC Micro it all made good sense and was a step up. But when looking at it from a corporate angle it wasn't quite what you'd expect from a workstation and it didn't run anything that you needed right there.
Did Oracle port any applications to it?
JavaOS was in ROM, on a module that can be removed (SIMM-style form factor). At one point people started to use BOOTP to run Linux compiled for SPARC as a replacement, as JavaOS was unpleasantly slow on JavaStation hardware.
So Sun Network Computers were JavaOS on SPARC, Oracle were NCOS (Acorn RiscOS derivative) on ARM – and I think IBM's had a similar tech stack to Oracle's...
were there any others?
I think it was just relative lack of apps in the end. Microsoft commodified the hardware so it became competitive and prices fell dramatically. Every other company stayed attached to their integrated designs and couldn't keep up on cost. Apple held on for a while because of the bigger US ecosystem and economy but nearly got wiped out also.
Also the RiscOS wasn't really backwards compatible with BBC apps and games, iirc. More like a clean-sheet design.
Acorn's CPU division is the most successful CPU design house in the world and sells around 10x more than all forms of Intel and Intel-compatible chips put together.
It was named after its first product, the Acorn RISC Machine: ARM. It is still called Arm Ltd. today.
Arm alone is one half of the entire CPU market.
https://morethanmoore.substack.com/p/arm-2025-q4-and-fy-fina...
An Acorn-compatible CPU is inside half of the processor-powered devices in the world.
How is that "dropping the ball"? It is the most successful processor design of all time, bar none.
It was such a pity. As a British schoolboy in the early 90s we had a mix of Acorns and PCs, and I had a BBC Model B at home and then a bit later also a PC. Very lucky in hindsight.
The Acorn machines were ridiculously better except for fewer games. At first I don't remember there being much of a gaming gap and there were plenty of games targeting the BBC Micros, but as games scaled up the bigger US economy started to matter much more and the app/game selection just wasn't as good.
But in terms of engineering the GUI was better than Windows, but more importantly the reliability was way higher. My primary school teachers (!) were constantly getting me to fix the computers or install new apps because they always broke. When an Acorn "broke" it was something like the printer being out of paper. When the PC "broke" it was always something much, much harder.
But the goal of a company is to survive, sell stuff, and make money. One division of Acorn survives, makes money, and dominates the industry, and the A in its name stands for Acorn.
(Some other bits survive inside Broadcom and things.)
It focussed on the successful bits and executed superbly. As the desktop PC industry consolidated on x86 and MS OSes it moved away. Good move. That's keeping your eye on the ball, in my book.
I can't think of any other company that did so well.
Sun, SGI, Cray, DEC, all either dead, or acquired, or sold on and split up, or sold off the divisions they were known for, and little or nothing of their tech lives on. IBM still makes POWER servers and workstations. That's about it. But not PCs.
Apple makes machines that use the Acorn ARM instruction set and can't run any binaries from their own PowerPC era kit, let alone 68k. It's doing great but by savagely chopping away legacy tech.
I think Acorn did great by comparison!
That's 4 years! The 386 came out in 1985 too.
I think Acorn did quite well and its legacy still lives on through ARM. Where's IBM in the desktop or CPU market?
The Beeb was a very well engineered machine, including the BASIC (which allowed in-line assembly and also allowed its functions to be called from assembly, ie other programs).
That’s my understanding as well. In the US the NES was huge in the late 80s, but in the UK home computers were dominant. The NES never sold well in the UK.
The 16-bit consoles did later on, though. So did the 8-bit Sega Master System, but not until the early 90s - it wasn’t a predecessor to the 16-bit machines, but a budget-friendly contemporary.
The Archimedes was powered by a 32-bit ARM 2 and it was awesome. :D
IMHO the GUI was better on RiscOS and being able to run video at 25fps in just software felt like magic. At the time, I never saw Amigas doing that without expensive hardware like the Video Toaster, even Amiga A3000s.
You could even get 12.5fps video off floppy disc which seemed crazy at the time.
And yet... Do you own a smartphone of any kind?
It has an Acorn-compatible CPU inside it. In fact, if it isn't 20+ years old, it has several: it has a multicore main CPU with several different Arm cores, and there are more in the Wifi controller, and more in the Bluetooth controller.
There is a pretty good chance that if you own an x86 machine with wifi, it includes multiple Arm cores too. Whatever OS you run, from Windows to BSD, if you were to search your SSD, you will find BLOBs of Arm code on it.
Is there any Amiga or ST derived tech in them? Not that I know of. But a company with "A" for Acorn in its name is in very nearly every device with a microprocessor.
The Apple II would be an example of the opposite.
The BBC Micro was 8 bit and a 6502 chip, that era had at least the following:
BBC Atom, Micro, Electron, Master
Commodore Pet, Vic32, Commodore 64
Atari 400/800 XL
Tandy TRS80
Oric Atmos
Sinclair ZX80, 81, Spectrum, QL
Amstrad CPC 464
Dragon 32/64
MSX machines
Edit: 8-bit data bus though, which I didn't know until reading up on the Motorola 68008 just now! Trust Uncle Clive to cheap-out as usual...
I cut my teeth on a ZX81 and even had a Spectrum +3 later on - that was the last gasp of the 8-bit Z80 Sinclair line, although the IP was owned by Amstrad by then.
You miscategorize most of the lines in this list.
> Commodore Amiga and Atari ST were 16-bit 68000 chips.
And the Mac which outsold both in the long run.
You missed:
Sinclair QL -- also a MC 680x0.
> The BBC Micro was 8 bit and a 6502 chip, that era had at least the following:
> BBC Atom, Micro, Electron, Master
> Commodore Pet, Vic32, Commodore 64
> Atari 400/800 XL
All 6502, yes. But you missed:
> Oric Atmos
Then you do not have a category for Zilog kit.
Powered by the Z80:
> Tandy TRS80
> Sinclair ZX80, 81, Spectrum, QL
Not the QL, no.
> Amstrad CPC 464
> MSX machines
Then another error. This line:
> Dragon 32/64
Is neither 6502 nor Z80. It is a Motorola 6809, along with 1 model of TRS-80.
Given this confusion I am not sure what you were trying to say.
The Amiga was a pretty impressive device with an OS that was fairly advanced. You could probably use it still today for word processing and sound design and not feel like you're missing much. The OS looks a lot like one of those super low-resource linux DE's.
Sophie Wilson and Steve Furber.
20 GOTO10
RUN
