ARM: "inspired by" 6502?

[BigEd]

From another discussion:

GARTHWILSON wrote:

Someone pointed to an article last year telling about the ARM being inspired by the 6502, but I didn't bookmark it and I can't remember the right search terms to find it again.

I thought I'd bring it over here to a new thread.

[BigEd]

Here's Tor's response from that other discussion:

Tor wrote:

GARTHWILSON wrote:

Someone pointed to an article last year telling about the ARM being inspired by the 6502, but I didn't bookmark it and I can't remember the right search terms to find it again.

I remembered having read that as well, and I remembered what I was searching for at the time - I was reading obituaries about Personal Computer World (I was a subscriber from the beginning back when I was 17. I learned English by reading that magazine. But I digress.) So I just went through the same chain of searches and ended up here: http://www.theregister.co.uk/2009/06/11/pcw/page2.html

Quote:

Inmos? Ask anybody in the street today: “Never heard of it.”

Let’s not list all the things that went wrong. There is one star survival, heritage, legacy, what you will, from the UK industry in the 80s. Sophie Wilson, the best 6502 programmer ever, became disappointed with what she could do with the BBC Micro, and went off on her own to design a RISC processor that would do all the good things she liked about the 6502, and all the other things which she wished the 6502 could do.

The chip was the Acorn Risc Machine – the ARM, which started out merely as “the chip inside the Acorn Archimedes”.

Edit: A less poetic reference is maybe: http://www.engineersgarage.com/articles ... processors
which merely refers to ".. with latencies as low as that of the 6502".
There's another one here, an interview with Steve Furber (the other designer of te ARM processor), where he refers to the 6502: http://queue.acm.org/detail.cfm?id=1716385

Quote:

What we found was that the 16-bit micros in the early 1980s had worse interrupt response time than the 6502.

-Tor

[BigEd]

And mine:

BigEd wrote:

On the business of 6502 inspiring the ARM, some of the story is elaborated on wikipedia, but Sophie maintains that "inspired by" isn't the right choice of words. Other people, myself included, also Sam, do use that word.

Cheers
Ed

[BigEd]

I found another source, with these phrases:

Quote:

...the design team decided to develop their own processor, which would provide an environment with some similarities to the familiar 6502 instruction set but lead Acorn and its products directly into the world of 32-bit computing.

Quote:

The design team worked in secret to create a chip which met their requirements. As described earlier, these were for a processor which retained the ethos of the 6502 but in a 32-bit RISC environment

Quote:

The original objective of the ARM design team was to produce a processor which provided a logical advance from the 6502 processor, and was suitable for use as the central processor of a business or home computer.

Quote:

The 6502, to which Acorn's designers looked when designing the original ARM, had a short and simple instruction set which lent itself well to RISC.

The web page is titled "The history of the ARM CPU" and declares as a footer:

Quote:

Taken from 'The ARM RISC Chip: A Programmers' Guide' by Carol Atack and Alex van Someren, published 1993 by Addison-Wesley

Edit: first chapter available on the internet archive here.

[ElEctric_EyE]

Very interesting history! I'm still reading... I was never aware that it all sprouted from the Acorn.

That book is $93 used on Amazon.com!

Can one purchase an ARM cpu, with address and data bus pins or are they always internally mated to an interface protocol? They're very cheap... If not I guess that ARM core on opencores is worth looking at (just out of curiousity's sake)...

[BitWise]

ElEctric_EyE wrote:

Very interesting history! I'm still reading... I was never aware that it all sprouted from the Acorn.

In 1985 I visited Acorn in Cambridge and met the folks sat at the back of their office designing the RISC chip. At the time I didn't pay it much attention I was too busy trying to get my database package running multi-user over ECONET.

[Tor]

This interview about the Acorn 6502 kit has a lot of funny or interesting facts: http://speleotrove.com/acorn/acornWilson.html

(including the one about the guy who assembled his Acorn kit with glue to avoid heat-damage from soldering..)

Several of the other links you can find out there (e.g. this one) will mention that the Acorn System 1 was based on a Sophie Wilson-designed 6502-powered cow feeder..

From that same link about the ARM development:

Quote:

Acorn's CEO at the time, Hermann Hauser, recalls that "while IBM spent months simulating their instruction sets on large mainframes, Sophie did it all in her brain."

-Tor

[GARTHWILSON]

Quote:

Acorn's CEO at the time, Hermann Hauser, recalls that "while IBM spent months simulating their instruction sets on large mainframes, Sophie did it all in her brain."

Now that is a real engineer!

[BigEd]

GARTHWILSON wrote:

Someone pointed to an article last year telling about the ARM being inspired by the 6502, but I didn't bookmark it and I can't remember the right search terms to find it again.

Found another interesting piece on ARM ("Great Microprocessors of the Past and Present", cpushack.com):

Quote:

A seven-member team created the first version in a year and a half, including four support chips. ... The instruction set is reminiscent of the 6502, used in Acorns earlier computers.

There's a nice description of the architectural features and comparisons to other processors too. Summary:

Quote:

These features make ARM code both dense (unlike most load-store processors) and efficient, despite the relatively low clock rate and short pipeline - it is roughly equivalent to a much more complex 80486 in speed.

[ChuckT]

BigEd wrote:

From another discussion:

GARTHWILSON wrote:

Someone pointed to an article last year telling about the ARM being inspired by the 6502, but I didn't bookmark it and I can't remember the right search terms to find it again.

I thought I'd bring it over here to a new thread.

Its been over 20 years since I read about this. What I remember is the 6502 requires other chips and ARM was designed to have the other chip components onboard. This is probably the part they forget to mention. I can't remember the particulars because it has been so long.

The 6502 wasn't fast enough for their graphical requirements so this is why they started their own design apart from licensing.

[GARTHWILSON]

Quote:

What I remember is the 6502 requires other chips and ARM was designed to have the other chip components onboard.

It sounds like you mean microcontrollers v. microprocessors. The 6502 by itself is a microprocessor. There have been various microcontrollers made with a 6502 inside them and then they have I/O and timers and memory etc. all on the same IC. The 6502 licensee that's running one over 200MHz is definitely doing this. I don't think you could get anything to run on a 6502-type bus at that speed if it were taken offboard, ie, outside the chip that has the processor in it. The timings are too tight for that. Maybe it could be done with BGAs and .002" traces and super-tight layout on a thin multilayer board with parts on both side and blind and burried vias. It would be fun but not be economically feasible. Unfortunately there are almost no general-purpose, off-the-shelf, 6502-based microcontrollers available today. WDC's 65134 and, with the '816, the 65265, are nice, but they don't allow you to program their ROM at your workbench, and, if you have outboard memory, you forfeit a lot of the I/O pins.

[BigDumbDinosaur]

GARTHWILSON wrote:

Unfortunately there are almost no general-purpose, off-the-shelf, 6502-based microcontrollers available today. WDC's 65134 and, with the '816, the 65265, are nice, but they don't allow you to program their ROM at your workbench, and, if you have outboard memory, you forfeit a lot of the I/O pins.

You precisely described why I built POC as a discrete MPU design. I didn't want to contend with someone else's notion on how it should work, or struggle with reconciling RAM requirements with available I/O paths.

[RichTW]

There are certainly some similarities in the instruction set design of the 6502 and ARM - whether the latter was 'inspired' by the former remains to be seen, but I'd be surprised if there wasn't at least a little influence.

The arithmetic flags (NVZC) are the same in each, and also used in exactly the same ways, for example:

• bitwise operations only affect N and Z, and leave the others untouched.
• C and V flags are used and set identically by adds/subtracts in both CPUs.
• Subtract operations use C as 'borrow' in both CPUs.

For what it's worth, the above is not true of (for example) the Z80.

Then - it's a small point - but the instruction set uses similar nomenclature to that of the 6502, for example:

• # to indicate an immediate operand
• EOR rather than XOR.
• Shifts called ASL, LSR and ROR (as well as ASR).
• Three character mnemonics for nearly everything (including the somewhat contrived ORR), like the 6502.
• Condition codes use 6502 type nomenclature, e.g. NE rather than NZ (as in BNE)

Either way, the ARM is a work of genius in terms of fitting such a rich symmetrical instruction set into a 32 bit word. It's a shame that more recent versions have lost this symmetry and become far more complicated - but that's progress...

[BigEd]

slightly off topic, but here's a pdf describing thumb2, ARM's second attempt to create a compact 16-bit instruction set, which seems to have got significantly better tradeoff of performance versus density:

For example, there's a multi-predication instruction (because there's no room for bits in every instruction) to setup the next four instructions to be conditional in various ways. Not sure how that plays with interrupts.

[bogax]

GARTHWILSON wrote:

Quote:

Acorn's CEO at the time, Hermann Hauser, recalls that "while IBM spent months simulating their instruction sets on large mainframes, Sophie did it all in her brain."

Now that is a real engineer!

There have been several usenet threads that touched on this topic.

I vaguely recall a post by Sophie Wilson, something to the effect
of them simulating the ARM across several BBC micros in BBC Basic.