GARTHWILSON wrote:
I have no experience with the 6809; but according to this post and this one, the 6809 had really no execution speed advantage over the 6502—and that's at a given clock rate; but the '02 went to much, much faster clock rates. If you read slightly further down that first thread, you'll notice that the issue in the post you reference was ported 6800 code not using the auto-increment addressing mode. The second one comes right out and says, "The results show that 6809 at 1.78 MHz matches 6502 at 2 MHz. The second accumulator gives a big advantage to 6809." And that's for math-heavy code that does not play to the 6809's strengths.
I considered bringing up that the 6502s were eventually clocked much higher than the 6809, but it seemed to be getting a bit off-topic. But what you need to consider there is that they clocked much higher because the 6809 was effectively abandoned much earlier than the 6502 was (if the 6502 can even be considered "abandoned" now.) By the early '80s Motorola (and Hitachi) really had no reason to continue 6809 development when they already had the 6811 for the low-end and the 68000 for the high end.
Quote:
Quote:
The "who can do more with fewer chips?" battle is not one you want to get into with Motorola, because they will win.
The only single-chip 65xx solutions you'll find today are in ASICs, in automotive, industrial, toy, appliance, and even life-support equipment, and since they're custom ICs with everything onboard and the processor at their heart being licensed by WDC, they won't be marked with anything you'd recognize, like having any "65" in the name. However, Mike Naberezny de-capped the controller IC in the control panel of his VW Jetta, in his project to make kind of an information center, and found that the processor in it was a 65c02.
https://wdc65xx.com/about-WDC/ says: "WDC and its licensees have shipped over eight billion CMOS processors; more than 1 per person on planet earth! This number is growing by hundreds of millions of units each year."
Sure, but 8 billion units over four decades is not a lot, for the microcontroller market. The total number of MCUs shipped in 2021 alone was
over 30 billion. I can't find stats for the 6811 (and the 6808, which I think it would be fair to include), and they may be impossible to get given that many different companies (Motorola, Hitachi, Maxim, Freescale, NXP, ...) have been shipping these, but given that the 6502 basically missed the automotive market, I'd be surprised if they'd shipped anywhere near the quantities that 6800, 6808 and 6811 cores have, which have been huge that market from the start. (Yes, I'm aware that WDC lists one automotive application on that page, a dashboard controller. But consider that your average car these days has 35 microcontrollers in it.)
We'll probably never know for sure, given that as you point out ASICs using a particular core are hard to see, and
soft cores are virtually invisible. But given that the 6502 started from far behind and is poorly suited for C code compared even to the 68HC11, much less more modern 8-bit CPU designs such as AVR, I think it's a pretty safe guess that it never caught up.
Note that this says nothing about the 6502's design, which was brilliant for the time, and in virtually all ways better than the original 6800. The main reasons Motorola came out so far ahead in the MCU market were being there first and, probably even more importantly, having the marketing and history within the automotive industry to sell it into that market. (Around the early '80s it may have helped a bit that Motorola's architecture allowed them to easily and compatibly fix some of their largest technical problems, such as by adding the Y register, but probably not all that much.)
Quote:
I had mentioned stack-relative addressing earlier, for stack frames, but this is also a good point which I was forgetting, regarding finding inlined data as well, and adjusting the return address to skip over the data....
Yup, that's another good example; there are probably various other techniques like that that I've forgotten to mention, too.
Login
Register
FAQ
Search
It wasn't long until I realized many 6502-type algorithms are either inefficient on or downright wrong for the 816. Once I got over that and learned, among other things, to use the stack as ephemeral workspace and not just a place to park the registers, I found myself writing smaller code that was getting a lot more done.
That they are often started and stoked by individuals with little-to-no practical experience with the 6502 and its assembly language comes as no surprise.
