http://6502.org/documents/datasheets/rockwell/rockwell_r650x_r651x.pdf

That data sheet has a number of different versions of the 6502. There’s a 28 pin one that does have IRQ, the R6504. It doesn’t say in the data sheet but I think that the missing address pins are just not accessible but are still there on the die, so the 8k range just repeats like if A15-A13 were not used on a normal 6502. So the vectors are just the same but they are also repeated 8 times in the address space. Same with stack and zero page etc.

Thanks for the link.

The same mirroring will apply to any RAM/ROM/IO included in the project using the 13 address lines that are brought out of the part. The address map of the Atari 2600 is quite complex because of the way the address lines were connected to the various devices: TIA, RIOT, and ROM cartridge. 4kB appear to be reserved for the cartridge: A12 = 1 appears to select the cartridge. This makes sense in context of the 2600, since that puts the reset vector under control of the game developer. The TIA and RIOT are selected using A12 = 0, and various other combinations of the the lower address lines like A9 and A7. It also appears that the TIA is not connected to RnW, but uses A3/A4 to control writes to some of its registers.

The mirroring of the address space in this project may present some issues, but it shouldn't be too hard to manage by using some care in programming.

[Note: I am definitely not a fan of incomplete address decoding. The mirroring of the various addresses spaces that result when incomplete address decoding is used can often result in some difficult to detect programming issues.]

_________________
Michael A.

How so? I've been doing this for 30 years and never had a problem. Not once. Complete address decoding requires more levels of glue-logic gates though, making a big reduction in your maximum speed.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

The 6507 does have a RDY pin, so it should be possible to rig that up as a precise "wait for signal" device mapped to some particular address. That would save the need for active polling; you only need to read some address, and the read stalls until the /IRQ line goes low.

As for rules, I think "just this 28-pin NMOS CPU" and "maximum 2KB ROM" and "use the RIOT somehow" are sufficiently limiting to define the contest. I can think of quite a few useful (or merely interesting) things that it should be feasible to build in that space, some of which would also fit comfortably within the 128 bytes of RAM provided by the RIOT, but some would not. And some would probably have trouble fitting within a particular component budget, though not for computing capability reasons.

Personally I wouldn't worry about limiting the project budget. It's by no means assured that someone spending $250 will be any more likely to finish their project, or produce something remarkable, than someone spending $25. Spend what you like, be sure to enjoy yourself, don't worry about winning, be happy to take part.

One of the things I had in mind was making the best of these anachronistic parts. That these parts would constitute the main-brain of the project. I'm not going to put a pending limit on it, but what I'd rather not see is a $10,000 FPGA on one port of the 6532, a Raspberry Pi-4B on the other with the 6507 acting as a bottleneck between them. That would not be the spirit of the concept.

Think: You are a late 1970's / early1980's engineer and you have to build the latest thingamajig, but it has to be built down to a price to give the most bang for the buck. KISS (Keep It Simple and Smart)

I think we are getting close to being able to put together the formal statement of the competition and the judging criteria. I'll take a first stab at that this coming Tuesday (too busy until then). I'll try to incorporate any other ideas that are generated between now and then too.Still need another judge though. Garth, BDD, Dr. Jefyll ... anyone?

_________________
Bill

Well, I think I have something in mind that does fit the spirit of the contest. The extra parts involved are not at all silly for 1980s technology.

I also think that you can take care of the cost argument by introducing it as a judging criterion. How much bang do you get for the buck? How much would you expect to pay for this sort of gadget on the high street (in appropriate era)? The FPGA and Raspberry Pi are already excluded, anyway, by the 2KB ROM limitation.

Are you sure you don't want to just let the community vote?

A community vote sounds good. To have appointed judges seems too stuffy for this forum -- it somehow doesn't feel right to me. It's not as if we're going to be receiving dozens of entries. There may only be a handful.

Going one step further, what if there were no voting or judging at all? Like this, maybe:

• Anyone who manages to get a design working and describe it becomes a winner. The prizes BillO has provided (thanks, Bill!) would go out chronologically, with the first successful applicant getting first choice, the second applicant getting second choice, and so on.

I know this changes the spirit of the the thing somewhat, and some folks may miss the excitement of having winners and losers. Be that as it may, there's one pure advantage I'm sure of. The whole thing becomes a lot easier to administer. That's because there's no need to have all the entries assembled by a certain deadline. Also need for the fun to end -- even years from now we may still get a straggler or two.

It's BillO's call, of course -- I'm only making a suggestion (and I'm unlikely to participate, unfortunately). I think the basic goals would still be satisfied, with lots of fun and some terrific entries. And I hardly suppose that anyone here would try to cheat by rushing in with a hasty and uninteresting project.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

I concur with Jeff. This is a case, methinks, of the journey being more important than the destination. As soon as it turns into a race it's no longer fun.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

That's exactly what it would be if prizes go out in the order people apply. The difference is quality (either according to judges or to the community) no longer plays a role. Just my humble 2 cents but the friendly competition aspect is really fun.

Yup, I agree... and I predict the friendly competition will still be there. It's just that winners would be recognized only tacitly. Each forum member can judge privately.

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

I agree with the no-judging suggestion as being a bit more friendly.

Perhaps an improvement: BillO announces a schedule for prizegiving, such as 6 months from now, and will give out prizes at that time. If as many projects as there are prizes have already declared they are ready at an earlier time, he gives out prizes at that time. If at 6 months there are still more prizes than projects, he announces a suitable extension, such as another 6 months.

Sorry, my workload is heavier than I had thought.

I like the idea of the project being friendly but not sure how the no judging thing will work. How would I determine who gets what?

It would be fairly easy to blink an LED or do some other trivial thing (not saying anyone would, but...) and not take a lot of time to design, build and debug it. Should that contestant, because they were one of the first, get equal access to the prizes?

As it is, there are certainly enough prizes to go around, and everyone that completes an entry will get a prize, but some are "better" than others - like having first choice.

Just a thought.

_________________
Bill

My inclination is to suggest not over-thinking the judging and prize-giving: a lottery is something everyone knows about, so you could make it random. Or you could apply some judgement - perhaps with the aid of a BUGS self-assessment - so that people get to choose their prize, in some order that's not quite arbitrary. But the judgement isn't strict - it never can be - two projects can have different strengths and weaknesses and not be comparable. (Much like two microprocessors or two text editors...)

I think the trick now is to get some people to show an interest in building something from your components. Maybe with a bit of PR elsewhere on the web - we might get some new members! (I think Ben Eater's videos might have brought in a few recent joiners and participants.)