This is not related to the 6502 at all, in fact, it doesn't even have to do with any processor: An amazing project with a BYO computer made only of ROM and RAM: http://2x-1.net/ob/gray1/ . Posted here because I'm guessing a lot of people will love the details of how this works. The creator is Olivier Bailleux, associate professor in computer science at Burgundy University. Via Hackaday.

Neat link. Thanks for posting it.

You're right. I love it!

That contraption should be named Big Mess O'ROM.

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

All functions can be replaced with a sufficiently large lookup table, including CPU functions.

_________________
WFDis Interactive 6502 Disassembler
AcheronVM: A Reconfigurable 16-bit Virtual CPU for the 6502 Microprocessor

It reminds me of some of the ingenious things people did with RAMs and ROMs when these were 'new'.

For example, some of the early radio gear with digital displays and switch-selectable programmable memories. Or things like TV Typewriter.

It's great to see people thinking this way instead of the usual "need Arduino," or "need Linux" mentality so prevalent today.

The host adapter of the original Lt. Kernal (LK) hard drive subsystem for the Commodore 64 used ROM for not only combinatorial logic it implemented a state machine for bus sequencing and timing. At the time of its development in 1984, the LK was using the old SASI (Shugart Associates System Interface) bus, for which no interface devices existed (SASI eventually morphed into SCSI, which was standardized in 1986). Ergo the LK's host adapter had to not only drive the SCSI bus (using the MC6821 PIA), it had to execute the bus protocol, which in addition to requiring that signals be manipulated in the correct order, had to done within strict timing deadlines. It worked well enough to produce a bus throughput of 38KB per second, which was pretty darn fast for a computer with a 1 MHz, eight bit MPU.

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

Fascinating, thanks for sharing.

I see it's filed under "Unusual Projects" - no kidding!

The idea of a double-clocked Gray-coded program counter, now that's very interesting.

Yup. But to deny himself the use of registers -- either transparent or edge-triggered -- is really pushing the envelope! This fellow has won my admiration -- very hardcore! Thanks for posting this, Scot. Folks who found this interesting may also enjoy the following project.

-- Jeff

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

Just like a 6502! It has a two-phase clock.

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

He got it to work without output flops or latches by using a gray code address
generator and a two phase clock on A0.
Dwight