Thank you for following my TMD-3: Turing Machine Demonstrator Mark 3 project Michael. 

Hi Michael,

I have posted a LogiSim for a 4 bir SubLEq if your interested. It is as simple as I can make it.

https://hackaday.io/project/181600-subleq-revisted

Worth a look if your still interested in SubLEq.

Regards AlanX

Hi AlanX, thanks for the reference. Yes I am still interested. Had the SubLeq board I have (80% complete) at a Hacker meetup and a few people were very interested, and I have not shelved the project, it has just stood still. I started making an animation of sorts of the block diagram to explain it to people. After I have "rebooted" my mancave/workshop (it's horribly overfilled at the moment) work will continue.

Incidentally my SubLeq is 8 bit (by nature of SubLeq you cant have different address and data widths...(?))

Hi Michael,

Yes, you can have different address/data width. You can use paging if the address width is wider. If the data width is wider or wider than the available memory, then the high bits can be used for opcodes. Usually, just the high bit, it is used to avoid the self-modifying code (something that basically does my head in each time I look at it) that is necessary for "pointer/indirect" move or "stack" type operations.

I used virtual ROM/RAM for one of my 8 bit TTA CPUs:

* RAM 0xC0-0xFF (Fixed RAM Page 0xFF: Used for registers/global data etc)

* RAM 0x80-0xB7 (User RAM Page 0x00-0xFE: Used for program/data etc)

* ROM 0x00-0x77 (User ROM Page 0x00-0xFF: Used for boot/IO/system etc)

The page registers are memory mapped. I could therefore access 16kb RAM and 32kb ROM but paging is messy.

Counting chips my version has ~15 chips for the 4 bit version and there after ~10 chips per 4 bits. Still, I am not building this one as it was an exercise to learn LogiSim and to try out an improved method of Hardware Control Logic.

Regards AlanX

Hi Micheal,
Thanks for following the CHUMP 4-Bit. I commented on your SubLEq project I while back.

Anyway, the CHUMP project looks to me to be one of he best simple CPU projects I have come across. The original CHUMP is not mine, but my version is somewhat extended. At 16 chips quite doable on bread board and surprisingly powerful.

I am currently teaching myself LogiSim, perhaps I should have done this before sending my PCB off for manufacture! Hopefully LogiSim will prove the schematic is viable.

Regards Alan