Not sure when I found this, but we don't seem to have mentioned it here yet.

Titan is an 8 bit TTL CPU by Marc Cleave, with 16 byte sized registers, six of which are paired to provide the program counter and two stack pointers.

Its instructions include add and subtract without carry/borrow, an add with carry, a shift right into carry. There's no rotate. We have AND, OR, XOR and NOT, also INC and DEC. Register clear and register transfer. All these operations on any register or two registers. Stack operations push, pop, and peek. The status bits include zero and negative, with a carry bit, but no overflow.

The addressing modes include indirecting through a pair of registers - not clear if that's any pair or just an aligned adjacent pair.

Sounds like the TTL implementation is working. Maybe 7 wirewrapped cards. There's an assembler. Other people have had a go at writing emulators.

The technique he's using for wire-wrap is other than what you might expect. Almost all the connections are on the component side.

It's a valid tradeoff for some situations, I suppose. To purchase actual wire-wrap IC sockets would have increased the cost considerably. Instead it seems he used ordinary IC sockets ("solder-tail" style) and added the long pins for wire-wrap afterward.

He's arranged it so the wire-wrap pins pass through the board, and that's why the wiring is up top with the IC's. Unfortunately, that means the wires are longer and more crowded than they might otherwise be, and AC performance will suffer as a result. OTOH having the pins pass through the board adds mechanical strength. The pins need to be fairly secure or else there'll be trouble when you try to wrap the wires on.

If he'd chosen protoboards that feature plate-through holes then securing the pins wouldn't be a problem, and the wiring could go on the bottom. But again there's a cost concern, since plate-through boards are more expensive. It looks to me as if he has considered his options carefully, and is managing rather well!

-- Jeff

I figured he put the wires on the component side to prevent the problem of trying to solder the next wires when previous wires are in the way. When they're in the way, moving them around several times before the project is done leads to broken wires, and the problem feeds on itself.

I'm not sure I understand your comment. It seems to me he could have the pins protruding on the wiring side. Power & ground would be soldered, then everything else could be wire-wrap.

Oh, you're right. I'm interested, but I have too much material to get through and I'm always in too big of a hurry. I was thinking the (wire-wrap) wires went through the holes (rather than being wrapped) and got soldered to the foils on the non-component side. This would mean that sometimes you have to re-heat solder to get an additional wire in the same hole in this case, but that's not as bad as the other situation where the wires are on the solder side. For having the posts there, thru-plated holes would indeed be much better.

Well spotted Jeff - this idea of component-side posts rings a bell. I wonder if we've talked about this design before... I see now that I did post it over at
http://forum.6502.org/viewtopic.php?f=1&t=2284
back in 2012, but we didn't discuss the construction at that time.