Nand gate computer.

Tony Duell ard at p850ug1.demon.co.uk
Sun Nov 29 13:00:04 CST 2009


> 
> Chuck Guzis wrote:
> 
> > Schmitt triggers aren't the only way (or even the best way) to 
> > debounce.  Why should that drive your entire parts choice?
> 
> The joke here, is I only use ONE part.

I think if I had to build a copmputer using only one type of chip, I'd 
pick an FPGA :-). OK, that's cheating... 

The VAX 11/730 was _largely_ built with PALs. There are some 2901s in 
there as well, and the 8085 stuff for the cosole processor and the 
microcdoe RAMs, of course, but almost all the logic is in 20 pin PALs, I 
always thought that was a nice hack.

If I had use just one type of chip from the TTL range, it'd be a toss-up 
betwee nthe 74150 and 74151. Both are multiplexers, the former is a 16 
input mux with 4 select inputs, the latter is an 8 input mux with 3 
select inputs. Using them you can make any 4 or 3 input logic gate 
(combinatorial function) and by adding a feedback connection you can make 
SR and D-type latches. Making D-type flip-flops (pulse rather than level 
triggered) is a little harder, but it can be done.

Although the '150 would let me make a more complicated function, it's a 24 
pin package (0.6" wide, at least all the ones I've seen) as against the 
16 pin (0.3" wide) pacakge for the '151. And I think the 151 exists in 
logic fanilies (e.g. 74F151) where the 150 doesn't.

Of coruse if you let me have inverters as well, I can make any 4 input 
function using the '151 and a single inverter (NOT gate). But then again 
a '150  and an inverter would let me make any 5 input fuction. 

Perhaps I should really cheat and get a custom 28 pin chip made 
containing the functionality of a '150 and a couple of NOT gates. Or a 
'150 and a 2 (or 3) input NAND gate.

-tony



More information about the cctalk mailing list