Sam Ismail <dastar(a)ncal.verio.com> wrote:
I'm getting this opinion straight from Ted Hoff himself. I don't know
where my notes are on this and I can't remember the details, but I think
there are some significant differences between the 4040 and 4004. The
8008 was supposed to have been based on the 4004 architecture, extending
it with more instructions and such. The 4040 was, I think, a whole
different design by a different team inside Intel.
Hi Sam
From the outside, the 4040 was quite similar to the 4004.
The 4040 was in a different 24 pin package but used the same multiplexed
4 bit bus and could use the 4001/4002 parts. There may have been
many differences internally. Code wise, the 4040 will run
all of the 4004's machine instructions and there were about
10 or so instructions added. There was a slight difference
about what was on the bus during the execution cycles. As
I recall the 4040 had the ALU available on the bus during
the last cycle. I have a data manual describing all of the
4004 and 4040 instructions. As was mentioned, the 8008
was quite a bit different.
When I worked at Intel in the late 70's, I worked on
the UPP ( Universal Prom Programmer ). I was involved in
system test. There were few tools made to deal with
debugging these early uP's. I built what I called a
"Slow Time Ice". One could remove the 4040 and plug this
unit in. I ran a continous clock on the 4040's buss and
would take input from a MDS800 or Series II. It allowed
one to read/write any of the ROM, RAM or I/O on the UPP.
Only one of these was ever built.
As for the questions of where I got the 4004, I found it
at a local electronic surplus store. I bought the entire
setup for $25, as I recall, about 10 years ago. I
knew it was a valuable item to save.
For those that don't have a 4004 to play with yet, they
can use my Assembler/Simulator to play with the instruction set.
It isn't quite real hardware but almost runs as fast.
Dwight