How hard is to build a processor?

Why not ?
Someone should be keeping the old craft alive. Good on him !
You could ask 'why build museums' ?
I think you agree.

wrote:


Thanks, Don.  I absolutely LOVE doing this kind of thing.  It
should be required work for anyone in embedded programming,
at least.  35 years ago I tried my hand at it and failed to
get much useful.  It wasn't until I later got access to a
logic simulator that I was finally able to complete a
successful processor design with 7400 logic, not using the
74181 ALU by the way, though I never built what seemed to
work so well.  Later, I picked up a Xilinx 4000 series demo
board and, for the very first time, learned to write VHDL and
finally coded up my first processor that worked!!!  (I didn't
do the floorplanning by hand -- way over my head at the
time.)

I learned so much doing this, though.

Anyway, thanks for the link!!

Jon

That reminds me that there are some excellent course videos by David
Culler of Berkeley showing how to build a CPU from logic gates. A bit
of searching for his webcasts for course 61CL Fall 2009 in H.264 video
should bring them up.

I had no idea until working through the lectures that *I* could build
a CPU using components I already knew about. As part of the course he
shows how to do so by working up from logic gates.

James

  wibbled on Friday 26 February 2010 10:52


I knew a chap (York University Computing Services) who claimed to have a
mate who built an elementary CPU from fruit machine relays[1]. Occupied a
bit of board about one square yard. He lost interest in building RAM from
more relays, so wedged a 1k RAM chip with suitable interfacing on the side.
Ran at about 1 IPS apparently...

[1] he got a box load from a surplus store, old stripped out ones. Had an
unusual contact configuration that made them quite suitable.

--
Tim Watts

Managers, politicians and environmentalists: Nature's carbon buffer.

Reminds me of the pdp11 look alike I worked on back in the early 70's... few
hundred ttl chips , 64 bit diode matrix microcode, 8K ferrite sotore driven
by loys of monostables to get the timing right....
I was used to poser some of the very first CNC machines in the UK- Herbert
BatchMAtic and in Germany - Boehringer Vxxx series lathes.
Those were the happy days......

Back in '74 Elektor published a design of a computer built with TTL-logic.
They really made it work though I doubt any subscriber built it too.

petrus bitbyter

Back in '73 I worked on a 16 bit mini computer built entirely out of
TTL, complete with a row of toggle switches on the front panel so you
could enter machine code by hand. It had a paper tape reader and along
with a paper tape punch you could use the two pass assembler - that's a
lot of paper.

Cheers

ian

@Ian, who did you work for in 73 ???

Cool!  Over on alt.folklore.computers there was a recent thread about
projects implementing old CPUs with an FPGA.  Very neat stuff out there
for the retrocomputing world.
--
As we enjoy great advantages from the inventions of others, we should
be glad of an opportunity to serve others by any invention of ours;
and this we should do freely and generously. (Benjamin Franklin)

Don McKenzie wrote:


It is a lost art. In the 70's I taught a course that students built
a small computer out of lab modules of TTL chip's. My first
personal computer was micro coded PDP-8 hand built.

Ram was 1K (bits) parts on a wirewrap board.

Walter..

On Fri, 26 Feb 2010 10:10:55 -0500, Walter Banks

Oh, my, Walter.  That would have been a wonderful class to
have taken.  I only wish...

Jon

Walter Banks wrote:

In 1961 a colleague told me of a machine in a lab at Cornell named
CADET, which reputedly stood for "can't add, doesn't even try".  But it
was a universal (Turing) machine.  Most everyone has stood on the
shoulders of software to extend behavior.  In another sense, cpu
development has also stood on the shoulders of software arts, for needs
drive real engineering, not possibilities, and software disciplines
provide the languages for expressing these needs.

Each to his own.  Some yearn to create a theory, some to create a
bridge, some to create an impression.  I still ponder the structure of a =

(programmable) computer of light and heavy marbles, gates, and a bunch
of elevators.

-- Bill Cooke

IBM 1620.  See http://en.wikipedia.org/wiki/IBM_1620


Another fun project!
--
As we enjoy great advantages from the inventions of others, we should
be glad of an opportunity to serve others by any invention of ours;
and this we should do freely and generously. (Benjamin Franklin)

Joe Pfeiffer wrote:

 processors

Hah!  My leg's been pulled.  and to think I'd just finished a year on
1401, 705 code!  I'd thought 'cadet' was a lab project, not a for-real
machine.  I've even read a 1620 manual, but never got to write for one.

-- Bill

On Fri, 26 Feb 2010 14:18:14 -0700, Bill Cooke


I can remember hearing that phrase from time to time when I
worked on the 1620.  It was a fun machine.  I used to swap
out the colored bezels on the control panel just to tease.

Jon

Jon Kirwan wrote:

I was working at a small shop that had a hand-me-down 1620.  They had a simple
Fortran compiler, "PDQ Fortran", but it lacked some features they wanted.  On my
own, I wrote a disassembler for the 1620, disassembled the compiler, studied the
code, figured out how to save some code space (memory was limited), then add
some features, basically enhanced write commands and formatting, all with no
external documentation.  When the code was almost ready I was so excited I
couldn't sleep, so went into work at 4 AM or such and got it working.  Fun days!

--
Thad

Bill Cooke wrote:


The IBM 1620 was the first computer I used. It had
some interesting features, a BCD machine with variable length
data words.

I ran a fibonacci series on it to 2000 terms it took 18 hours.

A lot of neat things could be done with the math tables which
could be set at run time. The math tables were also the target
some elaborate pranks in IBM 1620 labs

It was also the first computer that I wrote a compiler for. I
have a lot of good memories of the 1620.

That was a long time ago.

Regards,

Walter..
--
Walter Banks
Byte Craft Limited
http://www.bytecraft.com

Don McKenzie wrote:


It is trivial to build a processor. This type of project is a semester
work for a student. It is hard to make commercially viable processor,
though.

VLV

Don McKenzie wrote:


It is trivial to build a processor. This type of project is a semester
work for a student. It is hard to make commercially viable processor,
though.

VLV
But you have to add to that all the software to make it do something....
starting with a boot loader...