home .. forth .. machineforth mail list archive ..

[MachineForth] 25x Forth engine


Hi, all.

I hope this list is still alive!

I've just recently become aware of the x18 and 25x chips.  Having been deep
in the UNIX OS world followed by networking infrastructure, I was both
surprised and delighted to see Charles Moore still standing tall on the
principles of simplicity.  And I'm taking a break from the paycheck spin
cycle, so I have time to think about such things.

The 25x chip really sparks my imagination.  It's been clear for a while that
a starkly simple processor could be laid out in an array within a single
silicon die, and that this would create a radically new kind of CPU
horsepower.  I wouldn't at all mind pitching in funding to get something
like this off the ground.  But it seems like some intermediate steps would
be appropriate:

1. Spec out the CPU
	a. Instruction architecture
	b. Instruction set
	c. Bus(es) architecture
	d. Memory organization and hierarchy
	e. Inter-chip signalling
	f. Pin-out

2. Code up a simulation
	a. Accuracy versus planned CPU
	b. Tools (cross compiler, etc.)
	c. OS
	d. Word set for operating the CPU "farm"
	e. Iterate back to the CPU design as indicated by experience
	f. Finish with an instruction set verifier

3. Spec out an eval board for the CPU
	a. 25x interface
	b. Supporting peripherals
	c. Bootstrap
	d. PCB layout

Note that all of this (except, arguably OS and its supporting "farm" control
functions) is orthogonal to the target market.  But with something really
new like this, I don't believe you *can* target a market (read "The
Innovator's Dilemma" for more on this).  My own inclination is to start with
code breaking applications; my intuition is that the per-CPU memory size is
too thin to do a good job here, but that's what the simulation feedback to
CPU design is all about.  Once you tune that in simulation, my hope is that
the 25x's performance in silicon would get a lot of attention, which is how
you bootstrap from initial successes into wider adoption.

Sorry for carrying on in my initial message.  I hope others are thinking
along similar lines.

Regards,
Andy Valencia
vandys@xxxxxxxxx
------------------------

To Unsubscribe from this list, send mail to Mdaemon@xxxxxxxxxxxxxxxxxx with:
unsubscribe MachineForth
as the first and only line within the message body
Problems   -   List-Admin@xxxxxxxxxxxxxxxxxx
Main Machine Forth site   -   http://www.