Re: [colorforth] Hello - and where to begin?

["Jeff Fox" <fox@xxxxxxxxxxxxxxxxxxx>]

> Albert van der Horst wrote:
>> On Sat, Jan 19, 2008 at 12:20:41AM -0700, vaded@xxxxxxxxxxxxxx wrote:
>>> I realize the OLPC OX exists, but are we not doing a disservice by not
>>> pushing for a forth processor based equivalent?  Mr. Fox has spoken
>>> about producing a computer that could cost much, much less than
>>> an OLPC OX.

Making a computer that costs less than OLPC does require dealing
with the economy of scale.  Prices don't get low until you reach
a certain volume so a certain amount of funding is required.

I also would never refer to a Forth PC as the 'equivalent' of
something like OLPC because that would imply too much to most
people.  OLPC needs to have lots of backwards compatibility
with regular PCs and a Forth based PC would not.  To most people
PC doesn't mean just 'Personal Computer' any more, it means
certain legacy hardware standards and certain legacy software
from the C.

Chuck has always wanted to make a nice PC with only Forth chips,
only Forth OS, only Forth compiler etc. and IntellaSys will give
him the ability to do this. He has expressed interest in making
a colorforth PC out of SEAforth chips.  But people should
understand that when we say OS we don't mean Windows or Linux,
we mean Forth. And OS in Forth has a different meaning than
when refering to popular PC OS.

>> We do Forth a disservice by not pushing for OLPC right now.

It took a lot of work to create OLPC and it is available now.
Perhaps Forth processors will make Forth PC available at some
time in the future, but OLPC are available now.

> 3. I don't see how you can possibly make a *computer* that costs less
> than an XO using Forth chips.

The iTV web appliances and the UltraTechnology F21 workstation in a
mouse were examples of very low cost personal computers.  If you
want Linux or Windows you need considerable memory and if you want
backward compatibility with legacy PC there are lot of hardware
components needed.  OLPC integrates much of it into a few chips
but it ultimately comes down to a question of millions of
transistors needed for legacy PC compatility that are not needed
if your definition of 'Personal Computer' does not require
this backwards compatibility.

The idea was that the CPU, the memory controller, the keyboard
controller, the video controller, the serial port, the parallel
port, the real time clock, and other circuits in a PC were not
included in the $50-$200 central processor and raised the cost.
And considerable expensive cache memory and more normal memory
is required for large PC software compatibility.  While on the
Forth chips the CPU, the memory controller, the keyboard
conroller, the video card, the network card, the serial card,
the parallel card, and the real-time clock all come on the
same piece of silicon with a $.15 manufacture cost.

The bill of materials for a complete personal computer was
between $10 and $25 dollars.  Of course that was ten years
ago and things have become considerably smaller.  Now we
talk about processors needing much less silicon and fewer
pins because we will offer more of them.  Instead of
hardware memory controllers and keyboard controllers and
video controllers and realtime clocks etc this architecture
just uses Forth hardware and Forth software on multiple
nodes to do those things to keep costs low.

But the nature of PCs with Forth chips and Forth software
is very different than traditional PCs.  If Chuck makes
a PC it will run colorforth.  It won't come with the
30 languages that come on a traditional PC.  This is
the main reason why it could cost so much less.

But with parallel chips it doesn't make sense to compare
one 20k transistor processor to one 400M transitor processor
since cost, size, and power consumption are related to the
number of transistors.  And if you put 20k 20k transistor
cores together you have the same number of transistors as
one Pentium M that could do a couple billion Forth
instructions per second but you might be able to do
tens of thousand of billions of Forth instructions per
second.  But the programming of 20k processors and one
processor would be very different since they would be
small and parallel.  This is a different idea than
having one processor to run a big OS in one space.

And if you wanted a $200 personal computer you would
expect a lot of $.10 processors in it instead of one
like an OLPC. This would make the programming and
the operation of it very different than traditional
PC.

> The economics just aren't there.

That's the main issue.  You can make a handful of custom
Forth chips for tens of thousands of dollars each.  But
in quantity they might cost well under a dollar.  At
iTV there were requests for a hundred million
units as test samples before going into large quantitiy
in the billions.  Had that road been taken the
Windows and Linux PC would be following well behind
in the economics of scale.   And IntellaSys plans
are not so much to put 1000 Forth cores into
new personal computers for children as it is to
put them in every switch, lightbulb, in many places
in every house, office building, card, car,
and electronic gadget that could benefit by
having multiple almost free microscopic low
power high speed processors embedded in them.
Of course last time around most of the interest
came from military sources.

Given a fraction of the OLPC scale beating the OLPC
price would not be a problem.  But people who expect
an traditional PC style PC with lots of compatibility
would be dissapointed.   People who wanted something
that could perform hundreds of thousands of times
more Forth instructions per second might not be
dissapointed.

> First of
> all, the Geode in the XO, while only running at 433 MHz, is a full x86
> chip -- it has all the Pentium instructions plus MMX and 3Dnow!

Sure.  But a similarly clocked Forth chip with a built in video
generator and video accelerator would be a few cents.  For a few
bucks you could have a lot of them.  The Intel compatibility
is good if your goal is compatibility but from the perspective
of a Forth programmer it makes programming in Forth less
attractive than programming Forth chips.  So it is good if
promoting Intel is important while Forth chips are more about
promoting Forth.

> capabilities. It will run Linux -- just about any flavor of Linux as
> long as you have the kernel and drivers right. And it will run any Linux
> application that will fit in 256 MB of RAM.

If the goal is to run C, and Linux is written in C, then using a
processor designed to run C will be a good fit.  And if you need all
that software then you need big memory spaces for the C.

If the goal is to run Forth, then C and Linux are not particularly
important. It is after all the C and Linux that drive the price
up because of all the things that a multi-user mainframe style
operating system and compiler environment require.

> Second, there just isn't application software written for Forth chips.

That's true.  If you can't program and need to use off the shelf
software there is a lot of code in the C world on the shelf. This
is good for promoting C and keeping the C programmers working.
Forth programmers like the idea that it is fun to write software
in Forth.  And we tend to think that writing application is
easier when we don't have to deal with C being the biggest
factor driving the requirements.

> You can't sell a "computer" that doesn't have a word processor, browser,
> email, wireless, ebook reader, etc.

Possibly true, but things have changed a lot since cell phones
are moving to replace most PDA and many traditional personal
computers.  People use smaller, cheaper, more portable and more
integrated appliances today.  Programmers who get paid to work
on PCs still haul around big laptops but many people want smaller
electronics today.

The trend is away from monolithic mainframe style computing towards
more distributed processing with more machines.  OLPC is smaller
than most laptops but has to be a lot bigger than what you will
get in a $5 or $10 personal computer.

The iTV appliances had the browser, the email, and Forth script
that could deliver games or other applications that were not
web based much like cell phones can do those thing without
being as big as a personal computer.  But Forth programmers
can choose to see the need to create new applications in
Forth for new hardware as a good thing if they enjoy
programming and work in Forth.  If they perfer to work in
C there have other opportunities.

The $10 disposable PC for the third world was very different
idea than OLPC.  OLPC has a remarkably low price and remarkable
list of features considering that it is a C machine and it
can probably run Forth as well as a $10 Forth PC if someone
decides to produce those.  Until then OLPC is available and
will be a great Forth platform with great potential with
little competing in its niche.

Best Wishes



---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com