Re: [colorforth] DOES> How is colorForth different from other Forths?

[Mark Slicker <maslicke@xxxxxxxxxxx>]

Second part.

On Sat, 24 Jan 2004, Samuel A. Falvo II wrote:

> > > This suggests that there are important details to discuss.  The
> > > thing with ColorForth, MachineForth, and the whole concept of the
> > > MISC architecture is that they are *so simple* that there is
> > > *nothing* to discuss.  This is a very liberating thing.
> >
> > Does this mean the end of the colorForth mailing list?
>
> I didn't say there was nothing to discuss (I, for one, purely enjoy the
> philosophy-based discussions).  I said there was nothing *important* to
> discuss.  Performance issues are the only things that can even be
> remotely considered "important," and I can demonstrably prove that
> ColorForth is NOT the fastest programming environment out there.  But
> who here would even care?  It's "fast enough," for nearly everything
> anyone who uses it for, and that is all that is important.  So, even
> performance isn't a direly important issue.

I guess I can understand in the present day computing culture why you
think absolute speed is the only remotely important issue.

>
> > If this is the case, then how is there a market for any other simple
> > electrontic component? This same teenager could just as easily create
> > a ram with the same components.
>
> Yes, and I've done that too.  But 64 bits of RAM versus 64MB of RAM
> reveals something important: circuit replication can make building a
> circuit with discrete components tedious and expensive.

How about producing 1 CPU versus producing 1 million CPUs. Besides being
tedious and expensive to build, these TTL CPU will be no where near the
performance of mass produced VLSI desgined chips.

>  Last I checked,
> one needed only one CPU to make a truely useful circuit, and one based
> on MISC concepts will easily be an eye-opener for even the most die-hard
> RISC advocates.
>
> But you took my example the diametrically opposite direction than I had
> intended.  Consider, if a teenager with TTL design experience can design
> and implement an F21 clone with discrete components, who or what is to
> stop, say, Motorola or IBM (whose senior-level engineers often have
> doctorate-level degrees) from embedding these processors in their own
> chips, and not even considering giving Chuck a royalty?

Nothing, but then why would you embed a general purpose CPU of this kind
in a chip if the chip has no observable capability of interpreting
instructions?

>  He has been
> working hard to design these chips, and lo and behold, they turn out to
> be the most easily reverse-engineerable design for processors in the
> world.

They won't have to reverse engineer. They only need go to the patent
office's web site and look up Chuck's patents.

>  Chuck won't get a single cent of recompensation for his efforts,
> except for those companies that are truely altruistic.

Altruistic companies? Which planet do you live on? If Chuck has a decent
lawyer he can get compensation for patent infringement.

> If Chuck *were* rolling in the cash, he'd at least hire someone to
> renovate his website and maintain a list of current ColorForth events,
> news, articles, etc. if he couldn't find the time to do it himself.

Is this meant to prove MISC as not commercially viable?

>
> > If existing software artifacts/ideology had no bearing, there would be
> > no reluctance to use components which improve performance and lower
> > cost.
>
> But would you pay for such components?  Consider: to lower the cost of
> production of a product, you must reduce your expenses: this means
> paying LESS for something.  And you can't get much cheaper than FREE,
> and this is the point of my observation.

Sorry, I must have missed this observation, which company is producing
FREE CPUs?

Mark

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