Re: Re: Re(2): [colorforth] DOES> How is colorForth different from other

[<mark.bottomley@xxxxxxxxxx>]

> 
> From: <wtanksleyjr@xxxxxxx>
> Date: 2004/01/26 Mon PM 08:49:37 GMT
> To: colorforth@xxxxxxxxxxxxxxxxxx
> Subject: Re: Re(2): [colorforth] DOES> How is colorForth different from other
> 
> From: "Arthur W. Green" <goshawk@xxxxxxxxx>
> >>I posit, absolutely, that MISC can *never* catch 
> >>on commercially (sorry Chuck), because the concept is 
> >>so simple that a teenager with a year's experience 
> >>hacking TTL circuitry can replicate the CPU, in its 
> >>entirety, in discrete component TTL logic, based 
> >>*entirely* from online material accessible today.  
> >>E.g., the CPU si so simple, it's virtually public 
> >>domain.
> 
> >So, I submit, that *especially* because MISC is so 
> >simple that it *will* catch on commercially. 
> 
> I think you and Sam are actually in agreement -- but Sam's
> pointing out something different, namely that all the 
> pioneering work that Chuck's done could be wasted, since
> anyone can easily reproduce it (once they know that it
> could be done).
> 
> Why pay Chuck for MISC when you can get most of its 
> benefits just by hiring a mediocre hardware engineer to 
> read his webpage?
> 
> >>Likewise with ColorForth, and I believe this is why 
> >>Chuck released ColorForth to the public domain.  
> >>The software is just too simple.  
> 
> >To say it is "too simple" is trivializing Chuck's 
> >work, making it sound positively mundane.
> 
> Only if you think that's an insult. Sam's not using it
> that way -- it's a description, absent of value judgement.
> 
> It IS simple, and that makes Sam's conclusion necessary:
> you can't protect it once people know it's possible.
> 
> > -- Art
> 
> -Billy
> 
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
> For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
> Main web page - http://www.colorforth.com
> 
>

From what I can gather about Chuck's design, the "simplicity" is not easily reproducable because the design pushes the state of whatever manufacturing process is used and so close to the limits that one cannot use commercial tools at the VHDL/Verilog level to even come close to matching the speed and the low power operation. This is why Chuck's OKAD II and the simulation capability work at the polygon description level. He is aware of the thermal characteristics of each transistor and the modules are hand layed out to pack in a simple yet compact floorplan. The difference is obvious when you examine the F21 which executed non-memory-access instructions at 2ns in 0.8u design rules long before Intel was into the 500 MHz arena. The specs for the more recent X18 core in the X25 design executed at ~2400 MIPS. Even the several people implementing their own FPGA versions are clocking 25-100 MHz today at much more power. The other major challenge is that the design is asynchronous logic, which is a black art and generally not in the skill set of today's hardware engineers. This unique combination amazes me.

I admit that Chuck has exposed many ideas that people could try to use, but those who have tried to follow Chuck realize how far ahead he is.

Mark...


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