|
Message
From: "Jim Dempsey" <tapedisk@a...>
Date: Thu, 13 Feb 2003 11:18:13 -0600
Subject: Re: [oc] Beyond Transmeta...
----- Original Message -----
From: "Marko Mlinar" <markom@o...>
To: <cores@o...>
Sent: Thursday, February 13, 2003 1:32 AM
Subject: Re: [oc] Beyond Transmeta...
>
> > > Ok, I would agree, that function calls can be cutted out.
> > > But loops? I don't think so. Loops are actually a way of dinamically
> > > duplicating pieces of code/logic. You need some sort of "loops".
> >
> > You are falling into a common trap of assuming you code the same way
> > for both computing environments. (iow you are attempting to use the
> > same paradigm). This is an entirely different way of processing.
> >
> > Using this visualization tool:
> >
> > Consider the traditional CPU design as a six armed wonder worker
> > scurrying about a manufacturing plant at blazing speed to build a
> > complicated
> > product.
> >
> > Consider the bit streaming method as having 10,000 one armed workers
> > doing trivial tasks.
> Ok, I see where are you aiming -- you want to create biological system,
not
> brain like, but gene like (reproducing part of chains, etc).
> I thougth you wanted to create classical array of presented multi data
multi
> phase processors.
> Still I don't understand how would you manage huge amounts of
communciation.
> For example in cells you have shared liquid where molecules are exchanged;
> but everything happens so slowly.
> So, how did you wanted to solve communication -- it presents >90%
brainwork of
> your design.
>
I view this differently. I am confident that an elegant and simple solution
exist
and if given some assistance with the right people (team) that we could
produce
such a solution in a reasonable amount of time. This solution could be
emulated
in software on a desktop PC, then emulated using standard PLDs. Once
proof of concept is established then the new device would be built.
> > I would think that Alen Turing would find this a logical extension of
his
> > Turing
> > Machine concept. Instead of an infinite long tape that the "simple"
machine
> > traverses you dynamically snip and reconstitute the tape(s) and route
> > it(them)
> > through 1,000's - 1,000,000's of "simple" machines.
> I don't know if you noted but "dynamically snip" can be very hard problem,
I
> would say as hard as solving the problem itself.
> Maybe this new machine is not computationaly equivalent anymore if you do
not
> supply all the tape to each machine. Besides Turing machine requires
endless
> tape, which your machines do not have if you make fixed distributions.
>
The Turing/tape issue was a visualization tool. Turing only stated you could
solve
any problem given enough tape. For any paticular problem you could solve it
with a finite length of tape.
> I would rather think of your machine as an 'expert system' -- 1000
machines
> sharing the same tape and replacing (=calculating) parts they are able to.
>
Nothing expert about it. Just a plumbing problem.
Jim Dmepsey
|
 |
Browse
