Re: LZ: Re: A few questions, a few comments, blahb

Subject: Re: LZ: Re: A few questions, a few comments, blahb
From: Ben Shakal <shakal@ns.net>
Date: Tue, 9 Oct 1996 22:11:20 +0000
In-Reply-To: <LZ: Re: A few questions, a few comments, blahb>

> Hello?  I put out a proposition, so read that.
> 
> "Canada has no national identity other than our strongest policy:
> multiculturalism"
> -= Zenon@bbs.nexes.com =-


I meant the specifics.  From what you were saying, it sounded like 
you wanted to load each instruction straight from the expander.  This 
is impossible because:


  (1) It is too slow.  Instructions are executed much faster (I 
think a few hundred instructions per second, but correct me if I'm 
wrong.) than the link port can handle (a few thousand bytes per 
second).


  (2) It is not possible to snag the attention of the processor after 
each instruction, except to add a wire inside.  Even if this was 
done, the processor would be forced to run at the speed of the link 
port (see above).


  (3) Some other reasons I can't think of. :)


I think the general concensus (sp?) -- including Mel -- has been that 
we would load segments of the program into memory, execute those, and 
load new ones when necessary.  As a block transfer, it would be _much_ 
more efficient.  That might have been what you proposed, and if so, 
sorry, but I didn't understand your proposition. :)


Ben
shakal@ns.net