Path: utzoo!utgpu!attcan!uunet!lll-winken!lll-tis!daitc!jkrueger
From: jkrueger@daitc.ARPA (Jonathan Krueger)
Newsgroups: comp.arch
Subject: Re: Blit (was Re: Self-modifying code)
Message-ID: <150@daitc.ARPA>
Date: 2 Aug 88 19:56:50 GMT
References: <1152@ficc.UUCP> <3089@geac.UUCP>
Reply-To: jkrueger@daitc.UUCP (Jonathan Krueger)
Organization: Defense Applied Information Technology Center, Alexandria VA
Lines: 25

In article <3089@geac.UUCP> daveb@geac.UUCP (David Collier-Brown) writes:
>  If you will have a look at Foley & Van Dam[1], you will notice a
>cycle of main cpu software -> added hardware assist -> auxiliary
>special-purpose cpu -> auxiliary general-purpose cpu -> upgrading of
>the auxiliary cpu to a "main" cpu and then around the circle again.

Also noted in Bell, Mudge, & McNamara, "Computer Engineering: A DEC
view of hardware systems design", Digital Press: 1978, page 201:

	"An observation that display and other specialized processors
	evolve in a fashion called the `wheel of reincarnation'"

Which observation they attribute to Myer and Sutherland, "On the
Design of Display Processors",  Communications of the ACM,
11(6):410-414, June 1968.

How to divide up the work between general (central) and special
purpose hardware?  Depends on relative costs of computation versus
communication.  68020's made computation cheaper.  Blitters made
communications cheaper.  And all tradeoffs depend on where you are on
the wheel of reincarnation.  Ready for another spin, anyone?

-- 
Jonathan Krueger  uunet!daitc!jkrueger  jkrueger@daitc.arpa  (703) 998-4777

Inspected by: No. 15