Path: utzoo!attcan!uunet!ibmpa!lmb
From: lmb@ibmpa.UUCP (Larry Breed)
Newsgroups: comp.arch
Subject: Why 32 bits?
Summary: Blame it on checkers
Message-ID: <1710@ibmpa.UUCP>
Date: 12 Aug 89 00:41:15 GMT
References: <355@torsqnt.UUCP> <3490016@wdl1.UUCP>
Reply-To: lmb@ibmpa.UUCP (Larry Breed)
Organization: IBM AWD, Palo Alto
Lines: 20

In article <3490016@wdl1.UUCP> bobw@wdl1.UUCP (Robert Lee Wilson Jr) writes:
>And note that the tube type machines from which the 709x evolved were also
>36 bitters, and in turn that 36 is a nice multiple of 12 which is the
>number of rows in an "IBM" card. Since old machines like the 704/701 read
>cards as binary images into core, and translated bit patterns into
>(EBCDIC) characters via internal software rather than something in an I/O
>channel, this was not a coincidence!
>
Apocryphally, and perhaps even in truth, Herbert Simon and his checkers-
playing program influenced the choice of 36 bits.  When the 704 was
in the design stage (they didn't have "architecture" back then) there was
debate over choosing a 32- or a 36-bit word.  Simon represented a board
position by a boolean mask with one bit per red square.  (No need to
represent black squares since you can't move to them.)  That sounds like
32 bits, except that Simon's representation included an extra column off the
right edge of the board -- avoided special-casing his move generator at
the boundary.  That's 4 more red squares.  Simon urged a 36-bit word,
and that's what he got.

	Larry Breed