Path: utzoo!attcan!uunet!lll-winken!lll-lcc!ames!ll-xn!mit-eddie!killer!mic!d25001
From: d25001@mic.UUCP (Carrington Dixon)
Newsgroups: comp.misc
Subject: Re: "big endian" and "little endian" - first usage for computers
Message-ID: <203@mic.UUCP>
Date: 29 Dec 88 04:27:25 GMT
References: <2766@cbnews.ATT.COM> <10147@well.UUCP>
Reply-To: d25001@mic.UUCP (Carrington Dixon)
Organization: Micro Net
Lines: 26

In article <10147@well.UUCP> ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes:

>	Sometime back, I vaguely recall a rather heated discussion on which
>arrangement was better: big-endian or little-endian.  What was the outcome
>of this discussion?  Did anyone "win"?

      I think that the concensus was, "Be reasonable, do it MY way."  :-)


>	Least-significant-byte first has the advantage of being able to
>specify the low order byte or word of a longword in memory by specifying the
>same address for all three.  I.e. if you had the value 0xAABBCCDD stored at
>location <foo>, then you'd specify <foo> to get at the low order byte
>(0xDD), word (0xCCDD), and full longword.  No funny pointer arithmetic is
>needed.
>
     Uh,  why would you _want_ to do such a thing?  

     And conversely, the most-significant-byte-first has the advantage of
being able to specify the _high_ order byte or word of a longword in memory
be specifying the same address of all three.  One "advantage" seems about
as "significant" as the other ...


Carrington Dixon
UUCP: { convex, killer }!mic!d25001