home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!sun-barr!news2me.EBay.Sun.COM!exodus.Eng.Sun.COM!appserv.Eng.Sun.COM!appserv!limes
- From: limes@straylight.eng.sun.com (Greg Limes)
- Newsgroups: comp.arch
- Subject: Re: A theory for Big & Little Endian's origin
- Date: 8 Jan 93 17:29:06
- Organization: Sun Microsystems Computer Corporation (Mountain View, California,
- USA)
- Lines: 20
- Message-ID: <LIMES.93Jan8172906@straylight.eng.sun.com>
- References: <1iig7aINNtc@spim.mti.sgi.com> <1993Jan8.120225.8330@infodev.cam.ac.uk>
- <willmore.726511292@help.cc.iastate.edu> <16294@auspex-gw.auspex.com>
- NNTP-Posting-Host: straylight
- In-reply-to: wally@Auspex.COM's message of 8 Jan 93 20:02:07 GMT
-
- In article <16294@auspex-gw.auspex.com> wally@Auspex.COM (Wally Bass) writes:
- | Actually, I think that it's not the 'load' problem so much as it is
- | the 'add' problem. When you are going to do an add on a machine with
- | 1 byte adder, you want the low order bytes of both operands first,
- | and the high order bytes last, because that is the direction in
- | which the carry propagates.
-
- One can make an equivalent argument for the 'compare' problem. When
- you are going to do a compare on a machine with a 1 byte comparator,
- you want the high order bytes of both operands first, and the low
- order bytes last, because you can stop working sooner. One can even
- point out when sorting records that storing your numbers in big-endian
- format means you only need to know the offset and length of the key
- fields, not whether they are alpha or numeric.
-
- I believe that whether bytes were big-endian or little-endian was
- largely an arbitrary decision when the oldest ancestor of the ISA was
- being designed, based on reasons that made sense at the time but which
- may now be completely lost in the mist. All we can do now is deal with
- the results of the fact that about half the ISAs are "wrong" :-)
-