Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!purdue!mentor.cc.purdue.edu!l.cc.purdue.edu!cik
From: cik@l.cc.purdue.edu (Herman Rubin)
Newsgroups: comp.arch
Subject: Re: Byte ordering
Summary: What about floating point or fixed point "fractions"?
Keywords: byte sex
Message-ID: <1910@l.cc.purdue.edu>
Date: 6 Feb 90 12:38:20 GMT
References: <9656@spool.cs.wisc.edu> <1990Feb2.215421.24894@utzoo.uucp> <1991@osc.COM>
Organization: Purdue University Statistics Department
Lines: 29

In article <1991@osc.COM>, jgk@osc.COM (Joe Keane) writes:
> In article <1990Feb2.215421.24894@utzoo.uucp> henry@utzoo.uucp (Henry Spencer)
> writes:

			...........................

> >-- on a little-endian machine, they're not the same), which simplifies the
> >code for optimized raster operations.
> 
> Sorry, but little-endian is just as consistent.  The code is exactly the same,
> you just have to switch << and >> in the right places.  On say a MicroVAX, the
> leftmost bit on the screen is the least significant in its byte, word, page,
> or whatever.

It depends on the purpose.  If the part you want to keep after truncating
is the most significant part, such as is the case with floating point numbers
and fixed point fractions, big-endian is far easier to work with.  If you
want the low-order parts, little endian has the advantage.

On the VAX, in order to provide for this problem, floating point numbers
are little endian within words, but the words (16 bit) are arranged in a
big endian manner.  Packing and unpacking are quite difficult.

The raster problem corresponds to fixed point fractions; when truncating,
the most significant part is used.  This is easier in big endian.
-- 
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907
Phone: (317)494-6054
hrubin@l.cc.purdue.edu (Internet, bitnet, UUCP)