Path: utzoo!attcan!uunet!lll-winken!lll-ncis!helios.ee.lbl.gov!pasteur!ames!mailrus!uflorida!novavax!twwells!bill From: bill@twwells.uucp (T. William Wells) Newsgroups: comp.misc Subject: Re: "big endian" and "little endian" - first usage for computers Message-ID: <295@twwells.uucp> Date: 5 Jan 89 18:42:05 GMT References: <2766@cbnews.ATT.COM> <10147@well.UUCP> Reply-To: bill@twwells.UUCP (T. William Wells) Organization: None, Ft. Lauderdale Lines: 21 Summary: Expires: Sender: Followup-To: Distribution: Keywords: In article patterso@hardees.rutgers.edu (Ross Patterson) writes: : On the other hand, MSBF prevents the ridiculousness of performing multi-byte : arithmetic piecemeal, they way Intel 80[x]8x machines have to. It's : unnecessarily dificult to manage numbers when the bits aren't in a neat : sequence. On the other hand, ignorance lets people make statements like that. One of the *reasons* for LSBF was that it was cheaper to do multibyte arithmetic piecemeal. When you are making a microprocessor with an eight bit bus or word size but want to be able to do sixteen bit arithmetic, it is *lots* cheaper, in terms of code or chip real estate, to have LSBF. This was hardly news to chip designers; there are many chips out there which use LSBF. And what other reason might you imagine for the longs stored as B1,B0,B3,B2 (on a PDP-11, I think) where the convenient chunk was 16 bits but a cheap way of doing 32 bit arithmetic was wanted? --- Bill { uunet!proxftl | novavax } !twwells!bill