Article View: comp.arch.fpga
Article #4030Re: Big-Endian vs Little-Endian
From: tronsmith@my-dej
Date: Tue, 17 Nov 1998 00:00
Date: Tue, 17 Nov 1998 00:00
57 lines
2274 bytes
2274 bytes
In article <36507a30.14282068@news.alterdial.uu.net>, russmay@ditmco.com (Russell May) wrote: >> I do agree that byte memory dumps for structures larger than > bytes (especially for bit fields which span between bytes) > are generally less intelligible with Little-Endian types, but > that appears to be the only disadvantage of Little-Endian types. And they are only less intelligible because they are generally laid out backwards, i.e., the programmers who write monitors and such don't bother to order the dump: F E D C B A 9 8 7 6 5 4 3 2 1 0 instead of: 0 1 2 3 4 5 6 7 8 9 A B C D E F I got a chuckle out of an IBM document (part of some RS/6000 tech refs) trying to justify big-endianess, on the basis of crossover muxes being need to store and retrieve different size data types coherently from memory, when the muxes wouldn't be needed if their register naming/ordering conventions were reversed instead. > > yves@px.uk.com (Yves Tchapda) wrote: > >> > >> Hi Shail, > >> Technically speaking, there is no advantage of Big-Endianness over > >> Little-Endianness. Intel processors are predominantly Little-Endian, > >> while Motorola processors are Big-Endian. Choosing one over the other > >> depends primarily on the overall system. > > > >I'm not totally convinced. Why do Intel designers choose an Endianness that > >is, well, counter-intutive? Take a look at an x86 memory dump to know what I > >mean... See above. Personally I find little-endianness to be more 'intuitive'. It corresponds to the way I add, subtract, and live: start at the small end and work your way up. Are you sure there's no difference in the gate counts required to > >make a system Small or Big Endian? Maybe one of them requires much fewer > >gates to implement... Can you give an example? A down counter requires the same number of gates as an up counter, an adder has same gate count as a subtractor ... > > > >--Shail Bains > >Silicon Automation Systems Ltd. > > > >-----------== Posted via Deja News, The Discussion Network ==---------- > >http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own > - John -----------== Posted via Deja News, The Discussion Network ==---------- http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Message-ID:
<72slah$bq0$1@nnrp1.dejanews.com>
Path:
rocksolid-us.pugleaf.net!archive.newsdeef.eu!archive!mbox2nntp-comp.arch.fpga.mbox.zip!not-for-mail
References:
<364ff418.2246316251@firewall> <72pmo8$o2a$1@nnrp1.dejanews.com> <365072a3.12349032@news.alterdial.uu.net> <36507a30.14282068@news.alterdial.uu.net>