Am Mittwoch, den 19.08.2009, 07:59 +0000 schrieb freebsd-questions-requ...@freebsd.org: > On Tue, Aug 18, 2009 at 01:45:27PM -0400, Karl Vogel wrote: > > > >> On Mon, 17 Aug 2009 17:23:29 -0700, > > >> Walt Pawley <w...@wump.org> said: > > > > W> As speculation on my part, perhaps the six character limitation > is less > > W> a software issue than an early architecture issue - DEC's > PDP-6/10 > > W> design used 36-bit words and packed six characters (clearly from > a > > W> limited subset of the then current ASCII) per word, making simple > > W> searches very effective through symbol tables with a simple word > level > > W> compare loop. > > > > I'll second that. My first job for Uncle Sugar was on a DEC > 10/55 > > for the Air Force, and 36-bit words were a fact of life. There > were > > lots of programs around for conversion to/from 32-bit words, just > so > > we could talk to everybody else on Earth. > > CDC (Control Data) mainframe machines used 6 bit characters. > I believe the 3600 series had 36 bit words. > The 6000 series (6400, 6500, etc, plus 170/750) used 60 bit words > but still used 6 bit characters. So, everything was all upper case. > It had 12 bit 'peripheral processors' which tended the 60 bit main > processor[s] so later started to use 12 bit characters or sometimes 8 > in 12 to allow for upper/lower case. That was a Seymour Cray thing. > He designed their early mainframes before he bolted to make his > own companies (so he wouldn't have to conform to corporate control).
And I always thought it was 14 bit with 7 bit characters, perhaps this is why my outputs looked strange :) This was the last model I've used: http://www.cray-cyber.org/systems/cy960.php > Later CDC came out with their 180 series that used 64 bit words > and 8 bit bytes. It was kind of a nice system but it was too late for > them. The world was turning to clusters of cheap CPU chips running > UNIX > instead of massive mainframes running proprietary OSen and CDC didn't > jump on that bandwagon soon or strongly or cheaply enough. > > Anyway, in those earliest of days, 6 bits was the economical character > set. But it was an obstacle to upper/lower case characters without > using some shift code. IBM and DEC started doing 8 bit bytes - I > don't > know just when - and that allowed eash use of upper/lower characters > and > so quickly determined the standard character size for a long time. Didn't need lower case at this time. REAL PROGRAMMERS USED FORTRAN http://www.pbm.com/~lindahl/real.programmers.html The problem was, the programmers packed the string into integer arrays. 2 characters in 1 integer saved a lot of space, but the VAX didn't like this style. > Now > that 8 bit byte is a thorn in the side of those who want to create > and > universalize a character set that is international. > > ////jerry > Wasn't it just 3 or 4 releases ago FreeBSD went 8 bit clean ? _______________________________________________ firstname.lastname@example.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"