Date: Thu, 31 Oct 2002 14:37:38 -0800 From: Steve Kargl <sgk@troutmask.apl.washington.edu> To: "M. Warner Losh" <imp@bsdimp.com> Cc: tlambert2@mindspring.com, bde@zeta.org.au, rittle@labs.mot.com, rittle@latour.rsch.comm.mot.com, current@FreeBSD.ORG, dschultz@uclink.Berkeley.EDU Subject: Re: Lack of real long double support Message-ID: <20021031223738.GA17221@troutmask.apl.washington.edu> In-Reply-To: <20021031.151847.03097281.imp@bsdimp.com> References: <3DC0E0A7.290A57CA@mindspring.com> <20021031.013338.106483974.imp@bsdimp.com> <3DC17FC5.AF56552E@mindspring.com> <20021031.151847.03097281.imp@bsdimp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Oct 31, 2002 at 03:18:47PM -0700, M. Warner Losh wrote: > In message: <3DC17FC5.AF56552E@mindspring.com> > Terry Lambert <tlambert2@mindspring.com> writes: > : > Nope. The only difference between 53 bits and 64 bits of precision is > : > just that: precision. The number of bits for expoentent is > : > independent of this. > : > : .125 ^ 2 = 0.015625 > : .25 ^ 3 = 0.015625 > : > : So if I go from 3 digits of precision to 2 digits of precision for > : my mantissa, in order to represent the same number, I need a larger > : exponent. > > That's not how it works. The exponent is more like > > .1250000 * 2^3 > vs > .1249999 * 2^3 > > Both have exponent 3, but the differ by a bit or two in the mantissa. > Loren already posted a pointer to "What Every Scientist Should Know About Floating-Point Arithmetic" by David Goldberg. But, for Terry edification http://cch.loria.fr/documentation/IEEE754/ACM/goldberg.pdf This is only 1 of 66100 hits from a google search with keywords "floating point scientist". -- Steve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021031223738.GA17221>