Deutsch English Français Italiano |
<vq5go3$1hhe1$1@dont-email.me> View for Bookmarking (what is this?) Look up another Usenet article |
Path: news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail From: Pancho <Pancho.Jones@protonmail.com> Newsgroups: alt.folklore.computers,comp.os.linux.misc Subject: Re: evolution of bytes, The joy of FORTRAN Date: Tue, 4 Mar 2025 00:16:03 +0000 Organization: A noiseless patient Spider Lines: 18 Message-ID: <vq5go3$1hhe1$1@dont-email.me> References: <vpl91g$25q46$1@dont-email.me> <vpsu8r$ljl$1@gal.iecc.com> <vpt7uv$3r2n0$3@dont-email.me> <175819294.762482901.217276.peter_flass-yahoo.com@news.eternal-september.org> <vptp2b$1huf$1@gal.iecc.com> <vq1rmc$to24$1@paganini.bofh.team> <vq2j6n$v1q6$3@dont-email.me> <vq31t7$14njc$1@paganini.bofh.team> <1675186097.762702067.383557.peter_flass-yahoo.com@news.eternal-september.org> <vq5enq$1h3mg$11@dont-email.me> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Injection-Date: Tue, 04 Mar 2025 01:16:05 +0100 (CET) Injection-Info: dont-email.me; posting-host="a6b95ce1615e4c1646ae076e39d887cc"; logging-data="1623489"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+tSgXD+iWK2LUn3kx5U380opA1EW61qqU=" User-Agent: Mozilla Thunderbird Cancel-Lock: sha1:iBNALycvcxXUT59KO/ZNe4bkRx8= In-Reply-To: <vq5enq$1h3mg$11@dont-email.me> Content-Language: en-GB On 3/3/25 23:41, Lawrence D'Oliveiro wrote: > On Mon, 3 Mar 2025 06:54:31 -0700, Peter Flass wrote: > >> Some fractions that are exact in decimal are only approximate in binary. > > Base-ten has two prime divisors: 2 and 5. Base-two has only 2. So any > fraction that has a denominator that is the product of any integer powers > of those divisors can be represented exactly, while others cannot. > > The need to represent 1/3 exactly is also quite common. That’s why I think > the smallest place-system base that can cope with a reasonable range of > fractions is 30 -- it has 2, 3 and 5 as prime divisors, and so can cope > with fraction denominators made up arbitrary products and integer powers > of all of those. If you need to represent rational numbers exactly, why not represent them as a pair of integers? I find the idea that 1/3 needs to be exact, but 1/7 doesn't, contrived.