From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,c0fe3a04fa2416f1 X-Google-Attributes: gid103376,public From: kst@thomsoft.com (Keith Thompson) Subject: Re: Ada function sqrt(x) Date: 1996/09/26 Message-ID: #1/1 X-Deja-AN: 185483601 sender: news@thomsoft.com (USENET News Admin @flash) x-nntp-posting-host: pulsar references: <01bba50c$7c1a2760$dc014dc6@MountainNet> <51o7nh$5vl@cf01> organization: Thomson Software Products, San Diego, CA, USA newsgroups: comp.lang.ada originator: kst@pulsar Date: 1996-09-26T00:00:00+00:00 List-Id: In dewar@cs.nyu.edu (Robert Dewar) writes: [...] > In fact, the square root of 9 must be exactly 3. I can't find an RM reference for this requirement. For implementations that don't support the optional Numerics Annex (Annex G), the accuracy requirements on the mathematical functions are nearly nonexistent. Even Annex G only requires a relative error bound of 2.0*EF.Float_Type'Model_Epsilon (where EF.Float_Type is the floating-point type in question), which still allows Sqrt(9.0) to be slightly above or below 3.0. Of course a decent implementation of Sqrt *should* return exactly 3.0 for Sqrt(9.0). -- Keith Thompson (The_Other_Keith) kst@thomsoft.com <*> TeleSoft^H^H^H^H^H^H^H^H Alsys^H^H^H^H^H Thomson Software Products 10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2706 FIJAGDWOL