C++Talk.NET C++ language newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  Semantics of numeric_limits<double>::min()          C++Talk.NET Forum Index -> C++ Language (Moderated) View previous topic :: View next topic   Author Message Jonas Latt Guest Posted: Mon Aug 15, 2005 6:29 pm    Post subject: Semantics of numeric_limits<double>::min() Hi From its name, I expected numeric_limits<double>::min() to return a negative value as close as possible to -infinity. It does not. It actually returns a positive value very close to 0. The semantics for int are different. On my workstation, I get numeric_limits<int>::int() = -2147483648 (as tested on g++ and intel icc). This strikes me as odd. Does anybody know the actual definition of numeric_limits<double>::min() ? What do I do if I want to get the value of the datatype double that is closest to -infinity? Many thanks for your help, Jonas [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top E. Mark Ping Guest Posted: Tue Aug 16, 2005 11:08 am    Post subject: Re: Semantics of numeric_limits<double>::min() In article <4300bad9$1 (AT) nntp (DOT) unige.ch>, Jonas Latt <Jonas.Latt (AT) cui (DOT) unige.ch> wrote: Quote: Hi From its name, I expected numeric_limits<double>::min() to return a negative value as close as possible to -infinity. It does not. It actually returns a positive value very close to 0. Yes, that's why it's best not to guess what a function does. From the standard, 18.2.1.2/2: "For floating types with denormalization, returns the minimum positive normalized value." Quote: What do I do if I want to get the value of the datatype double that is closest to -infinity? that would be -std::numeric_limits<double>::max(). -- Mark Ping [email]emarkp (AT) soda (DOT) CSUA.Berkeley.EDU[/email] [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Greg Guest Posted: Tue Aug 16, 2005 11:09 am    Post subject: Re: Semantics of numeric_limits<double>::min() Jonas Latt wrote: Quote: Hi From its name, I expected numeric_limits<double>::min() to return a negative value as close as possible to -infinity. It does not. It actually returns a positive value very close to 0. The semantics for int are different. On my workstation, I get numeric_limits<int>::int() = -2147483648 (as tested on g++ and intel icc). This strikes me as odd. Does anybody know the actual definition of numeric_limits<double>::min() ? What do I do if I want to get the value of the datatype double that is closest to -infinity? Many thanks for your help, Jonas For floating point types the values in numeric_limits are always positive. To find a numeric limit for a floating point type within the range of negative numbers, simply negate the appropriate limit in the positive range. For example the value closest to negative infinity that a double can represent is its representable value closest to positive infinity, negated: -numeric_limits<double>::max(); Now it may seem asymmetric that the floating point limits are expressed as positive values, while the integral limits range from positive to negative. The difference I believe derives from the two different ways each manages their signedness. Integral types use a twos-complement format, while floating types have a single sign bit. In fact it is the presence of the sign bit for floating types that assures us that negating the positive limit as done above will produce the correct limit in the negative range. Greg [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Abhishek Pamecha Guest Posted: Tue Aug 16, 2005 11:11 am    Post subject: Re: Semantics of numeric_limits<double>::min() a sly trick would be to possibly use (- numeric_limits<double>::max() ) hth abhishek [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top kevin.hall@motioneng.com Guest Posted: Tue Aug 16, 2005 11:39 am    Post subject: Re: Semantics of numeric_limits<double>::min() Floating point numbers have a single bit that denotes sign. The largest negative number is just the largest positive number * -1. So you may use: -numeric_limits<double>::max() [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Rob Guest Posted: Tue Aug 16, 2005 2:42 pm    Post subject: Re: Semantics of numeric_limits<double>::min() Jonas Latt wrote: Quote: Hi From its name, I expected numeric_limits<double>::min() to return a negative value as close as possible to -infinity. It does not. It actually returns a positive value very close to 0. The semantics for int are different. On my workstation, I get numeric_limits<int>::int() = -2147483648 (as tested on g++ and intel icc). This strikes me as odd. Does anybody know the actual definition of numeric_limits<double>::min() ? It is try positive value closest to zero that can be represented by a double. Quote: What do I do if I want to get the value of the datatype double that is closest to -infinity? I would probably use -std::numeric_limits<double>::max() There is a complication that floting point types can sometimes include a representation of an infinite value (i.e. there is a reserved value that means "infinity"). Not all floating point types support that, and I assume it is not what you're looking for in any case. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Ben Hutchings Guest Posted: Tue Aug 16, 2005 6:01 pm    Post subject: Re: Semantics of numeric_limits<double>::min() Jonas Latt <Jonas.Latt (AT) cui (DOT) unige.ch> wrote: Quote: Hi From its name, I expected numeric_limits<double>::min() to return a negative value as close as possible to -infinity. It does not. It actually returns a positive value very close to 0. The semantics for int are different. On my workstation, I get numeric_limits<int>::int() = -2147483648 (as tested on g++ and intel icc). This strikes me as odd. Does anybody know the actual definition of numeric_limits<double>::min() ? It's the smallest strictly positive double. Quote: What do I do if I want to get the value of the datatype double that is closest to -infinity? Negate numeric_limits<double>::max(). A recent discussion of this oddity is archived at <http://groups.google.com/group/comp.lang.c++.moderated/browse_frm/thread/aea18fab432eee69> -- Ben Hutchings Having problems with C++ templates? Your questions may be answered by <http://womble.decadentplace.org.uk/c++/template-faq.html>. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First         C++Talk.NET Forum Index -> C++ Language (Moderated) All times are GMT Page 1 of 1   Jump to: Select a forum C/C++----------------C LanguageC Language (Moderated)C++ Language (Moderated)C++ language (comp.lang.c++)C++ language, library and standards Non-English C/C++ Newsgroups----------------C++ (German)C++ (French)  You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum Powered by phpBB © 2001, 2006 phpBB Group SEO toolkit © 2004-2006 webmedic.