Re: 0.0 ** 2 and pow(0.0,0.5) give different results

From: Louis Frayser (frayser@Earthlink.NET)
Date: Wed Apr 23 1997 - 11:46:31 MEST


Thats intersting (SING), the HP manpage says SING (or NaN) is printed 
whenever log(), log2(), or log10() are given 0.

Zero must be getting passed to one of the the log() functions.
There are some identies that probably reduce to taking the log of zero.

I'm running on Linux.  The manual here only says that log(A) is 
only valid for A > 0.

Thanks.

-Louis
--
On 23-Apr-97 "Jacek M. Holeczek" wrote:
>>> root [1] 0.0**2
>> (double)         NaN
>> root [2] (0.0)**2
>> (double)         NaN
>> root [3] pow(0.0,.5)
>> (double)0.000000000000e+00
>> root [4] 
>Which machine / os do you use ?
>On AIX 4.1 I get :
>------------------
>root [0] pow(0.0,.5)
>(double)0.000000000000e+00
>root [1] 0.0**2
>(double)0.000000000000e+00
>root [2] (0.0)**2
>(double)0.000000000000e+00
>------------------
>On HP-UX 9.1 I get :
>------------------
>root [0] pow(0.0,.5)
>(double)0.000000000000e+00
>root [1] 0.0**2
>log: SING error                              <-?
>(double)0.000000000000e+00
>root [2] (0.0)**2
>log: SING error                              <-?
>(double)0.000000000000e+00
>------------------
>Jacek.
>

   _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
  _/ mailto:frayser@earthlink.net     _/ LFrayser@msn.com   _/
 _/http://www.earthlink.net/~frayser _/ LFrayser@candle.com_/
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/



This archive was generated by hypermail 2b29 : Tue Jan 04 2000 - 00:26:18 MET