From: robin.vowels@gmail.com
Subject: Re: getting same output as gfortran, long_float
Date: Sun, 3 May 2015 17:51:57 -0700 (PDT)
Date: 2015-05-03T17:51:57-07:00 [thread overview]
Message-ID: <b477a0dd-63e1-4845-b525-bdc6d5b56061@googlegroups.com> (raw)
In-Reply-To: <mi0d1s$m8o$1@speranza.aioe.org>
On Saturday, May 2, 2015 at 3:27:29 AM UTC+10, Nasser M. Abbasi wrote:
> On 5/1/2015 10:39 AM, Waldek Hebisch wrote:
> > Nasser M. Abbasi <nnospam@12000.org> wrote:
> >>
> >> Thanks, this is very useful. I noticed small difference in output:
> >> side-by-side:
> >>
> >> Ada: 0.00182898948331047096479195244627343392775491540..
> >> Mathematica*: 0.00182898948331047096479195244627343392775491540..
> >> gfortran: 0.00182898948331047112025871115292829927
> >>
> >> at digit 18, gfortran result is different. But your Ada rational
> >> package gives same result as Mathematica. This tells me your
> >> result is the accurate one !
> >
> > Your previous gfortran result was:
> >
> > 1.82898948331047096479195244627343369E-3
> >
> > which agrees with others...
> >
>
> Thanks. My mistake. So the command line flag was needed after all!!
>
> >gfortran -Wall -fdefault-real-8 foo2.f90
> >./a.out
> 1.82898948331047096479195244627343369E-0003
>
> Without the flag, the digit is different
>
> >gfortran -Wall foo2.f90
> >./a.out
> 1.82898948331047112025871115292829927E-0003
>
> as you say, this is due to D0 in code using double, while
> the result "x" is quad. With the flag in, gfortran will
> treate all doubles as "quad", hence the correct result.
>
> I thought with -Wall it will catch all these things, but I think
> I need more flags to detect this user error
>
> >> for reference:
> >>
> >> -------------------
> >> PROGRAM foo
> >> IMPLICIT NONE
> >> REAL(KIND = 16) :: x !-- kind=16 tells it is double quad
> >> x = 12.0D0 * 0.0001D0/(1.0D0 * (1.0D0 - 0.1D0)**4 )
> >> PRINT *, x
> >> END PROGRAM
> >
> > AFAICS you have 34 digit x, but 17 digit constants. So
> > it seems that you assign 17 digit number to x, no
> > wonder there is no gain in accuracy.
> >
>
> Yes, needed the command line flag.
Not really. he correct way is to specify quad precision constants,
e.g.,
0.0001_quad
next prev parent reply other threads:[~2015-05-04 0:51 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-30 21:17 getting same output as gfortran, long_float Nasser M. Abbasi
2015-04-30 22:08 ` Dmitry A. Kazakov
2015-04-30 22:11 ` Dmitry A. Kazakov
2015-04-30 22:37 ` Nasser M. Abbasi
2015-04-30 22:53 ` Nasser M. Abbasi
2015-05-01 7:22 ` Jacob Sparre Andersen
2015-05-01 1:12 ` Nasser M. Abbasi
2015-05-01 6:52 ` Dmitry A. Kazakov
2015-05-01 7:32 ` Nasser M. Abbasi
2015-05-01 7:45 ` Dmitry A. Kazakov
2015-05-04 0:15 ` robin.vowels
2015-05-04 7:21 ` Dmitry A. Kazakov
2015-05-04 8:53 ` robin.vowels
2015-05-04 10:18 ` Dmitry A. Kazakov
2015-05-04 13:45 ` robin.vowels
2015-05-04 14:47 ` Dmitry A. Kazakov
2015-05-07 2:01 ` robin.vowels
2015-05-01 23:24 ` Dennis Lee Bieber
2015-05-04 0:09 ` robin.vowels
2015-05-01 7:01 ` Dmitry A. Kazakov
2015-05-04 0:42 ` robin.vowels
2015-04-30 22:12 ` Jeffrey R. Carter
2015-04-30 22:27 ` Qun-Ying
2015-05-01 0:59 ` Dennis Lee Bieber
2015-04-30 22:32 ` Nasser M. Abbasi
2015-05-01 1:16 ` Jeffrey R. Carter
2015-05-01 1:40 ` Nasser M. Abbasi
2015-05-01 7:47 ` Jacob Sparre Andersen
2015-05-01 15:39 ` Waldek Hebisch
2015-05-01 17:27 ` Nasser M. Abbasi
2015-05-01 18:03 ` Nasser M. Abbasi
2015-05-04 0:51 ` robin.vowels [this message]
2015-05-04 0:47 ` robin.vowels
2015-05-01 8:21 ` Simon Wright
2015-05-01 11:55 ` Georg Bauhaus
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox