Approximatrix Forums — problems with double precision functions

Re: problems with double precision functions

2013-06-11T18:42:34Z

It appears that a fresh installation of ver 1.44 in windows XP eliminated the problem.

Re: problems with double precision functions

2013-06-08T15:10:47Z

Okay, I'll have a look at it on Windows XP to see if I can replicate the issue.

Re: problems with double precision functions

2013-06-08T12:20:45Z

No. . I dont think so. . .However this program was recompile with no issues with this same compiler ( ver 44 ) in windows 7. The issues that I discuss occur under windows xp.

Re: problems with double precision functions

2013-06-07T22:33:19Z

Interesting... I didn't see the warning under 1.44 myself. Additionally, 1.43 and 1.44 should be using the same version of the Fortran compiler. I also don't see the issue under the upcoming 1.45, currently in testing.

I'll keep checking to see if I can replicate the issue. Did you set any interesting compiler flags by chance?

Re: problems with double precision functions

2013-06-07T21:13:36Z

no problems with build 1.43 1191. Program compiles 'cleanly'.

Re: problems with double precision functions

2013-06-07T18:29:46Z

i am using 1.44.

Re: problems with double precision functions

2013-06-07T18:28:27Z

In "Build Status" tab, it is noted that the arguments for dcos, dsin, datan2, and dsqrt are not 'double precision', even though they are declared as double precision (real*8).

Re: problems with double precision functions

2013-06-07T17:41:52Z

Where in the code are you seeing a problem? I spot-checked the results for dsqrt and datan2, and I didn't see a problem. I just want to understand what you meant in your original post.

Re: problems with double precision functions

2013-06-07T16:54:06Z

C Last change: SC 19 Feb 2013 6:11 pm

program Mis_Brg

implicit none
real*8 pi,second,distance,latitude,departure
real*8 radian_bearing,Missing_bearing,missing_distance
real*8 Sum_lat,sum_dep,s_mis
real*8 easting_start,northing_start,easting,northing

parameter (easting_start=5000.0d0,northing_start=5000.0d0)

integer*4 degree,minute,d_mis,m_mis

open(unit=9,file='brgdist.d95')
open(unit=11,file='results.d95')
open(unit=13,file='coordinates.d95')

sum_lat=0.0d0
sum_dep=0.0d0

pi=datan(1.0d0)*4.0d0

write(11,300)
300 format(///3x,'bearing',8x,'distance',11x,'latitude',11x
& ,'departure'/)

easting=easting_start
northing=northing_start

write(13,*)easting,northing

400 continue

read (9,*,end=500)degree,minute,second,distance

radian_bearing=(dble(degree)+dble(minute)/60.0d0+second/3600.0d0)
& *pi/180.0d0

latitude=distance*dcos(radian_bearing)
departure=distance*dsin(radian_bearing)

sum_lat=sum_lat+latitude
sum_dep=sum_dep+departure

easting=easting+departure

northing=northing+latitude

write(11,111)degree,minute,second,distance,
& latitude,departure

111 format(/i4,i3.2,f6.2,f12.3,f23.14,f20.14)

write(13,*)easting,northing

Go to 400

500 Missing_bearing=datan2(sum_dep,sum_lat)
if(Missing_bearing<0.0d0)
& Missing_bearing=Missing_bearing+2.0d0*pi

call rad_brg_to_dms_brg(pi,Missing_bearing,d_mis,m_mis,s_mis)

missing_distance=dsqrt(sum_lat**2+sum_dep**2)

WRITE(13,'(/)')
write(13,*)easting-easting_start,northing-northing_start

write(11,'(///f15.10)')Missing_bearing
write(11,'(i5,i3.2,f7.3)')d_mis,m_mis,s_mis
write(11,'(f15.6)')missing_distance
stop
end program Mis_Brg

subroutine rad_brg_to_dms_brg(pi,rad_brg,d,m,s)
implicit none
real*8 pi,s,rad_brg,dec_brg,minute
integer*4 d,m

dec_brg=rad_brg*180.0d0/pi

d=dec_brg
minute=(dec_brg-dble(d))*60.0d0
m=minute
s=(minute-dble(m))*60.0d0

return
end subroutine rad_brg_to_dms_brg

Re: problems with double precision functions

2013-06-07T16:31:08Z

Could you post a short snippet of offending code? I'm not seeing the issue with a simple example I've tried. Are you on the latest version of Simply Fortran (1.44)?

problems with double precision functions

2013-06-07T12:01:01Z

I have found that the double precision functions such as dsin, dcos ,dsqrt, etc does not recognise the double precision arguments even when the arguments are explicitly declared double precision.