Why is the output format different for two similar Fortran codes?

  • Context: Fortran 
  • Thread starter Thread starter adishpatel
  • Start date Start date
  • Tags Tags
    Fortran
Click For Summary

Discussion Overview

The discussion revolves around the differences in output formats generated by two similar Fortran codes. Participants explore the reasons behind the discrepancy in the organization of results produced by the two codes, focusing on the role of subroutines and write statements.

Discussion Character

  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant notes that the output from the first code appears to be generated by a subroutine called PATH, while the second code's output comes from explicit write statements.
  • Another participant highlights differences in variable assignments and loops between the two codes, suggesting that these may affect output but is unsure how they relate to the output format.
  • A later reply questions whether a variable passed to the PATH subroutine could control its output, indicating a need to examine the subroutine's implementation for clarity.
  • One participant asserts that the "STOP" statement in the second code may prevent the PATH subroutine from executing, thus affecting the output.
  • Another participant expresses skepticism about the relevance of certain code sections to the output, suggesting that the focus should remain on the PATH subroutine's functionality.

    • Areas of Agreement / Disagreement

    Participants generally agree that the output differences stem from the involvement of the PATH subroutine in the first code, while the second code relies on write statements. However, there is no consensus on the specific reasons for the output format differences or how the codes interact with the subroutine.

    Contextual Notes

    Participants mention the need to understand the PATH subroutine better, as its implementation details could clarify why the outputs differ. There are indications that certain variables or conditions may influence the output, but these aspects remain unresolved.

adishpatel
Messages
22
Reaction score
0
So, when I execute my code which is at http://sysden.com/dufing.f

I get the results which are very well organized:
Code: 
 PATH  ==========================================================================================

            N         INP       JCASE         JOB        JBIF        ILAM        IJAC          IW
            2           0           3           1           0           2           1         500


 ** ZPAR   **        EPSZ          HZ         ITZ       METHZ
               0.1000D-08  0.1000D-04          40           1


 ** GPAR   **        EPSG          HG         ITG
               0.1000D-04  0.5000D-03          15


 ** PPAR   **        HMIN          HP        HMAX      FACINC      FACDEC         ITP       NSTEP
               0.1000D-04  0.1000D-01  0.1000D-01  0.2300D+01  0.8000D+00          15          60


 ** EQSOL  **        IODE         LRW         TOL
                        2         100  0.1000D-09
 XBIFUR **** THE BIFURCATION POINT NOT BRACKETED
             AT C( 1) =    0.1498235665D+01   C( 2) =    0.1500000000D+01
             THE EIGENVALUE IS  LAMDA =    0.1018821933D+01    0.0000000000D+00

             AT C( 1) =    0.1498244342D+01   C( 2) =    0.1500000000D+01
             THE EIGENVALUE IS  LAMDA =    0.1018486738D+01    0.0000000000D+00

             AT C( 1) =    0.1500185162D+01   C( 2) =    0.1500000000D+01
             THE EIGENVALUE IS  LAMDA =    0.1075632960D+01    0.0000000000D+00


 PATH  ==========================================================================================


On the other hand, I have a very very similar code with simply some formulas modified, I get the following output (http://sysden.com/Trial.f) :

Code: 
 T =  0.1230D+02    X(1) =   -0.6995969146D-16    X(2) =   -0.8397826270D-14
 T =  0.1232D+02    X(1) =   -0.1296208881D-15    X(2) =    0.4397024319D-15
 T =  0.1233D+02    X(1) =   -0.6896299146D-16    X(2) =    0.8521701466D-14
 T =  0.1234D+02    X(1) =    0.4941227263D-16    X(2) =    0.8977390468D-14
 T =  0.1235D+02    X(1) =    0.1241519258D-15    X(2) =    0.1634297191D-14
 T =  0.1237D+02    X(1) =    0.9342308194D-16    X(2) =   -0.6885493191D-14
 T =  0.1238D+02    X(1) =   -0.1172392094D-16    X(2) =   -0.9706719735D-14
 T =  0.1239D+02    X(1) =   -0.1099822134D-15    X(2) =   -0.4513322049D-14
 T =  0.1240D+02    X(1) =   -0.1183246126D-15    X(2) =    0.4526377004D-14
 T =  0.1242D+02    X(1) =   -0.2783645508D-16    X(2) =    0.1001609073D-13
 T =  0.1243D+02    X(1) =    0.8740604234D-16    X(2) =    0.7396635050D-14
 T =  0.1244D+02    X(1) =    0.1334446529D-15    X(2) =   -0.1576740606D-14
 T =  0.1245D+02    X(1) =    0.6687026721D-16    X(2) =   -0.9388969273D-14
 T =  0.1247D+02    X(1) =   -0.5655341949D-16    X(2) =   -0.9486228884D-14
 T =  0.1248D+02    X(1) =   -0.1362274125D-15    X(2) =   -0.1479291222D-14
 T =  0.1249D+02    X(1) =   -0.1012353469D-15    X(2) =    0.7942125660D-14
 T =  0.1250D+02    X(1) =    0.1776983651D-16    X(2) =    0.1091033964D-13
 T =  0.1252D+02    X(1) =    0.1269133235D-15    X(2) =    0.4786738462D-14
 T =  0.1253D+02    X(1) =    0.1303159882D-15    X(2) =   -0.5510440636D-14
 T =  0.1254D+02    X(1) =    0.2519319926D-16    X(2) =   -0.1117597369D-13
 T =  0.1255D+02    X(1) =   -0.9953672094D-16    X(2) =   -0.7583895225D-14
 T =  0.1257D+02    X(1) =   -0.1393567280D-15    X(2) =    0.2126730762D-14


I want my output to be the way it is for the 1st file. Don't understand why the format differs?

I even commented the format bit from the Trail code and it still wouldn't give me the nicely organized results?

Code: 
      INDEX=1
      T=0.0D+00
   17 FORMAT(' T =',D12.4,'    X(1) =',D20.10,'    X(2) =',D20.10)
      DO 18,  K=1,1000
         TEND = K*TPER/500
         CALL ODE2(FCN,JAC,IJAC,N,T,X,TEND,INDEX,IER)
         IF (X(1).GT.1.0D+01) GOTO 18
         WRITE(6,17) T,X(11),X(12)
         WRITE(3,*) T,X(11),X(12)
   18 CONTINUE
      STOP
 
Technology news on Phys.org
The computer will only do what you tell it to do. In the first case, it appears that subroutine PATH gives some output. In the second case, the output is coming from the write statements you wrote.
 
In dufing.f it has

Code: 
      lamda = (0.1000d+01, 0.0000d+00)                                                                                          
c                                                                       
      tq(1) = tp(1)                                                     
      tq(2) = tp(2)                                                     
      tq(3) = tp(3)                                                     
      tq(4) = tp(4)                                                     
c                                                                       
      tp(1) = 0.0000d+00                                                
      tp(2) = 0.0000d+00                                                
      tp(3) = 0.0000d+00                                                
      tp(4) = 0.1000d+01

And Trial.f has the following instead:

Code: 
C
      DO J=1,N+2
         TQ(J) = TP(J)
      END DO
C
      DO J=1,N+1
         TP(J) = 0.0000d+00
      END DO
      TP(N+2) = 1.0000d+00
C

I even commented out the following from Trial.f to let call path do the job as it does in dufing.f and still the results come out in a different way?

Code: 
      INDEX=1
      T=0.0D+00
   17 FORMAT(' T =',D12.4,'    X(1) =',D20.10,'    X(2) =',D20.10)
      DO 18,  K=1,1000
         TEND = K*TPER/500
         CALL ODE2(FCN,JAC,IJAC,N,T,X,TEND,INDEX,IER)
         IF (X(1).GT.1.0D+01) GOTO 18
         WRITE(6,17) T,X(11),X(12)
         WRITE(3,*) T,X(11),X(12)
   18 CONTINUE
      STOP

Subroutine path? There are two instances where we define subroutine in dufing.f and 3 instances where we have "call path". It is the same for Trial.f as well. Also the code is no different than what I mentioned? Am I overlooking something here :O
 
adishpatel said:
In dufing.f it has

Code: 
      lamda = (0.1000d+01, 0.0000d+00)                                                                                          
c                                                                       
      tq(1) = tp(1)                                                     
      tq(2) = tp(2)                                                     
      tq(3) = tp(3)                                                     
      tq(4) = tp(4)                                                     
c                                                                       
      tp(1) = 0.0000d+00                                                
      tp(2) = 0.0000d+00                                                
      tp(3) = 0.0000d+00                                                
      tp(4) = 0.1000d+01

And Trial.f has the following instead:

Code: 
C
      DO J=1,N+2
         TQ(J) = TP(J)
      END DO
C
      DO J=1,N+1
         TP(J) = 0.0000d+00
      END DO
      TP(N+2) = 1.0000d+00
C
That's irrelevant. It doesn't concern output.

adishpatel said:
I even commented out the following from Trial.f to let call path do the job as it does in dufing.f and still the results come out in a different way?
I have no idea what the PATH subroutine is or does. It is possible that there is a certain variable passed to PATH that controls whether the subroutine produces any output. If you have the source file, you can check to see how PATH produces output. If it's in a library, you have to look at the documentation for that library.
 
I don't think the code shown could have printed the first output. It seems clear that PATH printed the first output. But the "STOP" after the "18 CONTINUE" line stops the program before PATH is executed. There are two write statements before the "STOP" that print the second output.
 
  • Like
Likes   Reactions: 1 person

Similar threads

Fortran How Do You Read and Manipulate Complex Number Data in Fortran?
  • · Replies 8 ·
Replies
8
Views
4K
MCNP Code Error -- Need to find the cause of the error
  • · Replies 7 ·
Replies
7
Views
2K
Fortran How are the initial X values chosen? Fortran
  • · Replies 1 ·
Replies
1
Views
2K
Why do I always get 0 values in the radiation part? (MCNP5)
  • · Replies 58 ·
2
Replies
58
Views
5K
Fortran Fortran 95 Reading in Multi-Dimensional Arrays
  • · Replies 4 ·
Replies
4
Views
9K
Fortran Segmentation error form the fortran code
  • · Replies 4 ·
Replies
4
Views
3K
Fortran Writing strings/arrays to files in FORTRAN 77
  • · Replies 3 ·
Replies
3
Views
4K
Fortran Data I/O problems in FORTRAN 77
  • · Replies 1 ·
Replies
1
Views
3K
Fortran [FORTRAN] Code misassigning values
  • · Replies 16 ·
Replies
16
Views
2K
Fortran Output of max consecutive values/days
  • · Replies 4 ·
Replies
4
Views
2K