10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
plo...@gmail.com <plo...@gmail.com> wrote:Thanks Nils,
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:very strange.
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact... It is
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:is very strange.
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact... It
Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Best regards
Andrey Nikitin
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:is very strange.
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact... It
Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Best regards
Andrey Nikitin
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:is very strange.
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact... It
Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:It is very strange.
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact...
some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem in
regards,Hi Andrey,
Andrey
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:It is very strange.
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact...
some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem in
I have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.FOR sourceregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:It is very strange.
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact...
some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem in
regards,Andrey
Andrey
On Wednesday, March 15, 2023 at 9:12:57 AM UTC-4, Андрей Никитин wrote:It is very strange.
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact...
some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem in
regards,Andrey
Andrey
Played with this: (on CP/M emulator -- haven't tried on "real" CP/M 2.2 yet):
: fred@llana prospero $; cpm profor
Pro Fortran Compiler - Version zz 1.2
Copyright (C) 1983 Prospero Software
Serial No: 63
Source filename - maze.for
Source listing ? (Y/N/.) Y
Map ? (Y/N/.) Y
Compact object code ? (Y/N/.) N
Restricted-width integer arithmetic ? (Y/N/.) N
Maintain source line numbers at run time ? (Y/N/.) Y
Range checks on subscripts ? (Y/N/.) N
Range checks on assignments ? (Y/N/.) N
Hold before .REL output ? (Y/N/.) Y
Accept only ANSI Standard Fortran ? (Y/N/.) N
Console output to .LOG file ? (Y/N/.) Y
NOTE: looks like just running the profor.com without command line arguments prompts for arguments.
Pass 1
Unit MAZE
17726 bytes free at end
Pass 2
Holding for REL - press any key to continue
Unit: Code: Data:
MAZE 3372 1385
Total source lines: 226
23378 bytes free at end
NOTE: Two compile passes! Yeah us!
: fred@llana prospero $; dir
copytx.for forlib.rel maze.for maze.prn profor2.com prolib.com
fcheck.com g1lib.mac maze.log maze.rel profor.com prolink.com
fconfig.com g2lib.mac maze.map prime.for profor.err read.me
NOTE: and we see maze.log, maze.map maze.prn and maze.rel have been generated. If we display maze.log:
: fred@llana prospero $; cat maze.log
Options: LMNHG
Pass 1
Unit MAZE
17726 bytes free at end
Pass 2
Holding for REL - press any key to continue
Unit: Code: Data:
MAZE 3372 1385
Total source lines: 226
23378 bytes free at end
�: fred@llana prospero $;
NOTE: We see that the compiler likely supports LMNHG switches.
Most likely one letter for each question... by running 10 times we could find out which switch is for each question!
NOTE: Now, we link. The compiler supports MAC files, so Microsoft M80, and comes with FORLIB.REL (not the same as Microsoft FORLIB.REL). So, I decided to try to link with L80:
: fred@llana prospero $; l80
Link-80 3.44 09-Dec-81 Copyright (c) 1981 Microsoft
*maze
Data 0103 1398 < 4757>
-$BMUL 095B -$DENT 06AC -$DEXI 06D5
-$DIN 06DE -$DSLIN 1394 -$EFMT 1383
-$EIRF 08BE -$EIWF 1386 -$ERFI1 08C5
-$ETRF 08C8 -$ETWF 1389 -$EUNIT 137C
-$EWFR 1339 -$GCGT 0E61 -$GINIT 06DB
-$HBNG 0BF7 -$HBPS 069A -$HBZR 12A9
-$HINVS 136D -$HST4T 12F1 -$LDEC 0B8E
-$LINC 1027 -$RFLT 06C6 -$RMUL 06C9
-$TRUNC 06CE -$WWMUL 1295 -RANDOM 06BA
27 Undefined Global(s)
48837 Bytes Free
*forlib/s
*maze/n/e
Data 0103 3248 <12613>
39888 Bytes Free
[06D7 3248 50]
NOTE: And we end up with MAZE.COM!!! Yeah us!
NOTE: And, the COM file is runnable:
: fred@llana prospero $; cpm maze
THIS PROGRAM WILL PRINT OUT A DIFFERENT MAZE EVERY TIME IT
IS RUN AND GUARANTEES ONLY ONE PATH THROUGH. YOU CAN
CHOOSE THE DIMENSIONS OF THE MAZE, I.E. THE NUMBER OF
SQUARES LONG AND THE NUMBER OF SQUARES WIDE.
A 25 BY 25 MAZE IS THE MAXIMUM, AND ANY DIMENSIONS UP TO
THESE LIMITS ARE O.K.
LENGTH: 10
...SNIP...
Notes: The PRN (listing) file correlates line number with object code address -- useful if
we want the code productions. The MAP file produces a variable map - name/kind/type/offset
These could be very useful if debugging.
Questions:
1 - Can Prospero be linked with Microsoft FORLIB.REL? If this is done, what is the result?
2 - Related to (1), Microsoft never supported COMPLEX, could this be the root of the problem?
3 - I do not see names for compiler aid functions that make sense (from a Microsoft perspective.
This leads me to think that Microsoft FORLIB.REL and Prospero FORLIB.REL are not compatible.
4 - From file PROFOR.ERR, we see:
223Illegal type mixture with COMPLEX
224COMPLEX expression in arithmetic IF
225COMPLEX operand in relational expression
234DATA: illegal occurrence of COMPLEX type
236Assignment: illegal use of COMPLEX type
320Error in COMPLEX constant
So,COMPLEX support appears somewhat special (error 223). *When/If* I get time, I will scratch further
(Note that I am not bothering with PROLINK or PROLIB)
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:.. It is very strange.
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is, intact.
in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem
source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints , itI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.FORregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:intact... It is very strange.
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is,
problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this
source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints , itI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.FORregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is,
in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this problem
source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints , itI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.FORregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is,
problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this
source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints , itI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.FORregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
On Thursday, March 16, 2023 at 3:23:43 PM UTC-4, Plo wrote:intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that is,
problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this
FOR source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints ,I have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.regards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
Strangely, the compile error "pass 2" DOES NOT APPEAR only when there is only one intrinsic function of a only complex variable i.e. Csin(z), Ccos(z), z**constant, etc. etc, when there are two or more complex operands, then the error appears!!!!So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
Il giorno giovedì 16 marzo 2023 alle 17:13:37 UTC-3 fridtjof.ma...@gmail.com ha scritto:is, intact... It is very strange.
On Thursday, March 16, 2023 at 3:23:43 PM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK! that
problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved this
FOR source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C and prints ,I have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) The COMPLEX.regards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeSo, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
Strangely, the compile error "pass 2" DOES NOT APPEAR only when there is only one intrinsic function of a only complex variable i.e. Csin(z), Ccos(z), z**constant, etc. etc, when there are two or more complex operands, then the error appears!!!!And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
Il giorno giovedì 16 marzo 2023 alle 20:32:48 UTC-3 Plo ha scritto:that is, intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 17:13:37 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 3:23:43 PM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK!
this problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved
COMPLEX.FOR source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C andI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) Theregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeStrangely, the compile error "pass 2" DOES NOT APPEAR only when there is only one intrinsic function of a only complex variable i.e. Csin(z), Ccos(z), z**constant, etc. etc, when there are two or more complex operands, then the error appears!!!!And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
Pe vineri, 17 martie 2023, la 01:39:01 UTC+2, Plo a scris:that is, intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 20:32:48 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 17:13:37 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 3:23:43 PM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK!
this problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved
COMPLEX.FOR source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C andI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) Theregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
Thank-You!!!So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
Perhaps this would help: I found a link for the PROSPERO PROFORTRAN manual : http://www.dilwyn.me.uk/language/ProsperoQlFortran77UM3.pdfI'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeStrangely, the compile error "pass 2" DOES NOT APPEAR only when there is only one intrinsic function of a only complex variable i.e. Csin(z), Ccos(z), z**constant, etc. etc, when there are two or more complex operands, then the error appears!!!!And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Why??????????????????????
Ladislau
Il giorno giovedì 16 marzo 2023 alle 20:32:48 UTC-3 Plo ha scritto:that is, intact... It is very strange.
Il giorno giovedì 16 marzo 2023 alle 17:13:37 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 3:23:43 PM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 16:19:41 UTC-3 Plo ha scritto:
Il giorno giovedì 16 marzo 2023 alle 15:22:40 UTC-3 fridtjof.ma...@gmail.com ha scritto:
On Thursday, March 16, 2023 at 10:32:44 AM UTC-4, Plo wrote:
Il giorno giovedì 16 marzo 2023 alle 08:22:41 UTC-3 Plo ha scritto:
Il giorno mercoledì 15 marzo 2023 alle 10:12:57 UTC-3 Андрей Никитин ha scritto:
среда, 15 марта 2023 г. в 03:23:54 UTC+3, plo...@gmail.com:
Il giorno sabato 11 marzo 2023 alle 16:39:13 UTC-3 Андрей Никитин ha scritto:
среда, 8 марта 2023 г. в 12:23:21 UTC+3, plo...@gmail.com:
Il giorno martedì 7 marzo 2023 alle 07:26:10 UTC-3 Nils M Holm ha scritto:
plo...@gmail.com <plo...@gmail.com> wrote:
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)It is probably unrelated, but I see two 2H with only one
subsequent blank here.
--Thanks Nils,
Nils M Holm < n m h @ t 3 x . o r g > http://t3x.org
But that's just my transcription error of the source. The strange thing is that in the package there is an executable file FCHECK.com which is used to check all the files in the package and when I run it, all the files are OK!
this problem in some other way, and generate an executable file. I will be happy to get any information on this matter.Is it possible to download this compiler from the Internet. If possible please send me a link to where I can do it.
Thanks for the link to the compiler.Best regardshttps://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=sharing
Andrey Nikitin
This compiler is less common among CP/M users and almost unavailable. It is a pity that there is no user manual for it.
I was able to compile the test programs and get object files, but I could not get executable files out of them. When I run the linker and specify an object file name and then a library name, the program crashes. Maybe you have solved
COMPLEX.FOR source generates an incomprehensible error during the PASS 2 of the compiler and if you always confirm to go to the end of the compilation and then do the LINK you will notice that when you run the complex.com executable which sums A+B=C andI have inserted the ProsFor.rar file which contains the complete compiler and linker files. To compile:>PROFOR and follow compiler prompts. For the LINKER :>Prolink and follow the linker requests (xxx1.rel,xxx2.rel....,FORLIB.rel) Theregards,Hi Andrey,
Andrey
I want to verify that the files I uploaded in the folder are correct...then I'll let you know...
I confirm your error:
Error C at address 1D5F
Continue ? (Y/N) y
FCHECK seems to like the compiler parts
: fred@llana prospero $; cpm fcheck
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
Is it possible that COMPLEX support is simply "not there"?
: fred@llana prospero $; cat complex.map
Pro Fortran Compiler - Version zz 1.2
Map of: complex.FOR
****************
* Unit COMPLE *
****************
Name Kind Type Area Offset
A Variable Comp Data 0008 (8)
B Variable Comp Data 0010 (16)
Z Variable Comp Data 0000 (0)
Sizes:-
Data 0018 (24)
The SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"So, the MAP file indicates that complex is there... 8 bytes is correct. I thought that maybe the IMPLICIT was interfering, and that's not it. Nor is it the variable >layout - commenting out Z = makes A 0, B 8 and Z 16 (offset). Same error.Why doesn't the compiler (PROFOR2) give an error with exponential complex function? In the algebraic sum Z=A+B it gives the error!! I can't understand this...
PROGRAM COMPLE
IMPLICIT REAL (A-H,O-Z)
COMPLEX A,B,Z
C Z = CMPLX(0.0, 0.0)
A = CMPLX(3.5, 3.5)
B = CMPLX(4.0, 5.0)
Z = B**2
WRITE(1,10) A, B, Z
10 FORMAT(1H ,2F11.3,2H ,2F11.3,2H ,2F11.3)
PAUSE
STOP
END
No compiler error!!!
++++++++++++++++++++++++++++++
Pass 2
Unit: Code: Data:
COMPLE 153 24
Total source lines 12
Prolink.........
Linking:
COMPLEX.REL
FORLIB.REL
Data:.......
...............
..............
Executable File: COMPLEX.COM
I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeStrangely, the compile error "pass 2" DOES NOT APPEAR only when there is only one intrinsic function of a only complex variable i.e. Csin(z), Ccos(z), z**constant, etc. etc, when there are two or more complex operands, then the error appears!!!!And, just for fun - the error is in both COMPLEX + COMPLEX *and* COMPLEX + REAL. Now, I am "nerd-sniped"C Z = CMPLX(0.0, 0.0) Comment in column 1 ==>>...... C Z=CMPLX(0.0, 0.0)COMPLEX3.500 3.500 4.000 5.000 -9.000 40.000
A complex B complex Z=B**2 EXACT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Why??????????????????????
I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m emulator+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link
Il giorno lunedì 20 marzo 2023 alle 04:34:31 UTC-3 ladislau szilagyi ha scritto:emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
Wonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version"
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr
8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$.....
Ladislau
Is it possible to disassemble Amstrad's PROFOR.COM and remove the version check? It would be a resounding success to be able to use it also on other platforms and versions of CP/M-80!!!
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
Wonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version"
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr 8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$.....
Ladislau
On 20.03.23 11:04, Plo wrote:emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
Il giorno lunedì 20 marzo 2023 alle 04:34:31 UTC-3 ladislau szilagyi ha scritto:
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
Hi,You can always replace at least the "jr c,l8ac8h" with two NOPs toWonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version" >>
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr >> 8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$..... >>
Ladislau
Is it possible to disassemble Amstrad's PROFOR.COM and remove the version check? It would be a resounding success to be able to use it also on other platforms and versions of CP/M-80!!!
neverever jump to the error message. BTDT(*)
Josef
(*) When I got my SB180FX, I wanted to add a hard disk (10MB!!!!!
Yeah!") which I bought second-hand from my employer. Unfortunately the formatting program refused to format the disk, so I assumed it was
defective and gave it back in exchange for another one, which ... also
did not work. An analysis of the formatting program revealed that it
first checked the existing format and refused to format the disk it it didn't have 512 byte sectors ;-) I then just over-wrote the check with
NOPs and ... voila ... the disk was successfully formatted to 512 bytes/sector.
Il giorno lunedì 20 marzo 2023 alle 08:08:50 UTC-3 ladislau szilagyi ha scritto:m emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
Pe luni, 20 martie 2023, la 12:15:10 UTC+2, Josef Möllers a scris:
On 20.03.23 11:04, Plo wrote:
Il giorno lunedì 20 marzo 2023 alle 04:34:31 UTC-3 ladislau szilagyi ha scritto:
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
that address you could correct the error... Since I'm ignorant of assembly language... and maybe therefore I'm talking nonsense... but I think that the 1D5F address could be the key to the solution to the problemYou can always replace at least the "jr c,l8ac8h" with two NOPs to neverever jump to the error message. BTDT(*)Wonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version"
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr
8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$.....
Ladislau
Is it possible to disassemble Amstrad's PROFOR.COM and remove the version check? It would be a resounding success to be able to use it also on other platforms and versions of CP/M-80!!!
Josef
(*) When I got my SB180FX, I wanted to add a hard disk (10MB!!!!! Yeah!") which I bought second-hand from my employer. Unfortunately the formatting program refused to format the disk, so I assumed it was defective and gave it back in exchange for another one, which ... also did not work. An analysis of the formatting program revealed that it first checked the existing format and refused to format the disk it it didn't have 512 byte sectors ;-) I then just over-wrote the check with NOPs and ... voila ... the disk was successfully formatted to 512 bytes/sector.Hi,
I tried to patch the PROFOR.COM, the message "This is Amstrad version" is not more displayed...
... but another message is displayed instead: "Bad copy of program".
It seems that the authors of the code used some sort of checksum control...
The only correct way to fix this is to disassemble all and remove the unwanted code... but this is not so simple as it looks.
LadislauSince it is the PROFOR2.com file that generates the error on algebraic operations with the COMPLEX data type i.e. Pass 2, and the error is always at the same address: Pass 2 error C at address 1D5F perhaps comparing the two versions of PROFOR2.com at
Pe luni, 20 martie 2023, la 12:15:10 UTC+2, Josef Möllers a scris:emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
On 20.03.23 11:04, Plo wrote:
Il giorno lunedì 20 marzo 2023 alle 04:34:31 UTC-3 ladislau szilagyi ha scritto:
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80 CP/m+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
Since it is the PROFOR2.com file that generates the error on algebraic operations with the COMPLEX data type i.e. Pass 2, and the error is always at the same address: Pass 2 error C at address 1D5F perhaps comparing the two versions of PROFOR2.com atYou can always replace at least the "jr c,l8ac8h" with two NOPs to neverever jump to the error message. BTDT(*)Wonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version"
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr
8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$.....
Ladislau
Is it possible to disassemble Amstrad's PROFOR.COM and remove the version check? It would be a resounding success to be able to use it also on other platforms and versions of CP/M-80!!!
Josef
(*) When I got my SB180FX, I wanted to add a hard disk (10MB!!!!!Hi,
Yeah!") which I bought second-hand from my employer. Unfortunately the formatting program refused to format the disk, so I assumed it was defective and gave it back in exchange for another one, which ... also
did not work. An analysis of the formatting program revealed that it
first checked the existing format and refused to format the disk it it didn't have 512 byte sectors ;-) I then just over-wrote the check with NOPs and ... voila ... the disk was successfully formatted to 512 bytes/sector.
I tried to patch the PROFOR.COM, the message "This is Amstrad version" is not more displayed...
... but another message is displayed instead: "Bad copy of program".
It seems that the authors of the code used some sort of checksum control...
The only correct way to fix this is to disassemble all and remove the unwanted code... but this is not so simple as it looks.
Ladislau
Il giorno lunedì 20 marzo 2023 alle 16:28:09 UTC-3 Plo ha scritto:CP/m emulator it tells me that the pc is not Amstrad PC!!!! and exits compilation. The files are the same as the generic CP/M-80 distribution. If you want to have fun, install the Amstrad Jocelyne emulator and after booting with the CP/M image disks and
Il giorno lunedì 20 marzo 2023 alle 08:08:50 UTC-3 ladislau szilagyi ha scritto:
Pe luni, 20 martie 2023, la 12:15:10 UTC+2, Josef Möllers a scris:
On 20.03.23 11:04, Plo wrote:
Il giorno lunedì 20 marzo 2023 alle 04:34:31 UTC-3 ladislau szilagyi ha scritto:
Pe luni, 20 martie 2023, la 00:43:00 UTC+2, Plo a scris:
Il giorno sabato 18 marzo 2023 alle 14:59:18 UTC-3 Mark Lougheed ha scritto:
There is also a version of Prospero PRO Fortran CP/m-80 specific for Amstrad CPC9512 which is distributed with the same files and which works perfectly without any compilation errors under CP/M-80. But if I try to use the compiler in my Z80+1 with regard to SuperSoft FORTRAN - For Z-80. I don't recall an 8080 targeted CP/M-80 FORTRAN compiler that had COMPLEX data types - certainly not Microsoft F-80 or Nevada FORTRAN.I'm going crazy!!!!! And I'm sorry because it's the only Fortran Compiler under CP/M-80 with COMPLEX data TypeThe SuperSoft FORTRAN (version 3.2) supports COMPLEX as well. But I would love to see the Prospero FORTRAN "brought back to life"
http://www.z80.eu/for-compiler.html
that address you could correct the error... Since I'm ignorant of assembly language... and maybe therefore I'm talking nonsense... but I think that the 1D5F address could be the key to the solution to the problemYou can always replace at least the "jr c,l8ac8h" with two NOPs to neverever jump to the error message. BTDT(*)Wonderful analysis!!!You can find everything in my previous link..... https://drive.google.com/drive/folders/1ms1yidXdQswVCvs0yU63sCnLTX6HhzKr?usp=share_link(@Plo)
Hi,
it seems that the Amstrad Fortran runs only on CPM 3...
A quick check of the code ( PROFOR.COM ) shows that there is code that checks this:
sub_8aadh:
ld c,00ch ;8aad GET CP/M VERSION
call 00005h ;8aaf
cp 030h ;8ab2 COMPARE WITH 3
jr c,l8ac8h ;8ab4 IF LESS, ISSUE ERROR MESSAGE: "This is Amstrad version"
l8ac8h:
ld c,009h ;8ac8
ld de,l8ad3h ;8aca SEE BELOW THE ERROR MESSAGE
call 00005h ;8acd
jp 00000h ;8ad0
8AD0 : C3 00 00 54 68 69 73 20 69 73 20 41 6D 73 74 72 : ...This is Amstr
8AE0 : 61 64 20 76 65 72 73 69 6F 6E 24 00 00 00 00 00 : ad version$.....
Ladislau
Is it possible to disassemble Amstrad's PROFOR.COM and remove the version check? It would be a resounding success to be able to use it also on other platforms and versions of CP/M-80!!!
Josef
(*) When I got my SB180FX, I wanted to add a hard disk (10MB!!!!! Yeah!") which I bought second-hand from my employer. Unfortunately the formatting program refused to format the disk, so I assumed it was defective and gave it back in exchange for another one, which ... also did not work. An analysis of the formatting program revealed that it first checked the existing format and refused to format the disk it it didn't have 512 byte sectors ;-) I then just over-wrote the check with NOPs and ... voila ... the disk was successfully formatted to 512 bytes/sector.Hi,
I tried to patch the PROFOR.COM, the message "This is Amstrad version" is not more displayed...
... but another message is displayed instead: "Bad copy of program".
It seems that the authors of the code used some sort of checksum control...
The only correct way to fix this is to disassemble all and remove the unwanted code... but this is not so simple as it looks.
LadislauSince it is the PROFOR2.com file that generates the error on algebraic operations with the COMPLEX data type i.e. Pass 2, and the error is always at the same address: Pass 2 error C at address 1D5F perhaps comparing the two versions of PROFOR2.com at
Or is this address referred to the MAP file of the relocatable obj file?Hi all, finally the problem is solved...The member of these two forums, Martin Hepperle, found a brilliant and perfectly working solution, here are the two links to the solution...
Hi all, finally the problem is solved...The member of these two forums, Martin Hepperle, found a brilliant and perfectly working solution, here are the two links to the solution...
https://forum.vcfed.org/index.php?threads/strange-compiler-error-prospero-fortran-cp-m-80.1241152/
https://forum.classic-computing.de/forum/index.php?thread/28785-strange-compiler-error-prospero-fortran-cp-m-80/&postID=406897#post407059
Greetings to all...
I don't know him personally, but I can tell you that he is a very knowledgeable and professional person in machine language programming and high level languages...Hi all, finally the problem is solved...The member of these two forums, Martin Hepperle, found a brilliant and perfectly working solution, here are the two links to the solution...Pardon me for the brief diversion...
https://forum.vcfed.org/index.php?threads/strange-compiler-error-prospero-fortran-cp-m-80.1241152/
https://forum.classic-computing.de/forum/index.php?thread/28785-strange-compiler-error-prospero-fortran-cp-m-80/&postID=406897#post407059
Greetings to all...
Martin Hepperle? THE Martin Hepperle of JavaFoil and JavaProp fame, legend and song?
My hat is off!
MDL
Il giorno lunedì 27 marzo 2023 alle 15:32:06 UTC-4 Mark Lougheed ha scritto:And I failed to attach: here it is - save as fcheck.hex then "load fcheck" to make new fcheck.com
Hi all, finally the problem is solved...The member of these two forums, Martin Hepperle, found a brilliant and perfectly working solution, here are the two links to the solution...Pardon me for the brief diversion...
https://forum.vcfed.org/index.php?threads/strange-compiler-error-prospero-fortran-cp-m-80.1241152/
https://forum.classic-computing.de/forum/index.php?thread/28785-strange-compiler-error-prospero-fortran-cp-m-80/&postID=406897#post407059
Greetings to all...
Martin Hepperle? THE Martin Hepperle of JavaFoil and JavaProp fame, legend and song?
My hat is off!
MDLI don't know him personally, but I can tell you that he is a very knowledgeable and professional person in machine language programming and high level languages...
Il giorno lunedì 27 marzo 2023 alle 15:32:06 UTC-4 Mark Lougheed ha scritto:With new new profor2.com, fcheck.com fails. Find attached a new fcheck.com that takes the new profor2.com
Hi all, finally the problem is solved...The member of these two forums, Martin Hepperle, found a brilliant and perfectly working solution, here are the two links to the solution...Pardon me for the brief diversion...
https://forum.vcfed.org/index.php?threads/strange-compiler-error-prospero-fortran-cp-m-80.1241152/
https://forum.classic-computing.de/forum/index.php?thread/28785-strange-compiler-error-prospero-fortran-cp-m-80/&postID=406897#post407059
Greetings to all...
Martin Hepperle? THE Martin Hepperle of JavaFoil and JavaProp fame, legend and song?
My hat is off!
MDLI don't know him personally, but I can tell you that he is a very knowledgeable and professional person in machine language programming and high level languages...
And as another note - the prospero fortran for cp/m is NOT fortran 77 -- it is fortran 66... pretty much the same as the microsoft, but with complex. Things like SAVE, PARAMETER, CHARACTER and BLOCK IF do not exist. Also, assigned goto requires astatement list (contents not important, needs the list). I am going to play with this compiler.
It uses z80 instructions... JR for one.
четверг, 30 марта 2023 г. в 01:44:53 UTC+3, fridtjof.ma...@gmail.com:statement list (contents not important, needs the list). I am going to play with this compiler.
And as another note - the prospero fortran for cp/m is NOT fortran 77 -- it is fortran 66... pretty much the same as the microsoft, but with complex. Things like SAVE, PARAMETER, CHARACTER and BLOCK IF do not exist. Also, assigned goto requires a
It uses z80 instructions... JR for one.Several files are missing for AMSTRAD. If you add the following files from the distribution for CP/M:
GLIB.MAC,
G2LIB.MAC,
PRIME.FOR,
COPYTX.FOR,
MAZE.FOR
then the original file FCHECK.COM checks all files and reports that they are normal. This suggests that these files are no different and can be used.
New file fcheck.com it is smaller in size than the original one and gives the following information at startup:
PROFOR2.COM - checks OK
PROFOR.ERR - does not compute
Stored values: 44 61509876
Calculated values: 45 62205943
FORLIB.REL - does not compute
Stored values: 332 492497947
Calculated values: 340 507591874
FCONFIG.COM - does not compute
Stored values: 56 93958051
Calculated values: 56 94334026
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
On Wednesday, March 29, 2023 at 6:52:06 PM UTC-4, Андрей Никитин wrote:statement list (contents not important, needs the list). I am going to play with this compiler.
четверг, 30 марта 2023 г. в 01:44:53 UTC+3, fridtjof.ma...@gmail.com:
And as another note - the prospero fortran for cp/m is NOT fortran 77 -- it is fortran 66... pretty much the same as the microsoft, but with complex. Things like SAVE, PARAMETER, CHARACTER and BLOCK IF do not exist. Also, assigned goto requires a
Perfect!It uses z80 instructions... JR for one.Several files are missing for AMSTRAD. If you add the following files from the distribution for CP/M:
GLIB.MAC,
G2LIB.MAC,
PRIME.FOR,
COPYTX.FOR,
MAZE.FOR
then the original file FCHECK.COM checks all files and reports that they are normal. This suggests that these files are no different and can be used.
New file fcheck.com it is smaller in size than the original one and gives the following information at startup:
PROFOR2.COM - checks OK
PROFOR.ERR - does not compute
Stored values: 44 61509876
Calculated values: 45 62205943
FORLIB.REL - does not compute
Stored values: 332 492497947
Calculated values: 340 507591874
FCONFIG.COM - does not compute
Stored values: 56 93958051
Calculated values: 56 94334026
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OKInteresting... here is what I get: I'll bundle up all the files I have... (into an lu library). Note that fcheck has size and checksum of each file within it. The issue then would be the actual compiler files...
: fred@llana p $; cpm fcheck.com
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
and, for compiler version:
: fred@llana p $; cpm profor
Pro Fortran Compiler - Version zz 1.2
Copyright (C) 1983 Prospero Software
Serial No: 63
Source filename -
Il giorno mercoledì 29 marzo 2023 alle 21:11:59 UTC-4 fridtjof.ma...@gmail.com ha scritto:statement list (contents not important, needs the list). I am going to play with this compiler.
On Wednesday, March 29, 2023 at 6:52:06 PM UTC-4, Андрей Никитин wrote:
четверг, 30 марта 2023 г. в 01:44:53 UTC+3, fridtjof.ma...@gmail.com:
And as another note - the prospero fortran for cp/m is NOT fortran 77 -- it is fortran 66... pretty much the same as the microsoft, but with complex. Things like SAVE, PARAMETER, CHARACTER and BLOCK IF do not exist. Also, assigned goto requires a
My corrected Prospero Fortran is atIt uses z80 instructions... JR for one.Several files are missing for AMSTRAD. If you add the following files from the distribution for CP/M:
GLIB.MAC,
G2LIB.MAC,
PRIME.FOR,
COPYTX.FOR,
MAZE.FOR
then the original file FCHECK.COM checks all files and reports that they are normal. This suggests that these files are no different and can be used.
New file fcheck.com it is smaller in size than the original one and gives the following information at startup:
PROFOR2.COM - checks OK
PROFOR.ERR - does not compute
Stored values: 44 61509876
Calculated values: 45 62205943
FORLIB.REL - does not compute
Stored values: 332 492497947
Calculated values: 340 507591874
FCONFIG.COM - does not compute
Stored values: 56 93958051
Calculated values: 56 94334026
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OKInteresting... here is what I get: I'll bundle up all the files I have... (into an lu library). Note that fcheck has size and checksum of each file within it. The issue then would be the actual compiler files...
: fred@llana p $; cpm fcheck.com
PROFOR2.COM - checks OK
PROFOR.ERR - checks OK
FORLIB.REL - checks OK
FCONFIG.COM - checks OK
G1LIB.MAC - checks OK
G2LIB.MAC - checks OK
PRIME.FOR - checks OK
COPYTX.FOR - checks OK
MAZE.FOR - checks OK
and, for compiler version:
: fred@llana p $; cpm profor
Pro Fortran Compiler - Version zz 1.2
Copyright (C) 1983 Prospero Software
Serial No: 63
Source filename -Perfect!
Thank's
Updated versions of STRING library and APU library (if I do that) will be put into the profor github repository.PROFOR FILE.FOR /OPTIONS
On Friday, March 31, 2023 at 9:53:55 AM UTC-4, fridtjof.ma...@gmail.com wrote:Perfect "autopsy" of the Prospero Fortran CP/M-80 version!... For CP/M-80 O.S. I believe it is the most complete among the various Fortran compilers, I played with all the other versions that I could find on the internet and the only two that support the
Updated versions of STRING library and APU library (if I do that) will be put into the profor github repository.PROFOR FILE.FOR /OPTIONS
Note space before the /
OPTIONS can be:
L List (PRN file)
M Map (MAP file)
C Compact (less space)
R Restrict (not validated. I think this does INTEGER*1, *2, *4 to just the specified width)
N Source line numbers (for debugging, takes extra code space)
I Subscript range checking
A Assignment range checking
H Hold before REL (why? maybe to allow disk switch)
S ANSI only. Eg. 'ABCD' instead of 4HABCD generates warning.
G Console to log (LOG file)
Note that ASSIGNed GOTO needs the statement list... Even a dummy list works. Eg
ASSIGN 10 TO I
GO TO I,(1)
1 PAUSE 4321
STOP
10 PAUSE 1234
STOP
does work. The list does not appear to be checked... It just needs to be there.
ANSI FORTRAN 66 code will go easier with Prospero than with Microsoft F80. The biggest issue is that
Microsoft uses INTEGER*2 as default and REAL*4 as default, so INTEGER and REAL do not
EQUIVALENCE according to the standard. Prospero uses INTEGER*4 as INTEGER and REAL*4
as REAL, so things "work out properly". And... Prospero supports COMPLEX (as COMPLEX*8, note
that DOUBLE PRECISION COMPLEX - COMPLEX*16 isn't there).
I want to get my virtc working with Prospero, to allow micro-timing benchmarks to compare PROFOR
and F80.
prolink prime
Hi,Kurt
unfortunately I can't report any success.
My knowledge of Fortran is so-so.
The program is processed with profor prime.for.
But Prolink needs to be called and I get this
Messages:
prolink prime
Pro Pascal Linker - Version zz 1.6
Copyright (C) 1982 Prospero Software
Serial No: 63
Linking:
PRIME.REL
Unsatisfied external(s):
$EFMT $EIRF $EIWF $ERFI4 $ETRF $ETWF
$EUNIT $EWFI2 $EWFI4 $GINIT $GIRST $HINVS
$LCMPR $LMODF $RADD $RFLT $TRUNC SQRT
Terminate ? (Y/N)
Hardware is SC114 von S.Cousins.
Kurt
type makfor.subprofor p:$1
Hi,
thanks to the specialist Fred.
I created a
makfor.sub.
type makfor.subprofor p:$1
l80 P:$1,forlib/s,p:$1/n/e
dir p:$1.*
It works.HI,
Before that I had worked with F80.
An attempt failed on KSAM80.
The KSAM80.REL etc for BASCOM don't work.
Prg crashes after CALL MKSAM .....
There is an F80 Fortran version. Not to be found.
Kurt
| Sysop: | Keyop |
|---|---|
| Location: | Huddersfield, West Yorkshire, UK |
| Users: | 496 |
| Nodes: | 16 (2 / 14) |
| Uptime: | 63:32:54 |
| Calls: | 9,762 |
| Calls today: | 3 |
| Files: | 13,744 |
| Messages: | 6,185,694 |