everybody，hi！
with GID 7.2 and some other files which are written in C language, it is possible to write the result file in binary format in order to compress the size of result file. I got some gidpost files which are written in C language from GID website. from readme.txt, i was informed that i can get a library file by compiling those resource files and with the library file, i can write the *.flavia.msh and *.flavia.res in binary format. but i cann't compile those resouce file successfully. i don't know if somebody try it or not.
i think it is very meaning to write the two result file in binary format because with ascii format the two files are always very big sometimes, especially you have a lot of results in your result files.
thanks a lot
[GiDlist] how to get the library
Moderator: GiD Team
-
Winfried Schepers
[GiDlist] Memory Limits of GiD 7.2
Dear GiD users,
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
-
Miguel A. de Riera Pasenau
[GiDlist] Memory Limits of GiD 7.2
Have you tried the latest beta:
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD7.3b-win.exe ?
There were some problems with the stack limitation imposed by
Microsoft Visual C++. Which is overcome in this version.
the bug of cutting hexahedras was also solved.
miguel
Winfried Schepers wrote:
Dear GiD users,
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
--
--------------------------------------------------------------------------------
Miguel A. Pasenau de Riera miguel at cimne.upc.es http://gid.cimne.upc.es
ftp://gid.cimne.upc.es/pub/gid_adds/Windows/GiD7.3b-win.exe ?
There were some problems with the stack limitation imposed by
Microsoft Visual C++. Which is overcome in this version.
the bug of cutting hexahedras was also solved.
miguel
Winfried Schepers wrote:
Dear GiD users,
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
--
--------------------------------------------------------------------------------
Miguel A. Pasenau de Riera miguel at cimne.upc.es http://gid.cimne.upc.es
-
Enrique Escolano
[GiDlist] how to get the library
What's your compilation problem?
You compile on Linux/Unix, Windows using MS Visual C++, ...?
You can compile gidpost.lib ?(read gidpost.html to compile)
Yon can compile the sample use testpost.c using this library, to create
testpost.exe?
This library is tested under Linux and Windows.
Enrique Escolano
----- Original Message -----
From: "Zhiwei Niu" nzwyc at sina.com
To: "gid" gidlist at gatxan.cimne.upc.es
Sent: Tuesday, April 29, 2003 3:29 AM
Subject: [GiDlist] how to get the library
everybody，hi！
with GID 7.2 and some other files which are written in C language, it is
possible to write the result file in binary format in order to compress the
size of result file. I got some gidpost files which are written in C
language from GID website. from readme.txt, i was informed that i can get a
library file by compiling those resource files and with the library file, i
can write the *.flavia.msh and *.flavia.res in binary format. but i cann't
compile those resouce file successfully. i don't know if somebody try it or
not.
i think it is very meaning to write the two result file in binary format
because with ascii format the two files are always very big sometimes,
especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
You compile on Linux/Unix, Windows using MS Visual C++, ...?
You can compile gidpost.lib ?(read gidpost.html to compile)
Yon can compile the sample use testpost.c using this library, to create
testpost.exe?
This library is tested under Linux and Windows.
Enrique Escolano
----- Original Message -----
From: "Zhiwei Niu" nzwyc at sina.com
To: "gid" gidlist at gatxan.cimne.upc.es
Sent: Tuesday, April 29, 2003 3:29 AM
Subject: [GiDlist] how to get the library
everybody，hi！
with GID 7.2 and some other files which are written in C language, it is
possible to write the result file in binary format in order to compress the
size of result file. I got some gidpost files which are written in C
language from GID website. from readme.txt, i was informed that i can get a
library file by compiling those resource files and with the library file, i
can write the *.flavia.msh and *.flavia.res in binary format. but i cann't
compile those resouce file successfully. i don't know if somebody try it or
not.
i think it is very meaning to write the two result file in binary format
because with ascii format the two files are always very big sometimes,
especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
-
Jorge Suit Perez Ronda
[GiDlist] how to get the library
Hi, have you compile gidpost? if you have not, wich are the problem you
have in the compilation?
Regarding your other comment, mesh (.msh) file should be written in
ascii format if you want to write it separately from the result file
(.res). You can include the mesh into the result file and take
adavantage of the binary format.
Zhiwei Niu wrote:
everybody，hi！
with GID 7.2 and some other files which are written in C language, it is possible to write the result file in binary format in order to compress the size of result file. I got some gidpost files which are written in C language from GID website. from readme.txt, i was informed that i can get a library file by compiling those resource files and with the library file, i can write the *.flavia.msh and *.flavia.res in binary format. but i cann't compile those resouce file successfully. i don't know if somebody try it or not.
i think it is very meaning to write the two result file in binary format because with ascii format the two files are always very big sometimes, especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
have in the compilation?
Regarding your other comment, mesh (.msh) file should be written in
ascii format if you want to write it separately from the result file
(.res). You can include the mesh into the result file and take
adavantage of the binary format.
Zhiwei Niu wrote:
everybody，hi！
with GID 7.2 and some other files which are written in C language, it is possible to write the result file in binary format in order to compress the size of result file. I got some gidpost files which are written in C language from GID website. from readme.txt, i was informed that i can get a library file by compiling those resource files and with the library file, i can write the *.flavia.msh and *.flavia.res in binary format. but i cann't compile those resouce file successfully. i don't know if somebody try it or not.
i think it is very meaning to write the two result file in binary format because with ascii format the two files are always very big sometimes, especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
-
Miguel A. de Riera Pasenau
[GiDlist] how to get the library
Which platform are you using?
Linux? windows?
Which are the errors that appears when you try
to compile the files?
miguel
Zhiwei Niu wrote:
everybody£¬hi£¡
with GID 7.2 and some other files which are written in C language, it is possible to write the result file in binary format in order to compress the size of result file. I got some gidpost files which are written in C language from GID website. from readme.txt, i was informed that i can get a library file by compiling those resource files and with the library file, i can write the *.flavia.msh and *.flavia.res in binary format. but i cann't compile those resouce file successfully. i don't know if somebody try it or not.
i think it is very meaning to write the two result file in binary format because with ascii format the two files are always very big sometimes, especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
--
--------------------------------------------------------------------------------
Miguel A. Pasenau de Riera miguel at cimne.upc.es http://gid.cimne.upc.es
Linux? windows?
Which are the errors that appears when you try
to compile the files?
miguel
Zhiwei Niu wrote:
everybody£¬hi£¡
with GID 7.2 and some other files which are written in C language, it is possible to write the result file in binary format in order to compress the size of result file. I got some gidpost files which are written in C language from GID website. from readme.txt, i was informed that i can get a library file by compiling those resource files and with the library file, i can write the *.flavia.msh and *.flavia.res in binary format. but i cann't compile those resouce file successfully. i don't know if somebody try it or not.
i think it is very meaning to write the two result file in binary format because with ascii format the two files are always very big sometimes, especially you have a lot of results in your result files.
thanks a lot
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
--
--------------------------------------------------------------------------------
Miguel A. Pasenau de Riera miguel at cimne.upc.es http://gid.cimne.upc.es
-
Enrique Escolano
[GiDlist] Memory Limits of GiD 7.2
GiD , in general, not have static memory limits.
It seems a physical memory limitation: what amount of RAM exists in your
machine?
Mour mesh is small, but exists too many time steps.
If your write only displacements (3 components) and stresses (6 components)
then GiD also store the vector modulus (1 component), and the eigenvalues (3
components) and eigenvectors (9 components) of your stress tensor.
For 4599 nodes, 2048 time steps, and stored as single precision (real*4
bytes), this is:
4599*2048*(3+6+1+3+9)*4=828850176 bytes (800 MB)
If your system not have 1GB or more of RAM, this overflow your physical
memory.
It is really needed to write too many time steps?
Enrique Escolano
----- Original Message -----
From: "Winfried Schepers" winfried.schepers at tu-berlin.de
To: gidlist at gatxan.cimne.upc.es
Sent: Sunday, April 27, 2003 7:38 PM
Subject: [GiDlist] Memory Limits of GiD 7.2
Dear GiD users,
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
It seems a physical memory limitation: what amount of RAM exists in your
machine?
Mour mesh is small, but exists too many time steps.
If your write only displacements (3 components) and stresses (6 components)
then GiD also store the vector modulus (1 component), and the eigenvalues (3
components) and eigenvectors (9 components) of your stress tensor.
For 4599 nodes, 2048 time steps, and stored as single precision (real*4
bytes), this is:
4599*2048*(3+6+1+3+9)*4=828850176 bytes (800 MB)
If your system not have 1GB or more of RAM, this overflow your physical
memory.
It is really needed to write too many time steps?
Enrique Escolano
----- Original Message -----
From: "Winfried Schepers" winfried.schepers at tu-berlin.de
To: gidlist at gatxan.cimne.upc.es
Sent: Sunday, April 27, 2003 7:38 PM
Subject: [GiDlist] Memory Limits of GiD 7.2
Dear GiD users,
are there any memory limits in GiD?
I have a mesh of 4599 nodes, 24 beams, and 3576 hexahedras. For
postprocessing I have a post.res file with 2048 time steps of
displacements and stresses.
First of all, I had to change the stack size of GiD.exe using EDITBIN,
otherwise it wasn't possible to open a file that large. However, there
seems to be a limit in GiD, since GiD crashes at some point before all
the time steps are read from the file. From the debugger I get the error
message "unhandled exception in gid.exe: access violation", which
usually means some array boundary violations of the program. I could
figure out that it happens when GiD allocated approximately 425 MB of
memory, even if I reduce the number of time steps or the amount of
postprocessing data.
What are these limits?
There is another strange behaviour of GiD: I want to cut volume sets of
the mesh above. GiD asks for two points, and then which side I want.
After pointing on the side GiD _always_ crashes with the same error
message from the debugger as above. This happens even when I haven't
read any post data, but only the mesh.
Does anyone have any ideas? I could send my data if requested.
I'm using GiD 7.2. I got this error on a Intel P4 as well as on an
Athlon 1000+, both running W2K.
Best regards
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
_______________________________________________
GiDlist mailing list
GiDlist at gid.cimne.upc.es
http://gid.cimne.upc.es/mailman/listinfo/gidlist
-
Winfried Schepers
[GiDlist] Memory Limits of GiD 7.2
Dear Enrique, dear Miguel,
indeed, the 7.3beta solved all the problems. Thanks for your help.
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
Enrique Escolano wrote:
GiD , in general, not have static memory limits.
It seems a physical memory limitation: what amount of RAM exists in your
machine?
Mour mesh is small, but exists too many time steps.
If your write only displacements (3 components) and stresses (6 components)
then GiD also store the vector modulus (1 component), and the eigenvalues (3
components) and eigenvectors (9 components) of your stress tensor.
For 4599 nodes, 2048 time steps, and stored as single precision (real*4
bytes), this is:
4599*2048*(3+6+1+3+9)*4=828850176 bytes (800 MB)
If your system not have 1GB or more of RAM, this overflow your physical
memory.
It is really needed to write too many time steps?
Enrique Escolano
indeed, the 7.3beta solved all the problems. Thanks for your help.
Winfried SCHEPERS
Technical University of Berlin
Soil Mechanics and Geotechnical Engineering Division
Enrique Escolano wrote:
GiD , in general, not have static memory limits.
It seems a physical memory limitation: what amount of RAM exists in your
machine?
Mour mesh is small, but exists too many time steps.
If your write only displacements (3 components) and stresses (6 components)
then GiD also store the vector modulus (1 component), and the eigenvalues (3
components) and eigenvectors (9 components) of your stress tensor.
For 4599 nodes, 2048 time steps, and stored as single precision (real*4
bytes), this is:
4599*2048*(3+6+1+3+9)*4=828850176 bytes (800 MB)
If your system not have 1GB or more of RAM, this overflow your physical
memory.
It is really needed to write too many time steps?
Enrique Escolano