Open anyonecan opened 2 years ago
tschoonj
commented
2 years ago Hi John,
Unfortunately due to time constraints, I am unable to continue the development of XMI-MSIM, and do not expect to be able to investigate this bug anytime soon.
My recommendation is to run XMI-MSIM on a Linux system, where I expect that things will run smoothly.
Thanks,
Tom
anyonecan
commented
2 years ago Hi Tom,
Thanks for the update. I understand your situation. I will look into using Linux, perhaps as a virtual machine on the Windows 11 box. I will post the results on the Github issue page.
Best regards,
John
On Wed, Mar 30, 2022 at 1:31 PM Tom Schoonjans @.***> wrote:
Hi John,
Unfortunately due to time constraints, I am unable to continue the development of XMI-MSIM, and do not expect to be able to investigate this bug anytime soon.
My recommendation is to run XMI-MSIM on a Linux system, where I expect that things will run smoothly.
Thanks,
Tom
— Reply to this email directly, view it on GitHub https://github.com/tschoonj/xmimsim/issues/86#issuecomment-1083478684, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACJWJPZUQ6FONKQLZ52JMBLVCSMYTANCNFSM5SAOXXNQ . You are receiving this because you authored the thread.Message ID: @.***>
Hello,
I have been using PyMca for a while and saw some references to a Monte Carlo plugin. I never got to work with that, but recently have been using XMI-MSIM in a stand-alone manner. It has been very interesting and useful in confirming and explaining some multilayer sample results.
There have been a few crashes that seem to be due to bad inputs or maybe having two instances of the program running, but there is a class of crashes that I can't figure out. I am running version 8.1 on three different computers:
i5 laptop with 12 GB of memory, Windows 10 i5 desktop with 8 GB of memory, Windows 10 i7 desktop with 32 GB of memory, Windows 11
If I run the simulation several times, either singly or in a batch, I will eventually get an access error like this:
00:00:00 Exception code=0xc0000005 flags=0x0 at 0x0000000066E8AF20. Access violation - attempting to read data at address 0x0000000006EE80F0 00:00:00 Simulating interactions 00:00:01 C:\Program Files\XMI-MSIM 64-bit\Bin\xmimsim.exe with process id 5720 exited with an error (code: -1073741819)
This is from the i7, the data address has been in the 0x6xxxxxx range for this set. Once it happens, it persists for quite some time. At first it seemed that resetting the PC fixed it, but just now, it fails on the first run after a power cycle.
I have just noticed that if "Simulate the non-radiative cascade effect" and "Enable variance reduction techniques" are not both selected, the simulation will run to completion.
I tried the newest nightly build and it would not run due to missing "libcsironn.dll".
I hope you can give me some suggestions to prevent this; I have attached some configuration files. I have been consistently using these tube and capillary files, but the error has occurred with this and several other .xmsi files.
Please let me know if you need more info. John
XMI-access error.zip