when i try to load a .xtc file from a 100ns MD my vmd crashes in the frame 9000, does anyone know how could i resolve this problem??
i am using vmd version 1.9.3 for linux, and my laptop has 8gb of ram
thanks beforehand
Without further knowledge I would suspect that you are running out of RAM. How many atoms are there in the system? 9000 frames with 74k atoms would use all your RAM, leaving nothing for the operating system, other programs and the rest of VMD etc.
Why do you need as much as 9000 frames? I would recommend gmx trjconv -timestep 100. That should give you 1000 frames, which, in most cases, would be enough for a 100 ns simulation.
Another solution would be to get more RAM for you laptop or another computer.
Thank you for your reply. My system has 5650 atoms, and I have checked the resources on my laptop. The RAM isn’t the issue; however, I notice that the program still crashes when two of my eight CPUs reach their maximum performance. Interestingly, the overall CPU usage isn’t very high during this process. If you have any suggestions to resolve this issue, I would greatly appreciate it.
Hello, I don’t know if this helps but I have also had this issue and I believe it’s a VMD problem. My laptop has 128 gb RAM and 16 cores (32 threads) so I’m 100% certain its not hardware use. Instead, VMD crashes reliably at a certain frame count (I have four similar systems at 1000 frames each - .5 ns frames, and any subset of them crashes after around 800-1000 frames) which also gives credence to it being a VMD issue. Furthermore I’m running these in NAMD, so it’s rather tangential to GROMACS regardless. I will reply here again if I can find a fix, though for right now, it may be best to wait for 2.0.0 to come out of alpha and then see if they’ve resolved the issue as my best guess is it’s an overflow on some counter.