im hitting the size limit of what datasets i can process and visualize on my little mac mini with 16gb ram. i tried opening a 10gb dcd trajectory but ChimeraX.app blew up to 24gb of memory usage. im not complaining, i accept that's a reasonable amount of resources to open a trajectory of that size. what im doing now is postprocessing 1 out of 10 (or 1 out of 100 frames) using mdanalysis in python then bringing that small sliced .dcd into chimerax on my home computer for visualization.
However, im trying to render a 2 minute movie for youtube at 60 frames/sec and 4k resolution. This is well beyond the capacity of my mac mini. I would guess im not the only user that's dealing with large datasets generated from simulations on clusters.
often provides unreliable or slow interactive performance when the rendering or display is not on the same computer that is running the application
not a problem when im on campus with a gigabit connection to the cluster(s).
We also do not support running Chimera or ChimeraX on a virtual machine, as configuring graphics drivers to use GPU acceleration on a virtual machine presents technical issues. The ChimeraX development team does not have the computer systems on which to test these unsupported configurations.
im willing to pitch in. i have access to cluster nodes with 256gb memory and four V100 GPUs with 32gb each. what are the technical issues ?
the main issue for me is actually requiring sudo access to install chimerax. ill never get root access to the cluster nodes (nor should i) i install all my stuff in ~/local.
im hitting the size limit of what datasets i can process and visualize on my little mac mini with 16gb ram. i tried opening a 10gb dcd trajectory but ChimeraX.app blew up to 24gb of memory usage. im not complaining, i accept that's a reasonable amount of resources to open a trajectory of that size. what im doing now is postprocessing 1 out of 10 (or 1 out of 100 frames) using mdanalysis in python then bringing that small sliced .dcd into chimerax on my home computer for visualization.
However, im trying to render a 2 minute movie for youtube at 60 frames/sec and 4k resolution. This is well beyond the capacity of my mac mini. I would guess im not the only user that's dealing with large datasets generated from simulations on clusters.
https://www.cgl.ucsf.edu/chimerax/system_requirements.html :
not a problem when im on campus with a gigabit connection to the cluster(s).
im willing to pitch in. i have access to cluster nodes with 256gb memory and four V100 GPUs with 32gb each. what are the technical issues ?
the main issue for me is actually requiring
sudo
access to install chimerax. ill never get root access to the cluster nodes (nor should i) i install all my stuff in ~/local.