The project I'm working on now basically takes an image and then displays the video using a blender from the command line. I am currently using Twisted to handle requests, but there is certainly something that I am doing wrong, as it does not work. I would like to. You can see the jist program here (I stripped nothing more).
Blender processing is done by creating a subprocess (I know that Twisted can process processes ) along with a python script to adjust the rendering and use the image provided as a texture.
The program should be able to handle as many connections as possible. At the moment, the subprocess does one render at a time, but ideally it will check the CPU / number of parallel renderings and adjust the number to the optimum. Each render is custom for the user, therefore, as soon as the custom rendering is complete, he should return his visualization (avi file).
My question is: is the right choice right? Are there any other options? If not, is my system implementation erroneous? I would appreciate any thoughts or opinions on this matter.
source
share