I’m having an issue using the swarm light builder over the network where the local machine is performing light calculations for up to 4 times as long as the remote machines. On the example below, it took a good 25 minutes longer than the other machines.
Is there something I’m doing which is causing the unequal distribution of lightmass jobs?
Hello TrentC,
I believe this to be expected behavior when building lighting across a network. What essentially occurs is that you are allocating tasks across the network. Once that task is done on that network machine it finishes it’s calculations and sends it back to the local machine. The local machine then compiles all of the calculations that is sent to it and displays the calculations locally.
The local machine will always take the longest because it is the source.
Yeah, that makes sense. It just seemed excessive with this particular level. I’m used to large overnight builds taking an extra few minutes on the local machine, but this one took significantly more for a much smaller bake. Are there any level specific issues that I should be looking into.
Hello TrentC,
You can look into optimization techniques that are posted on our documentation.
Link :
This could help you reduce build time and make it less taxing on all of your machines. If your project is not crashing or causing a freeze then I believe this will be a good start for profiling and optimizing your project.
Cheers for that! Checked the statistics panel for light build info and there was a single mesh with a massive overridden lightmap res that was taking longer than all the others combined. Its always the simple things.
Thanks again!
