Enscape software bottleneck?

REMINDER! If you encounter any issues with Enscape (e.g. crashes, installation problems) or your subscription please reach out to our dedicated support team directly through the Help Center or by using the Support button as detailed HERE! Thank you for your understanding.
  • Hello,


    Our office purchased a new machine through Puget Systems with a brand new RTX 3090 powered GPU. I wanted to put the card through a little benchmark and had attempted to render the same video clip with the same project on both the new machine and the old machine. The results were positive as the new machine was able to render the the same 32 second clip at 1080p/120fps 4x faster than the old machine. Below are the statistics between the two computers.


    Enscape Version: 3.0.0+39546

    Host Application: Revit 2020.2.4 Build: 20210420_1515(x64)


    OLD PC NEW PC
    CPU Intel Core i7-4820K (4 cores 8 threads) AMD Ryzen 9 5850X (16 cores 32 threads)
    Memory 32GB DDR3 1333MHz 64 GB GDDR4 3200MHz
    GPU Nvidia GTX 1080 Nvidia RTX 3090
    Render Time 27m:6.30s 6m:37.20s
    CPU Load 16-18% 3-7%
    GPU Average Load 96% 25%
    GPU Average Temperature (celcius) 80 65


    While the two machines were rendering the clip, I was monitoring the hardware and noticed something strange with the new pc. While it was able to render the clip 4 times faster, the GPU load with the RTX 3090 was not going above 25%. This can't be an issue due to thermal throttling since it was also running considerably cooler than the GTX 1080. I'm wondering if there might be a bottleneck with Enscape where it is not taking full advantage of the hardware resources available when rendering a video file? Would it be possible to squeeze any more performance out of the hardware?


    Any insight on this would be greatly appreciated!


    Thanks,

    Vadim

  • The 3090 is an RTX card, so that part of what the card can do can be handed over to the dedicated hardware it has.


    Whereas on the 1080, the main GPU also has to make these calculations.

    That would explain some of the difference in load I imagine

    • Official Post

    Vadim Maltsev , thanks a lot for your post - since the Render Time is substantially improved already, there should/will also be no further steps required to improve the performance further. You can check what happens when you export a video in 4k with 120FPS on Maximum quality, for example, the GPU should experience a bit more load. Also, may I ask how you're measuring the GPU load exactly? :)