Hiya, I'm wondering if there is any information or documentation on what effects performance when running Aximmetry and Unreal Engine.
I notice that my scene runs significantly differently if Aximmetry is also running (on the same machine) For instance, I can be getting 60fps at around 18ms per frame in a scene without Aximmetry running but when running through Aximmetry this drops to 25fps at 40ms, less than half the performance
I know the framerate is defined by Aximmetry but looking at my frame buffer and the GPU vizualizer I can't see a clear source of this extra frame time.
I'm planning on offloading the rendering to a second machine so we can isolate the server and compositing tasks of the Aximmetry machine but I want dig into it a bit first to make sure it's worthwhile doing so.
Are there any Cvars relating to Aximmetry that I could look at?
Cheers. Cooksey.
If you’re comparing performance when renderimg in Aximmetry and the UE5 Editor, first make sure you’re rendering at the same resolution in the editor. The UE5 Editor will by default and automatically lower the Screen Percentage, meaning the scene is rendered at a much lower resolution. Of course also make sure to play the scene in Immersive mode with a screen matching the resolution of your Aximmetry rendering settings.
I’ve found that there’s very little performance difference with just rendering the scene in Aximmetry if I’ve taken care to make sure it uses the exact same settings. Of course, rendering out to a Decklink card has a performance impact, especially in 4K (just as it does in Unreal). A 4K video input also has an impact and then you of course have the keyer taking up some GPU power as well as the light wrap, color correction, billboard rendering and so on.