Hi,

Both Xeon Silver 4215R and A6000 are much slower than 12900KF and 4080.
Scenes rendered by Unreal can only use one video card. Two video cards in SLI may even obstruct some of Unreal's processes.

To gain more out from the CPU-s, you could turn off the Hyper Threading ( SMT ) in both of them.
And in the case of 12900KF, you could also turn off the E-Cores (efficiency cores).

I am not sure if having two CPU-s gives any benefits or if it even slows down some processes. These processors and the multi-CPU setups are designed for server use. And not for rendering in real-time with rendering engines like Unreal or Aximmetry's own render engine.

Warmest regards,

Hi,Eifert,

I'd tested unbind SLI GPU, and set the one A6000 in Aximmetry Engine, but performance is bad too,

I also had turned off the SMT in bios, when I try cook and live sync,the difference of frame is 2-3ms

And in UE for Aximmetry, after add aximmetry camera ,the frame rate is 50ms,but when I play with PIE mode or cooked, the frame is 15ms in ax processorload

Can you give me some suggestion？

Best Regards，

Hi,

When you view your project in Unreal, it is rendered in the size of its viewport window which is usually much smaller than the resolution Aximmetry is set to. Also, without starting your project, it may save on several processes. So comparisons like that can be very inaccurate.

Instead, you should compare it to your 12900KF+4080 hardware. What kind of frame rates do you get with 12900KF+4080 with the same resolution in Aximmetry?

Maybe you have some Windows energy-saving settings turned on in your A6000 computer?

Warmest regards,