Intel Xe2 Lunar Lake Graphics Compute / OpenCL Performance Looking Great

Written by Michael Larabel in Graphics Cards on 26 November 2024 at 10:39 AM EST. Page 3 of 4. 18 Comments.
Blender benchmark with settings of Blend File: BMW27, Compute: Intel oneAPI. Core Ultra 7 256V Xe2 LNL was the fastest.
Blender benchmark with settings of Blend File: Classroom, Compute: Intel oneAPI. Core Ultra 7 256V Xe2 LNL was the fastest.
Blender benchmark with settings of Blend File: Fishy Cat, Compute: Intel oneAPI. Core Ultra 7 155H Xe MTL was the fastest.
Blender benchmark with settings of Blend File: Pabellon Barcelona, Compute: Intel oneAPI. Core Ultra 7 155H Xe MTL was the fastest.

With the Blender 4.3 performance using Cycles' Intel oneAPI backend, the Lunar Lake graphics were slightly faster than Meteor Lake except for some scenes where Meteor Lake managed to be faster. It looks like there may still be some Lunar Lake bugs to work out in the stack or where the higher graphics clock on Meteor Lake played a greater role.

Darktable benchmark with settings of Test: Masskrug, Acceleration: OpenCL. Core Ultra 7 256V Xe2 LNL was the fastest.
Darktable benchmark with settings of Test: Server Room, Acceleration: OpenCL. Core Ultra 7 256V Xe2 LNL was the fastest.
Darktable benchmark with settings of Test: Server Rack, Acceleration: OpenCL. Core Ultra 7 256V Xe2 LNL was the fastest.

There were nice wins for Lunar Lake with the Darktable RAW photography software leveraging OpenCL.

GpuOwl benchmark with settings of Exponent: 57885161. Core Ultra 7 256V Xe2 LNL was the fastest.
GpuOwl benchmark with settings of Exponent: 77936867. Core Ultra 7 256V Xe2 LNL was the fastest.

And some decent wins in GpuOwl for that Mersenne primality tester but not nearly as exciting as the FluidX3D numbers...

Related Articles