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 4 of 4. 18 Comments.
clpeak benchmark with settings of OpenCL Test: Global Memory Bandwidth. Core Ultra 7 256V Xe2 LNL was the fastest.
clpeak benchmark with settings of OpenCL Test: Double-Precision Compute. Core Ultra 7 256V Xe2 LNL was the fastest.
clpeak benchmark with settings of OpenCL Test: Transfer Bandwidth enqueueWriteBuffer. Core Ultra 7 256V Xe2 LNL was the fastest.
clpeak benchmark with settings of OpenCL Test: Transfer Bandwidth enqueueReadBuffer. Core Ultra 7 256V Xe2 LNL was the fastest.

The clpeak OpenCL benchmark was confirming some of the big speed-ups seen for FP64 / double precision compute as well as memory bandwidth intensive scenarios that were seen in some of the other benchmarks.

SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: Bus Speed Download. Core Ultra 7 256V Xe2 LNL was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: Bus Speed Readback. Core Ultra 7 256V Xe2 LNL was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: FFT SP. Core Ultra 7 256V Xe2 LNL was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: GEMM SGEMM_N. Core Ultra 7 256V Xe2 LNL was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: Reduction. Core Ultra 7 256V Xe2 LNL was the fastest.
SHOC Scalable HeterOgeneous Computing benchmark with settings of Target: OpenCL, Benchmark: S3D. Core Ultra 7 256V Xe2 LNL was the fastest.

Across the various SHOC compute benchmarks there were nice generational gains shown as well for the Xe2 Lunar Lake graphics on this latest upstream open-source driver stack.

ViennaCL benchmark with settings of Test: OpenCL BLAS, sCOPY. Core Ultra 7 256V Xe2 LNL was the fastest.
ViennaCL benchmark with settings of Test: OpenCL BLAS, sAXPY. Core Ultra 7 256V Xe2 LNL was the fastest.
ViennaCL benchmark with settings of Test: OpenCL BLAS, sDOT. Core Ultra 7 256V Xe2 LNL was the fastest.
ViennaCL benchmark with settings of Test: OpenCL BLAS, dCOPY. Core Ultra 7 256V Xe2 LNL was the fastest.
ViennaCL benchmark with settings of Test: OpenCL BLAS, dAXPY. Core Ultra 7 256V Xe2 LNL was the fastest.
ViennaCL benchmark with settings of Test: OpenCL BLAS, dGEMV-T. Core Ultra 7 256V Xe2 LNL was the fastest.

Some very nice wins for Lunar Lake with ViennaCL.

VkResample benchmark with settings of Upscale: 2x, Precision: Single. Core Ultra 7 256V Xe2 LNL was the fastest.
vkpeak benchmark with settings of fp64-vec4. Core Ultra 7 256V Xe2 LNL was the fastest.
vkpeak benchmark with settings of int16-scalar. Core Ultra 7 256V Xe2 LNL was the fastest.

Aside from OpenCL and the Intel Compute Runtime stack, with Vulkan compute benchmarks using the Mesq 25.0-devel stack with the ANV Vulkan driver were some very nice improvements too compared to prior generation Meteor Lake.

Geometric Mean Of All Test Results benchmark with settings of Result Composite, Intel GPU Compute Lunar Lake vs. Meteor Lake. Core Ultra 7 256V Xe2 LNL was the fastest.
CPU Power Consumption Monitoring Overview benchmark with settings of Accumulated CPU Power Consumption Monitoring.

Now that some of the early Linux issues with the Intel Core Ultra 200V have been addressed like the ASUS Zenbook platform performance issue and Intel Compute Runtime fixes and an annoying keyboard/touchpad issue, the Core Ultra 7 256V experience with the ASUS Zenbook S 14 is going much more smoothly now... And the Xe2 graphics with Lunar Lake beginning to show off quite nicely. Gaming and a fresh comparison to AMD RDNA3.5 graphics are up next now having the initial Lunar Lake OpenCL exploration out of the way.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.

Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.