intel_opencl_bench_basis Intel Xeon D-2796TE testing with a Kontron COMh-sdID E2 v1.0.0 (SDIDR112 BIOS) and NVIDIA AD104GL [RTX 4000 SFF Ada Generation] 20GB on Debian 12 via the Phoronix Test Suite. intel_opencl_bench_basis: Processor: Intel Xeon D-2796TE @ 3.10GHz (20 Cores / 40 Threads), Motherboard: Kontron COMh-sdID E2 v1.0.0 (SDIDR112 BIOS), Chipset: Intel Ice Lake IEH, Memory: 4 x 64GB DDR4-2933MT/s, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: NVIDIA AD104GL [RTX 4000 SFF Ada Generation] 20GB, Audio: NVIDIA AD104 HD Audio, Network: Intel I226-IT + 4 x Intel E823-C for backplane OS: Debian 12, Kernel: 6.1.0-26-amd64 (x86_64), Display Server: X Server, Display Driver: NVIDIA, OpenCL: OpenCL 3.0 CUDA 12.7.33, Compiler: GCC 12.2.0 + CUDA 12.6, File-System: btrfs ProjectPhysX OpenCL-Benchmark 1.2 Operation: Memory Bandwidth Coalesced Read GB/s > Higher Is Better intel_opencl_bench_basis . 260.87 |============================================ ProjectPhysX OpenCL-Benchmark 1.2 Operation: Memory Bandwidth Coalesced Write GB/s > Higher Is Better intel_opencl_bench_basis . 266.15 |============================================ ProjectPhysX OpenCL-Benchmark 1.2 Operation: FP64 Compute TFLOPs/s > Higher Is Better intel_opencl_bench_basis . 0.435 |============================================= ProjectPhysX OpenCL-Benchmark 1.2 Operation: FP32 Compute TFLOPs/s > Higher Is Better intel_opencl_bench_basis . 25.62 |============================================= ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT64 Compute TIOPs/s > Higher Is Better intel_opencl_bench_basis . 2.387 |============================================= ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT32 Compute TIOPs/s > Higher Is Better intel_opencl_bench_basis . 10.06 |============================================= ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT16 Compute TIOPs/s > Higher Is Better intel_opencl_bench_basis . 8.784 |============================================= ProjectPhysX OpenCL-Benchmark 1.2 Operation: INT8 Compute TIOPs/s > Higher Is Better intel_opencl_bench_basis . 6.657 |=============================================