ZLUDA Radeon Benchmarks

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2402096-NE-ZLUDA405505.

ZLUDA Radeon BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionRTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7700 XTRX 7900 XTRX 7900 XTXAMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS)AMD Device 14d82 x 16GB DRAM-6000MT/s G Skill F5-6000J3038F16G2000GB Samsung SSD 980 PRO 2TBNVIDIA GeForce RTX 3080 10GBNVIDIA GA102 HD AudioDELL U2723QEIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 22.046.2.0-26-generic (x86_64)GNOME Shell 42.9X Server 1.21.1.4NVIDIA 550.40.074.6.0OpenCL 3.0 CUDA 12.4.741.3.271GCC 11.4.0ext43840x2160NVIDIA GeForce RTX 4070 SUPER 12GBNVIDIA Device 22bcNVIDIA GeForce RTX 4080 16GBNVIDIA Device 22bbAMD Radeon RX 6800 16GB (2475/1000MHz)AMD Navi 21 HDMI AudioX Server 1.21.1.4 + Wayland4.6 Mesa 23.0.4-0ubuntu1~22.04.1 (LLVM 15.0.7 DRM 3.54)OpenCL 2.1 AMD-APP (3590.0)1.3.238AMD Radeon RX 6800 XT 16GB (2575/1000MHz)XFX AMD Radeon RX 7700 XT 12GB (2276/1124MHz)AMD Device ab30AMD Radeon RX 7900 XT 20GB (2025/1249MHz)AMD Radeon RX 7900 XTX 24GB (2304/1249MHz)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseProcessor Details- Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa601203Graphics Details- RTX 3080: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 94.02.20.00.07- RTX 4070 SUPER: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.04.69.00.01- RTX 4080: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 95.03.0e.00.04- RX 6800: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120900-101- RX 6800 XT: BAR1 / Visible vRAM Size: 16368 MB - vBIOS Version: 113-D4120500-101- RX 7700 XT: BAR1 / Visible vRAM Size: 12272 MB - vBIOS Version: 113-EXT90040-100- RX 7900 XT: BAR1 / Visible vRAM Size: 20464 MB - vBIOS Version: 113-D70401-00- RX 7900 XTX: BAR1 / Visible vRAM Size: 24560 MB - vBIOS Version: 113-D7020100-102OpenCL Details- RTX 3080: GPU Compute Cores: 8704- RTX 4070 SUPER: GPU Compute Cores: 7168- RTX 4080: GPU Compute Cores: 9728Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

ZLUDA Radeon Benchmarksnamd-cuda: ATPase Simulation - 327,506 Atomsblender: BMW27 - NVIDIA CUDAblender: Classroom - NVIDIA CUDAblender: Fishy Cat - NVIDIA CUDAblender: Barbershop - NVIDIA CUDAblender: Pabellon Barcelona - NVIDIA CUDAblender: BMW27 - Radeon HIPblender: Classroom - Radeon HIPRTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7700 XTRX 7900 XTRX 7900 XTX0.0849812.6825.7124.0597.8759.020.0843510.1719.3820.1278.9343.290.081057.8214.5015.6160.0132.310.1365225.8136.8838.95147.0982.6723.4539.8219.0931.8232.68122.4668.4219.0133.4720.9933.0530.320.0900411.4221.0320.2581.9845.1314.8824.250.0903110.0518.4418.0439.0012.9120.89OpenBenchmarking.org

NAMD CUDA

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsRTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 7900 XTRX 7900 XTX0.03070.06140.09210.12280.1535SE +/- 0.00017, N = 3SE +/- 0.00007, N = 3SE +/- 0.00046, N = 3SE +/- 0.00117, N = 3SE +/- 0.00055, N = 3SE +/- 0.00096, N = 50.084980.084350.081050.136520.090040.09031

Blender

Blend File: BMW27 - Compute: NVIDIA CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: BMW27 - Compute: NVIDIA CUDARTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7700 XTRX 7900 XTRX 7900 XTX612182430SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 3.44, N = 15SE +/- 0.02, N = 3SE +/- 3.60, N = 15SE +/- 0.00, N = 3SE +/- 0.03, N = 312.6810.177.8225.8119.0920.9911.4210.05

Blender

Blend File: Classroom - Compute: NVIDIA CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Classroom - Compute: NVIDIA CUDARTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7700 XTRX 7900 XTRX 7900 XTX816243240SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 325.7119.3814.5036.8831.8233.0521.0318.44

Blender

Blend File: Fishy Cat - Compute: NVIDIA CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Fishy Cat - Compute: NVIDIA CUDARTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7700 XTRX 7900 XTRX 7900 XTX918273645SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 324.0520.1215.6138.9532.6830.3220.2518.04

Blender

Blend File: Barbershop - Compute: NVIDIA CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Barbershop - Compute: NVIDIA CUDARTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7900 XT306090120150SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 297.8778.9360.01147.09122.4681.98

Blender

Blend File: Pabellon Barcelona - Compute: NVIDIA CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Pabellon Barcelona - Compute: NVIDIA CUDARTX 3080RTX 4070 SUPERRTX 4080RX 6800RX 6800 XTRX 7900 XTRX 7900 XTX20406080100SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 359.0243.2932.3182.6768.4245.1339.00

Blender

Blend File: BMW27 - Compute: Radeon HIP

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: BMW27 - Compute: Radeon HIPRX 6800RX 6800 XTRX 7900 XTRX 7900 XTX612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 323.4519.0114.8812.91

Blender

Blend File: Classroom - Compute: Radeon HIP

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0Blend File: Classroom - Compute: Radeon HIPRX 6800RX 6800 XTRX 7900 XTRX 7900 XTX918273645SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 339.8233.4724.2520.89

Phoronix Test Suite v10.8.5