AMD P-State Linux 6.3 Testing Benchmarks by Michael Larabel for a future article. amd_pstate_epp powersave balance_performance: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate_epp performance balance_performance: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate_epp powersave power: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate_epp performance performance: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate schedutil: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate performance: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate powersave: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 amd_pstate ondemand: Processor: 2 x AMD EPYC 7773X 64-Core @ 3.53GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 acpi_cpufreq schedutil: Processor: 2 x AMD EPYC 7773X 64-Core @ 2.20GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 acpi_cpufreq ondemand: Processor: 2 x AMD EPYC 7773X 64-Core @ 2.20GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 acpi_cpufreq performance: Processor: 2 x AMD EPYC 7773X 64-Core @ 2.20GHz (128 Cores / 256 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 512GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710 OS: Ubuntu 22.04, Kernel: 6.2.0-phx (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080 TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: GoogLeNet images/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 38.41 |====================== amd_pstate_epp performance balance_performance . 40.19 |======================= amd_pstate_epp powersave power ................. 31.64 |================== amd_pstate_epp performance performance ......... 40.39 |======================= amd_pstate schedutil ........................... 26.63 |=============== amd_pstate performance ......................... 40.89 |======================= amd_pstate powersave ........................... 8.74 |===== amd_pstate ondemand ............................ 37.03 |===================== acpi_cpufreq schedutil ......................... 39.24 |====================== acpi_cpufreq ondemand .......................... 40.00 |====================== acpi_cpufreq performance ....................... 40.42 |======================= OpenVKL 1.3.1 Benchmark: vklBenchmark Scalar Items / Sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 364 |======================== amd_pstate_epp performance balance_performance . 374 |======================== amd_pstate_epp powersave power ................. 346 |====================== amd_pstate_epp performance performance ......... 385 |========================= amd_pstate schedutil ........................... 327 |===================== amd_pstate performance ......................... 374 |======================== amd_pstate powersave ........................... 72 |===== amd_pstate ondemand ............................ 363 |======================== acpi_cpufreq schedutil ......................... 376 |======================== acpi_cpufreq ondemand .......................... 357 |======================= acpi_cpufreq performance ....................... 367 |======================== OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC Items / Sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 514 |======================== amd_pstate_epp performance balance_performance . 516 |======================== amd_pstate_epp powersave power ................. 477 |====================== amd_pstate_epp performance performance ......... 512 |======================== amd_pstate schedutil ........................... 468 |====================== amd_pstate performance ......................... 534 |========================= amd_pstate powersave ........................... 102 |===== amd_pstate ondemand ............................ 514 |======================== acpi_cpufreq schedutil ......................... 531 |========================= acpi_cpufreq ondemand .......................... 524 |========================= acpi_cpufreq performance ....................... 524 |========================= TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: AlexNet images/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 87.75 |===================== amd_pstate_epp performance balance_performance . 97.12 |======================= amd_pstate_epp powersave power ................. 65.63 |=============== amd_pstate_epp performance performance ......... 96.73 |======================= amd_pstate schedutil ........................... 36.50 |========= amd_pstate performance ......................... 97.50 |======================= amd_pstate powersave ........................... 16.54 |==== amd_pstate ondemand ............................ 75.69 |================== acpi_cpufreq schedutil ......................... 79.36 |=================== acpi_cpufreq ondemand .......................... 94.62 |====================== acpi_cpufreq performance ....................... 95.71 |======================= BRL-CAD 7.34 VGR Performance Metric VGR Performance Metric > Higher Is Better amd_pstate_epp powersave balance_performance ... 2853983 |==================== amd_pstate_epp performance balance_performance . 2905181 |==================== amd_pstate_epp powersave power ................. 2863951 |==================== amd_pstate_epp performance performance ......... 2980719 |===================== amd_pstate schedutil ........................... 2836504 |==================== amd_pstate performance ......................... 2815356 |==================== amd_pstate powersave ........................... 405614 |=== amd_pstate ondemand ............................ 2863742 |==================== acpi_cpufreq schedutil ......................... 3021171 |===================== acpi_cpufreq ondemand .......................... 2814269 |==================== acpi_cpufreq performance ....................... 2921572 |==================== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 17.37 |===== amd_pstate_epp performance balance_performance . 14.64 |==== amd_pstate_epp powersave power ................. 17.49 |===== amd_pstate_epp performance performance ......... 15.53 |===== amd_pstate schedutil ........................... 27.74 |======== amd_pstate performance ......................... 15.26 |===== amd_pstate powersave ........................... 75.98 |======================= amd_pstate ondemand ............................ 16.44 |===== acpi_cpufreq schedutil ......................... 17.91 |===== acpi_cpufreq ondemand .......................... 17.19 |===== acpi_cpufreq performance ....................... 15.18 |===== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Write TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 28873 |=================== amd_pstate_epp performance balance_performance . 35610 |======================= amd_pstate_epp powersave power ................. 28682 |=================== amd_pstate_epp performance performance ......... 32286 |===================== amd_pstate schedutil ........................... 19298 |============ amd_pstate performance ......................... 32858 |===================== amd_pstate powersave ........................... 6686 |==== amd_pstate ondemand ............................ 32486 |===================== acpi_cpufreq schedutil ......................... 28016 |================== acpi_cpufreq ondemand .......................... 29347 |=================== acpi_cpufreq performance ....................... 33019 |===================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Queries Per Minute, Geo Mean > Higher Is Better amd_pstate_epp powersave balance_performance ... 372.98 |==================== amd_pstate_epp performance balance_performance . 406.99 |====================== amd_pstate_epp powersave power ................. 353.17 |=================== amd_pstate_epp performance performance ......... 394.98 |===================== amd_pstate schedutil ........................... 251.84 |============== amd_pstate performance ......................... 406.41 |====================== amd_pstate powersave ........................... 76.66 |==== amd_pstate ondemand ............................ 310.66 |================= acpi_cpufreq schedutil ......................... 352.85 |=================== acpi_cpufreq ondemand .......................... 357.94 |=================== acpi_cpufreq performance ....................... 401.81 |====================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run Queries Per Minute, Geo Mean > Higher Is Better amd_pstate_epp powersave balance_performance ... 373.47 |==================== amd_pstate_epp performance balance_performance . 409.86 |====================== amd_pstate_epp powersave power ................. 349.30 |=================== amd_pstate_epp performance performance ......... 407.73 |====================== amd_pstate schedutil ........................... 254.40 |============== amd_pstate performance ......................... 404.95 |====================== amd_pstate powersave ........................... 75.86 |==== amd_pstate ondemand ............................ 306.46 |================ acpi_cpufreq schedutil ......................... 362.27 |=================== acpi_cpufreq ondemand .......................... 357.63 |=================== acpi_cpufreq performance ....................... 410.46 |====================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache Queries Per Minute, Geo Mean > Higher Is Better amd_pstate_epp powersave balance_performance ... 355.12 |==================== amd_pstate_epp performance balance_performance . 395.53 |====================== amd_pstate_epp powersave power ................. 339.68 |=================== amd_pstate_epp performance performance ......... 392.95 |====================== amd_pstate schedutil ........................... 252.79 |============== amd_pstate performance ......................... 394.32 |====================== amd_pstate powersave ........................... 66.81 |==== amd_pstate ondemand ............................ 297.77 |================= acpi_cpufreq schedutil ......................... 351.21 |==================== acpi_cpufreq ondemand .......................... 351.39 |==================== acpi_cpufreq performance ....................... 393.15 |====================== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 32.96 |===== amd_pstate_epp performance balance_performance . 29.82 |==== amd_pstate_epp powersave power ................. 33.13 |===== amd_pstate_epp performance performance ......... 29.00 |==== amd_pstate schedutil ........................... 66.39 |========== amd_pstate performance ......................... 28.72 |==== amd_pstate powersave ........................... 146.57 |====================== amd_pstate ondemand ............................ 41.99 |====== acpi_cpufreq schedutil ......................... 33.72 |===== acpi_cpufreq ondemand .......................... 34.06 |===== acpi_cpufreq performance ....................... 29.54 |==== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 24275 |==================== amd_pstate_epp performance balance_performance . 26830 |====================== amd_pstate_epp powersave power ................. 24173 |==================== amd_pstate_epp performance performance ......... 27596 |======================= amd_pstate schedutil ........................... 12397 |========== amd_pstate performance ......................... 27880 |======================= amd_pstate powersave ........................... 5465 |===== amd_pstate ondemand ............................ 19056 |================ acpi_cpufreq schedutil ......................... 23743 |==================== acpi_cpufreq ondemand .......................... 23493 |=================== acpi_cpufreq performance ....................... 27093 |====================== CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 512 ops/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 75876.4 |================== amd_pstate_epp performance balance_performance . 86214.8 |===================== amd_pstate_epp powersave power ................. 74380.0 |================== amd_pstate_epp performance performance ......... 86477.3 |===================== amd_pstate schedutil ........................... 72179.6 |================== amd_pstate performance ......................... 82071.4 |==================== amd_pstate powersave ........................... 16811.8 |==== amd_pstate ondemand ............................ 75334.7 |================== acpi_cpufreq schedutil ......................... 79962.2 |=================== acpi_cpufreq ondemand .......................... 81782.3 |==================== acpi_cpufreq performance ....................... 86045.2 |===================== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 0.390 |====== amd_pstate_epp performance balance_performance . 0.383 |====== amd_pstate_epp powersave power ................. 0.387 |====== amd_pstate_epp performance performance ......... 0.389 |====== amd_pstate schedutil ........................... 0.796 |============ amd_pstate performance ......................... 0.381 |====== amd_pstate powersave ........................... 1.571 |======================= amd_pstate ondemand ............................ 0.490 |======= acpi_cpufreq schedutil ......................... 0.389 |====== acpi_cpufreq ondemand .......................... 0.445 |======= acpi_cpufreq performance ....................... 0.389 |====== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Only TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 1288550 |===================== amd_pstate_epp performance balance_performance . 1306099 |===================== amd_pstate_epp powersave power ................. 1292798 |===================== amd_pstate_epp performance performance ......... 1286042 |===================== amd_pstate schedutil ........................... 636088 |========== amd_pstate performance ......................... 1311837 |===================== amd_pstate powersave ........................... 318480 |===== amd_pstate ondemand ............................ 1037703 |================= acpi_cpufreq schedutil ......................... 1291471 |===================== acpi_cpufreq ondemand .......................... 1124224 |================== acpi_cpufreq performance ....................... 1285905 |===================== Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 264.91 |=========== amd_pstate_epp performance balance_performance . 264.59 |=========== amd_pstate_epp powersave power ................. 263.30 |=========== amd_pstate_epp performance performance ......... 264.82 |=========== amd_pstate schedutil ........................... 267.54 |=========== amd_pstate performance ......................... 265.31 |=========== amd_pstate powersave ........................... 540.54 |====================== amd_pstate ondemand ............................ 262.40 |=========== acpi_cpufreq schedutil ......................... 263.63 |=========== acpi_cpufreq ondemand .......................... 264.80 |=========== acpi_cpufreq performance ....................... 263.63 |=========== Timed Gem5 Compilation 21.2 Time To Compile Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 157.10 |=== amd_pstate_epp performance balance_performance . 149.77 |=== amd_pstate_epp powersave power ................. 173.20 |=== amd_pstate_epp performance performance ......... 148.50 |=== amd_pstate schedutil ........................... 156.95 |=== amd_pstate performance ......................... 146.64 |=== amd_pstate powersave ........................... 1100.36 |===================== amd_pstate ondemand ............................ 153.90 |=== acpi_cpufreq schedutil ......................... 149.22 |=== acpi_cpufreq ondemand .......................... 152.99 |=== acpi_cpufreq performance ....................... 147.00 |=== VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 4K Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 11.63 |=================== amd_pstate_epp performance balance_performance . 13.37 |===================== amd_pstate_epp powersave power ................. 7.95 |============= amd_pstate_epp performance performance ......... 12.83 |===================== amd_pstate schedutil ........................... 3.06 |===== amd_pstate performance ......................... 13.32 |===================== amd_pstate powersave ........................... 1.86 |=== amd_pstate ondemand ............................ 8.47 |============== acpi_cpufreq schedutil ......................... 14.36 |======================= acpi_cpufreq ondemand .......................... 10.72 |================= acpi_cpufreq performance ....................... 14.17 |======================= Timed Linux Kernel Compilation 6.1 Build: allmodconfig Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 176.50 |=== amd_pstate_epp performance balance_performance . 172.88 |=== amd_pstate_epp powersave power ................. 184.47 |==== amd_pstate_epp performance performance ......... 172.86 |=== amd_pstate schedutil ........................... 182.77 |==== amd_pstate performance ......................... 173.64 |=== amd_pstate powersave ........................... 1094.37 |===================== amd_pstate ondemand ............................ 177.06 |=== acpi_cpufreq schedutil ......................... 172.61 |=== acpi_cpufreq ondemand .......................... 174.95 |=== acpi_cpufreq performance ....................... 173.63 |=== VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Fast Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 3.359 |=================== amd_pstate_epp performance balance_performance . 4.022 |======================= amd_pstate_epp powersave power ................. 2.471 |============== amd_pstate_epp performance performance ......... 4.023 |======================= amd_pstate schedutil ........................... 3.136 |================== amd_pstate performance ......................... 4.021 |======================= amd_pstate powersave ........................... 0.568 |=== amd_pstate ondemand ............................ 3.676 |===================== acpi_cpufreq schedutil ......................... 3.767 |===================== acpi_cpufreq ondemand .......................... 3.974 |======================= acpi_cpufreq performance ....................... 4.050 |======================= PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 0.641 |====== amd_pstate_epp performance balance_performance . 0.644 |====== amd_pstate_epp powersave power ................. 0.642 |====== amd_pstate_epp performance performance ......... 0.645 |====== amd_pstate schedutil ........................... 1.146 |========== amd_pstate performance ......................... 0.630 |====== amd_pstate powersave ........................... 2.522 |======================= amd_pstate ondemand ............................ 0.815 |======= acpi_cpufreq schedutil ......................... 0.636 |====== acpi_cpufreq ondemand .......................... 0.745 |======= acpi_cpufreq performance ....................... 0.648 |====== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 1248266 |===================== amd_pstate_epp performance balance_performance . 1241726 |===================== amd_pstate_epp powersave power ................. 1245914 |===================== amd_pstate_epp performance performance ......... 1240406 |===================== amd_pstate schedutil ........................... 704929 |============ amd_pstate performance ......................... 1268688 |===================== amd_pstate powersave ........................... 317360 |===== amd_pstate ondemand ............................ 997742 |================= acpi_cpufreq schedutil ......................... 1258066 |===================== acpi_cpufreq ondemand .......................... 1074500 |================== acpi_cpufreq performance ....................... 1234869 |==================== Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 Render Ratio > Higher Is Better amd_pstate_epp powersave balance_performance ... 2.059259 |================= amd_pstate_epp performance balance_performance . 2.348079 |==================== amd_pstate_epp powersave power ................. 1.801238 |=============== amd_pstate_epp performance performance ......... 2.346447 |==================== amd_pstate schedutil ........................... 0.811418 |======= amd_pstate performance ......................... 2.363070 |==================== amd_pstate powersave ........................... 0.301450 |=== amd_pstate ondemand ............................ 1.358549 |=========== acpi_cpufreq schedutil ......................... 2.346112 |==================== acpi_cpufreq ondemand .......................... 2.326770 |==================== acpi_cpufreq performance ....................... 2.367074 |==================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 45.51 |===== amd_pstate_epp performance balance_performance . 41.18 |===== amd_pstate_epp powersave power ................. 46.47 |===== amd_pstate_epp performance performance ......... 38.34 |==== amd_pstate schedutil ........................... 81.19 |========= amd_pstate performance ......................... 40.16 |==== amd_pstate powersave ........................... 201.10 |====================== amd_pstate ondemand ............................ 55.95 |====== acpi_cpufreq schedutil ......................... 44.36 |===== acpi_cpufreq ondemand .......................... 44.89 |===== acpi_cpufreq performance ....................... 39.60 |==== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 21978 |=================== amd_pstate_epp performance balance_performance . 24299 |===================== amd_pstate_epp powersave power ................. 21522 |=================== amd_pstate_epp performance performance ......... 26085 |======================= amd_pstate schedutil ........................... 12613 |=========== amd_pstate performance ......................... 24904 |====================== amd_pstate powersave ........................... 4973 |==== amd_pstate ondemand ............................ 17875 |================ acpi_cpufreq schedutil ......................... 22567 |==================== acpi_cpufreq ondemand .......................... 22279 |==================== acpi_cpufreq performance ....................... 25252 |====================== OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 3.51 |=== amd_pstate_epp performance balance_performance . 3.45 |=== amd_pstate_epp powersave power ................. 3.32 |=== amd_pstate_epp performance performance ......... 3.38 |=== amd_pstate schedutil ........................... 3.55 |=== amd_pstate performance ......................... 3.39 |=== amd_pstate powersave ........................... 24.55 |======================= amd_pstate ondemand ............................ 3.40 |=== acpi_cpufreq schedutil ......................... 3.41 |=== acpi_cpufreq ondemand .......................... 3.44 |=== acpi_cpufreq performance ....................... 3.72 |=== OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 35205.47 |=================== amd_pstate_epp performance balance_performance . 35601.83 |=================== amd_pstate_epp powersave power ................. 36975.34 |==================== amd_pstate_epp performance performance ......... 36507.11 |==================== amd_pstate schedutil ........................... 33751.58 |================== amd_pstate performance ......................... 36428.76 |==================== amd_pstate powersave ........................... 5175.96 |=== amd_pstate ondemand ............................ 35501.21 |=================== acpi_cpufreq schedutil ......................... 36248.38 |==================== acpi_cpufreq ondemand .......................... 35666.69 |=================== acpi_cpufreq performance ....................... 32979.74 |================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 0.828 |====== amd_pstate_epp performance balance_performance . 0.827 |====== amd_pstate_epp powersave power ................. 0.815 |====== amd_pstate_epp performance performance ......... 0.815 |====== amd_pstate schedutil ........................... 1.411 |========== amd_pstate performance ......................... 0.812 |====== amd_pstate powersave ........................... 3.203 |======================= amd_pstate ondemand ............................ 1.069 |======== acpi_cpufreq schedutil ......................... 0.808 |====== acpi_cpufreq ondemand .......................... 0.924 |======= acpi_cpufreq performance ....................... 0.832 |====== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 1207540 |==================== amd_pstate_epp performance balance_performance . 1209033 |===================== amd_pstate_epp powersave power ................. 1226429 |===================== amd_pstate_epp performance performance ......... 1226743 |===================== amd_pstate schedutil ........................... 714980 |============ amd_pstate performance ......................... 1230841 |===================== amd_pstate powersave ........................... 312245 |===== amd_pstate ondemand ............................ 955977 |================ acpi_cpufreq schedutil ......................... 1237871 |===================== acpi_cpufreq ondemand .......................... 1082794 |================== acpi_cpufreq performance ....................... 1202637 |==================== Blender 3.4 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 141.79 |=== amd_pstate_epp performance balance_performance . 140.18 |=== amd_pstate_epp powersave power ................. 143.04 |=== amd_pstate_epp performance performance ......... 140.36 |=== amd_pstate schedutil ........................... 147.46 |=== amd_pstate performance ......................... 140.35 |=== amd_pstate powersave ........................... 933.41 |====================== amd_pstate ondemand ............................ 141.91 |=== acpi_cpufreq schedutil ......................... 140.55 |=== acpi_cpufreq ondemand .......................... 141.98 |=== acpi_cpufreq performance ....................... 140.45 |=== CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 512 ops/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 44725.5 |=================== amd_pstate_epp performance balance_performance . 50217.2 |===================== amd_pstate_epp powersave power ................. 43287.2 |================== amd_pstate_epp performance performance ......... 49015.5 |==================== amd_pstate schedutil ........................... 41485.5 |================= amd_pstate performance ......................... 49340.4 |===================== amd_pstate powersave ........................... 11461.4 |===== amd_pstate ondemand ............................ 45780.3 |=================== acpi_cpufreq schedutil ......................... 47827.0 |==================== acpi_cpufreq ondemand .......................... 49459.9 |===================== acpi_cpufreq performance ....................... 49731.9 |===================== Timed Node.js Compilation 18.8 Time To Compile Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 111.95 |=== amd_pstate_epp performance balance_performance . 105.53 |=== amd_pstate_epp powersave power ................. 120.48 |==== amd_pstate_epp performance performance ......... 105.83 |=== amd_pstate schedutil ........................... 113.13 |=== amd_pstate performance ......................... 104.93 |=== amd_pstate powersave ........................... 752.99 |====================== amd_pstate ondemand ............................ 109.81 |=== acpi_cpufreq schedutil ......................... 106.47 |=== acpi_cpufreq ondemand .......................... 107.14 |=== acpi_cpufreq performance ....................... 105.54 |=== OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 41798 |=== amd_pstate_epp performance balance_performance . 41650 |=== amd_pstate_epp powersave power ................. 41532 |=== amd_pstate_epp performance performance ......... 41686 |=== amd_pstate schedutil ........................... 43705 |=== amd_pstate performance ......................... 41588 |=== amd_pstate powersave ........................... 332950 |====================== amd_pstate ondemand ............................ 41897 |=== acpi_cpufreq schedutil ......................... 42130 |=== acpi_cpufreq ondemand .......................... 42264 |=== acpi_cpufreq performance ....................... 41859 |=== OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 1304 | amd_pstate_epp performance balance_performance . 1299 | amd_pstate_epp powersave power ................. 1303 | amd_pstate_epp performance performance ......... 1302 | amd_pstate schedutil ........................... 1376 | amd_pstate performance ......................... 1300 | amd_pstate powersave ........................... 68548 |======================= amd_pstate ondemand ............................ 1322 | acpi_cpufreq schedutil ......................... 1306 | acpi_cpufreq ondemand .......................... 1322 | acpi_cpufreq performance ....................... 1306 | OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 1101 | amd_pstate_epp performance balance_performance . 1103 | amd_pstate_epp powersave power ................. 1102 | amd_pstate_epp performance performance ......... 1101 | amd_pstate schedutil ........................... 1166 | amd_pstate performance ......................... 1098 | amd_pstate powersave ........................... 67000 |======================= amd_pstate ondemand ............................ 1118 | acpi_cpufreq schedutil ......................... 1103 | acpi_cpufreq ondemand .......................... 1120 | acpi_cpufreq performance ....................... 1104 | OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 35438 |=== amd_pstate_epp performance balance_performance . 35573 |=== amd_pstate_epp powersave power ................. 35243 |=== amd_pstate_epp performance performance ......... 35210 |=== amd_pstate schedutil ........................... 36976 |=== amd_pstate performance ......................... 35186 |=== amd_pstate powersave ........................... 291743 |====================== amd_pstate ondemand ............................ 35440 |=== acpi_cpufreq schedutil ......................... 35505 |=== acpi_cpufreq ondemand .......................... 35628 |=== acpi_cpufreq performance ....................... 35406 |=== OpenVINO 2022.3 Model: Person Vehicle Bike Detection FP16 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 17.49 |= amd_pstate_epp performance balance_performance . 17.41 |= amd_pstate_epp powersave power ................. 17.53 |= amd_pstate_epp performance performance ......... 17.37 |= amd_pstate schedutil ........................... 32.42 |== amd_pstate performance ......................... 17.37 |= amd_pstate powersave ........................... 315.26 |====================== amd_pstate ondemand ............................ 35.39 |== acpi_cpufreq schedutil ......................... 17.44 |= acpi_cpufreq ondemand .......................... 17.65 |= acpi_cpufreq performance ....................... 17.48 |= OpenVINO 2022.3 Model: Person Vehicle Bike Detection FP16 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 3654.51 |===================== amd_pstate_epp performance balance_performance . 3670.78 |===================== amd_pstate_epp powersave power ................. 3645.60 |===================== amd_pstate_epp performance performance ......... 3679.31 |===================== amd_pstate schedutil ........................... 2062.04 |============ amd_pstate performance ......................... 3678.91 |===================== amd_pstate powersave ........................... 202.83 |= amd_pstate ondemand ............................ 1915.37 |=========== acpi_cpufreq schedutil ......................... 3663.48 |===================== acpi_cpufreq ondemand .......................... 3621.58 |===================== acpi_cpufreq performance ....................... 3655.41 |===================== VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Faster Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 6.326 |==================== amd_pstate_epp performance balance_performance . 7.397 |======================= amd_pstate_epp powersave power ................. 4.769 |=============== amd_pstate_epp performance performance ......... 7.352 |======================= amd_pstate schedutil ........................... 6.230 |=================== amd_pstate performance ......................... 7.306 |======================= amd_pstate powersave ........................... 1.157 |==== amd_pstate ondemand ............................ 6.646 |===================== acpi_cpufreq schedutil ......................... 7.347 |======================= acpi_cpufreq ondemand .......................... 7.373 |======================= acpi_cpufreq performance ....................... 7.320 |======================= Timed Linux Kernel Compilation 6.1 Build: defconfig Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 25.52 |==== amd_pstate_epp performance balance_performance . 22.62 |=== amd_pstate_epp powersave power ................. 30.83 |==== amd_pstate_epp performance performance ......... 22.53 |=== amd_pstate schedutil ........................... 25.39 |==== amd_pstate performance ......................... 22.52 |=== amd_pstate powersave ........................... 156.78 |====================== amd_pstate ondemand ............................ 24.52 |=== acpi_cpufreq schedutil ......................... 22.61 |=== acpi_cpufreq ondemand .......................... 23.44 |=== acpi_cpufreq performance ....................... 22.65 |=== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 167.41 |===== amd_pstate_epp performance balance_performance . 170.71 |===== amd_pstate_epp powersave power ................. 169.25 |===== amd_pstate_epp performance performance ......... 167.82 |===== amd_pstate schedutil ........................... 182.44 |===== amd_pstate performance ......................... 168.80 |===== amd_pstate powersave ........................... 773.03 |====================== amd_pstate ondemand ............................ 167.90 |===== acpi_cpufreq schedutil ......................... 166.48 |===== acpi_cpufreq ondemand .......................... 169.00 |===== acpi_cpufreq performance ....................... 164.04 |===== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 116.05 |=== amd_pstate_epp performance balance_performance . 114.87 |=== amd_pstate_epp powersave power ................. 116.28 |=== amd_pstate_epp performance performance ......... 116.19 |=== amd_pstate schedutil ........................... 127.86 |==== amd_pstate performance ......................... 116.11 |=== amd_pstate powersave ........................... 773.97 |====================== amd_pstate ondemand ............................ 119.63 |=== acpi_cpufreq schedutil ......................... 116.44 |=== acpi_cpufreq ondemand .......................... 120.10 |=== acpi_cpufreq performance ....................... 115.76 |=== CockroachDB 22.2 Workload: MoVR - Concurrency: 512 ops/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 641.8 |=================== amd_pstate_epp performance balance_performance . 761.0 |======================= amd_pstate_epp powersave power ................. 570.5 |================= amd_pstate_epp performance performance ......... 763.4 |======================= amd_pstate schedutil ........................... 195.9 |====== amd_pstate performance ......................... 766.8 |======================= amd_pstate powersave ........................... 126.8 |==== amd_pstate ondemand ............................ 574.8 |================= acpi_cpufreq schedutil ......................... 438.2 |============= acpi_cpufreq ondemand .......................... 698.0 |===================== acpi_cpufreq performance ....................... 765.2 |======================= nginx 1.23.2 Connections: 500 Requests Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 146738.16 |================== amd_pstate_epp performance balance_performance . 147764.25 |================== amd_pstate_epp powersave power ................. 145614.76 |================== amd_pstate_epp performance performance ......... 152654.86 |=================== amd_pstate schedutil ........................... 138510.00 |================= amd_pstate performance ......................... 151325.01 |=================== amd_pstate powersave ........................... 27774.70 |=== amd_pstate ondemand ............................ 122126.56 |=============== acpi_cpufreq schedutil ......................... 146083.16 |================== acpi_cpufreq ondemand .......................... 149876.52 |=================== acpi_cpufreq performance ....................... 147285.78 |================== OpenVINO 2022.3 Model: Vehicle Detection FP16-INT8 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 17.10 |==== amd_pstate_epp performance balance_performance . 17.06 |==== amd_pstate_epp powersave power ................. 17.14 |==== amd_pstate_epp performance performance ......... 17.07 |==== amd_pstate schedutil ........................... 46.49 |========== amd_pstate performance ......................... 17.05 |==== amd_pstate powersave ........................... 107.17 |====================== amd_pstate ondemand ............................ 36.35 |======= acpi_cpufreq schedutil ......................... 17.09 |==== acpi_cpufreq ondemand .......................... 17.33 |==== acpi_cpufreq performance ....................... 17.09 |==== OpenVINO 2022.3 Model: Vehicle Detection FP16-INT8 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 3738.04 |===================== amd_pstate_epp performance balance_performance . 3748.17 |===================== amd_pstate_epp powersave power ................. 3730.66 |===================== amd_pstate_epp performance performance ......... 3744.85 |===================== amd_pstate schedutil ........................... 1416.77 |======== amd_pstate performance ......................... 3749.69 |===================== amd_pstate powersave ........................... 596.04 |=== amd_pstate ondemand ............................ 1960.30 |=========== acpi_cpufreq schedutil ......................... 3741.06 |===================== acpi_cpufreq ondemand .......................... 3689.97 |===================== acpi_cpufreq performance ....................... 3741.35 |===================== OpenVINO 2022.3 Model: Machine Translation EN To DE FP16 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 239.05 |==== amd_pstate_epp performance balance_performance . 239.29 |==== amd_pstate_epp powersave power ................. 238.16 |==== amd_pstate_epp performance performance ......... 239.70 |==== amd_pstate schedutil ........................... 240.82 |==== amd_pstate performance ......................... 239.02 |==== amd_pstate powersave ........................... 1229.86 |===================== amd_pstate ondemand ............................ 228.75 |==== acpi_cpufreq schedutil ......................... 239.94 |==== acpi_cpufreq ondemand .......................... 241.17 |==== acpi_cpufreq performance ....................... 238.32 |==== OpenVINO 2022.3 Model: Machine Translation EN To DE FP16 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 267.15 |===================== amd_pstate_epp performance balance_performance . 266.96 |===================== amd_pstate_epp powersave power ................. 268.24 |===================== amd_pstate_epp performance performance ......... 266.56 |===================== amd_pstate schedutil ........................... 265.31 |===================== amd_pstate performance ......................... 267.30 |===================== amd_pstate powersave ........................... 51.52 |==== amd_pstate ondemand ............................ 279.33 |====================== acpi_cpufreq schedutil ......................... 266.14 |===================== acpi_cpufreq ondemand .......................... 264.82 |===================== acpi_cpufreq performance ....................... 268.04 |===================== Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 205.68 |==== amd_pstate_epp performance balance_performance . 205.42 |==== amd_pstate_epp powersave power ................. 205.70 |==== amd_pstate_epp performance performance ......... 205.30 |==== amd_pstate schedutil ........................... 212.63 |==== amd_pstate performance ......................... 205.41 |==== amd_pstate powersave ........................... 1124.30 |===================== amd_pstate ondemand ............................ 206.14 |==== acpi_cpufreq schedutil ......................... 205.49 |==== acpi_cpufreq ondemand .......................... 206.45 |==== acpi_cpufreq performance ....................... 205.28 |==== Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 310.30 |====================== amd_pstate_epp performance balance_performance . 310.43 |====================== amd_pstate_epp powersave power ................. 309.38 |====================== amd_pstate_epp performance performance ......... 310.74 |====================== amd_pstate schedutil ........................... 299.57 |===================== amd_pstate performance ......................... 310.30 |====================== amd_pstate powersave ........................... 56.66 |==== amd_pstate ondemand ............................ 309.60 |====================== acpi_cpufreq schedutil ......................... 310.09 |====================== acpi_cpufreq ondemand .......................... 308.68 |====================== acpi_cpufreq performance ....................... 310.61 |====================== Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed MB/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 1195.0 |====================== amd_pstate_epp performance balance_performance . 1203.6 |====================== amd_pstate_epp powersave power ................. 1186.0 |====================== amd_pstate_epp performance performance ......... 1203.3 |====================== amd_pstate schedutil ........................... 1191.5 |====================== amd_pstate performance ......................... 1171.2 |===================== amd_pstate powersave ........................... 140.1 |=== amd_pstate ondemand ............................ 1178.0 |===================== acpi_cpufreq schedutil ......................... 1203.2 |====================== acpi_cpufreq ondemand .......................... 1203.2 |====================== acpi_cpufreq performance ....................... 1209.1 |====================== Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed MB/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 8.91 |======================= amd_pstate_epp performance balance_performance . 9.28 |======================== amd_pstate_epp powersave power ................. 8.72 |====================== amd_pstate_epp performance performance ......... 9.26 |======================== amd_pstate schedutil ........................... 9.20 |======================== amd_pstate performance ......................... 9.31 |======================== amd_pstate powersave ........................... 1.06 |=== amd_pstate ondemand ............................ 9.13 |======================== acpi_cpufreq schedutil ......................... 9.29 |======================== acpi_cpufreq ondemand .......................... 9.21 |======================== acpi_cpufreq performance ....................... 9.12 |======================== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 68.92 |===== amd_pstate_epp performance balance_performance . 68.71 |===== amd_pstate_epp powersave power ................. 68.56 |===== amd_pstate_epp performance performance ......... 68.58 |===== amd_pstate schedutil ........................... 71.16 |===== amd_pstate performance ......................... 68.65 |===== amd_pstate powersave ........................... 295.49 |====================== amd_pstate ondemand ............................ 68.60 |===== acpi_cpufreq schedutil ......................... 68.55 |===== acpi_cpufreq ondemand .......................... 68.71 |===== acpi_cpufreq performance ....................... 68.77 |===== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 926.20 |====================== amd_pstate_epp performance balance_performance . 929.79 |====================== amd_pstate_epp powersave power ................. 931.60 |====================== amd_pstate_epp performance performance ......... 931.58 |====================== amd_pstate schedutil ........................... 897.64 |===================== amd_pstate performance ......................... 930.49 |====================== amd_pstate powersave ........................... 215.52 |===== amd_pstate ondemand ............................ 931.20 |====================== acpi_cpufreq schedutil ......................... 932.00 |====================== acpi_cpufreq ondemand .......................... 929.79 |====================== acpi_cpufreq performance ....................... 928.54 |====================== OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 4030.03 |=== amd_pstate_epp performance balance_performance . 4015.91 |=== amd_pstate_epp powersave power ................. 4033.63 |=== amd_pstate_epp performance performance ......... 4005.03 |=== amd_pstate schedutil ........................... 4262.29 |==== amd_pstate performance ......................... 4030.79 |=== amd_pstate powersave ........................... 24236.97 |==================== amd_pstate ondemand ............................ 4103.64 |=== acpi_cpufreq schedutil ......................... 4054.30 |=== acpi_cpufreq ondemand .......................... 4041.31 |=== acpi_cpufreq performance ....................... 4032.10 |=== OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 15.52 |======================= amd_pstate_epp performance balance_performance . 15.54 |======================= amd_pstate_epp powersave power ................. 15.50 |======================= amd_pstate_epp performance performance ......... 15.55 |======================= amd_pstate schedutil ........................... 14.63 |====================== amd_pstate performance ......................... 15.52 |======================= amd_pstate powersave ........................... 2.54 |==== amd_pstate ondemand ............................ 15.24 |======================= acpi_cpufreq schedutil ......................... 15.41 |======================= acpi_cpufreq ondemand .......................... 15.44 |======================= acpi_cpufreq performance ....................... 15.49 |======================= Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 587.74 |==== amd_pstate_epp performance balance_performance . 587.84 |==== amd_pstate_epp powersave power ................. 587.78 |==== amd_pstate_epp performance performance ......... 587.61 |==== amd_pstate schedutil ........................... 605.67 |==== amd_pstate performance ......................... 586.86 |==== amd_pstate powersave ........................... 3485.21 |===================== amd_pstate ondemand ............................ 589.69 |==== acpi_cpufreq schedutil ......................... 587.78 |==== acpi_cpufreq ondemand .......................... 588.61 |==== acpi_cpufreq performance ....................... 587.38 |==== Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 108.20 |====================== amd_pstate_epp performance balance_performance . 108.20 |====================== amd_pstate_epp powersave power ................. 108.15 |====================== amd_pstate_epp performance performance ......... 108.11 |====================== amd_pstate schedutil ........................... 104.89 |===================== amd_pstate performance ......................... 108.40 |====================== amd_pstate powersave ........................... 18.20 |==== amd_pstate ondemand ............................ 107.47 |====================== acpi_cpufreq schedutil ......................... 108.13 |====================== acpi_cpufreq ondemand .......................... 107.90 |====================== acpi_cpufreq performance ....................... 108.16 |====================== 7-Zip Compression 22.01 Test: Decompression Rating MIPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 716703 |===================== amd_pstate_epp performance balance_performance . 726521 |===================== amd_pstate_epp powersave power ................. 661420 |=================== amd_pstate_epp performance performance ......... 733882 |====================== amd_pstate schedutil ........................... 699228 |===================== amd_pstate performance ......................... 746376 |====================== amd_pstate powersave ........................... 104287 |=== amd_pstate ondemand ............................ 739016 |====================== acpi_cpufreq schedutil ......................... 748549 |====================== acpi_cpufreq ondemand .......................... 740039 |====================== acpi_cpufreq performance ....................... 742955 |====================== 7-Zip Compression 22.01 Test: Compression Rating MIPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 500471 |===================== amd_pstate_epp performance balance_performance . 522348 |====================== amd_pstate_epp powersave power ................. 459617 |=================== amd_pstate_epp performance performance ......... 514443 |===================== amd_pstate schedutil ........................... 394982 |================ amd_pstate performance ......................... 530689 |====================== amd_pstate powersave ........................... 100970 |==== amd_pstate ondemand ............................ 485387 |==================== acpi_cpufreq schedutil ......................... 475861 |==================== acpi_cpufreq ondemand .......................... 507039 |===================== acpi_cpufreq performance ....................... 528289 |====================== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 869.22 |==== amd_pstate_epp performance balance_performance . 868.87 |==== amd_pstate_epp powersave power ................. 868.92 |==== amd_pstate_epp performance performance ......... 868.26 |==== amd_pstate schedutil ........................... 895.13 |==== amd_pstate performance ......................... 867.90 |==== amd_pstate powersave ........................... 4864.70 |===================== amd_pstate ondemand ............................ 876.50 |==== acpi_cpufreq schedutil ......................... 868.82 |==== acpi_cpufreq ondemand .......................... 872.89 |==== acpi_cpufreq performance ....................... 869.19 |==== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 72.85 |======================= amd_pstate_epp performance balance_performance . 72.95 |======================= amd_pstate_epp powersave power ................. 72.80 |======================= amd_pstate_epp performance performance ......... 73.06 |======================= amd_pstate schedutil ........................... 70.75 |====================== amd_pstate performance ......................... 72.91 |======================= amd_pstate powersave ........................... 12.97 |==== amd_pstate ondemand ............................ 72.52 |======================= acpi_cpufreq schedutil ......................... 73.10 |======================= acpi_cpufreq ondemand .......................... 72.77 |======================= acpi_cpufreq performance ....................... 73.16 |======================= Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed MB/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 1278.3 |====================== amd_pstate_epp performance balance_performance . 1278.6 |====================== amd_pstate_epp powersave power ................. 1269.8 |====================== amd_pstate_epp performance performance ......... 1276.2 |====================== amd_pstate schedutil ........................... 1261.8 |====================== amd_pstate performance ......................... 1188.1 |==================== amd_pstate powersave ........................... 146.1 |=== amd_pstate ondemand ............................ 1280.5 |====================== acpi_cpufreq schedutil ......................... 1270.9 |====================== acpi_cpufreq ondemand .......................... 1264.7 |====================== acpi_cpufreq performance ....................... 1272.8 |====================== Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed MB/s > Higher Is Better amd_pstate_epp powersave balance_performance ... 16.00 |===================== amd_pstate_epp performance balance_performance . 17.60 |======================= amd_pstate_epp powersave power ................. 15.20 |==================== amd_pstate_epp performance performance ......... 17.50 |======================= amd_pstate schedutil ........................... 17.10 |====================== amd_pstate performance ......................... 17.60 |======================= amd_pstate powersave ........................... 1.97 |=== amd_pstate ondemand ............................ 17.20 |====================== acpi_cpufreq schedutil ......................... 17.40 |======================= acpi_cpufreq ondemand .......................... 17.50 |======================= acpi_cpufreq performance ....................... 17.50 |======================= Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 868.08 |==== amd_pstate_epp performance balance_performance . 867.93 |==== amd_pstate_epp powersave power ................. 867.02 |==== amd_pstate_epp performance performance ......... 866.98 |==== amd_pstate schedutil ........................... 893.43 |==== amd_pstate performance ......................... 866.96 |==== amd_pstate powersave ........................... 4868.11 |===================== amd_pstate ondemand ............................ 876.33 |==== acpi_cpufreq schedutil ......................... 866.62 |==== acpi_cpufreq ondemand .......................... 872.07 |==== acpi_cpufreq performance ....................... 867.83 |==== Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 73.02 |======================= amd_pstate_epp performance balance_performance . 72.94 |======================= amd_pstate_epp powersave power ................. 73.05 |======================= amd_pstate_epp performance performance ......... 73.14 |======================= amd_pstate schedutil ........................... 71.01 |====================== amd_pstate performance ......................... 73.27 |======================= amd_pstate powersave ........................... 13.00 |==== amd_pstate ondemand ............................ 72.41 |======================= acpi_cpufreq schedutil ......................... 73.03 |======================= acpi_cpufreq ondemand .......................... 72.78 |======================= acpi_cpufreq performance ....................... 73.25 |======================= Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 207.61 |==== amd_pstate_epp performance balance_performance . 207.44 |==== amd_pstate_epp powersave power ................. 207.33 |==== amd_pstate_epp performance performance ......... 207.19 |==== amd_pstate schedutil ........................... 214.47 |==== amd_pstate performance ......................... 207.67 |==== amd_pstate powersave ........................... 1143.06 |===================== amd_pstate ondemand ............................ 208.09 |==== acpi_cpufreq schedutil ......................... 207.27 |==== acpi_cpufreq ondemand .......................... 207.88 |==== acpi_cpufreq performance ....................... 207.58 |==== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 307.04 |====================== amd_pstate_epp performance balance_performance . 307.64 |====================== amd_pstate_epp powersave power ................. 307.71 |====================== amd_pstate_epp performance performance ......... 307.89 |====================== amd_pstate schedutil ........................... 297.14 |===================== amd_pstate performance ......................... 307.27 |====================== amd_pstate powersave ........................... 55.22 |==== amd_pstate ondemand ............................ 306.40 |====================== acpi_cpufreq schedutil ......................... 307.72 |====================== acpi_cpufreq ondemand .......................... 306.68 |====================== acpi_cpufreq performance ....................... 307.17 |====================== OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 1195.35 |=== amd_pstate_epp performance balance_performance . 1195.60 |=== amd_pstate_epp powersave power ................. 1199.87 |=== amd_pstate_epp performance performance ......... 1203.52 |=== amd_pstate schedutil ........................... 1260.49 |=== amd_pstate performance ......................... 1195.23 |=== amd_pstate powersave ........................... 7565.21 |===================== amd_pstate ondemand ............................ 1210.76 |=== acpi_cpufreq schedutil ......................... 1195.76 |=== acpi_cpufreq ondemand .......................... 1203.42 |=== acpi_cpufreq performance ....................... 1195.14 |=== OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 53.09 |======================= amd_pstate_epp performance balance_performance . 53.07 |======================= amd_pstate_epp powersave power ................. 52.91 |======================= amd_pstate_epp performance performance ......... 52.79 |======================= amd_pstate schedutil ........................... 50.35 |====================== amd_pstate performance ......................... 53.03 |======================= amd_pstate powersave ........................... 8.38 |==== amd_pstate ondemand ............................ 52.48 |======================= acpi_cpufreq schedutil ......................... 53.06 |======================= acpi_cpufreq ondemand .......................... 52.76 |======================= acpi_cpufreq performance ....................... 53.12 |======================= Timed Godot Game Engine Compilation 3.2.3 Time To Compile Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 41.90 |==== amd_pstate_epp performance balance_performance . 37.32 |=== amd_pstate_epp powersave power ................. 47.94 |==== amd_pstate_epp performance performance ......... 37.01 |=== amd_pstate schedutil ........................... 41.94 |==== amd_pstate performance ......................... 37.05 |=== amd_pstate powersave ........................... 258.87 |====================== amd_pstate ondemand ............................ 40.81 |=== acpi_cpufreq schedutil ......................... 37.58 |=== acpi_cpufreq ondemand .......................... 38.37 |=== acpi_cpufreq performance ....................... 37.38 |=== SVT-AV1 1.4 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 3.453 |=================== amd_pstate_epp performance balance_performance . 4.158 |======================= amd_pstate_epp powersave power ................. 2.443 |============= amd_pstate_epp performance performance ......... 4.181 |======================= amd_pstate schedutil ........................... 3.322 |================== amd_pstate performance ......................... 4.184 |======================= amd_pstate powersave ........................... 0.558 |=== amd_pstate ondemand ............................ 3.865 |===================== acpi_cpufreq schedutil ......................... 4.013 |====================== acpi_cpufreq ondemand .......................... 4.090 |====================== acpi_cpufreq performance ....................... 4.182 |======================= Blender 3.4 Blend File: Pabellon Barcelona - Compute: CPU-Only Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 45.99 |=== amd_pstate_epp performance balance_performance . 45.72 |=== amd_pstate_epp powersave power ................. 46.73 |=== amd_pstate_epp performance performance ......... 45.75 |=== amd_pstate schedutil ........................... 48.49 |==== amd_pstate performance ......................... 45.77 |=== amd_pstate powersave ........................... 297.67 |====================== amd_pstate ondemand ............................ 46.19 |=== acpi_cpufreq schedutil ......................... 45.95 |=== acpi_cpufreq ondemand .......................... 45.92 |=== acpi_cpufreq performance ....................... 45.85 |=== GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better amd_pstate_epp powersave balance_performance ... 11.079 |====================== amd_pstate_epp performance balance_performance . 11.179 |====================== amd_pstate_epp powersave power ................. 11.131 |====================== amd_pstate_epp performance performance ......... 11.074 |====================== amd_pstate schedutil ........................... 8.482 |================= amd_pstate performance ......................... 11.053 |====================== amd_pstate powersave ........................... 2.167 |==== amd_pstate ondemand ............................ 9.550 |=================== acpi_cpufreq schedutil ......................... 11.173 |====================== acpi_cpufreq ondemand .......................... 10.853 |===================== acpi_cpufreq performance ....................... 11.125 |====================== OpenVINO 2022.3 Model: Weld Porosity Detection FP16-INT8 - Device: CPU ms < Lower Is Better amd_pstate_epp powersave balance_performance ... 24.17 |=== amd_pstate_epp performance balance_performance . 24.16 |=== amd_pstate_epp powersave power ................. 24.24 |=== amd_pstate_epp performance performance ......... 24.17 |=== amd_pstate schedutil ........................... 25.43 |==== amd_pstate performance ......................... 24.17 |=== amd_pstate powersave ........................... 153.78 |====================== amd_pstate ondemand ............................ 24.48 |==== acpi_cpufreq schedutil ......................... 24.17 |=== acpi_cpufreq ondemand .......................... 24.36 |=== acpi_cpufreq performance ....................... 24.17 |=== OpenVINO 2022.3 Model: Weld Porosity Detection FP16-INT8 - Device: CPU FPS > Higher Is Better amd_pstate_epp powersave balance_performance ... 5291.45 |===================== amd_pstate_epp performance balance_performance . 5294.97 |===================== amd_pstate_epp powersave power ................. 5277.02 |===================== amd_pstate_epp performance performance ......... 5291.00 |===================== amd_pstate schedutil ........................... 5029.05 |==================== amd_pstate performance ......................... 5292.08 |===================== amd_pstate powersave ........................... 830.31 |=== amd_pstate ondemand ............................ 5224.83 |===================== acpi_cpufreq schedutil ......................... 5292.42 |===================== acpi_cpufreq ondemand .......................... 5250.15 |===================== acpi_cpufreq performance ....................... 5291.81 |===================== Blender 3.4 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 37.94 |=== amd_pstate_epp performance balance_performance . 37.68 |=== amd_pstate_epp powersave power ................. 38.35 |=== amd_pstate_epp performance performance ......... 37.57 |=== amd_pstate schedutil ........................... 40.04 |=== amd_pstate performance ......................... 37.73 |=== amd_pstate powersave ........................... 254.46 |====================== amd_pstate ondemand ............................ 37.85 |=== acpi_cpufreq schedutil ......................... 37.61 |=== acpi_cpufreq ondemand .......................... 37.92 |=== acpi_cpufreq performance ....................... 37.71 |=== PHPBench 0.8.1 PHP Benchmark Suite Score > Higher Is Better amd_pstate_epp powersave balance_performance ... 687682 |===================== amd_pstate_epp performance balance_performance . 697478 |====================== amd_pstate_epp powersave power ................. 671207 |===================== amd_pstate_epp performance performance ......... 700287 |====================== amd_pstate schedutil ........................... 711125 |====================== amd_pstate performance ......................... 701215 |====================== amd_pstate powersave ........................... 79570 |== amd_pstate ondemand ............................ 705341 |====================== acpi_cpufreq schedutil ......................... 698785 |====================== acpi_cpufreq ondemand .......................... 708218 |====================== acpi_cpufreq performance ....................... 699816 |====================== SVT-AV1 1.4 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 175.46 |===================== amd_pstate_epp performance balance_performance . 186.05 |====================== amd_pstate_epp powersave power ................. 155.77 |================== amd_pstate_epp performance performance ......... 173.31 |==================== amd_pstate schedutil ........................... 90.28 |=========== amd_pstate performance ......................... 182.36 |====================== amd_pstate powersave ........................... 36.65 |==== amd_pstate ondemand ............................ 171.88 |==================== acpi_cpufreq schedutil ......................... 167.61 |==================== acpi_cpufreq ondemand .......................... 180.33 |===================== acpi_cpufreq performance ....................... 180.86 |===================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 104.57 |==== amd_pstate_epp performance balance_performance . 104.76 |==== amd_pstate_epp powersave power ................. 104.66 |==== amd_pstate_epp performance performance ......... 104.62 |==== amd_pstate schedutil ........................... 108.66 |==== amd_pstate performance ......................... 104.73 |==== amd_pstate powersave ........................... 577.77 |====================== amd_pstate ondemand ............................ 105.15 |==== acpi_cpufreq schedutil ......................... 104.75 |==== acpi_cpufreq ondemand .......................... 105.02 |==== acpi_cpufreq performance ....................... 104.80 |==== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 610.70 |====================== amd_pstate_epp performance balance_performance . 609.61 |====================== amd_pstate_epp powersave power ................. 609.90 |====================== amd_pstate_epp performance performance ......... 610.32 |====================== amd_pstate schedutil ........................... 587.48 |===================== amd_pstate performance ......................... 609.96 |====================== amd_pstate powersave ........................... 109.77 |==== amd_pstate ondemand ............................ 607.33 |====================== acpi_cpufreq schedutil ......................... 609.76 |====================== acpi_cpufreq ondemand .......................... 608.30 |====================== acpi_cpufreq performance ....................... 609.04 |====================== Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 158.13 |==== amd_pstate_epp performance balance_performance . 157.55 |==== amd_pstate_epp powersave power ................. 158.08 |==== amd_pstate_epp performance performance ......... 157.70 |==== amd_pstate schedutil ........................... 164.20 |==== amd_pstate performance ......................... 157.80 |==== amd_pstate powersave ........................... 884.29 |====================== amd_pstate ondemand ............................ 158.95 |==== acpi_cpufreq schedutil ......................... 157.83 |==== acpi_cpufreq ondemand .......................... 158.05 |==== acpi_cpufreq performance ....................... 157.88 |==== Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 403.29 |====================== amd_pstate_epp performance balance_performance . 404.95 |====================== amd_pstate_epp powersave power ................. 403.59 |====================== amd_pstate_epp performance performance ......... 404.55 |====================== amd_pstate schedutil ........................... 388.64 |===================== amd_pstate performance ......................... 404.30 |====================== amd_pstate powersave ........................... 70.87 |==== amd_pstate ondemand ............................ 401.73 |====================== acpi_cpufreq schedutil ......................... 404.27 |====================== acpi_cpufreq ondemand .......................... 403.28 |====================== acpi_cpufreq performance ....................... 404.13 |====================== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better amd_pstate_epp powersave balance_performance ... 76.99 |==== amd_pstate_epp performance balance_performance . 76.98 |==== amd_pstate_epp powersave power ................. 77.02 |==== amd_pstate_epp performance performance ......... 76.91 |==== amd_pstate schedutil ........................... 78.87 |==== amd_pstate performance ......................... 76.91 |==== amd_pstate powersave ........................... 431.96 |====================== amd_pstate ondemand ............................ 76.57 |==== acpi_cpufreq schedutil ......................... 77.10 |==== acpi_cpufreq ondemand .......................... 77.01 |==== acpi_cpufreq performance ....................... 77.06 |==== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better amd_pstate_epp powersave balance_performance ... 830.10 |====================== amd_pstate_epp performance balance_performance . 830.01 |====================== amd_pstate_epp powersave power ................. 829.60 |====================== amd_pstate_epp performance performance ......... 830.32 |====================== amd_pstate schedutil ........................... 809.90 |===================== amd_pstate performance ......................... 830.85 |====================== amd_pstate powersave ........................... 147.69 |==== amd_pstate ondemand ............................ 834.46 |====================== acpi_cpufreq schedutil ......................... 828.90 |====================== acpi_cpufreq ondemand .......................... 829.17 |====================== acpi_cpufreq performance ....................... 828.82 |====================== x265 3.4 Video Input: Bosphorus 4K Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 18.56 |==================== amd_pstate_epp performance balance_performance . 20.55 |======================= amd_pstate_epp powersave power ................. 15.33 |================= amd_pstate_epp performance performance ......... 20.63 |======================= amd_pstate schedutil ........................... 16.36 |================== amd_pstate performance ......................... 20.74 |======================= amd_pstate powersave ........................... 4.37 |===== amd_pstate ondemand ............................ 19.29 |===================== acpi_cpufreq schedutil ......................... 20.67 |======================= acpi_cpufreq ondemand .......................... 19.90 |====================== acpi_cpufreq performance ....................... 20.96 |======================= PyBench 2018-02-16 Total For Average Test Times Milliseconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 1003 |=== amd_pstate_epp performance balance_performance . 998 |=== amd_pstate_epp powersave power ................. 1008 |=== amd_pstate_epp performance performance ......... 1010 |=== amd_pstate schedutil ........................... 987 |=== amd_pstate performance ......................... 1010 |=== amd_pstate powersave ........................... 8775 |======================== amd_pstate ondemand ............................ 1004 |=== acpi_cpufreq schedutil ......................... 999 |=== acpi_cpufreq ondemand .......................... 1006 |=== acpi_cpufreq performance ....................... 1002 |=== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 24.28 |===================== amd_pstate_epp performance balance_performance . 27.01 |======================= amd_pstate_epp powersave power ................. 17.33 |=============== amd_pstate_epp performance performance ......... 27.06 |======================= amd_pstate schedutil ........................... 12.25 |========== amd_pstate performance ......................... 26.89 |======================= amd_pstate powersave ........................... 3.97 |=== amd_pstate ondemand ............................ 24.58 |===================== acpi_cpufreq schedutil ......................... 21.21 |================== acpi_cpufreq ondemand .......................... 27.20 |======================= acpi_cpufreq performance ....................... 26.93 |======================= Embree 4.0 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 92.21 |====================== amd_pstate_epp performance balance_performance . 94.78 |======================= amd_pstate_epp powersave power ................. 93.73 |======================= amd_pstate_epp performance performance ......... 93.50 |======================= amd_pstate schedutil ........................... 88.80 |====================== amd_pstate performance ......................... 93.61 |======================= amd_pstate powersave ........................... 24.79 |====== amd_pstate ondemand ............................ 92.80 |======================= acpi_cpufreq schedutil ......................... 93.76 |======================= acpi_cpufreq ondemand .......................... 92.95 |======================= acpi_cpufreq performance ....................... 91.37 |====================== Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 50.67 |===================== amd_pstate_epp performance balance_performance . 56.14 |======================= amd_pstate_epp powersave power ................. 41.83 |================= amd_pstate_epp performance performance ......... 55.85 |======================= amd_pstate schedutil ........................... 26.54 |=========== amd_pstate performance ......................... 55.27 |======================= amd_pstate powersave ........................... 7.72 |=== amd_pstate ondemand ............................ 52.60 |====================== acpi_cpufreq schedutil ......................... 52.93 |====================== acpi_cpufreq ondemand .......................... 55.58 |======================= acpi_cpufreq performance ....................... 55.92 |======================= Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Super Fast Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 47.97 |==================== amd_pstate_epp performance balance_performance . 54.69 |======================= amd_pstate_epp powersave power ................. 45.56 |=================== amd_pstate_epp performance performance ......... 54.77 |======================= amd_pstate schedutil ........................... 26.21 |=========== amd_pstate performance ......................... 53.77 |====================== amd_pstate powersave ........................... 7.43 |=== amd_pstate ondemand ............................ 52.17 |====================== acpi_cpufreq schedutil ......................... 52.32 |====================== acpi_cpufreq ondemand .......................... 54.78 |======================= acpi_cpufreq performance ....................... 55.03 |======================= uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 41.13 |==================== amd_pstate_epp performance balance_performance . 47.25 |======================= amd_pstate_epp powersave power ................. 33.76 |================ amd_pstate_epp performance performance ......... 46.08 |====================== amd_pstate schedutil ........................... 21.65 |=========== amd_pstate performance ......................... 45.46 |====================== amd_pstate powersave ........................... 6.88 |=== amd_pstate ondemand ............................ 41.90 |==================== acpi_cpufreq schedutil ......................... 40.82 |==================== acpi_cpufreq ondemand .......................... 44.50 |====================== acpi_cpufreq performance ....................... 46.45 |======================= Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 29.42 |===================== amd_pstate_epp performance balance_performance . 31.99 |======================= amd_pstate_epp powersave power ................. 20.66 |=============== amd_pstate_epp performance performance ......... 32.30 |======================= amd_pstate schedutil ........................... 14.39 |========== amd_pstate performance ......................... 32.00 |======================= amd_pstate powersave ........................... 4.83 |=== amd_pstate ondemand ............................ 30.32 |====================== acpi_cpufreq schedutil ......................... 22.97 |================ acpi_cpufreq ondemand .......................... 32.36 |======================= acpi_cpufreq performance ....................... 32.09 |======================= uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 39.96 |==================== amd_pstate_epp performance balance_performance . 46.26 |======================= amd_pstate_epp powersave power ................. 33.79 |================= amd_pstate_epp performance performance ......... 46.31 |======================= amd_pstate schedutil ........................... 25.43 |============= amd_pstate performance ......................... 45.41 |======================= amd_pstate powersave ........................... 6.77 |=== amd_pstate ondemand ............................ 41.57 |===================== acpi_cpufreq schedutil ......................... 40.71 |==================== acpi_cpufreq ondemand .......................... 45.19 |====================== acpi_cpufreq performance ....................... 46.07 |======================= Blender 3.4 Blend File: Fishy Cat - Compute: CPU-Only Seconds < Lower Is Better amd_pstate_epp powersave balance_performance ... 19.17 |=== amd_pstate_epp performance balance_performance . 18.80 |=== amd_pstate_epp powersave power ................. 19.74 |=== amd_pstate_epp performance performance ......... 18.95 |=== amd_pstate schedutil ........................... 20.26 |==== amd_pstate performance ......................... 18.89 |=== amd_pstate powersave ........................... 124.48 |====================== amd_pstate ondemand ............................ 19.18 |=== acpi_cpufreq schedutil ......................... 18.90 |=== acpi_cpufreq ondemand .......................... 19.00 |=== acpi_cpufreq performance ....................... 19.01 |=== NAMD 2.14 ATPase Simulation - 327,506 Atoms days/ns < Lower Is Better amd_pstate_epp powersave balance_performance ... 0.22286 |=== amd_pstate_epp performance balance_performance . 0.22270 |=== amd_pstate_epp powersave power ................. 0.22450 |=== amd_pstate_epp performance performance ......... 0.22021 |=== amd_pstate schedutil ........................... 0.23420 |=== amd_pstate performance ......................... 0.22464 |=== amd_pstate powersave ........................... 1.40982 |===================== amd_pstate ondemand ............................ 0.22289 |=== acpi_cpufreq schedutil ......................... 0.22025 |=== acpi_cpufreq ondemand .......................... 0.22290 |=== acpi_cpufreq performance ....................... 0.22441 |=== SVT-AV1 1.4 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 170.27 |=================== amd_pstate_epp performance balance_performance . 194.10 |====================== amd_pstate_epp powersave power ................. 145.32 |================ amd_pstate_epp performance performance ......... 188.42 |===================== amd_pstate schedutil ........................... 72.22 |======== amd_pstate performance ......................... 191.05 |====================== amd_pstate powersave ........................... 32.27 |==== amd_pstate ondemand ............................ 161.78 |================== acpi_cpufreq schedutil ......................... 144.34 |================ acpi_cpufreq ondemand .......................... 180.34 |==================== acpi_cpufreq performance ....................... 189.13 |===================== Embree 4.0 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better amd_pstate_epp powersave balance_performance ... 122.50 |====================== amd_pstate_epp performance balance_performance . 123.15 |====================== amd_pstate_epp powersave power ................. 121.30 |====================== amd_pstate_epp performance performance ......... 122.87 |====================== amd_pstate schedutil ........................... 116.75 |===================== amd_pstate performance ......................... 123.02 |====================== amd_pstate powersave ........................... 19.67 |==== amd_pstate ondemand ............................ 121.86 |====================== acpi_cpufreq schedutil ......................... 122.71 |====================== acpi_cpufreq ondemand .......................... 120.80 |====================== acpi_cpufreq performance ....................... 123.05 |======================