AMD SME Benchmark Genoa
4th Gen AMD EPYC "Genoa" Secure Memory Encryption (SME) benchmarks by Michael Larabel for a future article.


No SME: 

	Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

	OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080

AMD SME Enabled: 

	Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe

	OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080


OpenFOAM 10
Input: drivaerFastback, Small Mesh Size - Mesh Time
Seconds < Lower Is Better
No SME .......... 25.07 |==================================================
AMD SME Enabled . 27.10 |======================================================


ASKAP 1.0
Test: tConvolve MPI - Degridding
Mpix/sec > Higher Is Better
No SME .......... 83598.3 |====================================================
AMD SME Enabled . 78718.7 |=================================================


PyHPC Benchmarks 3.0
Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State
Seconds < Lower Is Better
No SME .......... 0.884 |===================================================
AMD SME Enabled . 0.934 |======================================================


Appleseed 2.0 Beta
Scene: Emily
Seconds < Lower Is Better
No SME .......... 142.95 |==================================================
AMD SME Enabled . 150.92 |=====================================================


KTX-Software toktx 4.0
Settings: Zstd Compression 19
Seconds < Lower Is Better
No SME .......... 18.86 |===================================================
AMD SME Enabled . 19.88 |======================================================


DaCapo Benchmark 9.12-MR1
Java Test: H2
msec < Lower Is Better
No SME .......... 4807 |====================================================
AMD SME Enabled . 5050 |=======================================================


Graph500 3.0
Scale: 26
bfs median_TEPS > Higher Is Better
No SME .......... 1426480000 |=================================================
AMD SME Enabled . 1358510000 |===============================================


AOM AV1 3.5
Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No SME .......... 34.47 |======================================================
AMD SME Enabled . 33.12 |====================================================


ASKAP 1.0
Test: tConvolve MPI - Gridding
Mpix/sec > Higher Is Better
No SME .......... 93071.0 |====================================================
AMD SME Enabled . 89541.8 |==================================================


7-Zip Compression 22.01
Test: Compression Rating
MIPS > Higher Is Better
No SME .......... 917782 |=====================================================
AMD SME Enabled . 885135 |===================================================


x264 2022-02-22
Video Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No SME .......... 106.86 |=====================================================
AMD SME Enabled . 103.07 |===================================================


Graph500 3.0
Scale: 26
sssp median_TEPS > Higher Is Better
No SME .......... 593153000 |==================================================
AMD SME Enabled . 572510000 |================================================


libavif avifenc 0.11
Encoder Speed: 6
Seconds < Lower Is Better
No SME .......... 2.393 |====================================================
AMD SME Enabled . 2.469 |======================================================


Xmrig 6.18.1
Variant: Monero - Hash Count: 1M
H/s > Higher Is Better
No SME .......... 105141.7 |===================================================
AMD SME Enabled . 101932.1 |=================================================


PyHPC Benchmarks 3.0
Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing
Seconds < Lower Is Better
No SME .......... 1.691 |====================================================
AMD SME Enabled . 1.744 |======================================================


OpenVKL 1.3.1
Benchmark: vklBenchmark ISPC
Items / Sec > Higher Is Better
No SME .......... 1322 |=======================================================
AMD SME Enabled . 1286 |======================================================


Xsbench 2017-07-06

Lookups/s > Higher Is Better
No SME .......... 29806415 |===================================================
AMD SME Enabled . 29021428 |==================================================


Neural Magic DeepSparse 1.1
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 1962.10 |====================================================
AMD SME Enabled . 1911.40 |===================================================


Timed Godot Game Engine Compilation 3.2.3
Time To Compile
Seconds < Lower Is Better
No SME .......... 34.14 |=====================================================
AMD SME Enabled . 35.04 |======================================================


Neural Magic DeepSparse 1.1
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 48.83 |=====================================================
AMD SME Enabled . 50.11 |======================================================


OSPRay Studio 0.11
Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
ms < Lower Is Better
No SME .......... 22043 |=====================================================
AMD SME Enabled . 22614 |======================================================


Timed Gem5 Compilation 21.2
Time To Compile
Seconds < Lower Is Better
No SME .......... 138.64 |====================================================
AMD SME Enabled . 142.18 |=====================================================


Neural Magic DeepSparse 1.1
Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 617.03 |=====================================================
AMD SME Enabled . 601.93 |====================================================


Neural Magic DeepSparse 1.1
Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 155.10 |====================================================
AMD SME Enabled . 158.94 |=====================================================


Xmrig 6.18.1
Variant: Wownero - Hash Count: 1M
H/s > Higher Is Better
No SME .......... 126508.3 |===================================================
AMD SME Enabled . 123484.1 |==================================================


LULESH 2.0.3

z/s > Higher Is Better
No SME .......... 59069.41 |===================================================
AMD SME Enabled . 57686.09 |==================================================


nginx 1.23.2
Connections: 500
Requests Per Second > Higher Is Better
No SME .......... 201056.69 |==================================================
AMD SME Enabled . 196386.41 |=================================================


OpenVINO 2022.2.dev
Model: Vehicle Detection FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 6.44 |======================================================
AMD SME Enabled . 6.59 |=======================================================


OpenVINO 2022.2.dev
Model: Person Detection FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 1102.15 |===================================================
AMD SME Enabled . 1127.51 |====================================================


Neural Magic DeepSparse 1.1
Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 762.70 |=====================================================
AMD SME Enabled . 745.64 |====================================================


Neural Magic DeepSparse 1.1
Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 125.53 |====================================================
AMD SME Enabled . 128.40 |=====================================================


OpenVINO 2022.2.dev
Model: Person Detection FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 43.29 |======================================================
AMD SME Enabled . 42.33 |=====================================================


OpenVINO 2022.2.dev
Model: Vehicle Detection FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 7437.73 |====================================================
AMD SME Enabled . 7274.98 |===================================================


Neural Magic DeepSparse 1.1
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 1204.85 |====================================================
AMD SME Enabled . 1179.00 |===================================================


Neural Magic DeepSparse 1.1
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 79.49 |=====================================================
AMD SME Enabled . 81.23 |======================================================


Neural Magic DeepSparse 1.1
Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 111.46 |====================================================
AMD SME Enabled . 113.83 |=====================================================


Neural Magic DeepSparse 1.1
Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 858.47 |=====================================================
AMD SME Enabled . 840.94 |====================================================


oneDNN 3.0
Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU
ms < Lower Is Better
No SME .......... 22.68 |=====================================================
AMD SME Enabled . 23.14 |======================================================


Kvazaar 2.1
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Frames Per Second > Higher Is Better
No SME .......... 77.68 |======================================================
AMD SME Enabled . 76.22 |=====================================================


Rodinia 3.1
Test: OpenMP CFD Solver
Seconds < Lower Is Better
No SME .......... 5.938 |=====================================================
AMD SME Enabled . 6.043 |======================================================


OpenVINO 2022.2.dev
Model: Person Detection FP32 - Device: CPU
FPS > Higher Is Better
No SME .......... 42.76 |======================================================
AMD SME Enabled . 42.03 |=====================================================


Timed LLVM Compilation 13.0
Build System: Ninja
Seconds < Lower Is Better
No SME .......... 75.33 |=====================================================
AMD SME Enabled . 76.63 |======================================================


OpenVINO 2022.2.dev
Model: Person Detection FP32 - Device: CPU
ms < Lower Is Better
No SME .......... 1115.56 |===================================================
AMD SME Enabled . 1134.68 |====================================================


libavif avifenc 0.11
Encoder Speed: 2
Seconds < Lower Is Better
No SME .......... 34.69 |=====================================================
AMD SME Enabled . 35.26 |======================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM
eNb Mb/s > Higher Is Better
No SME .......... 415.1 |======================================================
AMD SME Enabled . 408.5 |=====================================================


Timed Linux Kernel Compilation 6.1
Build: defconfig
Seconds < Lower Is Better
No SME .......... 25.71 |======================================================
AMD SME Enabled . 25.30 |=====================================================


oneDNN 3.0
Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
No SME .......... 0.863726 |===================================================
AMD SME Enabled . 0.850418 |==================================================


Timed LLVM Compilation 13.0
Build System: Unix Makefiles
Seconds < Lower Is Better
No SME .......... 160.13 |====================================================
AMD SME Enabled . 162.63 |=====================================================


Renaissance 0.14
Test: In-Memory Database Shootout
ms < Lower Is Better
No SME .......... 4764.6 |====================================================
AMD SME Enabled . 4838.5 |=====================================================


KTX-Software toktx 4.0
Settings: Zstd Compression 9
Seconds < Lower Is Better
No SME .......... 2.734 |=====================================================
AMD SME Enabled . 2.776 |======================================================


Embree 3.13
Binary: Pathtracer ISPC - Model: Crown
Frames Per Second > Higher Is Better
No SME .......... 183.25 |=====================================================
AMD SME Enabled . 180.57 |====================================================


Timed Linux Kernel Compilation 6.1
Build: allmodconfig
Seconds < Lower Is Better
No SME .......... 146.33 |====================================================
AMD SME Enabled . 148.44 |=====================================================


OpenVINO 2022.2.dev
Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU
FPS > Higher Is Better
No SME .......... 165194.42 |=================================================
AMD SME Enabled . 167545.54 |==================================================


High Performance Conjugate Gradient 3.1

GFLOP/s > Higher Is Better
No SME .......... 88.39 |======================================================
AMD SME Enabled . 87.15 |=====================================================


OpenVINO 2022.2.dev
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 150792.42 |==================================================
AMD SME Enabled . 148736.04 |=================================================


RELION 3.1.1
Test: Basic - Device: CPU
Seconds < Lower Is Better
No SME .......... 128.66 |====================================================
AMD SME Enabled . 130.43 |=====================================================


Kvazaar 2.1
Video Input: Bosphorus 4K - Video Preset: Very Fast
Frames Per Second > Higher Is Better
No SME .......... 74.31 |======================================================
AMD SME Enabled . 73.32 |=====================================================


NAS Parallel Benchmarks 3.4
Test / Class: FT.C
Total Mop/s > Higher Is Better
No SME .......... 223096.07 |==================================================
AMD SME Enabled . 220214.75 |=================================================


NAMD 2.14
ATPase Simulation - 327,506 Atoms
days/ns < Lower Is Better
No SME .......... 0.12831 |===================================================
AMD SME Enabled . 0.12991 |====================================================


NWChem 7.0.2
Input: C240 Buckyball
Seconds < Lower Is Better
No SME .......... 1524.4 |====================================================
AMD SME Enabled . 1543.1 |=====================================================


Xcompact3d Incompact3d 2021-03-11
Input: input.i3d 193 Cells Per Direction
Seconds < Lower Is Better
No SME .......... 4.37420527 |================================================
AMD SME Enabled . 4.42424568 |=================================================


OSPRay 2.10
Benchmark: gravity_spheres_volume/dim_512/ao/real_time
Items Per Second > Higher Is Better
No SME .......... 43.85 |======================================================
AMD SME Enabled . 43.38 |=====================================================


Blender 3.4
Blend File: Barbershop - Compute: CPU-Only
Seconds < Lower Is Better
No SME .......... 80.77 |=====================================================
AMD SME Enabled . 81.57 |======================================================


Rodinia 3.1
Test: OpenMP LavaMD
Seconds < Lower Is Better
No SME .......... 16.51 |=====================================================
AMD SME Enabled . 16.67 |======================================================


WRF 4.2.2
Input: conus 2.5km
Seconds < Lower Is Better
No SME .......... 4077.19 |====================================================
AMD SME Enabled . 4116.62 |====================================================


NAS Parallel Benchmarks 3.4
Test / Class: SP.C
Total Mop/s > Higher Is Better
No SME .......... 255564.19 |==================================================
AMD SME Enabled . 253299.33 |==================================================


oneDNN 3.0
Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU
ms < Lower Is Better
No SME .......... 0.522052 |===================================================
AMD SME Enabled . 0.526628 |===================================================


Neural Magic DeepSparse 1.1
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 1133.31 |====================================================
AMD SME Enabled . 1143.01 |====================================================


x265 3.4
Video Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No SME .......... 23.48 |======================================================
AMD SME Enabled . 23.29 |======================================================


GPAW 22.1
Input: Carbon Nanotube
Seconds < Lower Is Better
No SME .......... 23.17 |======================================================
AMD SME Enabled . 22.98 |======================================================


Neural Magic DeepSparse 1.1
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
No SME .......... 1134.42 |====================================================
AMD SME Enabled . 1143.04 |====================================================


Neural Magic DeepSparse 1.1
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 84.25 |======================================================
AMD SME Enabled . 83.64 |======================================================


7-Zip Compression 22.01
Test: Decompression Rating
MIPS > Higher Is Better
No SME .......... 1160632 |====================================================
AMD SME Enabled . 1169038 |====================================================


OpenRadioss 2022.10.13
Model: Cell Phone Drop Test
Seconds < Lower Is Better
No SME .......... 18.45 |======================================================
AMD SME Enabled . 18.32 |======================================================


PostgreSQL 15
Scaling Factor: 100 - Clients: 250 - Mode: Read Only
TPS > Higher Is Better
No SME .......... 2951147 |====================================================
AMD SME Enabled . 2970869 |====================================================


TensorFlow 2.10
Device: CPU - Batch Size: 64 - Model: AlexNet
images/sec > Higher Is Better
No SME .......... 508.40 |=====================================================
AMD SME Enabled . 505.26 |=====================================================


srsRAN 22.04.1
Test: OFDM_Test
Samples / Second > Higher Is Better
No SME .......... 161733333 |==================================================
AMD SME Enabled . 162633333 |==================================================


Neural Magic DeepSparse 1.1
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
No SME .......... 84.28 |======================================================
AMD SME Enabled . 83.82 |======================================================


srsRAN 22.04.1
Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM
UE Mb/s > Higher Is Better
No SME .......... 94.4 |=======================================================
AMD SME Enabled . 94.9 |=======================================================


miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM2
Billion Interactions/s > Higher Is Better
No SME .......... 345.34 |=====================================================
AMD SME Enabled . 343.55 |=====================================================


miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM2
GFInst/s > Higher Is Better
No SME .......... 8633.54 |====================================================
AMD SME Enabled . 8588.75 |====================================================


OpenVINO 2022.2.dev
Model: Weld Porosity Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
No SME .......... 9.62 |=======================================================
AMD SME Enabled . 9.67 |=======================================================


Renaissance 0.14
Test: Finagle HTTP Requests
ms < Lower Is Better
No SME .......... 12286.3 |====================================================
AMD SME Enabled . 12347.5 |====================================================


OpenVINO 2022.2.dev
Model: Machine Translation EN To DE FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 967.90 |=====================================================
AMD SME Enabled . 963.14 |=====================================================


OpenVINO 2022.2.dev
Model: Weld Porosity Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
No SME .......... 19801.40 |===================================================
AMD SME Enabled . 19704.84 |===================================================


OpenVINO 2022.2.dev
Model: Machine Translation EN To DE FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 49.54 |======================================================
AMD SME Enabled . 49.78 |======================================================


GROMACS 2022.1
Implementation: MPI CPU - Input: water_GMX50_bare
Ns Per Day > Higher Is Better
No SME .......... 18.71 |======================================================
AMD SME Enabled . 18.62 |======================================================


Graph500 3.0
Scale: 26
bfs max_TEPS > Higher Is Better
No SME .......... 1533180000 |=================================================
AMD SME Enabled . 1526380000 |=================================================


oneDNN 3.0
Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
ms < Lower Is Better
No SME .......... 2011.15 |====================================================
AMD SME Enabled . 2002.43 |====================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM
eNb Mb/s > Higher Is Better
No SME .......... 413.9 |======================================================
AMD SME Enabled . 415.7 |======================================================


srsRAN 22.04.1
Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM
eNb Mb/s > Higher Is Better
No SME .......... 139.7 |======================================================
AMD SME Enabled . 139.1 |======================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM
eNb Mb/s > Higher Is Better
No SME .......... 444.0 |======================================================
AMD SME Enabled . 445.7 |======================================================


OpenVINO 2022.2.dev
Model: Person Vehicle Bike Detection FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 5.31 |=======================================================
AMD SME Enabled . 5.33 |=======================================================


OpenVINO 2022.2.dev
Model: Person Vehicle Bike Detection FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 9027.72 |====================================================
AMD SME Enabled . 8993.90 |====================================================


OpenVINO 2022.2.dev
Model: Face Detection FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 469.81 |=====================================================
AMD SME Enabled . 471.53 |=====================================================


Graph500 3.0
Scale: 26
sssp max_TEPS > Higher Is Better
No SME .......... 838505000 |==================================================
AMD SME Enabled . 835467000 |==================================================


OpenVINO 2022.2.dev
Model: Face Detection FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 101.90 |=====================================================
AMD SME Enabled . 101.55 |=====================================================


Zstd Compression 1.5.0
Compression Level: 19, Long Mode - Decompression Speed
MB/s > Higher Is Better
No SME .......... 3825.0 |=====================================================
AMD SME Enabled . 3837.3 |=====================================================


OSPRay 2.10
Benchmark: particle_volume/pathtracer/real_time
Items Per Second > Higher Is Better
No SME .......... 230.62 |=====================================================
AMD SME Enabled . 229.88 |=====================================================


NAS Parallel Benchmarks 3.4
Test / Class: BT.C
Total Mop/s > Higher Is Better
No SME .......... 496467.98 |==================================================
AMD SME Enabled . 494917.44 |==================================================


ONNX Runtime 1.11
Model: super-resolution-10 - Device: CPU - Executor: Standard
Inferences Per Minute > Higher Is Better
No SME .......... 5600 |=======================================================
AMD SME Enabled . 5583 |=======================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM
UE Mb/s > Higher Is Better
No SME .......... 172.7 |======================================================
AMD SME Enabled . 172.2 |======================================================


QuantLib 1.21

MFLOPS > Higher Is Better
No SME .......... 3052.8 |=====================================================
AMD SME Enabled . 3061.3 |=====================================================


oneDNN 3.0
Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
No SME .......... 0.916133 |===================================================
AMD SME Enabled . 0.918482 |===================================================


OpenFOAM 10
Input: drivaerFastback, Small Mesh Size - Execution Time
Seconds < Lower Is Better
No SME .......... 22.08 |======================================================
AMD SME Enabled . 22.13 |======================================================


Blender 3.4
Blend File: Classroom - Compute: CPU-Only
Seconds < Lower Is Better
No SME .......... 20.99 |======================================================
AMD SME Enabled . 20.95 |======================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM
UE Mb/s > Higher Is Better
No SME .......... 166.0 |======================================================
AMD SME Enabled . 165.7 |======================================================


OpenRadioss 2022.10.13
Model: Bumper Beam
Seconds < Lower Is Better
No SME .......... 79.85 |======================================================
AMD SME Enabled . 79.97 |======================================================


ASTC Encoder 4.0
Preset: Exhaustive
MT/s > Higher Is Better
No SME .......... 11.82 |======================================================
AMD SME Enabled . 11.84 |======================================================


ACES DGEMM 1.0
Sustained Floating-Point Rate
GFLOP/s > Higher Is Better
No SME .......... 70.37 |======================================================
AMD SME Enabled . 70.28 |======================================================


miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM1
GFInst/s > Higher Is Better
No SME .......... 7281.28 |====================================================
AMD SME Enabled . 7290.64 |====================================================


miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM1
Billion Interactions/s > Higher Is Better
No SME .......... 291.25 |=====================================================
AMD SME Enabled . 291.63 |=====================================================


ASTC Encoder 4.0
Preset: Thorough
MT/s > Higher Is Better
No SME .......... 106.42 |=====================================================
AMD SME Enabled . 106.56 |=====================================================


Liquid-DSP 2021.01.31
Threads: 256 - Buffer Length: 256 - Filter Length: 57
samples/s > Higher Is Better
No SME .......... 10332000000 |================================================
AMD SME Enabled . 10344666667 |================================================


OpenVINO 2022.2.dev
Model: Face Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
No SME .......... 246.95 |=====================================================
AMD SME Enabled . 247.23 |=====================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM
eNb Mb/s > Higher Is Better
No SME .......... 445.2 |======================================================
AMD SME Enabled . 444.8 |======================================================


OpenVINO 2022.2.dev
Model: Weld Porosity Detection FP16 - Device: CPU
FPS > Higher Is Better
No SME .......... 9997.68 |====================================================
AMD SME Enabled . 9990.19 |====================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM
UE Mb/s > Higher Is Better
No SME .......... 157.8 |======================================================
AMD SME Enabled . 157.7 |======================================================


OpenVINO 2022.2.dev
Model: Face Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
No SME .......... 193.95 |=====================================================
AMD SME Enabled . 193.83 |=====================================================


srsRAN 22.04.1
Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM
UE Mb/s > Higher Is Better
No SME .......... 165.8 |======================================================
AMD SME Enabled . 165.9 |======================================================


Liquid-DSP 2021.01.31
Threads: 384 - Buffer Length: 256 - Filter Length: 57
samples/s > Higher Is Better
No SME .......... 10346000000 |================================================
AMD SME Enabled . 10350000000 |================================================


OpenVINO 2022.2.dev
Model: Vehicle Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
No SME .......... 11180.63 |===================================================
AMD SME Enabled . 11184.76 |===================================================


NAS Parallel Benchmarks 3.4
Test / Class: EP.C
Total Mop/s > Higher Is Better
No SME .......... 16457.94 |===================================================
AMD SME Enabled . 16462.35 |===================================================


OpenRadioss 2022.10.13
Model: INIVOL and Fluid Structure Interaction Drop Container
Seconds < Lower Is Better
No SME .......... 80.88 |======================================================
AMD SME Enabled . 80.90 |======================================================


OpenVINO 2022.2.dev
Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU
ms < Lower Is Better
No SME .......... 0.36 |=======================================================
AMD SME Enabled . 0.36 |=======================================================


OpenVINO 2022.2.dev
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 0.55 |=======================================================
AMD SME Enabled . 0.55 |=======================================================


OpenVINO 2022.2.dev
Model: Weld Porosity Detection FP16 - Device: CPU
ms < Lower Is Better
No SME .......... 4.79 |=======================================================
AMD SME Enabled . 4.79 |=======================================================


OpenVINO 2022.2.dev
Model: Vehicle Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
No SME .......... 4.28 |=======================================================
AMD SME Enabled . 4.28 |=======================================================


Intel Open Image Denoise 1.4.0
Run: RTLightmap.hdr.4096x4096
Images / Sec > Higher Is Better
No SME .......... 1.66 |=======================================================
AMD SME Enabled . 1.66 |=======================================================


oneDNN 3.0
Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
No SME .......... 3.89299 |====================================================
AMD SME Enabled . 3.92313 |====================================================


SVT-AV1 1.4
Encoder Mode: Preset 13 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No SME .......... 248.33 |====================================================
AMD SME Enabled . 251.44 |=====================================================


Zstd Compression 1.5.0
Compression Level: 19, Long Mode - Compression Speed
MB/s > Higher Is Better
No SME .......... 52.9 |=======================================================
AMD SME Enabled . 49.9 |====================================================