EPYC 7502 AOCC 2.3 Compiler Comparison

AMD EPYC 7502 testing of various benchmarks under AMD AOCC 2.3, GCC 10.2, LLVM Clang 11. CFLAGS/CXXFLAGS of "-O3 -march=znver2" throughout. Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2012080-HA-EPYC7502A97
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

AV1 2 Tests
Chess Test Suite 2 Tests
C/C++ Compiler Tests 21 Tests
Compression Tests 2 Tests
CPU Massive 19 Tests
Creator Workloads 17 Tests
Cryptography 2 Tests
Database Test Suite 3 Tests
Encoding 7 Tests
Game Development 2 Tests
HPC - High Performance Computing 6 Tests
Imaging 4 Tests
Common Kernel Benchmarks 3 Tests
Machine Learning 4 Tests
Multi-Core 13 Tests
Programmer / Developer System Benchmarks 2 Tests
Renderers 2 Tests
Server 5 Tests
Server CPU Tests 12 Tests
Single-Threaded 6 Tests
Texture Compression 2 Tests
Video Encoding 6 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
GCC 10.2
December 07 2020
  3 Hours, 49 Minutes
LLVM Clang 11
December 08 2020
  6 Hours, 56 Minutes
AMD AOCC 2.3
December 07 2020
  3 Hours, 58 Minutes
Invert Hiding All Results Option
  4 Hours, 54 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

EPYC 7502 AOCC 2.3 Compiler Comparison AMD EPYC 7502 testing of various benchmarks under AMD AOCC 2.3, GCC 10.2, LLVM Clang 11. CFLAGS/CXXFLAGS of "-O3 -march=znver2" throughout. Benchmarks by Michael Larabel for a future article. GCC 10.2: Processor: AMD EPYC 7502 32-Core @ 2.50GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 280GB INTEL SSDPED1D280GA, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel I350 OS: Ubuntu 20.10, Kernel: 5.8.0-31-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1080 LLVM Clang 11: Processor: AMD EPYC 7502 32-Core @ 2.50GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 280GB INTEL SSDPED1D280GA, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel I350 OS: Ubuntu 20.10, Kernel: 5.8.0-31-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, Compiler: Clang 11.0.0-2Target:, File-System: ext4, Screen Resolution: 1920x1080 AMD AOCC 2.3: Processor: AMD EPYC 7502 32-Core @ 2.50GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 280GB INTEL SSDPED1D280GA, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel I350 OS: Ubuntu 20.10, Kernel: 5.8.0-31-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, Compiler: Clang 11.0.0, File-System: ext4, Screen Resolution: 1920x1080 NCNN 20200916 Target: CPU - Model: blazeface ms < Lower Is Better GCC 10.2 ...... 3.99 |========================================================= LLVM Clang 11 . 2.84 |========================================= AMD AOCC 2.3 .. 2.19 |=============================== NCNN 20200916 Target: CPU - Model: mnasnet ms < Lower Is Better GCC 10.2 ...... 8.95 |========================================================= LLVM Clang 11 . 6.26 |======================================== AMD AOCC 2.3 .. 5.01 |================================ C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better GCC 10.2 ...... 18.92 |================================ LLVM Clang 11 . 30.65 |==================================================== AMD AOCC 2.3 .. 33.28 |======================================================== oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 254.61 |======================================================= LLVM Clang 11 . 172.51 |===================================== AMD AOCC 2.3 .. 147.52 |================================ NCNN 20200916 Target: CPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better GCC 10.2 ...... 8.88 |========================================================= LLVM Clang 11 . 6.87 |============================================ AMD AOCC 2.3 .. 5.20 |================================= NCNN 20200916 Target: CPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better GCC 10.2 ...... 9.37 |========================================================= LLVM Clang 11 . 6.88 |========================================== AMD AOCC 2.3 .. 5.62 |================================== NCNN 20200916 Target: CPU - Model: efficientnet-b0 ms < Lower Is Better GCC 10.2 ...... 11.32 |======================================================== LLVM Clang 11 . 8.77 |=========================================== AMD AOCC 2.3 .. 6.94 |================================== dav1d 0.7.0 Video Input: Chimera 1080p 10-bit FPS > Higher Is Better GCC 10.2 ...... 143.02 |======================================================= LLVM Clang 11 . 92.56 |==================================== AMD AOCC 2.3 .. 122.39 |=============================================== Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image Test Cases Per Minute > Higher Is Better GCC 10.2 ...... 18452.62 |===================================================== LLVM Clang 11 . 11946.57 |================================== AMD AOCC 2.3 .. 13720.16 |======================================= LibRaw 0.20 Post-Processing Benchmark Mpix/sec > Higher Is Better GCC 10.2 ...... 52.54 |======================================================== LLVM Clang 11 . 36.98 |======================================= AMD AOCC 2.3 .. 38.47 |========================================= SVT-AV1 0.8 Encoder Mode: Enc Mode 0 - Input: 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 0.103 |======================================== LLVM Clang 11 . 0.145 |======================================================== AMD AOCC 2.3 .. 0.146 |======================================================== GraphicsMagick 1.3.33 Operation: Sharpen Iterations Per Minute > Higher Is Better GCC 10.2 ...... 434 |========================================================== LLVM Clang 11 . 384 |=================================================== AMD AOCC 2.3 .. 316 |========================================== OpenSSL 1.1.1 RSA 4096-bit Performance Signs Per Second > Higher Is Better GCC 10.2 ...... 7395.4 |======================================================= LLVM Clang 11 . 5412.5 |======================================== AMD AOCC 2.3 .. 5413.5 |======================================== Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster Test Cases Per Minute > Higher Is Better GCC 10.2 ...... 890.71 |======================================================= LLVM Clang 11 . 674.01 |========================================== AMD AOCC 2.3 .. 678.51 |========================================== SVT-AV1 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 6.775 |============================================ LLVM Clang 11 . 8.591 |======================================================== AMD AOCC 2.3 .. 8.657 |======================================================== SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 55.71 |============================================ LLVM Clang 11 . 70.23 |======================================================== AMD AOCC 2.3 .. 70.70 |======================================================== GraphicsMagick 1.3.33 Operation: Resizing Iterations Per Minute > Higher Is Better GCC 10.2 ...... 2092 |========================================================= LLVM Clang 11 . 1827 |================================================== AMD AOCC 2.3 .. 1653 |============================================= NCNN 20200916 Target: CPU - Model: googlenet ms < Lower Is Better GCC 10.2 ...... 20.85 |======================================================== LLVM Clang 11 . 18.88 |=================================================== AMD AOCC 2.3 .. 16.51 |============================================ LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better GCC 10.2 ...... 8.798 |============================================ LLVM Clang 11 . 10.078 |================================================== AMD AOCC 2.3 .. 11.026 |======================================================= oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 0.532944 |================================================= LLVM Clang 11 . 0.575864 |===================================================== AMD AOCC 2.3 .. 0.464235 |=========================================== oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 1.67503 |===================================================== LLVM Clang 11 . 1.72247 |====================================================== AMD AOCC 2.3 .. 1.40879 |============================================ TNN 0.2.3 Target: CPU - Model: MobileNet v2 ms < Lower Is Better GCC 10.2 ...... 324.17 |============================================= LLVM Clang 11 . 392.20 |======================================================= AMD AOCC 2.3 .. 365.39 |=================================================== ASTC Encoder 2.0 Preset: Medium Seconds < Lower Is Better GCC 10.2 ...... 7.00 |========================================================= LLVM Clang 11 . 6.05 |================================================= AMD AOCC 2.3 .. 5.81 |=============================================== NCNN 20200916 Target: CPU - Model: resnet50 ms < Lower Is Better GCC 10.2 ...... 23.70 |======================================================== LLVM Clang 11 . 22.23 |===================================================== AMD AOCC 2.3 .. 19.70 |=============================================== NCNN 20200916 Target: CPU - Model: squeezenet ms < Lower Is Better GCC 10.2 ...... 17.29 |======================================================== LLVM Clang 11 . 15.35 |================================================== AMD AOCC 2.3 .. 14.38 |=============================================== GraphicsMagick 1.3.33 Operation: Enhanced Iterations Per Minute > Higher Is Better GCC 10.2 ...... 590 |======================================================= LLVM Clang 11 . 627 |========================================================== AMD AOCC 2.3 .. 526 |================================================= NCNN 20200916 Target: CPU - Model: vgg16 ms < Lower Is Better GCC 10.2 ...... 30.89 |=============================================== LLVM Clang 11 . 36.60 |======================================================== AMD AOCC 2.3 .. 32.03 |================================================= ASTC Encoder 2.0 Preset: Thorough Seconds < Lower Is Better GCC 10.2 ...... 9.61 |========================================================= LLVM Clang 11 . 8.32 |================================================= AMD AOCC 2.3 .. 8.16 |================================================ WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression Encode Time - Seconds < Lower Is Better GCC 10.2 ...... 8.861 |======================================================== LLVM Clang 11 . 7.655 |================================================ AMD AOCC 2.3 .. 7.546 |================================================ oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 1.95695 |====================================================== LLVM Clang 11 . 1.69390 |=============================================== AMD AOCC 2.3 .. 1.66772 |============================================== oneDNN 1.5 Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 1.17556 |====================================================== LLVM Clang 11 . 1.03137 |=============================================== AMD AOCC 2.3 .. 1.00926 |============================================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better GCC 10.2 ...... 35.78 |================================================ LLVM Clang 11 . 41.66 |======================================================== AMD AOCC 2.3 .. 37.76 |=================================================== oneDNN 1.5 Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 13.88 |======================================================== LLVM Clang 11 . 13.46 |====================================================== AMD AOCC 2.3 .. 11.92 |================================================ NCNN 20200916 Target: CPU - Model: mobilenet ms < Lower Is Better GCC 10.2 ...... 19.46 |======================================================== LLVM Clang 11 . 17.85 |=================================================== AMD AOCC 2.3 .. 17.02 |================================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better GCC 10.2 ...... 1007283 |================================================ LLVM Clang 11 . 1143642 |====================================================== AMD AOCC 2.3 .. 1138442 |====================================================== VP9 libvpx Encoding 1.8.2 Speed: Speed 0 Frames Per Second > Higher Is Better GCC 10.2 ...... 6.27 |====================================================== LLVM Clang 11 . 5.96 |=================================================== AMD AOCC 2.3 .. 6.65 |========================================================= ASTC Encoder 2.0 Preset: Exhaustive Seconds < Lower Is Better GCC 10.2 ...... 73.46 |======================================================== LLVM Clang 11 . 67.17 |=================================================== AMD AOCC 2.3 .. 66.04 |================================================== Basis Universal 1.12 Settings: UASTC Level 2 + RDO Post-Processing Seconds < Lower Is Better GCC 10.2 ...... 755.52 |================================================== LLVM Clang 11 . 837.54 |======================================================= AMD AOCC 2.3 .. 833.54 |======================================================= VP9 libvpx Encoding 1.8.2 Speed: Speed 5 Frames Per Second > Higher Is Better GCC 10.2 ...... 19.03 |===================================================== LLVM Clang 11 . 18.23 |=================================================== AMD AOCC 2.3 .. 20.12 |======================================================== NCNN 20200916 Target: CPU - Model: resnet18 ms < Lower Is Better GCC 10.2 ...... 13.12 |======================================================== LLVM Clang 11 . 13.19 |======================================================== AMD AOCC 2.3 .. 11.99 |=================================================== oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 2.17957 |====================================================== LLVM Clang 11 . 2.05539 |=================================================== AMD AOCC 2.3 .. 1.99141 |================================================= Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping Test Cases Per Minute > Higher Is Better GCC 10.2 ...... 874.54 |=================================================== LLVM Clang 11 . 945.32 |======================================================= AMD AOCC 2.3 .. 923.68 |====================================================== Stockfish 12 Total Time Nodes Per Second > Higher Is Better GCC 10.2 ...... 58347386 |================================================== LLVM Clang 11 . 62434784 |===================================================== AMD AOCC 2.3 .. 62375605 |===================================================== LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed MB/s > Higher Is Better GCC 10.2 ...... 45.57 |===================================================== LLVM Clang 11 . 48.40 |======================================================== AMD AOCC 2.3 .. 48.52 |======================================================== oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 1.94380 |===================================================== LLVM Clang 11 . 1.98927 |====================================================== AMD AOCC 2.3 .. 1.87485 |=================================================== SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 348.24 |==================================================== LLVM Clang 11 . 365.57 |======================================================= AMD AOCC 2.3 .. 368.17 |======================================================= GraphicsMagick 1.3.33 Operation: Swirl Iterations Per Minute > Higher Is Better GCC 10.2 ...... 1295 |====================================================== LLVM Clang 11 . 1323 |======================================================= AMD AOCC 2.3 .. 1368 |========================================================= x265 3.4 Video Input: Bosphorus 4K Frames Per Second > Higher Is Better GCC 10.2 ...... 22.83 |====================================================== LLVM Clang 11 . 22.44 |===================================================== AMD AOCC 2.3 .. 23.69 |======================================================== WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression Encode Time - Seconds < Lower Is Better GCC 10.2 ...... 44.38 |======================================================== LLVM Clang 11 . 42.38 |===================================================== AMD AOCC 2.3 .. 42.70 |====================================================== RNNoise 2020-06-28 Seconds < Lower Is Better GCC 10.2 ...... 21.72 |======================================================== LLVM Clang 11 . 21.54 |======================================================== AMD AOCC 2.3 .. 20.76 |====================================================== NCNN 20200916 Target: CPU - Model: yolov4-tiny ms < Lower Is Better GCC 10.2 ...... 29.47 |====================================================== LLVM Clang 11 . 30.70 |======================================================== AMD AOCC 2.3 .. 29.38 |====================================================== PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency ms < Lower Is Better GCC 10.2 ...... 0.096 |======================================================== LLVM Clang 11 . 0.094 |======================================================= AMD AOCC 2.3 .. 0.092 |====================================================== Zstd Compression 1.4.5 Compression Level: 19 MB/s > Higher Is Better GCC 10.2 ...... 109.9 |====================================================== LLVM Clang 11 . 111.2 |====================================================== AMD AOCC 2.3 .. 114.6 |======================================================== SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 279.73 |===================================================== LLVM Clang 11 . 286.24 |====================================================== AMD AOCC 2.3 .. 291.22 |======================================================= LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed MB/s > Higher Is Better GCC 10.2 ...... 9459.69 |==================================================== LLVM Clang 11 . 9838.54 |====================================================== AMD AOCC 2.3 .. 9780.39 |====================================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better GCC 10.2 ...... 2759.35 |====================================================== LLVM Clang 11 . 2673.79 |==================================================== AMD AOCC 2.3 .. 2779.62 |====================================================== SQLite Speedtest 3.30 Timed Time - Size 1,000 Seconds < Lower Is Better GCC 10.2 ...... 75.14 |====================================================== LLVM Clang 11 . 77.64 |======================================================== AMD AOCC 2.3 .. 78.07 |======================================================== Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis Seconds < Lower Is Better GCC 10.2 ...... 94.28 |======================================================== LLVM Clang 11 . 93.84 |======================================================== AMD AOCC 2.3 .. 90.76 |====================================================== PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only TPS > Higher Is Better GCC 10.2 ...... 521332 |===================================================== LLVM Clang 11 . 530684 |====================================================== AMD AOCC 2.3 .. 541314 |======================================================= x264 2019-12-17 H.264 Video Encoding Frames Per Second > Higher Is Better GCC 10.2 ...... 149.35 |====================================================== LLVM Clang 11 . 146.43 |===================================================== AMD AOCC 2.3 .. 151.92 |======================================================= dav1d 0.7.0 Video Input: Summer Nature 1080p FPS > Higher Is Better GCC 10.2 ...... 567.42 |===================================================== LLVM Clang 11 . 584.65 |======================================================= AMD AOCC 2.3 .. 588.62 |======================================================= PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average Latency ms < Lower Is Better GCC 10.2 ...... 0.268 |======================================================== LLVM Clang 11 . 0.265 |======================================================= AMD AOCC 2.3 .. 0.259 |====================================================== PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write TPS > Higher Is Better GCC 10.2 ...... 3739 |======================================================= LLVM Clang 11 . 3772 |======================================================== AMD AOCC 2.3 .. 3865 |========================================================= LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed MB/s > Higher Is Better GCC 10.2 ...... 44.72 |======================================================= LLVM Clang 11 . 44.30 |====================================================== AMD AOCC 2.3 .. 45.76 |======================================================== SVT-VP9 0.1 Tuning: VMAF Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 354.98 |===================================================== LLVM Clang 11 . 363.72 |======================================================= AMD AOCC 2.3 .. 366.61 |======================================================= x265 3.4 Video Input: Bosphorus 1080p Frames Per Second > Higher Is Better GCC 10.2 ...... 49.05 |====================================================== LLVM Clang 11 . 50.02 |======================================================= AMD AOCC 2.3 .. 50.60 |======================================================== PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average Latency ms < Lower Is Better GCC 10.2 ...... 0.035 |======================================================== LLVM Clang 11 . 0.035 |======================================================== AMD AOCC 2.3 .. 0.034 |====================================================== libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput Megapixels/sec > Higher Is Better GCC 10.2 ...... 171.89 |===================================================== LLVM Clang 11 . 174.99 |====================================================== AMD AOCC 2.3 .. 176.87 |======================================================= Crypto++ 8.2 Test: Unkeyed Algorithms MiB/second > Higher Is Better GCC 10.2 ...... 305.90 |====================================================== LLVM Clang 11 . 314.39 |======================================================= AMD AOCC 2.3 .. 312.64 |======================================================= PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only TPS > Higher Is Better GCC 10.2 ...... 28919 |======================================================= LLVM Clang 11 . 28921 |======================================================= AMD AOCC 2.3 .. 29645 |======================================================== TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 ms < Lower Is Better GCC 10.2 ...... 305.13 |====================================================== LLVM Clang 11 . 311.68 |======================================================= AMD AOCC 2.3 .. 304.18 |====================================================== NGINX Benchmark 1.9.9 Static Web Page Serving Requests Per Second > Higher Is Better GCC 10.2 ...... 30658.67 |==================================================== LLVM Clang 11 . 30676.68 |==================================================== AMD AOCC 2.3 .. 31368.86 |===================================================== dav1d 0.7.0 Video Input: Summer Nature 4K FPS > Higher Is Better GCC 10.2 ...... 269.72 |====================================================== LLVM Clang 11 . 275.41 |======================================================= AMD AOCC 2.3 .. 274.20 |======================================================= GraphicsMagick 1.3.33 Operation: Rotate Iterations Per Minute > Higher Is Better GCC 10.2 ...... 535 |========================================================== LLVM Clang 11 . 527 |========================================================= AMD AOCC 2.3 .. 525 |========================================================= dav1d 0.7.0 Video Input: Chimera 1080p FPS > Higher Is Better GCC 10.2 ...... 564.96 |====================================================== LLVM Clang 11 . 572.19 |======================================================= AMD AOCC 2.3 .. 575.22 |======================================================= oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 3.68073 |===================================================== LLVM Clang 11 . 3.72354 |====================================================== AMD AOCC 2.3 .. 3.66096 |===================================================== Basis Universal 1.12 Settings: UASTC Level 3 Seconds < Lower Is Better GCC 10.2 ...... 25.52 |======================================================== LLVM Clang 11 . 25.32 |======================================================== AMD AOCC 2.3 .. 25.21 |======================================================= Basis Universal 1.12 Settings: UASTC Level 2 Seconds < Lower Is Better GCC 10.2 ...... 16.51 |======================================================== LLVM Clang 11 . 16.49 |======================================================== AMD AOCC 2.3 .. 16.32 |======================================================= PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write TPS > Higher Is Better GCC 10.2 ...... 3453 |========================================================= LLVM Clang 11 . 3443 |========================================================= AMD AOCC 2.3 .. 3413 |======================================================== PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency ms < Lower Is Better GCC 10.2 ...... 14.49 |======================================================= LLVM Clang 11 . 14.53 |======================================================== AMD AOCC 2.3 .. 14.66 |======================================================== Zstd Compression 1.4.5 Compression Level: 3 MB/s > Higher Is Better GCC 10.2 ...... 7849.6 |====================================================== LLVM Clang 11 . 7866.2 |======================================================= AMD AOCC 2.3 .. 7937.8 |======================================================= WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless Encode Time - Seconds < Lower Is Better GCC 10.2 ...... 20.80 |======================================================== LLVM Clang 11 . 20.72 |======================================================== AMD AOCC 2.3 .. 20.90 |======================================================== Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better GCC 10.2 ...... 291925951.64 |================================================= LLVM Clang 11 . 292874904.54 |================================================= AMD AOCC 2.3 .. 294314450.62 |================================================= NCNN 20200916 Target: CPU - Model: alexnet ms < Lower Is Better GCC 10.2 ...... 9.46 |================================================= LLVM Clang 11 . 10.83 |======================================================== AMD AOCC 2.3 .. 8.94 |============================================== NCNN 20200916 Target: CPU - Model: shufflenet-v2 ms < Lower Is Better GCC 10.2 ...... 10.59 |======================================================== LLVM Clang 11 . 7.35 |======================================= AMD AOCC 2.3 .. 6.25 |================================= Redis 6.0.9 Test: SET Requests Per Second > Higher Is Better GCC 10.2 ...... 1369757.13 |=============================================== LLVM Clang 11 . 1483322.31 |=================================================== AMD AOCC 2.3 .. 1446872.35 |================================================== Redis 6.0.9 Test: GET Requests Per Second > Higher Is Better GCC 10.2 ...... 1809976.83 |=========================================== LLVM Clang 11 . 2122749.43 |=================================================== AMD AOCC 2.3 .. 1874175.91 |============================================= Redis 6.0.9 Test: LPUSH Requests Per Second > Higher Is Better GCC 10.2 ...... 1212068.06 |============================================= LLVM Clang 11 . 1304842.23 |================================================ AMD AOCC 2.3 .. 1380024.29 |=================================================== oneDNN 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better GCC 10.2 ...... 79.25 |======================================================== LLVM Clang 11 . 64.99 |============================================== AMD AOCC 2.3 .. 30.72 |====================== Geometric Mean Of All Test Results Result Composite - EPYC 7502 AOCC 2.3 Compiler Comparison Geometric Mean > Higher Is Better GCC 10.2 ...... 113.07 |=================================================== LLVM Clang 11 . 115.39 |==================================================== AMD AOCC 2.3 .. 121.37 |======================================================= Number Of First Place Finishes Wins - 89 Tests Wins AMD AOCC 2.3 .. 61 |=========================================================== GCC 10.2 ...... 17 |================ LLVM Clang 11 . 11 |=========== Number Of Last Place Finishes Losses - 89 Tests Losses GCC 10.2 ...... 56 |=========================================================== LLVM Clang 11 . 23 |======================== AMD AOCC 2.3 .. 10 |===========