Antares: an automatic engine for multi-platform kernel generation and optimization. Supporting CPU, CUDA, ROCm, DirectX12, GraphCore, SYCL for CPU/GPU, OpenCL for AMD/NVIDIA, Android CPU/GPU backends.

Darknet on OpenCL Convolutional Neural Networks on OpenCL on Intel & NVidia & AMD & Mali GPUs for macOS & GNU/Linux & Windows & FreeBSD

Fix OpenCL detection and performance issues on AMD Radeon GPU's by registering Installable Client Driver (ICD) files.

Implementing CNN code in CUDA and OpenCL to evaluate its performance on NVIDIA GPUs, AMD GPUs, and an FPGA platform.

OpenCL Lyra2REv2/v3 miner for AMD GCN GPUs

OpenCL Miner for Autolykos v2 (Ergo) for AMD GPUs

Amoveo mining with OpenCL for NVidia and AMD GPU.

Working Modified Caffe for AMD GPU's OpenCL for Windows 10 .

RL Forex Bot with PPO AI-powered trading bot using PPO, dynamic position sizing, AMD GPU acceleration (DirectML/OpenCL), and MetaTrader 5 integration.

Heterogeneous Ethereum Miner with support for AMD, Intel and Nvidia GPUs using SYCL, OpenCL and CUDA backends

⚡ AV2 video encoding accelerated 3-5x via SYCL/CUDA/OpenCL — runs on NVIDIA, AMD & Intel GPUs

A high quality real time re-sampler DSP for Foobar2000. Based on band limited interpolation theory and using a lookup table to speed up the calculation of Sinc function. GPGPU programming is based on OpenCL and defaultly for AMD GPU but can be quickly port to other GPU. THD+N is about 0.0004% while converting from 44.1K to 96K.

Wolf's OpenCL XMR Miner for AMD GPUs

wolf-xmr-miner is Wolf's OpenCL XMR Miner for AMD GPUs

Data Science: AMD/OpenCL GPU Deep Learning: Setup Python + Caffe/XGBoost + 1.7x RAM

CURRENNT is a machine learning library for Recurrent Neural Networks (RNNs). Original code is CUDA enabled. It's now able to run with OpenCL (support Nvidia, AMD, Intel GPU)

Enablement for Caffe (OpenCL) on AMD GPUs

FASTMiner-Equi is miner for equihash currently support only equihash 150/5 beamhash 2, miner with focus on AMD GPUs (OpenCL based).

My instructions for installing OpenCL development environments for all CPUs and AMD GPUs on Linux distros, as I've found a lot of existing online instructions to not be that reliable for properly setting up a OpenCL development on a Linux distro such as the Ubuntu one which I am running.

An efficient routine was developed and implemented to calculate the cross correlation among the radio waves samples captured by the array of telescopes. The GPU used was AMD HD6990 on which using openCL, a performance of about 600GFlops was obtained

GPU加速检测工具，本工具是一套用于检测计算机系统中GPU加速能力的Python程序集合，适用于从普通用户到专业开发者的各个层次用户。它能全面检测系统中可用的GPU加速方案，包括NVIDIA CUDA、AMD ROCm、Apple Metal、OpenCL、Intel oneAPI等多种技术。

21:55:54: Open Broadcaster Software v0.65b - 32bit (´・ω・｀) 21:55:54: ------------------------------- 21:55:54: CPU Name: AMD FX(tm)-4170 Quad-Core Processor 21:55:54: CPU Speed: 4200MHz 21:55:54: Physical Memory: 4095MB Total, 4095MB Free 21:55:54: stepping id: 2, model 1, family 21, type 0, extmodel 0, extfamily 3, HTT 1, logical cores 4, total cores 2 21:55:54: monitor 1: pos={0, 0}, size={1920, 1080} 21:55:54: Windows Version: 6.2 Build 9200 21:55:54: Aero is Enabled 21:55:54: ------------------------------- 21:55:54: OBS Modules: 21:55:54: Base Address Module 21:55:54: 01320000 OBS.exe 21:55:54: 59AF0000 OBSApi.dll 21:55:54: 5D440000 DShowPlugin.dll 21:55:54: 59900000 GraphicsCapture.dll 21:55:54: 598D0000 NoiseGate.dll 21:55:54: 5D3D0000 PSVPlugin.dll 21:55:54: 598B0000 scenesw.dll 21:55:54: ------------------------------------------ 21:55:54: Adapter 1 21:55:54: Video Adapter: AMD Radeon R9 200 Series 21:55:54: Video Adapter Dedicated Video Memory: 2125631488 21:55:54: Video Adapter Shared System Memory: 2169270272 21:55:54: Video Adapter Output 1: pos={0, 0}, size={1920, 1080}, attached=true 21:55:54: =====Stream Start: 2015-03-16, 21:55:54=============================================== 21:55:54: Multithreaded optimizations: Off 21:55:54: Base resolution: 1920x1080 21:55:54: Output resolution: 1280x720 21:55:54: ------------------------------------------ 21:55:54: Loading up D3D10 on AMD Radeon R9 200 Series (Adapter 1)... 21:55:55: ------------------------------------------ 21:55:55: Audio Format: 44100 Hz 21:55:55: ------------------------------------------ 21:55:55: Audio Channels: 2 Ch 21:55:55: Playback device Default 21:55:55: ------------------------------------------ 21:55:55: Using desktop audio input: Headset Earphone (ASTRO MixAmp Pro ) 21:55:55: Global Audio time adjust: 0 21:55:55: ------------------------------------------ 21:55:55: Audio Encoding: AAC 21:55:55: bitrate: 96 Warning -- DShowPlugin: Could not create device filter 21:55:55: Using directshow input 21:55:55: Scene buffering time set to 700 21:55:55: Using custom x264 settings: "opencl=true" 21:55:58: ------------------------------------------ 21:55:58: Video Encoding: x264 21:55:58: fps: 25 21:55:58: width: 1280, height: 720 21:55:58: preset: veryfast 21:55:58: profile: main 21:55:58: keyint: 50 21:55:58: CBR: yes 21:55:58: CFR: no 21:55:58: max bitrate: 6000 21:55:58: buffer size: 6000 21:55:58: ------------------------------------------ Warning -- DShowPlugin: Could not create device filter 21:56:00: Using directshow input Warning -- DShowPlugin: Could not create device filter 21:56:03: Using directshow input 21:56:17: ------------------------------------------ 21:56:17: device: Elgato Game Capture HD, 21:56:17: device id {39F50F4C-99E1-464A-B6F9-D605B4FB5918}, 21:56:17: chosen type: UYVY, usingFourCC: false, res: 1280x720 - 1280x720, frameIntervals: 166833-166833 21:56:17: use buffering: true - 0, fourCC: 'UYVY' 21:56:17: 21:56:17: device audio info - bits per sample: 16, channels: 2, samples per sec: 48000, block size: 4 21:56:17: Elgato Game Capture: force buffering with 1 msec 21:56:17: Using directshow input 21:56:24: Total frames encoded: 626, total frames duplicated: 12 (1.92%) 21:56:24: Total frames rendered: 637, number of late frames: 1 (0.16%) (it's okay for some frames to be late) 21:56:32: 21:56:32: Profiler time results: 21:56:32: 21:56:32: ============================================================== 21:56:32: video thread frame - [100%] [avg time: 2.279 ms] [children: 47.2%] [unaccounted: 52.8%] 21:56:32: | scene->Preprocess - [7.59%] [avg time: 0.173 ms] 21:56:32: | GPU download and conversion - [39.6%] [avg time: 0.902 ms] [children: 39.7%] [unaccounted: -0.0878%] 21:56:32: | | flush - [4.12%] [avg time: 0.094 ms] 21:56:32: | | CopyResource - [0.219%] [avg time: 0.005 ms] 21:56:32: | | conversion to 4:2:0 - [35.3%] [avg time: 0.805 ms] [children: 34.7%] [unaccounted: 0.658%] 21:56:32: | | | Convert444toNV12 - [34.7%] [avg time: 0.79 ms] 21:56:32: encoder thread frame - [100%] [avg time: 1.349 ms] [children: 0.445%] [unaccounted: 99.6%] 21:56:32: | sending stuff out - [0.445%] [avg time: 0.006 ms] 21:56:32: ============================================================== 21:56:32: 21:56:32: 21:56:32: Profiler CPU results: 21:56:32: 21:56:32: ============================================================== 21:56:32: video thread frame - [cpu time: avg 1.079 ms, total 687.5 ms] [avg calls per frame: 1] 21:56:32: | scene->Preprocess - [cpu time: avg 0.171 ms, total 109.375 ms] [avg calls per frame: 1] 21:56:32: | GPU download and conversion - [cpu time: avg 0.907 ms, total 578.125 ms] [avg calls per frame: 1] 21:56:32: | | flush - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 21:56:32: | | CopyResource - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 21:56:32: | | conversion to 4:2:0 - [cpu time: avg 0.936 ms, total 578.125 ms] [avg calls per frame: 1] 21:56:32: | | | Convert444toNV12 - [cpu time: avg 0.936 ms, total 578.125 ms] [avg calls per frame: 1] 21:56:32: encoder thread frame - [cpu time: avg 1.048 ms, total 656.25 ms] [avg calls per frame: 1] 21:56:32: | sending stuff out - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 21:56:32: ============================================================== 21:56:32: 21:56:32: =====Stream End: 2015-03-16, 21:56:32================================================= 21:56:54: =====Stream Start: 2015-03-16, 21:56:54=============================================== 21:56:54: Multithreaded optimizations: Off 21:56:54: Base resolution: 1920x1080 21:56:54: Output resolution: 960x540 21:56:54: ------------------------------------------ 21:56:54: Loading up D3D10 on AMD Radeon R9 200 Series (Adapter 1)... 21:56:54: ------------------------------------------ 21:56:54: Audio Format: 44100 Hz 21:56:54: ------------------------------------------ 21:56:54: Audio Channels: 2 Ch 21:56:54: Playback device Default 21:56:54: ------------------------------------------ 21:56:54: Using desktop audio input: Headset Earphone (ASTRO MixAmp Pro ) 21:56:54: Global Audio time adjust: 0 21:56:54: ------------------------------------------ 21:56:54: Audio Encoding: AAC 21:56:54: bitrate: 96 21:56:54: ------------------------------------------ 21:56:54: device: Elgato Game Capture HD, 21:56:54: device id {39F50F4C-99E1-464A-B6F9-D605B4FB5918}, 21:56:54: chosen type: UYVY, usingFourCC: false, res: 1280x720 - 1280x720, frameIntervals: 166833-166833 21:56:54: use buffering: true - 0, fourCC: 'UYVY' 21:56:54: 21:56:54: device audio info - bits per sample: 16, channels: 2, samples per sec: 48000, block size: 4 21:56:54: Elgato Game Capture: force buffering with 1 msec 21:56:54: Using directshow input 21:56:54: Scene buffering time set to 700 21:56:54: Using custom x264 settings: "opencl=true" 21:56:57: ------------------------------------------ 21:56:57: Video Encoding: x264 21:56:57: fps: 25 21:56:57: width: 960, height: 540 21:56:57: preset: veryfast 21:56:57: profile: main 21:56:57: keyint: 50 21:56:57: CBR: yes 21:56:57: CFR: no 21:56:57: max bitrate: 6000 21:56:57: buffer size: 6000 21:56:57: ------------------------------------------ 21:56:59: Using RTMP service: Twitch 21:56:59: Server selection: rtmp://live-lhr.twitch.tv/app 21:56:59: Interface: Realtek PCIe FE Family Controller (ethernet, 100 mbps) 21:57:00: Completed handshake with rtmp://live-lhr.twitch.tv/app in 848 ms. 21:57:00: SO_SNDBUF was at 65536 21:57:01: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 131072 (buffer: 0 / 780288) 21:57:03: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 262144 (buffer: 0 / 780288) 21:57:05: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 524288 (buffer: 0 / 780288) 21:57:07: Audio timestamp for device 'Elgato Game Capture HD' was behind target timestamp by 10 21:57:07: Audio timestamp for device 'Elgato Game Capture HD' was behind target timestamp by 100 21:57:07: Audio timestamp for device 'Elgato Game Capture HD' was behind target timestamp by 110 21:57:15: Audio timestamp for device 'Elgato Game Capture HD' was behind target timestamp by 50 21:57:15: Audio timestamp for device 'Elgato Game Capture HD' was behind target timestamp by 80 21:57:36: No Intel graphics adapter visible in QSVHelper.exe, Optimus problem? 21:57:36: Failed loading CUDA dll 21:58:00: x264: bitrate 6000->3500, buffer size 6000->3500 21:58:00: Settings::Encoding: Changing bitrate from 6000kb/s to 3500kb/s 21:58:03: Total frames encoded: 1624, total frames duplicated: 36 (2.22%) 21:58:03: Number of frames skipped due to encoder lag: 6 (0.37%) 21:58:03: Total frames rendered: 1637, number of late frames: 3 (0.18%) (it's okay for some frames to be late) 21:58:03: ~RTMPPublisher: Packet flush completed in 146 ms 21:58:03: ~RTMPPublisher: Send thread terminated in 1 ms 21:58:03: RTMPPublisher::SocketLoop: Graceful loop exit 21:58:03: ~RTMPPublisher: Socket thread terminated in 0 ms 21:58:07: 21:58:07: Profiler time results: 21:58:07: 21:58:07: ============================================================== 21:58:07: video thread frame - [100%] [avg time: 2.23 ms] [children: 63.8%] [unaccounted: 36.2%] 21:58:07: | scene->Preprocess - [27.9%] [avg time: 0.622 ms] 21:58:07: | GPU download and conversion - [35.9%] [avg time: 0.801 ms] [children: 34.9%] [unaccounted: 0.987%] 21:58:07: | | flush - [11.3%] [avg time: 0.253 ms] 21:58:07: | | CopyResource - [0.224%] [avg time: 0.005 ms] 21:58:07: | | conversion to 4:2:0 - [23.4%] [avg time: 0.521 ms] [children: 22.7%] [unaccounted: 0.673%] 21:58:07: | | | Convert444toNV12 - [22.7%] [avg time: 0.506 ms] 21:58:07: encoder thread frame - [100%] [avg time: 2.067 ms] [children: 14.2%] [unaccounted: 85.8%] 21:58:07: | sending stuff out - [14.2%] [avg time: 0.293 ms] 21:58:07: ============================================================== 21:58:07: 21:58:07: 21:58:07: Profiler CPU results: 21:58:07: 21:58:07: ============================================================== 21:58:07: video thread frame - [cpu time: avg 1.393 ms, total 2281.25 ms] [avg calls per frame: 1] 21:58:07: | scene->Preprocess - [cpu time: avg 0.477 ms, total 781.25 ms] [avg calls per frame: 1] 21:58:07: | GPU download and conversion - [cpu time: avg 0.773 ms, total 1265.63 ms] [avg calls per frame: 1] 21:58:07: | | flush - [cpu time: avg 0.124 ms, total 203.125 ms] [avg calls per frame: 1] 21:58:07: | | CopyResource - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 21:58:07: | | conversion to 4:2:0 - [cpu time: avg 0.638 ms, total 1031.25 ms] [avg calls per frame: 1] 21:58:07: | | | Convert444toNV12 - [cpu time: avg 0.58 ms, total 937.5 ms] [avg calls per frame: 1] 21:58:07: encoder thread frame - [cpu time: avg 0.336 ms, total 546.875 ms] [avg calls per frame: 1] 21:58:07: | sending stuff out - [cpu time: avg 0.105 ms, total 171.875 ms] [avg calls per frame: 1] 21:58:07: ============================================================== 21:58:07: 21:58:07: =====Stream End: 2015-03-16, 21:58:07================================================= 21:58:13: ~RTMPPublisher: Final socket shutdown completed in 10183 ms 21:58:13: Average send payload: 5897 bytes, average send interval: 7 ms 21:58:13: Number of times waited to send: 0, Waited for a total of 0 bytes 21:58:13: Number of b-frames dropped: 0 (0%), Number of p-frames dropped: 0 (0%), Total 0 (0%) 21:58:13: Number of bytes sent: 46606952 21:58:34: =====Stream Start: 2015-03-16, 21:58:34=============================================== 21:58:34: Multithreaded optimizations: Off 21:58:34: Base resolution: 1920x1080 21:58:34: Output resolution: 960x540 21:58:34: ------------------------------------------ 21:58:34: Loading up D3D10 on AMD Radeon R9 200 Series (Adapter 1)... 21:58:34: ------------------------------------------ 21:58:34: Audio Format: 44100 Hz 21:58:34: ------------------------------------------ 21:58:34: Audio Channels: 2 Ch 21:58:34: Playback device Default 21:58:34: ------------------------------------------ 21:58:34: Using desktop audio input: Headset Earphone (ASTRO MixAmp Pro ) 21:58:34: Global Audio time adjust: 0 21:58:34: ------------------------------------------ 21:58:34: Audio Encoding: AAC 21:58:34: bitrate: 96 21:58:34: ------------------------------------------ 21:58:34: device: Elgato Game Capture HD, 21:58:34: device id {39F50F4C-99E1-464A-B6F9-D605B4FB5918}, 21:58:34: chosen type: UYVY, usingFourCC: false, res: 1280x720 - 1280x720, frameIntervals: 166833-166833 21:58:34: use buffering: true - 0, fourCC: 'UYVY' 21:58:34: 21:58:34: device audio info - bits per sample: 16, channels: 2, samples per sec: 48000, block size: 4 21:58:34: Elgato Game Capture: force buffering with 1 msec 21:58:34: Using directshow input 21:58:34: Scene buffering time set to 700 21:58:34: ------------------------------------------ 21:58:34: Video Encoding: x264 21:58:34: fps: 25 21:58:34: width: 960, height: 540 21:58:34: preset: veryfast 21:58:34: profile: main 21:58:34: keyint: 50 21:58:34: CBR: yes 21:58:34: CFR: no 21:58:34: max bitrate: 3500 21:58:34: buffer size: 3500 21:58:34: ------------------------------------------ 21:58:36: Using RTMP service: Twitch 21:58:36: Server selection: rtmp://live-lhr.twitch.tv/app 21:58:36: Interface: Realtek PCIe FE Family Controller (ethernet, 100 mbps) 21:58:37: Completed handshake with rtmp://live-lhr.twitch.tv/app in 938 ms. 21:58:38: SO_SNDBUF was at 65536 21:58:39: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 131072 (buffer: 0 / 459776) 21:58:46: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 262144 (buffer: 61851 / 459776) 21:58:46: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 524288 (buffer: 0 / 459776) 21:59:55: Total frames encoded: 2000, total frames duplicated: 14 (0.70%) 21:59:55: Total frames rendered: 2020, number of late frames: 2 (0.10%) (it's okay for some frames to be late) 21:59:55: ~RTMPPublisher: Packet flush completed in 140 ms 21:59:55: ~RTMPPublisher: Send thread terminated in 0 ms 21:59:55: RTMPPublisher::SocketLoop: Graceful loop exit 21:59:55: ~RTMPPublisher: Socket thread terminated in 1 ms 22:00:00: 22:00:00: Profiler time results: 22:00:00: 22:00:00: ============================================================== 22:00:00: video thread frame - [100%] [avg time: 1.936 ms] [children: 61.4%] [unaccounted: 38.6%] 22:00:00: | scene->Preprocess - [23.7%] [avg time: 0.459 ms] 22:00:00: | GPU download and conversion - [37.7%] [avg time: 0.73 ms] [children: 36.5%] [unaccounted: 1.19%] 22:00:00: | | flush - [10%] [avg time: 0.194 ms] 22:00:00: | | CopyResource - [0.258%] [avg time: 0.005 ms] 22:00:00: | | conversion to 4:2:0 - [26.2%] [avg time: 0.508 ms] [children: 25.3%] [unaccounted: 0.981%] 22:00:00: | | | Convert444toNV12 - [25.3%] [avg time: 0.489 ms] 22:00:00: encoder thread frame - [100%] [avg time: 1.059 ms] [children: 8.4%] [unaccounted: 91.6%] 22:00:00: | sending stuff out - [8.4%] [avg time: 0.089 ms] 22:00:00: ============================================================== 22:00:00: 22:00:00: 22:00:00: Profiler CPU results: 22:00:00: 22:00:00: ============================================================== 22:00:00: video thread frame - [cpu time: avg 1.229 ms, total 2484.38 ms] [avg calls per frame: 1] 22:00:00: | scene->Preprocess - [cpu time: avg 0.348 ms, total 703.125 ms] [avg calls per frame: 1] 22:00:00: | GPU download and conversion - [cpu time: avg 0.796 ms, total 1609.38 ms] [avg calls per frame: 1] 22:00:00: | | flush - [cpu time: avg 0.154 ms, total 312.5 ms] [avg calls per frame: 1] 22:00:00: | | CopyResource - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 22:00:00: | | conversion to 4:2:0 - [cpu time: avg 0.633 ms, total 1265.63 ms] [avg calls per frame: 1] 22:00:00: | | | Convert444toNV12 - [cpu time: avg 0.586 ms, total 1171.88 ms] [avg calls per frame: 1] 22:00:00: encoder thread frame - [cpu time: avg 0.132 ms, total 265.625 ms] [avg calls per frame: 1] 22:00:00: | sending stuff out - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 22:00:00: ============================================================== 22:00:00: 22:00:00: =====Stream End: 2015-03-16, 22:00:00================================================= 22:00:07: ~RTMPPublisher: Final socket shutdown completed in 11251 ms 22:00:07: Average send payload: 6122 bytes, average send interval: 13 ms 22:00:07: Number of times waited to send: 0, Waited for a total of 0 bytes 22:00:07: Number of b-frames dropped: 0 (0%), Number of p-frames dropped: 0 (0%), Total 0 (0%) 22:00:07: Number of bytes sent: 33946699 22:00:32: =====Stream Start: 2015-03-16, 22:00:32=============================================== 22:00:32: Multithreaded optimizations: Off 22:00:32: Base resolution: 1920x1080 22:00:32: Output resolution: 960x540 22:00:32: ------------------------------------------ 22:00:32: Loading up D3D10 on AMD Radeon R9 200 Series (Adapter 1)... 22:00:32: ------------------------------------------ 22:00:32: Audio Format: 44100 Hz 22:00:32: ------------------------------------------ 22:00:32: Audio Channels: 2 Ch 22:00:32: Playback device Default 22:00:32: ------------------------------------------ 22:00:32: Using desktop audio input: Headset Earphone (ASTRO MixAmp Pro ) 22:00:32: Global Audio time adjust: 0 22:00:32: ------------------------------------------ 22:00:32: Audio Encoding: AAC 22:00:32: bitrate: 96 22:00:32: ------------------------------------------ 22:00:32: device: Elgato Game Capture HD, 22:00:32: device id {39F50F4C-99E1-464A-B6F9-D605B4FB5918}, 22:00:32: chosen type: UYVY, usingFourCC: false, res: 1280x720 - 1280x720, frameIntervals: 166833-166833 22:00:32: use buffering: true - 0, fourCC: 'UYVY' 22:00:32: 22:00:32: device audio info - bits per sample: 16, channels: 2, samples per sec: 48000, block size: 4 22:00:32: Elgato Game Capture: force buffering with 1 msec 22:00:32: Using directshow input 22:00:32: Scene buffering time set to 700 22:00:32: ------------------------------------------ 22:00:32: Video Encoding: x264 22:00:32: fps: 30 22:00:32: width: 960, height: 540 22:00:32: preset: veryfast 22:00:32: profile: high 22:00:32: keyint: 60 22:00:32: CBR: yes 22:00:32: CFR: no 22:00:32: max bitrate: 3500 22:00:32: buffer size: 3500 22:00:32: ------------------------------------------ 22:00:34: Using RTMP service: Twitch 22:00:34: Server selection: rtmp://live-lhr.twitch.tv/app 22:00:34: Interface: Realtek PCIe FE Family Controller (ethernet, 100 mbps) 22:00:35: Completed handshake with rtmp://live-lhr.twitch.tv/app in 782 ms. 22:00:35: SO_SNDBUF was at 65536 22:00:36: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 131072 (buffer: 0 / 459776) 22:00:43: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 262144 (buffer: 0 / 459776) 22:00:43: RTMPPublisher::SocketLoop: Increasing send buffer to ISB 524288 (buffer: 0 / 459776) 22:01:05: FlushBufferedVideo: Flushing 5 packets over 134 ms 22:01:05: Total frames encoded: 976, total frames duplicated: 28 (2.87%) 22:01:05: Total frames rendered: 994, number of late frames: 2 (0.20%) (it's okay for some frames to be late) 22:01:06: ~RTMPPublisher: Packet flush completed in 134 ms 22:01:06: ~RTMPPublisher: Send thread terminated in 0 ms 22:01:06: RTMPPublisher::SocketLoop: Graceful loop exit 22:01:06: ~RTMPPublisher: Socket thread terminated in 1 ms 22:01:10: 22:01:10: Profiler time results: 22:01:10: 22:01:10: ============================================================== 22:01:10: video thread frame - [100%] [avg time: 3.361 ms] [children: 69.7%] [unaccounted: 30.3%] 22:01:10: | scene->Preprocess - [39.9%] [avg time: 1.341 ms] 22:01:10: | GPU download and conversion - [29.8%] [avg time: 1 ms] [children: 28.4%] [unaccounted: 1.31%] 22:01:10: | | flush - [12%] [avg time: 0.404 ms] 22:01:10: | | CopyResource - [0.149%] [avg time: 0.005 ms] 22:01:10: | | conversion to 4:2:0 - [16.3%] [avg time: 0.547 ms] [children: 15.8%] [unaccounted: 0.446%] 22:01:10: | | | Convert444toNV12 - [15.8%] [avg time: 0.532 ms] 22:01:10: encoder thread frame - [100%] [avg time: 1.719 ms] [children: 7.56%] [unaccounted: 92.4%] 22:01:10: | sending stuff out - [7.56%] [avg time: 0.13 ms] 22:01:10: ============================================================== 22:01:10: 22:01:10: 22:01:10: Profiler CPU results: 22:01:10: 22:01:10: ============================================================== 22:01:10: video thread frame - [cpu time: avg 2.247 ms, total 2234.38 ms] [avg calls per frame: 1] 22:01:10: | scene->Preprocess - [cpu time: avg 1.288 ms, total 1281.25 ms] [avg calls per frame: 1] 22:01:10: | GPU download and conversion - [cpu time: avg 0.817 ms, total 812.5 ms] [avg calls per frame: 1] 22:01:10: | | flush - [cpu time: avg 0.047 ms, total 46.875 ms] [avg calls per frame: 1] 22:01:10: | | CopyResource - [cpu time: avg 0 ms, total 0 ms] [avg calls per frame: 1] 22:01:10: | | conversion to 4:2:0 - [cpu time: avg 0.757 ms, total 734.375 ms] [avg calls per frame: 1] 22:01:10: | | | Convert444toNV12 - [cpu time: avg 0.757 ms, total 734.375 ms] [avg calls per frame: 1] 22:01:10: encoder thread frame - [cpu time: avg 0.257 ms, total 250 ms] [avg calls per frame: 1] 22:01:10: | sending stuff out - [cpu time: avg 0.064 ms, total 62.5 ms] [avg calls per frame: 1] 22:01:10: ============================================================== 22:01:10: 22:01:10: =====Stream End: 2015-03-16, 22:01:10================================================= 22:01:17: ~RTMPPublisher: Final socket shutdown completed in 11457 ms 22:01:17: Average send payload: 5260 bytes, average send interval: 11 ms 22:01:17: Number of times waited to send: 0, Waited for a total of 0 bytes 22:01:17: Number of b-frames dropped: 0 (0%), Number of p-frames dropped: 0 (0%), Total 0 (0%) 22:01:17: Number of bytes sent: 13219848 22:01:21: Terminating 0x284c 22:01:21: Terminating 0x26b0 22:01:21: Terminating 0x256c Last game capture log: 2014-09-24, 20:46:45: we're booting up: 20:46:45: (half life scientist) everything.. seems to be in order 20:46:45: D3D9 Present 22:01:21: Memory Leaks Were Detected. 22:01:21:

Wolf's OpenCL AEON Miner for AMD GPUs

PKGBUILD for proprietary OpenCL drivers on AMD GPUs on Arch Linux

OpenCL Hello World examples demonstrating libclc math functions across CPU, Nvidia GPU, and AMD GPU

iChess.io on OpenCL UCI Chess Engine on Intel & NVidia & AMD & Mali GPUs for macOS & GNU/Linux & Windows & FreeBSD

This is a Pyrit Version that comes with CAL ( Compute Abstraction Layer) support out of the box. Pyrit-CAL is much faster than Pyrit-opencl on AMD Radeon GPUs (Redwood/Capilano cores and beyond).

AMD GPU benchmark using OpenCL

OpenCL neural networks framework for AMD GPUs

An AMD R5xxxx series gpu driver written in python.