- Видео 721
- Просмотров 16 503 612
Arm Software Developers
Великобритания
Добавлен 27 июл 2020
Welcome to Arm Software Developers. We are all about the latest and greatest features in software development and how it powers the world around us. We focus on all kinds of content like getting started guides, unboxing of the latest Arm-based hardware, tutorials, and demos covering IoT, machine learning, cloud-native development and graphics. You can also join us for a chat in our Innovation Coffee live streams to see what is new in the Arm developer community.
Connect with us on Twitter: ArmSoftwareDev
Check out our resources for developers : developer.arm.com
Connect with us on Twitter: ArmSoftwareDev
Check out our resources for developers : developer.arm.com
.Net 9 and SVE, and VectorCamps's simd.info | Innovation Coffee
Kunal Pathak from Microsoft covers .Net 9 and Arm's Scalable Vector Extensions, and Konstantinos Margaritis and Georgios Mermigkis from VectorCamp cover their new simd.info, an online searchable reference tool for C intrinsics of all major SIMD engines.
0:00 Intro
01:25 Kunal Pathak - .Net 9 and SVE
13:51 Konstantinos Margaritis and Georgios Mermigkis - VectorCamp's SIMD.info
37:03 Recent Learning Paths with Joe Stech
Discord: discord.gg/ArmSoftwareDev
Arm Developer Hub: arm.com/developer-hub
Arm Developer Program: www.arm.com/resources/developer-program
reddit: www.reddit.com/r/ArmSoftwareDev/
Twitter: armsoftwaredev
Learning Paths: learn.arm.com
Arm Tech Talks: www.arm.com/partners/te...
0:00 Intro
01:25 Kunal Pathak - .Net 9 and SVE
13:51 Konstantinos Margaritis and Georgios Mermigkis - VectorCamp's SIMD.info
37:03 Recent Learning Paths with Joe Stech
Discord: discord.gg/ArmSoftwareDev
Arm Developer Hub: arm.com/developer-hub
Arm Developer Program: www.arm.com/resources/developer-program
reddit: www.reddit.com/r/ArmSoftwareDev/
Twitter: armsoftwaredev
Learning Paths: learn.arm.com
Arm Tech Talks: www.arm.com/partners/te...
Просмотров: 174
Видео
Real-Time Twitter/X Sentiment Analysis on Arm Neoverse-Based CPUs demo at KubeCon NA 2024
Просмотров 77День назад
Pranay Bakre explains his real-time Twitter/X sentiment analysis program that uses a StanfordNLP model entirely on an Arm Neoverse-based CPU. Shown off at the Arm booth at KubeCon CloudNativeCon North America 2024. Arm Developer Program: www.arm.com/resources/developer-program Discord: discord.gg/ArmSoftwareDev Arm Developer Hub: arm.com/developer-hub reddit: www.reddit.com/r/ArmSoftwareDev/ Tw...
Join the AI Software Revolution | Arm Developer Program
Просмотров 24021 день назад
Ready to Change the Future of Tech? The Arm Developer Program connects you with a global community of experts and provides essential tools, insights, and resources to supercharge your software projects across devices-from smartphones to cloud platforms. Join today and get the support you need to bring your brilliant ideas to life. 👉 Join now: bit.ly/40s4bdM
AI from Cloud to Edge: PyTorch Conference 2024 Recap, Demos, and Learning Paths | Innovation Coffee
Просмотров 362Месяц назад
Catch an abridged version of Gian Marco Iodice’s presentation on Democratizing AI: Powering the Future with Arm's Global Compute Ecosystem at PyTorch Conference 2024. Explore KleidiAI-accelerated demos shown at our booth, discover the latest in AI innovations from cloud to edge, and get a brief recap of the event. And as always find out about the latest Learning Paths. 0:00 Intro 01:16 Gian Mar...
AI Performance Boost with KleidiAI and PyTorch on AWS Graviton 4 Demo | PyTorch Conference 2024
Просмотров 5142 месяца назад
Nobel Chowdary Mandepudi demonstrates how KleidiAI enhances AI performance in the cloud. Using AWS Graviton 4 instances, this side-by-side demo compares PyTorch inference with and without KleidiAI optimizations. The demo shows the significant boost in efficiency and speed, as demonstrated using the Llama 3.1 8B model. KleidiAI leverages cutting-edge Arm instructions to accelerate generative AI ...
Edge AI Acceleration with KleidiAI and ExecuTorch Demo | PyTorch Conference 2024
Просмотров 1802 месяца назад
Learn how Arm’s KleidiAI, integrated with ExecuTorch, accelerates AI inference on edge devices. In this demo, presented by Gian Marco Iodice, we showcase real-time inference of the Llama 3.1 8B parameter model on a mobile phone. KleidiAI uses highly optimized micro-kernels to leverage advanced Arm CPU instructions, such as SMMLA and FMLA, to deliver powerful, efficient AI performance on edge de...
Demos on Vulkan SDK, LLM Inference, and Advancing Mobile Graphics | Innovation Coffee
Просмотров 6703 месяца назад
The episode features information on Vulkan SDK from Karen Ghavam and Richard Wright Jr from LunarG, a demo of the Learning Path LLM inference on Android with KleidiAI, MediaPipe, and XNNPACK from Jakub Sujak, a segment on advancing mobile graphics with Arm from Sam Martin, and find out about the latest Learning Paths from Pareena Verma. LunarG Vulkan SDK Announcement: www.lunarg.com/news-insigh...
Create Computer Vision Applications with OpenCV on Android Devices
Просмотров 8023 месяца назад
Learn the basics of building a computer vision application on Android using OpenCV. The is based off the Learning Path by Dawid Borycki and serves as a great entry into working with OpenCV. The Learning Path this follows: learn.arm.com/learning-paths/smartphones-and-mobile/android_opencv_camera/ developer.android.com/studio developer.android.com/studio/run/device 0:00 Intro 0:43 OpenCV descript...
Unleashing the Potential of the Arm Architecture - Write Once, Deploy Anywhere
Просмотров 3634 месяца назад
Discover how Arm helps developers overcome software challenges, explore our role in the tech ecosystem, and learn about essential tools to help you perfect your Arm-based applications. Presented by Andrew Wafaa, Senior Director Software Communities at Arm. 🔗 Learn about Arm Kleidi: www.arm.com/products/development-tools/embedded-and-software/kleidi 🔗 Get Arm Performance Studio: developer.arm.co...
Make an Orca Mini and Phi 3 chatbot with Python running locally on a Raspberry Pi 5 | Learning Path
Просмотров 7054 месяца назад
Learn how to build a chatbot using Python with both the Orca Mini 3B and the Phi 3 Mini 4K Instruct models, entirely running on the Raspberry Pi 5, no internet connection or subscription required. learn.arm.com The Learning Path this video follows: learn.arm.com/learning-paths/embedded-systems/llama-python-cpu/ Orca Mini 3B model: huggingface.co/pankajmathur/orca_mini_3b Phi 3 Mini 4K Instruct ...
Arm at Open Sauce, Microsoft Sandbox, Learning Paths, and Confidential Compute | Innovation Coffee
Просмотров 2514 месяца назад
We're here with July's Innovation Coffee episode, covering our time at Open Sauce 2024, two learning paths on Microsoft Sandbox on Arm, information on Confidential Computing, our website for Learning Paths, and some bonus information about our Developer Program and Developer Hub 0:00 Intro 0:35 Welcome to Innovation Coffee 1:39 Open Sauce 12:47 Microsoft Sandbox Learning Paths 31:02 Did You Kno...
GitHub Arm-hosted Runners with Larissa Fortuna and Diego Russo | Innovation Coffee
Просмотров 5235 месяцев назад
Come watch Larissa Fortuna from GitHub and Diego Russo from Arm talk about the new Arm-based Linux and Windows runners for GitHub Actions and how they are now available in Public Beta. Blog about Arm64 on GitHub Actions: github.blog/2024-06-03-arm64-on-github-actions-powering-faster-more-efficient-build-systems/ Learning Path on how to create a new Arm-hosted runner: learn.arm.com/learning-path...
Arm at Gitex Africa 2024
Просмотров 1125 месяцев назад
Gitex Africa is the continent's biggest tech expo, it's the pulse of Africa's digital transformation, setting the scene for AI's pivotal role in the future. Arm is enabling #AI everywhere and it is clear that there are significant opportunities for #African-based developers to build innovative #AI computing solutions on Arm. Our efforts towards Developer Education continued at #GitexAfrica as w...
Windows on Arm with David Whaley and Jeremy Sinclair
Просмотров 2916 месяцев назад
Microsoft MVP and newest Arm Ambassador of the Arm Developer Program, Jeremy Sinclair, joins us for an exclusive Innovation Coffee right before Microsoft Build to share his insights on porting Windows applications to Arm and the ecosystem support available. We cover Windows on Arm on PC and in the cloud, how to easily port applications to Arm64 using Microsoft’s Visual Studio for best performan...
Maximize Android Game Performance: Arm Tools for Monitoring, Reporting, and Optimization
Просмотров 2186 месяцев назад
Maximize Android Game Performance: Arm Tools for Monitoring, Reporting, and Optimization
Realistic Mobile Graphics with Optimized Ray Tracing
Просмотров 4906 месяцев назад
Realistic Mobile Graphics with Optimized Ray Tracing
Steel Arms: Bringing Lumen to Immortalis
Просмотров 2076 месяцев назад
Steel Arms: Bringing Lumen to Immortalis
Supporting INT8 quantized networks with Unity Sentis
Просмотров 4136 месяцев назад
Supporting INT8 quantized networks with Unity Sentis
SmartGI Evolution: Adaptive NanoMesh on Mobile
Просмотров 6626 месяцев назад
SmartGI Evolution: Adaptive NanoMesh on Mobile
Arm Developer Hub - Take Your Development to the Next Level
Просмотров 687 месяцев назад
Arm Developer Hub - Take Your Development to the Next Level
Weekly Dose on Arm #13: April 19th to April 25th, 2024
Просмотров 1097 месяцев назад
Weekly Dose on Arm #13: April 19th to April 25th, 2024
Weekly Dose on Arm #12: April 12th to April 18th, 2024
Просмотров 997 месяцев назад
Weekly Dose on Arm #12: April 12th to April 18th, 2024
Weekly Dose on Arm #11: April 5th - April 11th
Просмотров 1197 месяцев назад
Weekly Dose on Arm #11: April 5th - April 11th
Steel Arms demo from Arm at GDC 2024
Просмотров 1967 месяцев назад
Steel Arms demo from Arm at GDC 2024
Arm Optimized Ray Tracing demo at GDC 2024
Просмотров 3727 месяцев назад
Arm Optimized Ray Tracing demo at GDC 2024
Arm Generative AI on Mobile demo at GDC 2024
Просмотров 1237 месяцев назад
Arm Generative AI on Mobile demo at GDC 2024
Weekly Dose on Arm #10: Mar 22 - Mar 28, 2024
Просмотров 2508 месяцев назад
Weekly Dose on Arm #10: Mar 22 - Mar 28, 2024
Unlocking Depth: Supporting INT8 Quantized Networks with Unity Sentis by Yusuf Duman
Просмотров 1178 месяцев назад
Unlocking Depth: Supporting INT8 Quantized Networks with Unity Sentis by Yusuf Duman
Transforming Light: Exploring AI Powered Image Relighting by Yusuf Duman
Просмотров 958 месяцев назад
Transforming Light: Exploring AI Powered Image Relighting by Yusuf Duman
g68 mc4 don't have drivers?
Can u help me with Expresso IDE
result on my snapdragon 680: but neon 60ms neon 110ms I don’t know why, but it’s worse with neon
keil vscode stm32cubemx
Thank you for creating such an enlightening video. Could you please provide your LinkedIn profile? I'm eager to stay updated by following you there.
Thanks for sharing this! Is there any ARM documentation I can see the signal composing the external coprocessor interface for Armv8 architecture?
it would have been helpful to see more of the setup of k3s and how you connect gitlab to it.. ie which agent or runner type did you go with.. I can only see the first job where you build using what looks to be a Docker Runner.. but for k3s I thought you'd need to go with a kubernetes type agent?
I have a presentation on multicore CPUs and after two weeks of searching this video helped me immensely. thank you so much
I definitely advocate for a community standard core set of ops that higher level ops have a decomposition into, for the proliferation of operators and incompatibilities between libraries is unmaintainable. 38:58 NHWC🤨? Given that NCHW is canonical in ML and is the default for most libraries (PyTorch, OneDNN, CoreML, DirectML...) *except* for TensorFlow/XNNPack, NHWC it seems a weird choice.
give me a simple example code to write into artemis so that i can see how to communicate with the processor
Awesome video, thanks!!!
video are not helpful. after clicking given website not getting project code link
usbConnect error: Failed to execute 'open' on 'USBDevice': Access denied------------ I am getting this error , please help
The Orange Pi 800 all in one keyboard system uses an ARM Mali-T860MP4. Someone should make a learning DIY video on making GPU Python-IDE projects for learning. 😎Thanks.
nice may i know the gsm gprs board that supported to your sparkfun board
You have failed to optimize it for call of duty mobile smh
Hi sir what camera module are you using:
Are you saying that if a framework is fully TOSA compliant, I only have to know TOSA rather than having to learn the framework? So, if TensorFlow become fully TOSA compliant then I do not have to care about what TensorFlow is actually doing because it supports all TOSA operations? I only have to know how to use TOSA.
Hi. Thanks for reaching out. TOSA doesn't replace the frameworks for training of models, you will author and train your model in the framework. TOSA relates to the construction and behavior of a neural network. Once you export your neural networks from the framework, you can then test it for TOSA compliance to ensure that it will run on TOSA compliant devices. If you compile your neural network with an MLIR based compiler, there is infrastructure within the MLIR 'toolkit' that will support the lowering of your neural network model (from various framework formats) to TOSA and lower to the codegen for the target device. For further queries you can raise a ticket at: discuss.mlplatform.org/
Do you have a git repo we can use as a test to follow your demo..
So offline one would use Vela to compile the Tensorflow's tflite file, to something optimized for the microNPU (helium enabled Cortex-M55/Ethos-N or U65)? Optimizations include things like Single Instruction Multiple Data aspect, and shape of tensors (to avoid data re-arrangement). I understand Arm has an open source CMSIS-NN library that can speed-up TFL for Microcontrollers (TFLu) / tinyML applications on Cortex-M devices. uTensor is also a TF framework optimized for Arm.
Question: Where can I get more info about mobile render tiles and there particular sizes on different graphics hardware? At 10:30 he had mentioned you don't want triangles covering multiple tiles, but to topoligize assets to fit that constraint I assume I would need to know how many tiles the screen is divided into.
Using TrustZone technology. Guide to run Linux and Secure OS: ruclips.net/video/m8WlSanGDwI/видео.html ruclips.net/video/CWG2WuOJtIc/видео.html ruclips.net/video/FthDUSW3yQY/видео.html
I Like that arm developer hat I want to buy one.
Приветствую! Почему-то ощущается, что автор русскоязычный, возможно ошибаюсь
Not working
👍
Hello, I have a question about the SAMD21 Break Out Board (DEV-13664). can you please tell me the boot-up time for this board. This shall be the time from power-on till execution of first instruction when programmed in Arduino IDE. I am looking for MCU boards for a project that will boot-up within 100 milli secs.
Informative👍
Cool
Why is this an ad?
ruclips.net/video/0yinmORwcbM/видео.html
Hi, presentation ends at 20:30 you can cut off the rest
Outch, not the best possible example as dot product ends with horizontal sum.. 4 times the computing power yields 7% performance improvement is a bit underwhelming as an example to demo the power of NEON SIMD. =/
Why G78 sucks in exynos 2100. BGMI/PUBG still not optimized. Fortnite in Exynos 2100 sucks and can not maintain stable FPS. BGMI/PUBG is not available in HDR graphics with Extreme fps in graphics menu for exynos 2100. Any suggestions on these points?
Game developers are lazy to optimize games to run on Mali, not ARM's problem.
MaliG77 in Mediatek Dimensity is decent....Exynos are just bad to create SoC due to their Shit Custom cores
Mediatek does good job even has less Gpu cores than Exynos... Maybe its impractical to have more Gpu cores
@@messier8379 Actually it's because of samsung's inferior fabrication process that the exynos was built on while the mediatek was built on TSMC's more superior and advanced process. If the exynos was built on TSMC's fab, it would have been more powerful and still more efficient.
@@fidelisitor8953 i know this....but its Weird why Samsung soc dint improve very much as TSMC
Do you have Wake on motion interrupt example for Artemis?
Really do need a higher resolution when there is text, I tried reading what was happening on screen and just couldnt
How does this perform compared to an M1 for example, or an i7? I'm contemplating replacing my desktop with this (for software development) - should it perform comparably?
The LX2160a in the HoneyComb is more comparable to something in the Xeon Silver / Gold class than a modern high end desktop CPU. It is quite good at power / performance on parallel workloads, but single core performance is not its strong suit. There are many developers happily using HoneyComb as their primary workstation and are very happy with the higher end options you won't find in other "desktop" chipsets, like multiple 10Gbps networking, PCIe bifurcation, uBMC remote management port. Overall the most important feature of the HoneyComb is the SystemsReady standard and open source firmware that it provides for the users.
Good work sir
Pls defeat Adreno gpus
Great episode. Question for Eben Upton; Why did he focus on Linux as the main Operating System for the very first Pi when there were other ARM Operating System's already available?
It was at the time (and I think this is still true), the most mature and widely supported operating system for ARM.
This has an incredible value and shows another way of optimizing (other than just using things like Renderdoc). Thank you very much, I will definitely take a look at Arm studio now!
I enjoyed the show and the coffee.
This is Legendary...... less than 1,000 views, and less than 1 million subs.... When the masses find out.. it will be too late. Imagine people on Work Release receiving a pebble tracker for work conditions... Or home health aids required to use pebble tracker and a body cam for proof of working.... Or workers compensation insurance requirement.... So many features.... let alone, the games to achieve real world features, like a city wide scavenger hunt! that's related to seasons, or temperatures...... Then you have food delivery services to prove the food was in clean temperature controlled environment. Don't forget they can develop a feature for field trips, so our children don't get lost or human traffic!!! Protection from kidnapping here folks!!!! This is AWESOME!!!!!!
IOTEX is moving the world!
Hi Pranay Bakre! Great video explaining on gitlab intergration. May i have the source demo codes for app as well? Thanks and cheers!
Thanks! Glad you liked the video! You can find source code files on Pranay's GitHub: github.com/pbk8s/gitlab-eks Hope this helps!
Hey intruder would ya please hold still for sma second, my security system is booting up. 😂 but honestly I like the concept of a standby pi using a pico tho
really cool
Is the slides available for download? Thanks.
Hi Wei, unfortunately the slides are not available for this session.
Waw, I cannot wait for hardware ray tracing on mali GPUs
Thank You
Thanks for the guide, including the parts list. Very helpful!