What's new in Metal?
Generated on 8/5/2024
1 search
What's New in Metal?
Machine Learning Enhancements
- MPS Graph Improvements: Enhancements to Metal Performance Shaders (MPS) and MPS Graph, focusing on improving the efficiency of transformer models. This includes better GPU synchronization and memory control, which can be particularly useful for applications that already use Metal (Accelerate machine learning with Metal).
- FFT-Based ML Models: New features to accelerate Fast Fourier Transform (FFT) based machine learning models (Accelerate machine learning with Metal).
- Training Frameworks: Support for popular machine learning frameworks like PyTorch, TensorFlow, and JAX, with improvements in distributed training and mixed precision to boost performance (Train your machine learning and AI models on Apple GPUs).
Gaming Enhancements
- Metal Shader Converter: A tool to convert shaders to Metal, supporting all shader stages including ray tracing and mesh shaders. This tool helps in porting shaders from other platforms to Metal, significantly reducing development time (Port advanced games to Apple platforms).
- Unified Metal Shaders: Compile shaders once and deploy them across macOS and iOS, simplifying the development process (Port advanced games to Apple platforms).
- Resource Residency: Improvements in resource residency, allowing better management of GPU memory and resources (Port advanced games to Apple platforms).
- MetalFX: A new feature called MetalFX, which scales lower resolution images to higher resolutions, improving game performance (Port advanced games to Apple platforms).
General Enhancements
- Low-Level Mesh and Textures: New APIs for low-level mesh and textures, offering improved control over app visuals, especially for dynamic models and textures (Platforms State of the Union).
- Unified Gaming Platform: With the advancements in Metal and Apple silicon, Apple devices are now capable of running console-class games, creating a unified gaming platform across Macs, iPads, and iPhones (Platforms State of the Union).
Relevant Sessions
- Accelerate machine learning with Metal
- Port advanced games to Apple platforms
- Platforms State of the Union
- Train your machine learning and AI models on Apple GPUs
These sessions provide a comprehensive overview of the latest advancements in Metal, covering both machine learning and gaming enhancements.
Accelerate machine learning with Metal
Learn how to accelerate your machine learning transformer models with new features in Metal Performance Shaders Graph. We’ll also cover how to improve your model’s compute bandwidth and quality, and visualize it in the all new MPSGraph viewer.
Platforms State of the Union
Discover the newest advancements on Apple platforms.
Train your machine learning and AI models on Apple GPUs
Learn how to train your models on Apple Silicon with Metal for PyTorch, JAX and TensorFlow. Take advantage of new attention operations and quantization support for improved transformer model performance on your devices.
Port advanced games to Apple platforms
Discover how simple it can be to reach players on Apple platforms worldwide. We’ll show you how to evaluate your Windows executable on Apple silicon, start your game port with code samples, convert your shader code to Metal, and bring your game to Mac, iPhone, and iPad. Explore enhanced Metal tools that understand HLSL shaders to validate, debug, and profile your ported shaders on Metal.