Mali-G72 High Performance GPU

Mali-G72 Block Diagram.

About Mali-G72

Arm Mali-G72 is the second generation high performance GPU based on the Bifrost architecture. Benefitting from advanced technologies such as claused shaders and full system coherency, Mali-G72 adds increased tile buffer memory supporting up to 16 x Multi-Sample Anti-Aliasing at minimal performance cost. Arithmetic optimizations tailored to complex Machine Learning and High Fidelity Mobile Gaming use cases provide 25% higher energy efficiency, 20% better performance density and 40% greater overall performance than devices based on previous generation Bifrost GPU.

Contact Us

Specification

Features Value Description
Anti-Aliasing 4x MSAA
8x MSAA
16x MSAA
Hardware-implemented Full-Scene Multiple Sample Anti-Aliasing. 
API Support OpenGL®ES 1.1, 2.0, 3.0, 3.1, 3.2
Vulkan1.0*
OpenCL™ 1.1, 1.2, 2.0
RenderScript
Full support for next-generation and legacy2D/3D graphics applications.
Bus Interface AMBA®4 ACE, ACE-LITE and AXI Compatible with a wide range of bus interconnect and peripheral IP.
L2 Cache Configurable 128kB-2048kB From 1-4 slices, each configurable from 128kB – 512kB.
Memory System Virtual Memory Built-in Memory Management Unit (MMU) to support virtual memory.
Multi-Core Scaling 1 to 32 cores Optimized for high energy efficiency to address the high-end mobile and consumer device requirements.   
Adaptive Scalable Texture Compression

Low dynamic range (LDR) and high dynamic range (HDR).
Supports both 2D and 3D images
ASTC offers a number of advantages over previous texture compression schemes by improving image quality, while reducing memory bandwidth and thus energy use.
Arm Frame Buffer Compression 4x4 pixel block size AFBC is a lossless image compression format that provides random access to pixel data to a 4x4 pixel block granularity. It is employed to reduce memory bandwidth both internally within the GPU and externally throughout the SoC.
Transaction Elimination 16x16 pixel block size Transaction Elimination spots the identical pixel blocks between two consecutive render targets and performs a partial update to the frame buffer with the changed pixel blocks only, which reduces memory bandwidth and thus energy.
Smart Composition 16x16 pixel block size Smart Composition extends the concept of Transaction Elimination to every stage of UI composition. Identical pixel blocks of input surfaces are not read, not processed for composition and not written to final frame buffer.

Performance - Mali-G72 (MP32)

Features Value Description
Frequency 850 MHz in 16nm (16 FinFET)
Throughput 850 Mtri/s, 27.2 Gpix/s in 16nm (16 FinFET)

  • A desktop, a folder, 3D shapes etc.
  • Development Tools for Graphics and Compute Applications

    A range of development tools to assist in the deployment of graphics applications and content on Mali GPU based systems.

  • A phone, a tablet, game console etc.
  • Mali Developer Centre

    An online portal for a growing community of developers, technology partners, software vendors and content companies to create a thriving community around Mali embedded graphics IP.

    Learn more

Get Support

Community Blogs

Community Forums

Answered Why does the CPU perform better than the GPU when blending small areas? 0 votes 1833 views 1 replies Latest 12 days ago by Peter Harris Answer this
Suggested answer E/MaliEGL: FindAppHintInFile open fail errno=0x2 (No such file or directory)
  • Mali OpenGL ES SDK for Android
  • Android Gralloc
  • rendering
0 votes 3733 views 2 replies Latest 24 days ago by crescent72 Answer this
Answered Zero Copy Buffers using cl_arm_import_memory extension in OpenCL 1.2 - arm mali midgard GPUs.
  • Midgard
  • Mali GPU (Midgard Architecture)
  • Mali OpenCL SDK
0 votes 5203 views 7 replies Latest 25 days ago by Kévin Petit Answer this
Suggested answer Optimised OpenCL SGEMM implementation for ARM Mali Midgard GPUs.
  • High Performance Computing (HPC)
  • OpenCL
  • High-Performance Computing (HPC)
  • Mali GPU (Midgard Architecture)
  • Mali OpenCL SDK
0 votes 6385 views 1 replies Latest 27 days ago by Kévin Petit Answer this
Answered Irregular behaviour of vectors in OpenCL(1.2) kernels
  • OpenCL
  • Mali GPU (Midgard Architecture)
  • Mali OpenCL SDK
  • Linux OpenCL
0 votes 9259 views 1 replies Latest 27 days ago by Kévin Petit Answer this
Answered Map/Unmap operations with Zero copy buffer.
  • OpenCL
  • High-Performance Computing (HPC)
  • Mali GPU (Midgard Architecture)
0 votes 6462 views 1 replies Latest 27 days ago by Kévin Petit Answer this
Answered Why does the CPU perform better than the GPU when blending small areas? Latest 12 days ago by Peter Harris 1 replies 1833 views
Suggested answer E/MaliEGL: FindAppHintInFile open fail errno=0x2 (No such file or directory) Latest 24 days ago by crescent72 2 replies 3733 views
Answered Zero Copy Buffers using cl_arm_import_memory extension in OpenCL 1.2 - arm mali midgard GPUs. Latest 25 days ago by Kévin Petit 7 replies 5203 views
Suggested answer Optimised OpenCL SGEMM implementation for ARM Mali Midgard GPUs. Latest 27 days ago by Kévin Petit 1 replies 6385 views
Answered Irregular behaviour of vectors in OpenCL(1.2) kernels Latest 27 days ago by Kévin Petit 1 replies 9259 views
Answered Map/Unmap operations with Zero copy buffer. Latest 27 days ago by Kévin Petit 1 replies 6462 views