Article Index

Looking back at two years of Graphics Core Next

ATI, later AMD endured several major evolutions in its graphics architecture which mirrored industry developments and major revolutions in the PC and graphics industry. At times Radeon graphics was used by Microsoft for the development of their DirectX APU. All roads lead to GCN, which encompasses AMD's vision for a graphics core that can scale top to bottom, for high end gaming graphics cards, consoles and tablets.

AMD Radeon HD 7000 seres graphics gpu evolution



Fixed Function

  • 3dfx VooDoo Graphics series
  • ATI Rage series
  • ATI Radeon 7000 series
  • MATROX G-Series
  • NVIDIA GeForce, GeForce 2

ATI Rage Pro Turbo


Simple Shaders


  • ATI Radeon 8000-9000 series
  • ATI Radeon X1000 series
  • MATROX P-Series
  • NVIDIA GeForce 3 - 7
ATI Radeon 8500

Graphics Parallel Core

  • ATI/AMD Radeon HD 2000-8000 series
  • AMD Radeon R7 series
  • AMD Radeon R9 series
  • NVIDIA GeForce 8000-9000 series
  • NVIDIA GeForce 200-700 series
ATI Radeon HD 2600 Pro

AMD Radeon HD 7000 seres graphics core next architecture

The design goals for the original GCN at launch sound remarkably familiar to those for R9, with the exception being heterogeneous computing which is still somewhat undeveloped and Fusion which is now rebranded as HSA.

AMD Radeon HD 7000 series tahiti block diagram

Tahiti resembles a 'stack of Lego bricks' which, Compared to Hawaii

AMD Radeon R9 290 series hawaii block diagram

Hawaii's GCN '2.0's hardware resources are organised into is organised into four units called "Shader Engines" which allows resources to be scaled and shared more effectively. This effectively mirror's NVIDIAs approach with Kepler expect their name for the topology is SMX units.

AMD Radeon R9 290 graphics core next shader engine

This allows for GPUs to be scaled down more easily by disabling an entire Shader engine (or SMX units) without re-spinning the entire chip to reduce the number of cores or fusing off clusters of cores. There is still some resource sharing within resources contained in each shader engine such as renderers and caches.

Each Shader Engine contains1 rasteriser and 1 Geometry Unit which can load balance, 1 Shader Engine is sufficient to operate the entire GPU.

This diagram shows a simplified overview of the graphics pipeline

  • The Graphics command processor overseas operations across load balanced resources
  • Geometry is setup and tessellated in The Geometry processors. Data can be exchanged with the compute units if needed or sent to Rasteriser directly.
  • The compute units execute pixel shaders or perform GPU computing on the scene
  • Pixel data is then passed onto the rasterisers which handle assignment or partitioning of pixels on the screen as well as Hierarchal Z sorting, i.e. the pixels depth in the scene
  • Finally, the Render Back Ends handle Pixel Depth Testing as well as stencilling and colour operations

AMD Radeon R9 290 graphics core next gpu pipeline and geometry processing

Further to do actual processing, Each of Hawaii's Shader Engines contain 11 Compute Units. The Compute Unit is the smallest physical processing block of the GPU containing all of the necessary low level building blocks that a compute processor needs to fetch, decode and execute instructions.

AMD Radeon R9 290 graphics core next compute unit

The final stage are the Render Back Ends which handle operations relating to the scene's Z(Depth), Stenciling and Color.

AMD Radeon R9 290 graphics core next render backend

That is all the graphics and compute processing pipelines explained, but a GPU many processors in parallel, which need to be fed tasks and be directed.

We need a means of scheduling and dispatching to allow the GPU to perform multi-tasking across its parallel computing units. This is where the Asynchronous Compute Units come in, which Hawaii has 8 of which are independent of the Shader engines. The ACE units queue, store and share data for use in GPU computing across the entire GPU. Graphics specific commands are issued by a separate command unit.

AMD Radeon R9 290 graphics core next aysnc compute unit

So in summary the layout of Graphics Core Next Architecture, 'version 2' as used in the 290X is essentially a scaled up version of Tahiti.

GCN '2.0' supports:

  • 1-8 Asynchronous Compute Engines
  • 1-4 Shader Engines
  • 1-11 Compute Unit per Shader Engine, giving 64 to 704 shaders per Shader Engine

AMD Radeon R9 290 graphics core next hawaii block simplified diagram

Tahiti v Hawaii – spec comparison


AMD Graphics Core Next

AMD Graphics Core Next


Compute Units/ IEEE-2008 Compliant  Shaders

32 / 2048

44 / 2816


Geometry Processors




Render Back-Ends
Color Operations
Depth/Stencil Operations




L2 Cache




Memory Bus

384 bit wide GDDR5 264 GB/sec

512-bit wide GDDR5 320GB/s


In addition to the increased GPU resources, Hawaii adds updating display controllers for Eyefinity, AMD TrueAudio and a new version of CrossFire.

GCN v Kepler Architecture Performance & Efficiency – spec comparison


AMD Radeon
HD 7970 GHz Edition

AMD Radeon
R9 290X


NVIDIA GeForce GTX 780



Geometry Processing

2.1 billion primitives/sec

4 billion primitives/sec









Texture Fill Rate

134.4 Gtexels/sec

176 Gtexels/sec


166 Gtexels/sec

188 Gtexels/sec

Pixel Fill Rate

33.6 Gpixels/sec

64 Gpixels/sec


41.4 Gpixels/sec

40.2 Gpixels/sec

Peak Bandwidth

264 GB/sec



288 GB/sec

288 GB/sec

Die Area

352 mm^2

438 mm^2


561 mm^2

561 mm^2

Peak GFLOPS/mm^2






While Peak raw power and computing have not significantly increased, the GPU’s horsepower within its engines is much stronger with almost 2x throughput available for 3D Graphics intensive tasks such as pixel shaders and geometry at only a 25% increase in die size. On paper 290X is more efficient, thanks to its ‘higher horsepower’ design at a smaller die size than the Kepler GK110 based GTX 780.

On paper, 290X provides a good step-up from the previous generation HD 7970. The lower compute performance for NVIDIA GeForce is expected as this is a hallmark of their consumer oriented GPU.

2016-08-29 | LG puts OLED in all the things - Integrate/CEDIA AUS 2016 Booth Tour

2016-08-12 | What is Software Defined Networking with Doug Murray, CEO Big Switch Networks.

2016-06-17 | NVIDIA DGX-1 launches in Australia - NVIDIA ANZ Interviewed

2016-05-01 | Build a kickass Dolby ATMOS Home Theatre - The Pros show us how

2016-04-01 | The Panasonic Toughbook CF-20 Story

2016-03-23 | Seagate 8TB Family, 10TB Helium & Mobile HDD 2016 Hard Disk Update

2016-03-21 | Panasonic Toughbook 20 : Fully Rugged detachable Laptop launch - Water, Drop & Weight Tested

2016-03-23 | ASUS ROG Gaming House Tour - Sydney, Australia

2016-03-01 | ASUS ROG GX700 Liquid Cooled Notebook Preview & Interview

2016-03-23 | League of Legends Legacy eSports Team Interview

2015-12-09 | Microsoft Store SYDNEY - Grand Opening Press Conference

2015-11-11 | HP 6th Gen Core Intel Skylake PC & Jet Intelligence Printing Launch - Melbourne Experience Centre

2015-11-11 | HP Experience Centre Melbourne Tour -Part 2

2015-11-11 | HP Experience Centre Melbourne Tour -Part 1

2015-08-28 | Parrot Interviewed - Flower Power Wireless Plant Monitor

2015-08-28 | Parrot Interviewed - ZiK 2.0 Headphones

2015-08-28 | Parrot Interviewed - AR. Drone, MiniDrones, Bebop Drone

2015-08-23 | MOTO G 3rd Gen IPX7 Water Proofing Demo & Moto X Style/Play/G Q&A

2015-08-21 | Motorola Mobility Interviewed - 3rd Gen Moto G, Moto X Play, Moto X Style

2015-08-20 | NVIDIA GEFORCE GTX 950 Launch Ep.1 | Gamestream Co-Op - A New way to share games

2015-08-20 | NVIDIA GEFORCE GTX 950 Launch Ep.2 | MOBA/DOTA 2 Latency and Smoothness

2015-08-08 | Intel SKYLAKE i5-6600K LN2 overclock at 6.33 GHz with ASUS Z170 ROG Hero

2015-08-07 | Intel Skylake i7-6700K @ 4.8 GHz & DDR4 @ 3.5GHz on ASUS Z170 Deluxe Overclocking Guide

2015-08-05 | ASUS Z170 Deluxe Motherboard for Intel Skylake 6th Gen Core CPU Overview

2015-08-05 | Review - ASUS Installation Tool for Intel SKYLAKE 6th Gen Core CPU

2015-08-02 | HP Pavilion and Envy 2015 Consumer Notebook First Look

2015-07-25 | HP Envy Phoenix 2015 Overclockable Gaming Tower First Look

2015-07-22 | Ashes Of The Singularity - DirectX 12 Hands-On Demo with AMD FURY X

2015-07-14 | Media Q&A with AMD and Richard Huddy - AMD Fury X ‘Fiji’ Graphics comes to Australia

2015-07-14 | MSI Interview - AMD Radeon R9 300 Gaming, Armour and Lightning GPUs

2015-07-14 | AMD's Richard Huddy Interviewed - AMD Radeon Fury X ‘Fiji’ Graphics comes to Australia

2015-07-14 | ASUS Interview - AMD Radeon R9 300 STRIX, Auto Extreme Technology and ROG MATRIX cards

2015-07-14 | Simplivity Omnicube Server Accelerator - Hyperconvergence Live Demo and Technical Interview

2015-07-14 | Simplivity Interviewed - What is Hyperconvergence & the Omnicube Server Accelerator

2015-06-21 | Media Q&A - NVIDIA GeForce GTX 980 Ti, Mobile G-SYNC & Gameworks VR Launch

2015-06-21 | Mobile G-SYNC explained - NVIDIA Mobile G-SYNC & Gameworks VR Launch

2015-06-21 | Gameworks VR explained - NVIDIA Mobile G-SYNC & Gameworks VR Launch

2015-06-21 | DirectX 12 Explained - NVIDIA Mobile G-SYNC & Gameworks VR Launch

2015-06-05 | NVIDIA Mobile G-SYNC & Gameworks VR - First Look, Q&A and Demos

2015-04-09 | HP Zvr 3D Virtual Reality zSpace Monitor First Look and Hands-On Demos

2015-04-09 | HP 2015 Zbook Workstation Ultrabooks and Displays First Look

2015-03-30 | EPIC Unreal Engine 4 Kite Demo on NVIDIA GEFORCE GTX TITAN X

2015-03-08 | ASUS Transformer Book Chi T300 Australian Launch

2015-03-08 | Intel RealSense 3D Camera Demos & Games - ASUS Chi T300 Launch

2015-03-07 | Dolby Conference Phone and BT MeetMe Service with Dolby Voice - Demo and Q&A

2015-03-07 | Who is Dolby and What is Dolby Voice

2015-03-07 | The Science and Psychology of Dolby Voice

2015-03-07 | The Great ASUS Comedy Debate - Laptop v Tablet Toilet Humor

2015-03-03 | Dual Intel XEON E5-2643V2 Benchmarked - Cinebench & Blender

2015-02-25 | 2015 HP Pro Slate, Pro Tablet EE & Elitepad 1000 First Look