PRODUCT OVERVIEW
The AMD Embedded G-Series processor is the world’s rst
integrated circuit to combine a low-power CPU and a discrete-level
GPU into a single embedded Accelerated Processing Unit (APU).
This unprecedented level of graphics integration builds a new
foundation for high performance multi-media content delivery in a
small form factor and power efcient platform for a broad range of
embedded designs. Based on a power-optimized core, the
AMD Embedded G-Series platform delivers levels of performance
in a compact BGA package that is ideal for low power designs in
embedded applications such as Digital Signage, x86 Set-Top Box
(xSTB), IP-TV, Thin Client, Information Kiosk, Point-of-Sale, Casino
Gaming, Media Servers, and Industrial Control Systems.
LOW POWER, EXCEPTIONAL PERFORMANCE
The AMD G-Series platform is a exible solution that in the same
package offers scalable choices for the level of CPU performance,
power efciency, and visual experience. The AMD G-T16R APU with
an average power of only 2.3 Watts2 enables very small form factor,
fan-less and portable applications. APU congurations are
available with single or dual x86 cores, at 4.5W, 9W or 18W
thermal design power (TDP), and two levels of graphics and video
performance. Each APU supports single or dual-independent high
resolution displays and exceptional multi-media capability with
hardware decode support for H.264, VC-1, MPEG2, WMV, DivX
and Adobe® Flash. When paired with the AMD A50M controller hub
with support for advanced interfaces such as 6Gb/s SATA,
Generation 2 PCI Express®, and HD Audio, the AMD G-Series
platform delivers a low power, value-oriented solution for
applications requiring a better balance of CPU and multimedia
performance. For high-end, full- featured platforms, the AMD A55E
controller hub is an alternative pairing choice with added features of
Gigabit Ethernet MAC, RAID (0/1/5/10) support with FIS-based
switching, and PCI local bus support.
HEADLESS EMBEDDED SYSTEMS
For deeply embedded headless systems, the AMD Embedded
G-Series CPU models T48L and T30L enable whisper-quiet
media servers and model T24L enables fanless design for factory
control systems.
FEATURES AND BENEFITS
A exible platform that in one package offers many choices
for power efciency, visual experience with high resolution,
and fully connected Internet experience with maximum
multi-media capability. Delivering unprecedented high denition
visual experience while still maintaining a low power design.
> DirectX® 11 support lets you enjoy awesome graphics performance,
stunning 3D visual effects and dynamic interactivity
> Advanced discrete-level GPU with OpenGL 4.0 and OpenCL™
1.1 support in an integrated device provides support to build the
designs of tomorrow, today
> Unprecedented graphics performance/per watt thru advanced
graphics and hardware acceleration delivering over 3X
performance per watt of previous generation1.
> Selective models, T56N and T40N, have additional boost
capability enabled by AMD Turbo CORE technology without
additional power draw.
Enabling Innovative Small Form Factor Designs - Smaller foot
print and low power solution reduces overall system costs.
> The integration of the APU reduces the foot print of a traditional
three-chip platform to two chips, the APU and its companion
controller hub. This simplies the design, requiring fewer
board layers and a smaller power supply, thereby driving down
system costs.
Reduced Costs over the Product Life Cycle - A single, scalable
platform design that allows OEMs to reduce development costs,
optimize solutions and increase product stability.
> A single APU architecture with an array of performance
options, the AMD Embedded G-Series platform lets OEMs utilize
a single board design to enable solutions that span from entry
level to high end. The commonality of the scalable platform
design across multiple product variants can help reduce
development costs, simplify the supply chain and
reduce operational complexity.
> AMD’s embedded lifecycle, with planned availability of at
least 5 years, helps to ensure a long life for an OEM design.
1. When comparing the AMD Embedded G-Series T44R platform to the AMD Sempron 210U/780E platform running 3DMark 06. Internal testing of current vs. previous generation AMD processor-based
embedded systems as of October 13, 2010 showed more than 3X graphics performance-per-watt advantage for the current generation. Current system: AMD G-T44R APU (9W TDP), Inagua development
platform, 2 GB RAM, Windows® 7. Previous generation: AMD Sempron 2010U processor (15W TDP), MSI 9858 motherboard, 2 GB RAM, Windows 7. Graphics performance-per-watt calculated based on
3DMark 06 benchmark divided by solution thermal design power (TDP).
2. While running a Winbench 99 business graphics benchmark the AMD G-T16R APU consumed an average of 2.284W. I/O Controller Hub power is estimated based on the measured average power drawn
by the I/OCH of .965W during a run of 3DMark 06. System Conguration: AMD G-T16R APU (DVT) at 30°C, “Inagua” Development Board, 4GB 1.35V DDR3, Windows 7 Ultimate. EMB-26
The world’s first combination of low-power
CPU and advanced GPU integrated into
a single embedded device.
AMD EMBEDDED G-SERIES PLATFORM
EMBEDDED
SOLUTIONS
APU FEATURES
> High speed device interconnect
> High performance integrated x86 core(s)
> Integrated cutting edge graphics processor
> High-bandwidth, low-latency integrated
memory controller
> Low-latency platform interface
X86 CORE ARCHITECTURE
> Single or dual x86 Processor
> Advanced Branch Prediction
> Out-of-Order Instruction Execution
- Full OOO Instruction Execution
- Full OOO Load/Store Engine
> High Performance Floating Point
> AMD64 64-bit ISA
> SSE1,2,3, SSSE3 ISA, SSE4A, MMX™
> Secure Advanced Virtualization
GPU CORE ARCHITECTURE
> Dedicated graphics memory controller
- High efciency ring bus memory controller
- Direct connection to memory
> 2D Acceleration
- Highly-optimized 128-bit engine, capable
of processing multiple pixels per clock
> 3D Acceleration
- Full DirectX® 11 support, including full speed
32-bit oating point per component operations
- Shader Model 5
- OpenCL™ 1.1 support
- OpenGL 4.0 support
> Motion Video Acceleration
- Dedicated hardware (UVD 3) for H.264,
VC-1 and MPEG2 decode
- HD HQV and SD HQV support: noise removal,
detail enhancement, color enhancement,
cadence detection, sharpness, and advanced
de-interlacing
- Super up-conversion for SD to HD resolutions
INTEGRATED DDR3 MEMORY CONTROLLER
> Support for solder-down, SODIMM and DIMM
memory, two slots, non-ECC type
> 64-bit DDR3 SDRAM controller operating at
frequencies up to 1333MT/s (667MHz)
LARGE HIGH-PERFORMANCE ON-CHIP
CACHE
> 32KB I-Cache, 32KB D-Cache
> 512KB L2 per core
ENABLING FAN-LESS, HIGH RELIABILILTY,
LOW COST SOLUTIONS
> The two low power models, dual-core T40E and
single-core T40R, deliver up to 40% power
savings compared to the other models in the
AMD Embedded G-Series product family. They
expand the fanless design options for system
architects and design engineers who want to
build a more reliable, lower cost solution with
smaller foot print without compromising on
features for video, graphics and support for
multiple displays.
AMD VIRTUALIZATION TECHNOLOGY (AMD-V)
> SVM lock and unlock
> Nested paging
> Next RIP
> LBR virtualization
> 8 Address space identiers
> Performance counter guest/host bit
> Nested page table fault info
I/O
> 4x1 or 1x4 PCI Express® interface on APU
> See controller hub table for detail I/O features of
A50M and A55E
> Full ISA bus support available, please see Delivering
Full ISA Support with the AMD Embedded G-Series
Platform and the ITE Tech, ITE888 PCI to ISA Bridge
posted on the AMD Embedded Developers Web site
PID (51762)
PACKAGE
> APU
- 413-pin lidless micro BGA
- 19mm x 19mm
> Controller Hub
- 605-pin lidless
2. Low voltage (1.35V) DDR3 is assumed for the 9W TDP processors. The use of 1.5V DDR3 will incur a power adder. Always refer to the processor/chipset data sheets for
technical specications. Feature information in this document is provided for reference only.
3. Models enabled by AMD Turbo CORE technology, up to 10% clock speed increase is planned. For CPU boost, only one processor core of a dual-core has boost enabled.
Key Architecture Benefits
The AMD Embedded G-Series APU family supports dual independent display with different display
technologies and is ideal for Thin Client, Digital Signage, Smart Kiosk/Point-of-Sales, Casino Gaming,
Internet Ready IP-TV and Set-Top Box. Depending on the model and the respective thermal TDP, the
GPU is capable of different resolution support.
AMD Embedded G-Series platform with dual display options
Display One Resolution Extended Display Resolution1
DisplayPort / eDP2
2560 x 1600 at 60Hz
(1920 x 1200 at 60Hz)
DisplayPort 2560 x 1600 at 60Hz (1900 x 1200 at 60Hz)
HDMI1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
Single-link DVI 1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
VGA 2048 x 1536 at 60Hz (1900 x 1200 at 60Hz)
Single-link LVDS3,4
1440 x 900 or 1400 x 1050 at 60Hz
(1024 x 600 at 60Hz)
DisplayPort 2560 x 1600 at 60Hz (1900 x 1200 at 60Hz)
HDMI1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
Single-link DVI 1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
VGA 2048 x 1536 at 60Hz (1900 x 1200 at 60Hz)
Single-link DVI
1920 x 1200 at 60Hz
(1920 x 1200 at 60Hz)
HDMI1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
Single-link DVI 1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
VGA 2048 x 1536 at 60Hz (1900 x 1200 at 60Hz)
VGA
2048 x 1536 at 60Hz
(1920 x 1200 at 60Hz)
HDMI™ 1920 x 1200 at 60Hz (1900 x 1200 at 60Hz)
Notes:
1. Extended displays connected to any interface of the AMD Radeon HD 6290/6250 graphics support a maximum resolution of 1920 x 1200 at 60Hz.
2. Internal LCD panel.
3. Internal LCD panel, DP0 only.
4. 24-bit, dual-channel, LVDS support option available. See the AMD FT1 Processor Display Option Schematic (PID 48601) on the AMD Embedded Developer Support Web Site (wwwd.amd.com/dev).
AMD G-Series Accelerated Processing Units
Model OPN Package CPU
Cores
L2
Cache
/Core
Memory
Interface
CPU Core
Frequency
Discrete Class
Graphics
GPU Core
Frequency
Hardware
Video
Acceleration12
Graphics Display
Outputs13
Display
Resolutions
Thermal
Design
Power
Tdie Product
Release
T56N GET56NGBB22GVE
FT1 413-pin
µBGA 19mm2
2512KB DDR3-1333,
Unbuffered
1.65GHz17 AMD Radeon
HD 6320
500MHz UVD 3 for H.264,
VC-1 and MPEG2
with Blu-ray
support
DirectX® 11
Shader Model 5
OpenCL 1.1
OpenGL 4.0
Dual independent
display controllers
2 active outputs
from:
1x VGA
2x single link DVI
1x single link LVDS
2x DisplayPort 1.1a
1x HDMI
1x DVO
VGA: 2560x1600
(HD 6320, HD 6310)
or 1920x1200 (HD
6290, HD 6250)
30 bpp
Single link DVI:
1920x1200 24 bpp
HDMI: 1920x1080p
36 bpp Single link
LVDS14: 1400x1050
18 bpp
DisplayPort 1.1a:
2560x1600 (HD 6310)
or 1920x1200 (HD
6250) @ 30 bpp
18W 90°C Q2-11
T56E GET56EGBB22GVE 2512KB DDR3-1333,
Unbuffered
1.65GHz17 AMD Radeon
HD 6250
275MHz UVD 3 for H.264,
VC-1 and MPEG2
18W 90°C Q2-12
T52R GET52RGBB12GVE 1512KB DDR3-1333,
Unbuffered
1.5GHz AMD Radeon
HD 6310
500HMz UVD 3 for H.264,
VC-1 and MPEG2
with Blu-ray
support
18W 90°C Q2-11
T48E GET48EGBB22GVE 2512KB DDR3-1066,
Unbuffered
1.4GHz AMD Radeon
HD 6250
280MHz UVD 3 for H.264,
VC-1 and MPEG2
18W 90°C Q2-12
T44R GET44RFPB12GVE 1512KB DDR3-1066,15
Unbuffered
1.2GHz AMD Radeon
HD 6250
280MHz
UVD 3 for H.264,
VC-1 and MPEG2
9W 90°C Q2-11
T40N GET40NFPB22GVE 2512KB DDR3-1066,15
Unbuffered
1.0GHz17 AMD Radeon
HD 6290
280MHz 9W 90°C Q2-11
T48N GET48NGBB22GVE 2512KB DDR3-1066,
Unbuffered
1.4GHz AMD Radeon
HD 6310
520MHz 18W 90°C Q2-11
T40E GET40EFSB22GVE 2512KB DDR3-1066,15
Unbuffered
1.0GHz AMD Radeon
HD 6250
280MHz 6.4W 90°C Q2-11
T40R GET40RFSB12GVE 1512KB DDR3-1066,15
Unbuffered
1.0GHz AMD Radeon
HD 6250
280MHz 5.5W 90°C Q2-11
T16R
GET16RFWB12GVE
1512KB LVDDR3-1066 615MHz AMD Radeon
HD 6250
276MHz VGA: 1920x1200
(HD 6290, HD 6250)
30 bpp
Single link DVI:
1920x1200 24 bpp
DisplayPort 1.1a:
1920 x 1200
(HD 6250) @ 30 bpp
4.5W 90°C Q2-12
T48L
GET48LGBB22GVE
2512KB DDR3-1066,
Unbuffered
1.4GHz N/A N/A N/A N/A N/A N/A 18W 90°C Q2-11
T30L
GET30LGBB12GVE
1512KB DDR3-1066,
Unbuffered
1.4GHz N/A N/A N/A N/A N/A N/A 18W 90°C Q2-11
T24L
GET24LFQB12GVE
1512KB DDR3-1066,15
Unbuffered
1.0GHz N/A N/A N/A N/A N/A N/A 5W 90°C Q2-11
AMD G-Series Platform Controller Hubs
Model OPN CPU
Interface Package PCI
Express®PCI SATA FIS-Based
Switching
Ethernet
MAC USB HD Audio
LPC
SPI
SMBus
Max
GPIOs
APU Fan
Control
APU Clock
Gen Power16
A50M
100-CG2198
1x4 Gen 1 FCBGA
605-pin
lidless µBGA
2mm2
4x1 Gen 2 No 6x 6Gb/s No No 14 v2.0
2 v1.1
Up to
4-channels
Yes 102 Yes Yes 2.7W-
5.9W
A55E
100-CG2293
1x4 Gen 2 FCBGA
605-pin
lidless µBGA
2mm2
4x1 Gen 2 33MHz
4 Slots
6x 6Gb/s
with RAID
0,1,5,10
Yes Yes
(non-EEE
PHY only)
14 v2.0
2 v1.1
Up to
4-channels
Yes 102 Yes Yes 2.7W-
5.9W
12. Refer to the Brazos Platform Minimum System Recommendations for HD Video Playback, order# 48826 to view the minimum system congurations required to enable HD playback and the maximum resolution supported for
each advanced video quality feature.
13. Other resolutions available that do not oversubscribe link bandwidth. Display resolutions limited by available memory bandwidth.
14. eDP translator option supporting 2048x1536 (HD 6310) or 1920x1200 (HD 6250) at 18 or 24 bpp.
15. Low voltage (1.35V) DDR3 is assumed for the 9W TDP processors. The use of 1.5V DDR3 will incur a power adder.
16. Conguration dependent. See product databook for congurations.
17. Models enabled by AMD Turbo CORE technology, up to 10% clock speed increase is planned. For CPU boost, only one processor core of a dual-core has boost enabled.
www.amd.com/embedded
©2012 AMD, the AMD Arrow logo, AMD Sempron, Radeon and combinations thereof, are trademarks of Advanced Micro Devices. DirectX and Windows are registered trademarks of Microsoft Corporation in the U.S.
and/or other jurisdictions. HDMI is a trademark of HDMI Licensing, LLC. OpenCL is a trademark of Apple Inc. used by permission by Khronos. Adobe is a registered trademark of Adobe Systems Incorporated. PCIe and
PCI Express are registered trademarks of PCI-SIG. 3DMark is a trademark of Futuremark Corporation. Other names are for informational purposes only and may be trademarks of their respective owners. PID 49282I
AMD G-Series Platform Block Diagram