A Basic Overview of GPU HD Video Processing

Feb 13 2010

videodecodeheader.jpg

CHARTS UPDATED January 7, 2011!

With all kinds of questions about what GPU can accelerate what codec? and how much acceleration does it offer? I've been working on helping to answer these questions in various revisions of this guide off and on since 2008...

Updated with Flash 10.1 and HDMI 1.4 information!

The Table and Notes

UPDATED January 7, 2011

Current GPUs

Graphics processor

VC-1
decode ability

H.264 decode ability

MPEG-2 decode ability

Adobe Flash 10.1 acceleration

Advanced HD Deinterlacing

Other Notes

AMD/ATI 780G/785G IGP [Radeon HD 3200/4200]

880G/890GX IGP [Radeon HD 4250/4290]

 full  full  advanced (780G) / full (785G or higher)  Yes
(Catalyst 9.11 or higher)
 A Phenom architecture based CPU is required to do advanced deinterlacing.

785G and higher add UVD Gen. 2 for dual-stream decode acceleration.

**None of these chipsets support multichannel LPCM or bitstreaming HD audio over HDMI**

AMD/ATI Radeon HD 4550 or higher  full  full  full  Yes
(Catalyst 9.11 or higher)
 Yes -- current driver issues prevent cards under a 4670 from doing Vector Adaptive.

 UVD Gen.2 -- Offers dual stream decode for hardware accelerated PiP

 8-channel LPCM support over HDM

AMD/ATI Radeon HD 5450 or higher

 full full  full   Yes
(Catalyst 9.11 or higher)
 Yes -- the 5450 is right on the line of just having enough power to do VA when used with a display/monitor no higher than 1080p.

 UVD Gen. 2 -- Offers dual stream decode for hardware accelerated PiP

 5000 series adds proper HDMI 1.3 TrueHD/DTS-HD bit streaming

AMD/ATI Radeon HD 6850 or higher

 full full  full   Yes  Yes

UVD Gen. 3

Supports HDMI 1.3 TrueHD/DTS-HD bit streaming and Blu-ray 3D MVC via HDMI 1.4

NVIDIA GeForce 9400GT / GT 210  advanced  full  advanced  Yes
(driver version 195.62 or higher)
 Video sources only. Fails inverse 3:2 of 1080i sources  Memory starved sibling of GeForce 9500GT / GT 220
NVIDIA GeForce 9500 or higher and GTS/GTX 200 series  advanced  full  advanced  Yes
(driver version 195.62 or higher)
 Yes  Offers dual stream decode for hardware accelerated PiP

NVIDIA GeForce 8200/8300/9300[ION]/9400 IGP

 full  full  advanced  Yes
(driver version 195.62 or higher)

 Yes**

NOTE: A Phenom architecture based CPU is required for AMD IGPs to do advanced deinterlacing.

 8-channel LPCM support over HDMI

 Working 1080p/24 support

**May be broken in drivers past 182.5

NVIDIA GeForce GT 220 or better

 full  full  advanced  Yes
(driver version 195.62 or higher)
 Yes

  8-channel LPCM support over HDMI

GT 240 or better support HDMI 1.4 and the H.264 MVC codec used in 3D Blu-ray discs.

NVIDIA GT 430 or better

 full  full  advanced  Yes  Yes

8-channel LPCM support over HDMI

460/450/430 supports HDMI 1.3 bit streaming

480/470 only support DD/DTS bit streaming and 8-channel LPCM

Supports HDMI 1.4 and the H.264 MVC codec used in 3D Blu-ray discs.

NVIDIA GTX 570/580

 full  full  advanced  Yes  Yes

8-channel LPCM support over HDMI

580/570 only support DD/DTS bit streaming and 8-channel LPCM

Supports HDMI 1.4 and the H.264 MVC codec used in 3D Blu-ray discs.
Intel HD Graphics
in Intel Core i5 6xx / Core i3 5x0 family (aka Clarkdale)
 full  full  full   Yes
(driver 15.16.5.2021 or higher)
 Yes

 An evolution to the X4500 IGP, now integrated into the CPU package of Clarkdale based Core i5 and Core i3 processors.

 Offers dual-stream decode for hardware accelerated PiP.

 Adds proper HDMI 1.3 TrueHD/DTS-HD bitstreaming in addition to 8-channel LPCM.

Intel Processor Graphics in second generation Intel Core i7/i5/i3 2xxx family (aka Sandy Bridge)
 full  full  full   Yes  Yes

The next generation of Intel graphics with the CPU and GPU integrated on the same die.

Offers dual-stream decode for hardware accelerated PiP and HDMI 1.4 (3D MVC).

Includes HDMI 1.3 TrueHD/DTS-HD bitstreaming in addition to 8-channel LPCM.

Chart Legend

Moderate = Most of the steps in the decocde pipeline are offloaded, however the last steps (which can be some of the most intensive) in the decode process are not. These GPUs lack acceleration for the IDCT [Frequency Transform] and the VLD or CAVLC/CABAC [Entropy Decode] portions of the decode process, they only offer Motion Compensation [Pixel Prediction] and Deblocking.

Advanced = These GPUs accelerate the IDCT portion of the codec, but stop short of full hardware acceleration of the entire decode process.

Full = As the name implies the decode process is fully accelerated, leaving only minimal overhead for the system's processor to handle.

Advanced HD Deinterlacing = this refers to two technqiues for 1080i sources:
1. An enhanced motion adaptive deinterlacing for video sources.
2. Proper inverse telecine (3:2 cadence) detection for film based sources.

Advanced Deinterlacing Product Names:

  • ATI has two modes of advanced deinterlacing: the highest quality is 'Vector Adaptive' followed by 'Motion Adaptive'. Vector Adaptive (VA) is the level needed for proper adaptive deinterlacing of all types of 1080i content.
  • NVIDIA's advanced deinterlacing is referred to as 'Spatial-Temporal Deinterlacing' in various PureVideo documentation.
  • Intel's advanced motion-adaptive deinterlacing is folded into their 'Clear Video' moniker.

If deinterlacing for a type of content is not working it is noted in the column. Driver updates may enable advanced deinterlacing features at a future time.

Legacy GPUs

 Graphics processor

VC-1 
decode ability
 

 H.264 decode ability 

 MPEG-2 decode ability 

 Adobe Flash 10.1 acceleration

 Advanced HD Deinterlacing

Other Notes 

  ATI Radeon X1x00 series [includes Radeon HD 2900]  advanced  advanced  advanced  N/A  Depends on shader power, needs a higher end card to do it.   Decode done mostly in shader programs
 ATI/AMD 690G/740G IGP [Radeon 2100]  moderate  moderate  advanced  N/A  Video sources only  GPU is derived from Radeon X700.

 740G is a 55nm fabricated version of the 690G.

 ATI Radeon HD 2400 series / Radeon HD 34x0 series  full  full  advanced  N/A

  Video sources only with 2400 Pro, possibly film sources with 2400XT/3450 - not enough data to be certain.

3470 seems to be just powerful enough to do Vector Adaptive deinterlacing for both film and video sources. However the ability seems to change from driver version to driver version.

 UVD Gen. 1 adds dedicated silicon for full H.264 and VC-1 decode
 ATI Radeon HD 2600 series and HD 3650 and higher  full  full  advanced  N/A  Yes  
 NVIDIA GeForce 7-series [includes GeForce 8800GTS/GTX (G80) and GeForce 7050PV IGP]  moderate  moderate  advanced   N/A  GeForce 7600GT and higher have the power to do inverse telecine on 1080i sources.  Decode mostly done in shader programs.
 NVIDIA GeForce 8400GS (G86)  advanced  full  advanced  Yes (use latest drivers)  Too underpowered  
  NVIDIA GeForce 8500GT, 8600GT/GTS, 8800GT/GS  advanced  full  advanced  Yes (use latest drivers)  Yes   GeForce 8500 and newer add dedicated silicon for full H.264 and partial VC-1 decode
 Intel G35
[GMA X3500]
 moderate  N/A  advanced   

 Video sources only, fails inverse 3:2 of 1080i sources

  8-channel LPCM support over HDMI

 Intel G45
[GMA X4500 HD]

 full  full  full  Yes
(driver 15.16.5.2021 or higher)

 Yes

  8-channel LPCM support over HDMI.

Performance and Misc. Notes 

Standard Definition notes:

Generally, most common graphics chips available today are more then capable of decoding 480i material, deinterlacing it, and post-processing it. Even integrated video chipsets like Intel's G35 scores near perfect on the HQV tests, ATI's 690G scores in the 80s, and the newer 780G scores towards the top with a score of 113 [out of 130].

MPEG-2 Decoding Notes:

With MPEG-2 being pretty easy to decode on modern CPUs neither NVIDIA nor ATI were in a hurry fully accelerated the MPEG-2 decoding process. Both vendors did the acceleration in their pixel shaders up until very recently. With introduction of the ATI Radeon HD 4000 series and its new UVD 2 engine did ATI move to dedicated silicon to handle MPEG2 decode much like the H.264 and VC1 support the earlier Radeon HD cards had.

GeForce 7-series H.264 Decode Notes:

In the GeForce 7-series H.264 decode performance was all done in the pixel shaders and so it is highly dependant on GPU speed, ~500MHz core clock or faster is best for 1080p content.

Post-procesing Notes:

Both ATI and NVIDIA appear to do advanced deinterlacing and 3:2 pulldown detection in the pixel shaders. This means that underpowered graphics cards may not be able to utilize the advanced features. There is some level of correlation between shader power, memory bandwidth, and the ability to deinterlace 1080i properly and/or layer on additional post-processing such as noise reduction. For example the Radeon HD 2400 Pro was widely known to not have enough horsepower to deinterlace 1080i properly. 

Noise reduction and image sharpening, while less useful in HD sources, are tested by the HD HQV test disc. NVIDIA's processing is done in the shaders while ATI's Radeon HD 2x00 and newer have dedicated silicon borrowed from ATI's portfolio of Xileon digital TV products.

NOTE: NVIDIA currently doesn't have noise reduction/image sharpening of HD sources enabled in their Windows XP drivers and likely never will.

More Information:

Wikipedia now has a very through article on NVIDIA PureVideo and the different generations of the hardware and what decoding they support, it also discusses Linux support for video acceleration

http://en.wikipedia.org/wiki/Nvidia_PureVideo

Website design by Yammm Software
Powered by Drupal