Factors to consider when choosing the right codec for your ... › download › get › ?path=... · H.264 H.263 VP8 AVC THEORA DIRAC JPEG AVC-INTRA MPEG-2 H.261 MPEG-4 VC-1 34 Feature

H.264

H.263

VP8

AVC

THEORA

DIRAC

AVC-INTRAJPEG

MPEG-2

H.261

MPEG-4

VC-1

www.avsystemsdesign.com 35 34 AV SYSTEMS DESIGN Summer 2011Feature StorY

E very day, video over IP technology reaches farther

into our professional and personal lives. As networks

expand and streaming technologies mature, we find

ourselves with greater access to information and an

improved ability to communicate using video. Some streaming

applications like videoconferencing have been commonplace in

AV systems for many years. New applications continue to emerge

every day, as streaming technologies provide new capabilities

or greater economy to existing applications. As the variety of

streaming products increases, AV professionals will be challenged

with the task of selecting the ideal streaming products and

compression codecs for their applications.

The term “codec” has several meanings. It is the combination

of two words, encoder and decoder. The word “codec” is used

as a generic name for a hardware or software-based product,

such as a videoconferencing codec, which encodes and

decodes video simultaneously. It is also used to describe the

analog-digital-analog conversion process. As we explore the

Selecting a Streaming CodecFactors to consider when choosing the right codec for your streaming application

By Karl Johnson

avsd_summer2011.indd 34 6/6/11 7:12 PM

H.264

H.263

VP8

AVC

THEORA

DIRAC

AVC-INTRAJPEG

MPEG-2

H.261

MPEG-4

VC-1

www.avsystemsdesign.com 35 34 AV SYSTEMS DESIGN Summer 2011 Feature StorY

Low High

Convenience

Mission critical availability

Price Sensitivity

Network bandwidth abundance

Network quality of service

Network management and security

Low communication delay

Range of adjustments and controls

Asset management: Metadata, search tools

Scalability

Resolution and image quality

Simplicity of user interface

Relative ExpectationsIndustrial Consumer

variety of codecs that are available for streaming AV media and

consider the rapid advancement of technology in this area, many

questions emerge, such as:

• Why are there so many different codecs?

• What are the pros and cons of these codecs?

• Is there one codec that is ideal for every application?

• What differences in technical performance exist between

codecs, and in what application environments might they

be important?

• Standards based and proprietary codecs exist. When do

I use one versus the other?

Consumer and Business UseExperiences at home with personal electronics or computers can

shape expectations in industrial applications. It’s not uncommon

to hear customers express their frustration comparing a tried

and true office videoconferencing product with a similar

communication tool they use at home such as Skype. I’ve heard

individuals say, “Why is it that when I’m at home, I can Skype

with my friends and see and hear them so clearly, yet when I’m

in the office with all this expensive equipment, I get such poor

quality picture and sound?”

A lot of factors can contribute to this experience. For starters,

the individual may be using a dated, standard-definition

videoconferencing system in the office. The system may use a

low bandwidth communication link that delivers a fraction of

the resolution that is available on the large, flat panel display

it is used with. At home they may be operating Skype in a

small application window on a moderately sized desktop flat

panel. The quality experienced at home is perceived to be higher

because the video is presented at a higher pixel density in an

application window that matches the streaming resolution. The

Skype video would also appear grainy and pixelated if it was

presented on a large flat panel. Streaming quality is often a

matter of perspective, and gauging “good enough” for a specific

application is important. Regardless of the specific conditions,

customers can have experiences at home that raise expectations

for the office.

Furthermore, applications used on a personal computer

at home will not support the same mission-critical reliability

or provide the security that industrial equipment must.

Industrial and consumer solutions are faced with a different set

of requirements.

The impact and excitement created from the rapid uptake of

tablet PCs and video communication on mobile phones is likely to

expand an expectation gap as users desire to connect professional

AV systems with the convenience of mobile computing devices

and intelligent mobile phones.

Codec StandardsMany codecs have been developed by international or industry

standards organizations. Some codecs have been developed by

independent organizations with intentions for use in open or

closed application environments. Of the codecs that have been

developed independently, some have made a transition from

proprietary to becoming a standard.

The most commonly known codecs have originated from

standards organizations. The Joint Photographic Experts

Group – JPEG, developed the JPEG and JPEG2000 for still image

compression, and these have also been applied in motion video

applications. JPEG2000, the most recent standard, has been

adopted as the compression standard used in the digital cinema

industry for playback of motion pictures from hard disk players.

The Motion Picture Experts Group, or MPEG, of the

International Standardization Organization – ISO developed

the first MPEG-1 codec. It continued with MPEG-2 to provide

increased image quality and support for high definition video.

MPEG-2 has been used in a wide range of products from DVDs,

cable and satellite distribution, as well as high quality broadcast

contribution applications.

The International Telecommunication Union - ITU

recommended the H.323 audiovisual communication protocols,

which included the H.261, H.263, and H.264 video codecs intended

for specific use in interactive videoconferencing applications used

on communication links such as ISDN, T1, E1, or the Internet.

The ITU and ISO came together to form the Joint Video Team

– JVT and developed H.264, MPEG-4 Part 10, Advanced Video

Coding or AVC codec. More commonly referred to as H.264 these

Business applications require greater attention to a broader set of capabilities

than consumer ones, particularly reliability and security.

Comparison of expectations for Industrial and Consumer Communications applications



Codecs

JPEG, JPEG 2000 MPEG-1, MPEG-2, MPEG-4

NOTE: Google aquired On2 Technologies in 2010.

H.261, H.262, H.264

H.264, MPEG-4 Part 10/AVC

VC-1, SMPTE VC-1 Dirac, Dirac Pro SMPTE VC-2 AVC-Intra

Theora VPx, VP6, VP8 (WebM)

Organization

Developer

CODECS AND THE DEVELOPER ORGANIZATION

ISO

JPEG

ISO

MPEG

ITU

H.323 Recommendation

ISO and ITU

JVT

Microsoft

Microsoft

BBC

BBC Research

Panasonic

Panasonic

Xiph. Org Foundation


Google

On2 Technologies

TransportEncoder

HD-SDIor

DVI

PresentationSystem

Production System

Decoder

VIDEO

Live MediaCollection

HD-SDI

DVI

VIDEO VIDEO

MediaExchangePoint

days, this codec has replaced MPEG-2 as the most commonly

used codec applied in new streaming media hardware encoders.

In the broadcast television industry, the Society of Motion

Picture and Television Engineers – SMPTE and the European

Broadcasting Union – EBU have a vested interest in defining

recommended platforms for broadcast video, interchange, and

interoperability for transmission and production systems. They

recommend and define standards for their industry, and their

decisions extend into other industries.

Both Microsoft and Panasonic have produced video codecs

independently, targeted at specific applications. Microsoft’s VC-1

was developed to support a more efficient and higher quality

compression of interlaced video, making it more attractive to

broadcast applications where use of interlaced video is common.

VC-1 started as a proprietary codec, but was later established as

a standard by SMPTE. Microsoft has also applied the VC-1 codec

technology into the WMV3, WMVA, and WVC1 codecs used in

Windows Media Player. The Panasonic AVC-Intra codec is fully

compliant with the H.264 MPEG-4/AVC standard and SMPTE

recommended practices. It is targeted for capturing production

quality 10-bit video collected in camcorders at bit rates typically

used in Electronic News Gathering applications. The format is

also used in storage and archive applications. Development of this

codec aligns with the performance needs of related equipment

that Panasonic manufactures: cameras, decks, and camcorders.

As you can see, both of these organizations developed a codec

to fulfill targeted applications. One, VC-1 started as proprietary,

while the other, AVC-Intra is branded as unique, but is based

on standards. As illustrated with these real world examples,

applications drive codec development.

Open and Closed EnvironmentsMany streaming applications apply codecs in closed network

systems or on managed connections where control exists over the

hardware or software used at the endpoints. Hardware encoding

and decoding solutions will use an electronic signal interface such

as HD-SDI, DVI, or other formats to transfer the media. If software

decoding is used, installation of plug-ins can be managed in the

closed system. Here, the initial concerns for the AV integrator are

interfacing the correct signal format to the encoder input and the

decoder output, and establishing and maintaining the required

communications link or network connection. Beyond that, the

quality, bit rate requirements, and technical performance of the

codec are assessed individually for each project.

Open network environments, such as the Internet or even large

enterprises, will have very large numbers of potential endpoints.

Use of decoding hardware at every potential endpoint will not be

Streaming solutions using an electronic signal for media exchange.

electronic Image Signal Delivered from managed Decoder



Codecs

JPEG, JPEG 2000 MPEG-1, MPEG-2, MPEG-4

NOTE: Google aquired On2 Technologies in 2010.

H.261, H.262, H.264

H.264, MPEG-4 Part 10/AVC

VC-1, SMPTE VC-1 Dirac, Dirac Pro SMPTE VC-2 AVC-Intra

Theora VPx, VP6, VP8 (WebM)

Organization

Developer

CODECS AND THE DEVELOPER ORGANIZATION

ISO

JPEG

ISO

MPEG

ITU

H.323 Recommendation

ISO and ITU

JVT

Microsoft

Microsoft

BBC

BBC Research

Panasonic

Panasonic



Google

On2 Technologies

TransportEncoder

HD-SDIor

DVI

Video Bitstream1001011011

Video Bitstream1001011011

Live MediaCollection

MediaExchangePoint

Decoding on PC

Decoder

Flat Panel Display

VIDEO

VIDEO

practical, and supporting software at each endpoint may not be

practical either. If we consider streaming to PCs or decoders that

are not managed, the media exchange for the streamed content

needs to be as a video bitstream. The endpoints must support

codecs, protocols, or applications that are compatible with the

video bitstream.

Codecs in the Computing EnvironmentProfessional AV systems pull together and manage presentation of

media from three different environments 1) Telecommunications –

videoconferencing systems, 2) Video based on broadcast standards,

and 3) PC originated media. The PC environment creates another

variable for the world of codecs.

Standards in this environment are frequently established

based on market forces and the strength, weakness, or ubiquity

established by commercial organizations or coalitions. The most

dominant players establish de facto standards. Industry giants

such as Microsoft, Apple, Adobe, and Google each pursue different

product and technology paths to support their customers, and as

commercial organizations, they make investments in technology

for the purpose of establishing competitive barriers and achieving

financial objectives.

In the PC streaming environment, a broader set of variables

must often be considered beyond the codec, including the

container format, transport method, media player, Digital Rights

Management, operating systems, and computing devices.

Today, Adobe’s Flash Player delivers the vast majority of

online video. It uses a proprietary container format – Flash Video

installed as a plug-in that is used in Web browsers for viewing

video on PCs. The ubiquity of Flash represents a competitive

advantage for Adobe products, but many developer solutions

in the PC environment would prefer to use an open-source

technology rather than conform to the operational requirements

of using Flash.

A great deal of attention has been paid to the very public

battle between Adobe and Apple concerning use of the Flash

Player on Apple’s iOS operating system used on iPhone and iPod

Video Bitstream Delivered in open environment

Streaming of video to unmanaged or unknown endpoints requires exchange of

media as a video bitstream.

what is a “de facto” standard?a de facto standard is a product, specification, industry convention, or

method that has achieved dominant use in an industry based on market

forces. users, suppliers, and developers all follow the de facto standard

in order to conform to the solution environment. Defacto standards are

not established by standards bodies like ISo or Itu.



Touch. Adobe had been pursuing a programming environment

that crossed platforms and developed software that converted

Flash applications into native iPhone apps. Apple then changed

its developer agreement blocking Flash-derived apps. Adobe

chose to add support for Apple’s HTTP Live Streaming in Flash

Media Server and introduced this capability at the 2011 National

Association of Broadcasters Show, adapting to the restriction Apple

placed on their platform. Both Apple and Adobe pursued their

own interests in hopes of establishing a competitive advantage.

An industry battle such as this creates risks for customers

making investments in software platforms, applications, and

endpoint devices.

A great deal of news has also been generated over the past

year concerning use of codecs in HTML5 - the new Internet

browser standard. HTML5 represents the first opportunity to

add video embedded into the Web experience and workflow as

an object just like text or graphics. Members of the World Wide

Web Consortium HTML Working Group could not agree on a

common codec for use in HTML5. The ideal situation would be

for the group to select a single codec providing interoperability

for all endpoint applications. But this is not what occurred. A

whole series of codec choices have taken place. Initially Theora

and Ogg container, developed by Xiph.Org Foundation, appeared

to be the codec of choice. It was royalty free. H.264 was favored

by Microsoft and it provided strong compression performance.

Members such as Mozilla and Google could not agree to use

H.264 due to licensing requirements and in 2010, Google had

acquired On2 Technologies, developer of the WebM format. This

format includes the VP8 codec, and Google elected to use it with

their Chrome browser. Mozilla also chose to use WebM in Firefox,

as did Opera. VP8 has also been gaining greater use by YouTube.

In 2010, Microsoft announced Internet Explorer 9 support

for H.264 and a new Windows Media Player HTML5 Extension

for Chrome, allowing Windows 7 Chrome users to view H.264

encoded content. This means operating system suppliers can offer

H.264 plug-ins for use in browsers that don’t support it. Likewise,

Google can offer plug-ins for WebM. Most recently, in May 2011,

Microsoft, who has been committed to using H.264, purchased

Skype, which uses the VP8 codec.

Nearly all digital video that is delivered to end users is

developed as an end to end solution. Most of the concerns

about codec standards for video distribution are focused on the

development of content and delivery through connections across

the Internet, cable, satellite, or physical media. Even if the video

content is encoded with a standards based codec, it will likely use

a proprietary container, transport protocol, media player, or Digital

Rights Management scheme. In order to view standards-based

content, the endpoint device must use a browser based plug-in or

a mini application embedded into an appliance or other media

accessory device. Here, proprietary influence has been exerted by

the PC industry members, creating a barrier to the interoperability

that a standards-based codec is envisioned to deliver.

The standards war played out by PC industry giants creates

interesting news; however, this pursuit of competitive advantage

prohibits use of a single codec, and simple interoperability for

Internet streaming. It also demonstrates how popular video

streaming will be on PCs and other devices, and how critical

competitive barriers are to their future profitability.

That “Standards” Topic, AgainStandards are established by industry, national, or international

working groups. They can also be established by unofficial

consortiums or organizations. Standards are essential to

interoperability in communications technologies, particularly

one-to-many or many-to-many applications. Examples of

standards include:

• IEEE 100BaseT – Ethernet networking interface

• NTSC or ATSC - Video broadcast standards

• ITU H.323 – Videoconferencing standard

Products that employ these standards are truly

interoperable. You plug an NTSC signal into a display with

a BNC connector labeled NTSC, and you get a video picture.

what is a container format?a container or wrapper format holds the encoded audio and video

media as well as certain information pertaining to the media such

as: the compression codec, number and types of streams, subtitles,

metadata, and sync information. a container is more important to media

playback files than live streamed content, but media streamed in the pC

environment must factor in the container and media player applications,

that are used to view them. examples of media container formats include:

FLV and F4V used with adobe Flash Video, aSF and aVI from microsoft,

Quicktime from apple, the mpeG-2 transport stream, and mp4 that is

based on ISo’s mpeG-4 part 12.

h.264 Profiles and Levelsthe h.264 compression standard includes 17 profiles and 16 levels for

encoding video for different applications. each profile uses different

techniques with different complexity, requiring different processing power

from decoding devices. each level specifies the data rate and resolution.

Different profiles and levels can be used to target performance specific

to different classes of streaming applications. the h.264 compression

standard is used widely in many different streaming encoders; however,

the variety of profiles and levels available means it is important to select

devices applying a profile and level appropriate for the application, as well

as confirm that encoding and decoding device performance is matched.



The same applies to Ethernet communications or the H.323

videoconferencing standard.

Codec standards typically apply to the data format and

the decoding, leaving opportunities for different product

implementations to optimize or simplify the encoding for a

specific application. Standards are not established for quality;

they are established with a priority to facilitate interoperability.

However, “best solutions” are solicited from the research

community and they make their way into the standard.

As discussed earlier, the AV industry uses broadcast video

standards every day, but there is also a vast array of connections,

resolutions, signal formats, digital rights management, and

encryption methods used by computer-video and consumer video

equipment, which are based on specifications or de facto standards.

As these are not “ratified” standards, the specifications for these

interfaces may not always be followed by manufacturers to a

degree that will deliver the interoperable performance a “ratified”

standard offers. Integration of this type of equipment represents

an ongoing challenge for AV manufacturers and integrators.

Standards based codecs are more likely to exploit the cost

curve of popular technology. For example, increased use of

MPEG-2 or H.264 codecs will create more suppliers, increasing

competition, resulting in a greater variety of products. The

combined effort of many users of the standard will also advance

and improve the capability of the technology over time. However,

use of standard codec does not guarantee interoperability

between products. The processing capability of the streaming

devices, transport protocols, or container formats used are often

proprietary. Compatibility between products may need to be

audited or proven in field tests.

Interoperability is highly desirable in applications with open,

unmanaged endpoints, with one-to-many or many-to-many

delivery paradigms or production workflows where the video

must be delivered as a digital file or bitstream in a common user

Extron’s H.264 Streaming Technology – A Platform

Supporting the Broad requirements of AV Systems

In 2011, extron introduces the Sme 100 Streaming

media encoder, its first streaming product based on the

mpeG-4/aVC h.264 compression standard. the Sme 100

interfaces commonly used DVI, rGB, and standard definition

and high definition video and audio, providing advanced

interfacing and signal processing features common to extron

signal processing products. It features the high quality scaling

technology found in many extron products and offers a wide

range of audio and video compression controls providing

the ability to stream at a variety of resolutions and frame

rates. the combination of compression and bit rate controls

and the flexibility to choose from a variety of streaming

protocols make the product flexible for use in a variety of

network environments.

Selection of the h.264 standard as the codec for aV

streaming products supports an open technology environment,

offering a high degree of interoperability and compatibility for

the Sme 100 with media servers as well as the flexibility to

decode live streams on hardware or software platforms.

the Sme 100 is used in Corporate, education,

Government, and other enterprise applications. It can be used

to stream aV System sources to pCs for monitoring purposes.

presentations or videoconferences can be streamed to

desktop pCs or overflow rooms, extending the reach of aV

systems. It can also be integrated with Streaming media

Servers or Content Delivery networks to provide scalable

distribution across an enterprise network or the Internet.

Extron’s h.264 Streaming Technology

Sme 100

model Version Part no.

Sme 100 h.264 Streaming media encoder 60-1061-01



format. A natural conclusion may be to always select a standard.

However, arbitrarily selecting a commonly used codec because

it is “a standard” may or may not be the best choice for every

application due to technical or performance reasons.

Image Transforms and CodecsVideo codecs use one of two primary transforms. The Discrete

Cosine Transform – DCT and the Discrete Wavelet Transform

– DWT. These transforms convert the video data from a spatial

domain into a frequency domain where the image data may be

compressed more easily. Each transform has different strengths

and weaknesses. Use of DCT is partly traditional. Limited

computer processing power at the time of its creation made use of

the DCT desirable because it consumed minimal processing power.

The DWT is used in many applications but has seen broader use

with the advent of the JPEG2000 compression standard.

The DCT is used broadly in codecs employing temporal

compression, using data processed across a Group of Pictures –

GOP. This includes codecs used to stream video on the Internet

such as: H.264, Theora, and VP8. These codecs achieve bit rates

below 10 and even 5 Mbps streaming high definition video, and

below 1 Mbps for standard definition video or lower resolutions,

such as 176 × 144, QCIF – Quarter Common Intermediate Format.

The DWT on the other hand, provides very high efficiency in

still image compression for a more continuous or graceful image

output as compression ratios are increased on still frames.

Applied in JPEG2000, it has utility streaming high definition video

on networks that can support bit rates of 100 Mbps or higher,

maintaining very high image qualities and low encode-to-decode

latency below 100 ms.

Exceptions to this summary include : 1) four H.264 High profiles

using intra-compression, which can be applied to low delay

applications providing comparable qualities to JPEG2000, 2) the

Dirac codec, which includes a temporal compression applying a

Group of Pictures scheme, and 3) the PURE3 codec from Extron,

which uses the DWT, but offers a unique form of temporal

compression capable of achieving bit rates from 10 Mbps down

to 1 Mbps with high resolution computer-video inputs in

certain applications.

Proprietary CodecsProprietary codecs are used every day in industrial and consumer

applications. Situations that motivate development of a proprietary

codec include:

1. A developer or manufacturer has established user

requirements that cannot be fulfilled with available

standards. An investment is made to develop a codec

that fulfills a unique need. This is more likely to occur in

closed applications.

2. An organization maintains a degree of control over a

significant user group and produces a unique codec,

container format, or viewer application from which it can

manage the quality delivered to endpoints. This situation is

more likely to exist in broad use applications such as PCs.

3. A developer or manufacturer creates a codec to provide

a competitive advantage and barrier to entry for

alternative products.

4. A company creates a codec to avoid licensing fees or legal

risk associated with patents.

Enterprise applications with defined, manageable endpoints

are strong candidates to consider proprietary codecs. The delivery

interface, endpoints, economics, and expansion of the system can

be planned. Specific performance or quality requirements may be

based on bit rate targets or error resilience operating on certain

types of networks. They may be based on support for a specific

input or output signal format, resolution, or picture quality. They

may also be based on very low encoding and decoding delay,

below 100ms for instance.

When applied to enterprise applications, dedicated hardware

designed alongside a proprietary video codec can provide

superior performance relative to standards-based video codecs

that often rely on a blend of intellectual property or technology

from different suppliers. Proprietary encoding algorithms and

dedicated hardware can often be optimized and tailored to special

requirements found in unique applications. Standards-based

hardware and software applications may be fixed or preconfigured

dCT and dwT ComparisondCT dwT

Spatial Compression efficiency moderate high

Degradation with increased spatial compression

Discrete Continuous

applied with temporal compression using – Gop

Common rare

applied in low delay applications Common Common

Transform use in different CodecsdCT dwT

JpeG, mpeG-1, mpeG-2, h.264, Vp6, Vp8, VC-1

JpeG2000, Dirac, Dirac-pro, pure3



for delivery of specific video content such as entertainment or

video teleconferencing at bandwidths targeted below 5 Mbps.

These low bandwidth solutions may not be suitable for the image

quality or low latency requirements of enterprise level applications

such as remote video contribution links, telemedicine, command

and control, surveillance, or simulation environments.

Proprietary codecs can also provide greater security when the

technology platform is not easily accessible to the public at large.

The foundation for security in any communication application

starts with physical security, communications architecture and

network policies for authentication, and encryption. Use of a

proprietary codec provides opportunity for an additional layer

of protection, since the codec will not be easily accessible to

unintended users.

The PUrE3 Codec – Fulfilling the Most Challenging

Streaming requirements

the extron pure3 codec has been developed for customers

operating on private networks with real-time, mission-critical

applications. It has been developed to support the most

demanding image quality requirements streaming computer

or video inputs, including maintaining native source resolution

and frame rate for virtually any computer or video input up

to 1920×1200 or 1080p resolutions. Visually lossless image

quality and 4:4:4 color quantization are retained ensuring that

video delivered to production systems or large projection

displays will not exhibit compression artifacts. the codec

supports use in real-time, interactive, collaborative applications

where dispersed participants can be confident they are acting

on identical visual information and bidirectional communication

will not be hampered by delay of any significance.

the pure3 codec provides highly efficient compression

and is intended for use on commonly used Lan and Wan

infrastructures. an error concealment system in the pure3

codec makes it highly immune to network errors, preserving

high image quality even under conditions of heavy packet loss

without requiring additional delay or bandwidth used by error

correction technology.

the pure3 codec was developed because the performance

available from mpeG-2 or JpeG2000 technology was unable

to support this combination of demanding requirements.

JpeG2000 was capable of supporting the quality, but at high

bit rates; mpeG-2 supported more efficient bit rates but would

not maintain the quality with low encoding delays. the pure3

codec has been implemented in extron’s Vn matrix 225

codecs, which support rGB or DVI signals and digital audio,

and the Vn matrix 300 codec, which supports SDI, hD-SDI,

and 3G-SDI video formats and embedded audio. Vn-matrix

products employing the pure3 codec are in use by customers

in the following quality-critical applications:

• Video Contribution & Collaboration - For broadcast,

post-production, scientific, military, product design, and

oil/gas exploration

• Control rooms - For broadcast, surveillance, and

command & control applications

• training, education & Documentation - For visualization

and simulation environments

Extron’s PuRE3 Codec

model Version Part no.

VnC 225 DVI Codec for DVI-I, audio/Keyboard/mouse 60-1118-02

Vne 225 DVI encoder for DVI-I & Digital audio 60-1119-02

VnD 225 DVI Decoder for DVI-I & Digital audio 60-1120-02

VnC 225 DVI



A real-world example of vulnerability using an open standard

was experienced in 2008 when Iraqi insurgents used SkyGrabber,

an openly available $26 satellite snooping software program,

to intercept and monitor video from US Predator drones. A

proprietary codec would have made it considerably more difficult

for the insurgents to extract this information.

Network and User EnvironmentsIt’s valuable to consider the network environment in which

your streaming product will be deployed. Let’s examine three

network environments:

• Private – Private networks can be designed and

configured to support streaming traffic requirements.

Their performance can be measured and managed.

Where the network infrastructure is applied in shared

use data, voice, and video, greater attention will be made

of bandwidth use.

• Public – Streaming delivered across the Internet presents

more challenges as it is a publicly shared network. Limits

to Quality of Service – QoS and available bandwidth at

endpoints limit the types of streaming applications which

can be served. Security is also a concern.

• Virtual Private Networks or VPNs – VPNs provide

encryption and greater security for connections crossing

the Internet.

Private, managed networks can be designed to support use

of codecs requiring bandwidths of 50 Mbps or higher. Public

networks or shared use private networks will typically have lower

bandwidth available, making streaming at bit rates below 5 Mbps

or lower desirable. Fewer guarantees for QoS on the Internet

may require error correction systems or transport and streaming

protocols designed to deliver reliable performance. For more

information concerning transport and streaming protocols, read

“Different Methods for Streaming Media” in the Spring 2011 issue

of AV Systems Design.

Deciding on a Codec – Standards-Based or Proprietary?The discussion thus far has presented a broad set of topics

including a variety of codecs targeted at different applications

both proprietary and standards-based, open and closed systems,

different network environments, and summary performance

delivered by codecs using different transforms. Some streaming

applications require delivery of electronic signals to presentation

or production systems, and others require direct delivery of a

video bitstream to a PC or hardware decoders in open systems.

Finally, the PC world is currently subject to a dynamic playing

field influenced by powerful organizations. Where codecs are

concerned, it’s clear that:

• There are many codecs in use, some of which have

targeted different applications and requirements

• Certain applications requiring interoperability for

communications or media exchange use codecs, which

have been ratified or endorsed into standards by official

standards bodies

• Audiovisual applications must provide solutions for media

sources, which often do not offer a standard resolution,

interface, or interoperability

• Use of a standard offers the opportunity to support many

endpoints through interoperability and transfer of digital

media in live streaming or production workflows

• Proprietary solutions can offer unique performance and

potentially increased security

• Plug-ins and conversion tools exist for many codecs and

proprietary container formats providing compatibility for

different decoding devices

A decision criteria for selecting a codec is presented on the

facing page. It considers: 1) the user and system environment,

2) exchange format, 3) delivery paradigm, 4) endpoints, and 5)

performance. This decision criteria is not absolute, but it provides

a guidepost for selecting a codec in a manner that avoids

over-simplified thinking, marketing hype, and the politics of

industrial markets.

A final word of wisdom - stay informed. The Joint Video Team

of ITU and ISO are currently working on the High Efficiency

Video Coding - HEVC as a successor to H.264. The goal is to

cut the bit rate of H.264 in half for low complexity applications.

Expect continued editorial discussion, white papers, and

promotion of different codecs in the future. Rapid growth and

advancements in both consumer and industrial streaming video

applications will continue to bring change and the emergence

of new codecs.

Karl Johnson is Director of product marketing at extron electronics for streaming

technologies and videowall processing systems. he worked at electrosonic for over 20 years,

most recently as General manager of the electrosonic product division.

how many could many-to-many be?It could be thousands to millions of endpoints.

how many could few-to-few be?It could be hundreds of endpoints.



Personal computers or consumer devices

Video bitstream, software decoding

One-to-many or many-to-many

Hardware decoding or managed PCs

Electronic signal or software decoding

One-to-one or few-to-few

Endpoint device?

Exchange format?

Delivery paradigm?

Endpoint positions?

Performance?

PROPRIETARY

Defined endpointsUnique performance

Open, Unmanaged Endpoints Closed, Managable Endpoints

STANDARDS-BASED

A B COpen computing environment

De facto or real standards

Defined endpoints

Unique performance

Variable endpoints

Standard performance

use a Standards-Based CodecEvaluate Standards-Based

Codecs and Proprietary Codecs use a Proprietary Codec

For applications with:

1. an open network environment delivering media to pCs or consumer devices

2. the delivery paradigm is one-to-many or many-to-many

a standards based codec should be selected which originates from a standards body such as ISo. however, certain applications may target use of endpoints that are restricted to use a proprietary codec or container and media players. Best practice would be to not let the endpoint device drive the technology selection. a solution using a standards based codec will offer a future that includes greater interoperability with new media systems and endpoint devices.

If the application is:

1. a closed system or the opportunity exists to manage the endpoints

2. applied to a private or managed connection

3. uses an electronic signal interface such as hD-SDI or DVI, or decoder plug-ins can be used on decoding devices

4. the delivery paradigm is one-to-one or few-to-few

If endpoints are not managed or part of a closed system and interoperability is required, then a standards-based codec is recommended. Compatibility between encoders and decoders must still be confirmed based on the class of application served and use of common protocols.

a proprietary codec should be considered for applications that are:

1. Closed systems2. reside on a private network or offer

the ability to manage endpoints3. uses an electronic signal interface such

as hD-SDI, DVI, or decoder plug-ins can be used on decoding devices

4. the delivery paradigm is one-to-one or few-to-few

5. unique streaming performance is required and is not available from standards-based codecs

When security is an important consideration, a proprietary codec may contribute additional protection for an application.

A B C


Factors to consider when choosing the right codec for your ... › download › get › ?path=... · H.264 H.263 VP8 AVC THEORA DIRAC JPEG AVC-INTRA MPEG-2 H.261 MPEG-4 VC-1 34 Feature

Documents