June 2005

MPEG-4 Part 10 AVC (H.264) Video Encoding

Abstract

H.264 has the potential to revolutionize the industry as it eases the bandwidth burden of service delivery and opens the service provider market to new players. This document describes the process of H.264 encoding and transmission over IP.

Table of Contents

Introduction..... 1 H.264 Overview..... 1 H.264 Technical Description... 3 Organization of the Bit Stream... 3 Intra Prediction and Coding.... 6 Inter Prediction and Coding.... 7 Block Sizes..... 7 Motion Estimation Accuracy.... 8 Multiple Reference Picture Selection... 8 De-blocking (Loop) Filter.... 8 Integer Transform.... 9 Quantization and Transform Coefficient Scanning... 9 Entropy Coding.... 10 UVLC/CAVLC.... 10 H.264 Profiles.... 11 Baseline Profile: Specific Features.... 11 Main Profile: Specific Features... 12 Extended Profile.... 13 Fidelity Range Extensions (FRExt)... 13 High Profile: Specific Features... 13 The Main Impacts of the High Profile... 13 High Profile Details.... 14 Transport over IP.... 16 IP Layer.... 16 UDP Layer.... 16 MPEG-2 Transport Stream.... 17 Optional RTP Layer.... 17 Conclusion.... 17 Appendix A - Comparison of H.264 and MPEG-2... 18 Appendix B - The D9154 - Scientific-Atlantas H.264 Encoding Solution.. 19

Introduction

The advent of H.264 (MPEG-4 part 10) video encoding technology has been met with great enthusiasm in the video industry. H.264 has video quality similar to that of MPEG-2, but is more economical with its use of bandwidth. Being less expensive to distribute, H.264 is a natural choice for broadcasters who are trying to find cost effective ways of distributing High Definition Television (HDTV) channels and reducing the cost of carrying conventional Standard Definition channels. In fact, the use of bandwidth has been reduced to the point that it has captured the interest of telephone and data services providers, whose bandwidth limited link to the subscriber had previously not allowed for delivery of bandwidth thirsty television services. H.264 has the potential to revolutionize the industry as it eases the bandwidth burden of service delivery and opens the service provider market to new players. This document describes the process of H.264 encoding and transmission in detail. An overview of Scientific-Atlantas encoding solution can be found in Appendix B.

H.264 Overview

The main objective behind the H.264 project was to develop a high-performance video coding standard by adopting a back to basics approach with simple and straightforward design using well known building blocks. The ITU-T Video Coding Experts Group (VCEG) initiated the work on the H.264 standard in 1997. Towards the end of 2001, and witnessing the superiority of video quality offered by H.264-based software over that achieved by the (existing) most optimized MPEG-4 part 10 based software, ISO/IEC MPEG joined ITU-T VCEG by forming a Joint Video Team (JVT) that took over the H.264 project of the ITU-T. The JVT objective was to create a single video coding standard that would simultaneously result in a new part (i.e., Part 10) of the MPEG-4 family of standards and a new ITU-T (i.e., H.264) recommendation. The H.264 standard has a number of advantages that distinguish it from existing standards, while at the same time, sharing common features with other existing standards. The following are some of the key advantages of H.264: 1. Up to 50% in bit rate savings: Compared to MPEG-2 or MPEG-4 Simple Profile, H.264 permits a reduction in bit rate by up to 50% for a similar degree of encoder optimization at most bit rates. 2. High quality video: H.264 offers consistently good video quality at high and low bit rates. 3. Error resilience: H.264 provides the tools necessary to deal with packet loss in packet networks and bit errors in error-prone wireless networks. 4. Network friendliness: Through the Network Adaptation Layer, that is the same as for MPEG-2, H.264 bit streams can be easily transported over different networks.

7007887 Rev B

The above advantages make H.264 an ideal standard for offering TV services over bandwidth restricted networks, such as DSL networks, or for HDTV. Figure 1 shows a block diagram of the H.264 encoding engine.
Figure 1. Block Diagram of the H.264 Encoder
MPEG-4 Part 10 (H.264) Video Encoding
H.264 Technical Description
The main objective of the emerging H.264 standard is to provide a means to achieve substantially higher video quality compared to what could be achieved using any of the existing video coding standards. Nonetheless, the underlying approach of H.264 is similar to that adopted in previous standards such as MPEG-2 and MPEG-4 part 2, and consists of the following four main stages: a. Dividing each video frame into blocks of pixels so that processing of the video frame can be conducted at the block level. b. Exploiting the spatial redundancies that exist within the video frame by coding some of the original blocks through spatial prediction, transform, quantization and entropy coding (or variable-length coding). c. Exploiting the temporal dependencies that exist between blocks in successive frames, so that only changes between successive frames need to be encoded. This is accomplished by using motion estimation and compensation. For any given block, a search is performed in the previously coded one or more frames to determine the motion vectors that are then used by the encoder and the decoder to predict the subject block. d. Exploiting any remaining spatial redundancies that exist within the video frame by coding the residual blocks, i.e., the difference between the original blocks and the corresponding predicted blocks, again through transform, quantization and entropy coding. On the motion estimation/compensation side, H.264 employs blocks of different sizes and shapes, higher resolution 1/4-pel motion estimation, multiple reference frame selection and complex multiple bi-directional mode selection. On the transform side, H.264 uses an integer based transform that approximates roughly the Discrete Cosine Transform (DCT) used in MPEG-2, but does not have the mismatch problem in the inverse transform. In H.264, entropy coding can be performed using either a combination of a single Universal Variable Length Codes (UVLC) table with an Context Adaptive Variable Length Codes (CAVLC) for the transform coefficients or using Context-based Adaptive Binary Arithmetic Coding (CABAC).

Organization of the Bit Stream
A given video picture is divided into a number of small blocks referred to as macroblocks. For example, a picture with QCIF resolution (176x144) is divided into 99 16x16 macroblocks as indicated in Figure 2. A similar macroblock segmentation is used for other frame sizes. The luminance component of the picture is sampled at these frame resolutions, while the chrominance components, Cb and Cr, are down-sampled by two in the horizontal and vertical directions. In addition, a picture may be divided into an integer number of slices, which are valuable for resynchronization should some data be lost.
Figure 2. Subdivision of a QCIF picture in 16 x 16 macroblock
A H.264 video stream is organized in discrete packets, called NAL units (Network Abstraction Layer units). Each of these packets can contain a part of a slice, that is, there may be one or more NAL units per slice. But not all NAL units contain slice data. There are also NAL unit types for other purposes, such as signaling, headers and additional data. The slices, in turn, contain a part of a video frame. In normal bit streams, each frame consists of a single slice whose data is stored in a single NAL unit. Nevertheless, the possibility to spread frames over an almost arbitrary number of NAL units can be useful if the stream is transmitted over an error prone medium. The decoder may resynchronize after each NAL unit instead of skipping a whole frame if a single error occurs. H.264 also supports optional interlaced encoding. In this encoding mode, a frame is split into two fields. Fields may be encoded using spatial or temporal interleaving. To encode color images, H.264 uses the YCbCr color space like its predecessors, separating the image into luminance (or luma, brightness) and chrominance (or chroma, color) planes. It is, however, fixed at 4:2:0 sub-sampling, i.e. the chroma channels each have half the resolution of the luma channel. H.264 defines five different slice types: I, P, B, SI and SP. I slices or Intra slices describe a full still image, containing only references to itself. A video stream may consist only of I slices, but this implementation is typically not used. The first frame of a sequence always needs to be built out of I slices. P slices or Predicted slices use one or more recently decoded slices as a reference (or prediction) for picture construction. The prediction is usually not exactly the same as the actual picture content, so a residual may be added. B slices or Bi-Directional Predicted slices work like P slices with the exception that former and future I or P slices (in playback order) may be used as reference pictures. For this to work, B slices must be decoded after the following I or P slice. SI and SP slices or Switching slices may be used for transitions between two different H.264 video streams. This is a very uncommon feature. The Sequence Parameter Set (abbreviated SPS) and Picture Parameter Set (PPS) contain the basic stream headers. Each of these parameter sets is stored in its own NAL unit, usually occupying only a few bytes. Both parameter sets have their own ID values so that multiple video streams can be transferred in only one H.264 elementary stream.

The most important fields of a sequence parameter set are: A profile and level indicator signaling conformance to a profile/level combination specified in H.264 Annex A. Information about the decoding method of the picture order. The number of reference frames. The frame size in macroblocks as well as the interlaced encoding flag. Frame cropping information for enabling non-multiple-of-16 frame sizes. Video Usability Information (VUI) parameters, such as aspect ratio or color space details. The most important fields of a picture parameter set are: A flag indicating which entropy coding mode is used. Information about slice data partitioning and macroblock reordering. The maximum reference picture list index. Flags indicating the usage of weighted (bi)prediction. The initial quantization parameters as well as the luma/chroma quantization parameter offset. A flag indicating whether inter-predicted macroblocks may be used for intra prediction or not (constrained intra prediction).
Intra Prediction and Coding
Intra coding refers to the case where only spatial redundancies within a video picture are exploited. The resulting frame is referred to as an I-picture. I-pictures are typically encoded by directly applying the transform to the different macroblocks in the frame. Consequently, encoded I-pictures are large in size since a large amount of information is usually present in the frame, and no temporal information is used as part of the encoding process. In order to increase the efficiency of the intra coding process in H.264, spatial correlation between adjacent macroblocks in a given frame is exploited. The idea is based on the observation that adjacent macroblocks tend to have similar properties. Therefore, as a first step in the encoding process for a given macroblock, one may predict the macroblock of interest from the surrounding macroblocks (typically the ones located on top and to the left of the macroblock of interest, since those macroblocks would have already been encoded). The difference between the actual macroblock and its prediction is then coded, which results in fewer bits to represent the macroblock of interest compared to when applying the transform directly to the macroblock itself. In order to perform the intra prediction mentioned above, H.264 offers nine modes for prediction of 4x4 luminance blocks, including DC prediction (Mode 2) and eight directional modes, labeled 0, 1, 3, 4, 5, 6, 7, and 8 in Figure 3.

Figure 3. Intra prediction modes for 4x4 luminance blocks.
Pixels A to M from neighboring blocks have already been encoded and may be used for prediction. For example, if Mode 0 (Vertical prediction) is selected, then the values of the pixels a to p are assigned as follows: a, e, i and m are equal to A, b, f, j and n are equal to B, c, g, k and o are equal to C, and d, h, l and p are equal to D.
For regions with less spatial detail (i.e., flat regions), H.264 supports 16x16 intra coding, in which one of four prediction modes (DC, Vertical, Horizontal and Planar) is chosen for the prediction of the entire luminance component of the macroblock. In addition, H.264 supports intra prediction for the 8x8 chrominance blocks also using four prediction modes (DC, Vertical, Horizontal and Planar). Finally, the prediction mode for each block is efficiently coded by assigning shorter symbols to more likely modes, where the probability of each mode is determined based on the modes used for coding the surrounding blocks.
Inter Prediction and Coding
Inter prediction and coding is based on using motion estimation and compensation to take advantage of the temporal redundancies that exist between successive frames, hence, providing very efficient coding of video sequences. As stated in section 2.1, when a selected reference frame(s) for motion estimation is a previously encoded frame(s), the frame to be encoded is referred to as a P-picture. When both a previously encoded frame and a future frame are chosen as reference frames, then the frame to be encoded is referred to as a B-picture. Motion estimation in H.264 supports most of the key features adopted in earlier video standards, but its efficiency is improved through added flexibility and functionality. In addition to supporting P-pictures (with single and multiple reference frames) and Bpictures, H.264 supports a new inter-stream transitional picture called an SP-picture. The inclusion of SP-pictures in a bit stream enables efficient switching between bit streams with similar content encoded at different bit rates, as well as random access and fast playback modes.

Block Sizes

Motion compensation for each 16x16 macroblock can be performed using a number of different block sizes and shapes. These are illustrated in Figure 4. Individual motion vectors can be transmitted for blocks as small as 4x4, so up to 16 motion vectors may be transmitted for a single macroblock.
Figure 4. Different modes of dividing a macroblock for motion estimation in H.264
Block sizes of 16x8, 8x16, 8x8, 8x4, and 4x8 are also supported as shown. The availability of smaller motion compensation blocks improves prediction in general, and in particular, the small blocks improve the ability of the model to handle fine motion detail and result in better subjective viewing quality because they do not produce large blocking artifacts. Moreover, through the recently adopted tree structure segmentation method, it is possible to have a combination of 4x8, 8x4, or 4x4 sub-blocks within an 8x8 sub-block. Figure 5 shows an example of such a configuration for a 16x16 macroblock.

Figure 5. Example of 16x16 macroblock
Motion Estimation Accuracy
The prediction capability of the motion compensation algorithm in H.264 is further improved by allowing motion vectors to be determined with higher levels of spatial accuracy than in existing standards. Quarter-pixel accurate motion compensation is the lowest-accuracy form of motion compensation in H.264 (in contrast with prior standards based primarily on half-pel accuracy, with quarter-pel accuracy only available in the newest version of MPEG-4).
Multiple Reference Picture Selection
The H.264 standard offers the option of having multiple reference frames in inter-picture coding, resulting in better subjective video quality and more efficient coding of the video frame under consideration. Moreover, using multiple reference frames helps make the H.264 bit stream more error resilient. However, from an implementation point of view, there would be additional processing delays and higher memory requirements at both the encoder and decoder.
De-blocking (Loop) Filter
H.264 specifies the use of an adaptive de-blocking filter that operates on the horizontal and vertical block edges within the prediction loop in order to remove artifacts caused by block prediction errors. The filtering is generally based on 4x4 block boundaries, in which two pixels on either side of the boundary may be updated using a different filter. The rules for applying the de-blocking filter are intricate and quite complex, however, its use is optional for each slice (loosely defined as an integer number of macroblocks). Nonetheless, the improvement in subjective quality often more than justifies the increase in complexity.

Integer Transform

The information contained in a prediction error block resulting from either intra prediction or inter prediction is then re-expressed in the form of transform coefficients. H.264 is unique in that it employs a purely integer spatial transform (a rough approximation of the DCT) which is primarily 4x4 in shape, as opposed to the usual floating-point 8x8 DCT specified with rounding-error tolerances as used in earlier standards. The small shape helps reduce blocking and ringing artifacts, while the precise integer specification eliminates any mismatch issues between the encoder and decoder in the inverse transform.
Quantization and Transform Coefficient Scanning
The quantization step is where a significant portion of data compression takes place. In H.264, the transform coefficients are quantized using scalar quantization with no widened dead-zone. Fifty-two different quantization step sizes can be chosen on a macroblock basis this being different from prior standards. Moreover, in H.264 the step sizes are increased at a compounding rate of approximately 12.5%, rather than increasing it by a constant increment. The fidelity of chrominance components is improved by using finer quantization step sizes compared to those used for the luminance coefficients, particularly when the luminance coefficients are coarsely quantized.

Figure 6. Scan pattern for frame coding in H.264
The quantized transform coefficients correspond to different frequencies, with the coefficient at the top left hand corner in Figure 6 representing the DC value, and the rest of the coefficients corresponding to different nonzero frequency values. The next step in the encoding process is to arrange the quantized coefficients in an array, starting with the DC coefficient. A single coefficient-scanning pattern is available in H.264 (Figure 6) for frame coding, and another one is being added for field coding. The zigzag scan illustrated in Figure 6 is used in all frame-coding cases, and it is identical to the conventional scan used in earlier video coding standards. The zigzag scan arranges the coefficient in an ascending order of the corresponding frequencies.

Entropy Coding

The last step in the video coding process is entropy coding. Entropy coding is based on assigning shorter codewords to symbols with higher probabilities of occurrence, and longer codewords to symbols with less frequent occurrences. Some of the parameters to be entropy coded include transform coefficients for the residual data, motion vectors and other encoder information. Two types of entropy coding have been adopted. The first method represents a combination of Universal Variable Length Coding (UVLC) and Context Adaptive Variable-Length coding (CAVLC). The second method is represented by Context-Based Adaptive Binary Arithmetic Coding (CABAC).

UVLC/CAVLC

In some video coding standards, symbols and the associated codewords are organized in look-up tables, referred to as variable length coding (VLC) tables, which are stored at both the encoder and decoder. In MPEG-2, a number of VLC tables are used, depending on the type of data under consideration (e.g., transform coefficients, motion vectors). H.264 offers a single Universal VLC (UVLC) table that is to be used in entropy coding of all symbols in the encoder except for the transform coefficients. Although the use of a single UVLC table is simple, is has a major disadvantage, which is that the single table is usually derived using a static probability distribution model, which ignores the correlations between the encoder symbols. In H.264, the transform coefficients are coded using Context Adaptive Variable Length Coding (CAVLC). CAVLC is designed to take advantage of several characteristics of quantized 4x4 blocks. First, non-zero coefficients at the end of the zigzag scan are often equal to +/- 1. CAVLC encodes the number of these coefficients (trailing 1s) in a compact way. Second, CAVLC employs run-level coding efficiently to represent the string of zeros in a quantized 4x4 block. Moreover, the numbers of non-zero coefficients in neighboring blocks are usually correlated. Thus, the number of non-zero coefficients is encoded using a look-up table that depends on the numbers of non-zero coefficients in neighboring blocks. Finally, the magnitude (level) of non-zero coefficients increase near the DC coefficient and decrease around the high-frequency coefficients. CAVLC takes advantage of this by making the choice of the VLC look-up table for the level adaptive where the choice depends on the recently coded levels.

H.264 Profiles

H.264 describes two popular profiles: Baseline, mainly for video conferencing and telephony/mobile applications, and Main, primarily for broadcast video applications. Figure 7 shows the common features between the Baseline and Main profiles as well as the additional specific features for each. The Baseline profile allows the use of Arbitrary Slice Ordering (ASO) to reduce the latency in real-time communication applications, as well as the use of Flexible Macroblock Ordering (FMO) and redundant slices to improve error resilience in the coded bit stream. The Main profile enables additional reduction in bandwidth over the Baseline profile through mainly sophisticated Bi-directional prediction (B-pictures), Context Adaptive Binary Arithmetic Coding (CABAC) and weighted prediction.
Figure 7. Features for the Baseline and Main profiles
Baseline Profile: Specific Features

Arbitrary Slice Ordering

Arbitrary slice ordering allows the decoder to process slices in an arbitrary order as they arrive to the decoder. Hence the decoder does not have to wait for all the slices to be properly arranged before it starts processing them. This reduces the processing delay at the decoder, resulting in less overall latency in real-time video communication applications.
Flexible Macroblock Ordering (FMO)
Macroblocks in a given frame are usually coded in a raster scan order. With FMO, macroblocks are coded according to a macroblock allocation map that groups, within a given slice, macroblocks from spatially different locations in the frame. Such an arrangement enhances error resilience in the coded bit stream since it reduces the interdependency that would otherwise exist in coding data within adjacent macroblocks in a given frame. In the case of packet loss, the loss is scattered throughout the picture and can be easily concealed.

Redundant Slices

Redundant slices allow the transmission of duplicate slices over error-prone networks to increase the likelihood of the delivery of a slice that is free of errors.

Main Profile: Specific Features

B Pictures

B-pictures provide a compression advantage as compared to P-pictures by allowing a larger number of prediction modes for each macroblock. Here, the prediction is formed by averaging the sample values in two reference blocks, generally, but not necessarily using one reference block that is forward in time and one that is backward in time with respect to the current picture. In addition, "Direct Mode" prediction is supported, in which the motion vectors for the macroblock are interpolated based on the motion vectors used for coding the co-located macroblock in a nearby reference frame. Thus, no motion information is transmitted. By allowing so many prediction modes, the prediction accuracy is improved, often reducing the bit rate by 5-10%.

Weighted Prediction

This allows the modification of motion compensated sample intensities using a global multiplier and a global offset. The multiplier and offset may be explicitly sent, or implicitly inferred. The use of the multiplier and the offset aims at reducing the prediction residuals due, for example, to global changes in brightness, and consequently, leads to enhanced coding efficiency for sequences with fades, lighting changes, and other special effects.
Context Adaptive Binary Arithmetic Coding (CABAC) makes use of a probability model at both the encoder and decoder for all the syntax elements (transform coefficients, motion vectors, etc). To increase the coding efficiency of arithmetic coding, the underlying probability model is adapted to the changing statistics within a video frame, through a process called context modeling. The context modeling provides estimates of conditional probabilities of the coding symbols. Utilizing suitable context models, given inter-symbol redundancy can be exploited by switching between different probability models according to already coded symbols in the neighborhood of the current symbol to encode. The context modeling is responsible for most of CABACs 10% savings in bit rate over the VLC entropy coding method (UVLC/CAVLC).

Interlace Support

Interlaced video has two half pictures (fields) in a frame or full picture and they are at different times. The Main profile copes with this by supporting field coding and picture or macroblock adaptive switching between frame and field coding.

Extended Profile

This profile supports all features of the Baseline profile, with the addition of B slices, weighted prediction, field coding and picture or macroblock adaptive switching between frame and field coding. Furthermore it is the only profile to support the SP/SI slice data portioning. It does not support CABAC.
Fidelity Range Extensions (FRExt)
New tools and 4 new profiles have been added to address more demanding applications. The DVD forum and BlueRay disc are very interested in these new profiles. Tools New 8x8 transform Adaptive macro block-level transform (4x4,8x8) switching Encoder-specified quant scaling matrices Encoder-specified separate control of chroma quant parameter Profiles High Profile (HP) 8 bit video, 4:2:0 chroma sampling High 10 Profile (Hi10P) up to 10 bit video, 4:2:0 chroma sampling High 4:2:2 Profile (H422P) up to 10 bit video, 4:2:2 chroma sampling High 4:4:4 Profile (H444P) up to 10 bit video, 4:4:4 chroma sampling
High Profile: Specific Features
High profile contains the Main profile, a switchable 8x8 transform for residual coding and Scaling matrices for subjective quality optimization. High profile improves objective compression quality (significantly for some video, especially HD video). It also improves subjective compression capability with support for quantization weighting matrices. High profile includes Main profile as a subset, so there is little risk involved in the implementation.
The Main Impacts of the High Profile
Coding efficiency impact (measured as average bit-rate reduction):
HD Film: 12% HD Video (progressive): 12% HD Video (interlace): 4% (only 2 test clips) SD Video (interlace): 6%

Complexity Impact

Implementation beyond Main Profile affects Intra prediction, transform, de-blocking filter control, CABAC decoding. There is very little increase in computational requirements and only a slight increase in memory requirements (CABAC, transform).

High Profile Details

Integer 8x8 Transform (luma only)
Per 8x8 block, same number of adds (64) and 4 extrashifts (20 vs 16) compared with four 4x4 transforms.

Figure 8. 8x8 Transform

Scanning of the Matrix
Two scans for the 4x4 transform, switched for frame/field coding to support interlaced video.
Figure 9. Frame and Field Scan

Intra Prediction

There are nine Intra 8x8 prediction modes similar to the nine modes for the intra 4x4 blocks as stated in section on Intra Prediction and Coding earlier in this document.
Figure 10. Intra Prediction

Scaling Matrices

Scaling matrices are similar to MPEG-2 video. A matrix can be transmitted in the SPS as well as in the PPS. There is a separate matrix for 4x4 and 8x8 transforms and a separate matrix for Inter frame and Intra frame.
De-blocking Filters, CABAC and Signaling
For the de-blocking filter, only control of the filter is adjusted. It does not filter 4x4 blocks. Furthermore, there is no change to the filter operation itself compared to the Main Profile. The CABAC, mentioned in the main profile section, contains 61 new contexts and corresponding initialization values, but no change to the CABAC engine. The High Profile adds signaling in the form of the 8x8 transform on/off flag at the PPS level and an 8x8 transform on/off flag per macroblock to allow for adaptive use.

Transport over IP

H.264 bandwidth efficiencies enable video to be compressed to the point that it can be transferred through networks using a small amount of bandwidth. This has created an opportunity for voice and data service providers to offer television services as part of their product portfolio. In order to carry H.264 services to the customer premises, the existing IP infrastructure is used, requiring a method of encapsulating transport streams over IP.
Figure 11. MPEG over IP For telco type applications, H.264 video and the accompanying audio are put into an MPEG-2 transport stream, which can be sent directly over the User Datagram Protocol (UDP over IP) [IETF RFC 768]. Another method involves the use of the Real-time Transmission Protocol (RTP) [IETF RFC 1889] as an intermediate layer between UDP and the MPEG-2 transport stream.

IP Layer

At the network layer, routing video to the end user from the source can be done through unicast or multicast IP packets. Unicast packets are sent from the source to one destination, whereas multicast packets are sent from one source to all destinations that are part of the multicast group. The latter is useful for routing packets to multiple subscribers from a single encoder source. IP networks inherently introduce delay and jitter to the services they carry. If the delays are somewhat predictable, decoders can compensate for small amounts of jitter by queuing data before it is processed. In order to maximize quality and minimize decoding delays, the IP network connections must be provisioned with acceptable guarantees on quality of service.

UDP Layer

Working on top of the IP layer, UDP is used to provide a connectionless transport from the source to the destination(s). The UDP layer introduces logical communication ports, allowing multiple UDP Figure 12. MPEG-2 Ethernet Frame sessions to be handled between source and destination equipment. Unlike TCP, UDP does not provide an assured data link, so dropped packets are not re-transmitted. The provider of the IP connection must ensure that the quality of the IP link is high in order to minimize or eliminate dropped packets entirely.

MPEG-2 Transport Stream

The MPEG-2 transport stream that contains the H.264 video can be mapped into the UDP packet payload, or optionally into an RTP payload. More than one MPEG-2 transport stream frame is generally mapped into a single UDP or RTP packet. This is done for reasons of bandwidth efficiency. MPEG-2 transport stream frames are generally 188 or 204 bytes in length and the length of the combined Ethernet, IP and UDP overhead is 54 bytes. Including an Figure 13. MPEG-2 in Ethernet Composition RTP layer adds at least another 12 bytes. Maximizing the number of MPEG-2 transport stream frames you can fit in one UDP or RTP packet improves the efficiency of transporting video with IP.

Optional RTP Layer

The MPEG-2 transport stream frames can optionally be inserted into an RTP packet before insertion into UDP. RTP adds a sequence number that allows the detection of missing packets and allows the decoder to reorder packets that arrive out of order, as can happen in IP networks with multiple routes between source and destination. It also adds a timestamp that is useful for jitter measurement and synchronization. RTP adds additional overhead to the transportation of MPEG over IP.

Conclusion

H.264 technology reduces the bandwidth requirement for video service distribution without impacting video quality, making delivery of SDTV and HDTV signals less expensive. Service providers can exploit the ubiquity of IP networks to deliver lowbandwidth, high-quality video services to subscribers.
Appendix A - Comparison of H.264 and MPEG-2
Algorithm Characteristic General MPEG-2 Motion compensated predictive, residual, transformed, entropy coded None I, B,P 8x8 DCT 16x16 Multiple VLC Tables +/- 1 Pixel (MPEG-2) 1/4 Pixel (MPEG-4) None H.264 Same basic structure as MPEG
Intra Prediction Coded Image Types Transform Motion Estimation Blocks Entropy Coding Frame Distance for Prediction Fractional Motion Estimation De-blocking Filter
Multi-direction, Multi-pattern I, B, P, SP, SI 4x4 DCT-like Integer Transform 16x16, 16x8, 8x16, 8x8, 8x4, 4x8, 4x4 Arithmetic Coding and adaptive VLC Tables Unlimited forward/backward 1/4 Pixel Dynamic edge filters
Figure 14. Overview between H.264 and MPEG-2
Appendix B - The D9154 - Scientific-Atlantas H.264 Encoding Solution

Scientific-Atlantas D9154 Advanced Compression Encoder (ACE) provides high quality H.264 video encoding Figure 15. D9154 Encoder Front View for Telco and Broadcast applications using the H.264 Main profile. The D9154 is implemented with high performance DSPs and state of the art FPGAs, providing an encoding platform that can be upgraded with additional functionality in the field.

Bandwidth Efficiency

The D9154 encoders highly optimized implementation realizes the significant bit rate reductions promised by H.264 technology while achieving video quality that is comparable to MPEG-2 video. Encoded bit rate savings are on average approximately 50%, allowing broadcasters to fit as many as twice the number of channels into a single satellite transponder for distribution. On a per video channel basis, the reduction in encoded bit rate opens the television service provider market to telcos where the bandwidth restrictions imposed by last-mile copper links are no longer a limiting factor.

Pre-processing Filters

D9154 encoder users can make use of pre-processing filters to reduce noise in the signal being encoded. The D9154 applies motion-compensated filtering to reduce noise in successive video frames, reducing noise in the temporal direction while taking into account motion in successive video frames. This pre-processing not only achieves good performance in terms of reducing the occurrence of artifacts, but it also results in encoded bit rate savings, leading to better overall video quality. This is especially useful for telcos and MSOs who are encoding channels at Central Office and headend sites where the source may contain some noise.

Encoder Output

The D9154 encoder is equipped with 10/100 BaseT Ethernet IP outputs as well as traditional ASI outputs. An MPEG-2 transport stream can be transmitted to both types of network simultaneously, allowing distribution over traditional ASI networks as well as IP networks. The D9154 supports multicast and unicast of H.264 encoded video and audio in an MPEG-2 transport stream over a choice of IP/UDP or IP/UDP/RTP.
Figure 16. D9154 Advanced Compression Encoder Rear Panel
Scientific-Atlanta, Inc. 770.236.5000 www.scientificatlanta.com Scientific-Atlanta, the Scientific-Atlanta logo, and PowerVu are registered trademarks of Scientific-Atlanta, Inc. All other trademarks shown are trademarks of their respective owners. Product and service availability are subject to change without notice. 2005 Scientific-Atlanta, Inc. All rights reserved. June 2005 Printed in USA Part Number 7007887 Rev B

Cite as: 552 U. S. ____ (2008) Syllabus
action against most aiders and abettors and thereby undermining Congress determination that this class of defendants should be pur sued only by the SEC. The practical consequences of such an expan sion provide a further reason to reject petitioners approach. The ex tensive discovery and the potential for uncertainty and disruption in a lawsuit could allow plaintiffs with weak claims to extort settle ments from innocent companies. See, e.g., Blue Chip, supra, at 740 741. It would also expose to such risks a new class of defendants overseas firms with no other exposure to U. S. securities laws thereby deterring them from doing business here, raising the cost of being a publicly traded company under U. S. law, and shifting securi ties offerings away from domestic capital markets. Pp. 813. (d) Upon full consideration, the history of the 10(b) private right of action and the careful approach the Court has taken before proceed ing without congressional direction provide further reasons to find no liability here. The 10(b) private cause of action is a judicial con struct that Congress did not direct in the text of the relevant stat utes. See, e.g., Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350, 358359. Separation of powers provides good reason for the now-settled view that an implied cause of action exists only if the underlying statute can be interpreted to disclose the intent to create one, see, e.g., Alexander v. Sandoval, 532 U. S. 275, 286287. The decision to extend the cause of action is thus for the Congress, not for this Court. This restraint is appropriate in light of the PSLRA, in which Congress ratified the implied right of action af ter the Court moved away from a broad willingness to imply such private rights, see, e.g., Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U. S. 353, 381382, and n. 66. It is appropriate for the Court to assume that when PSLRA 104 was enacted, Congress ac cepted the 10(b) private right as then defined but chose to extend it no further. See, e.g., Alexander, supra, at 286287. Pp. 1315. 443 F. 3d 987, affirmed and remanded. KENNEDY, J., delivered the opinion of the Court, in which ROBERTS, C. J., and SCALIA, THOMAS, and ALITO, JJ., joined. STEVENS, J., filed a dissenting opinion, in which SOUTER and GINSBURG, JJ., joined. BREYER, J., took no part in the consideration or decision of the case.

Cite as: 552 U. S. ____ (2008) Opinion of the Court
NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Wash ington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

_________________

No. 0643
STONERIDGE INVESTMENT PARTNERS, LLC,
PETITIONER v. SCIENTIFIC-ATLANTA,

INC., ET AL.

ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF
APPEALS FOR THE EIGHTH CIRCUIT

[January 15, 2008]

JUSTICE KENNEDY delivered the opinion of the Court. We consider the reach of the private right of action the Court has found implied in 10(b) of the Securities Ex change Act of 1934, 48 Stat. 891, as amended, 15 U. S. C. 78j(b), and SEC Rule 10b5, 17 CFR 240.10b5 (2007). In this suit investors alleged losses after purchasing com mon stock. They sought to impose liability on entities who, acting both as customers and suppliers, agreed to arrangements that allowed the investors company to mislead its auditor and issue a misleading financial state ment affecting the stock price. We conclude the implied right of action does not reach the customer/supplier com panies because the investors did not rely upon their statements or representations. We affirm the judgment of the Court of Appeals. I This class-action suit by investors was filed against Charter Communications, Inc., in the United States Dis trict Court for the Eastern District of Missouri. Ston eridge Investment Partners, LLC, a limited liability com
SCIENTIFIC-ATLANTA, INC. Opinion of the Court
pany organized under the laws of Delaware, was the lead plaintiff and is petitioner here. Charter issued the financial statements and the securi ties in question. It was a named defendant along with some of its executives and Arthur Andersen LLP, Char ters independent auditor during the period in question. We are concerned, though, with two other defendants, respondents here. Respondents are Scientific-Atlanta, Inc., and Motorola, Inc. They were suppliers, and later customers, of Charter. For purposes of this proceeding, we take these facts, alleged by petitioner, to be true. Charter, a cable operator, engaged in a variety of fraudulent practices so its quar terly reports would meet Wall Street expectations for cable subscriber growth and operating cash flow. The fraud included misclassification of its customer base; delayed reporting of terminated customers; improper capitaliza tion of costs that should have been shown as expenses; and manipulation of the companys billing cutoff dates to inflate reported revenues. In late 2000, Charter execu tives realized that, despite these efforts, the company would miss projected operating cash flow numbers by $15 to $20 million. To help meet the shortfall, Charter decided to alter its existing arrangements with respondents, Sci entific-Atlanta and Motorola. Petitioners theory as to whether Arthur Andersen was altogether misled or, on the other hand, knew the structure of the contract arrange ments and was complicit to some degree, is not clear at this stage of the case. The point, however, is neither controlling nor significant for our present disposition, and in our decision we assume it was misled. Respondents supplied Charter with the digital cable converter (set top) boxes that Charter furnished to its customers. Charter arranged to overpay respondents $20 for each set top box it purchased until the end of the year, with the understanding that respondents would return the

overpayment by purchasing advertising from Charter. The transactions, it is alleged, had no economic substance; but, because Charter would then record the advertising purchases as revenue and capitalize its purchase of the set top boxes, in violation of generally accepted accounting principles, the transactions would enable Charter to fool its auditor into approving a financial statement showing it met projected revenue and operating cash flow numbers. Respondents agreed to the arrangement. So that Arthur Andersen would not discover the link between Charters increased payments for the boxes and the advertising purchases, the companies drafted docu ments to make it appear the transactions were unrelated and conducted in the ordinary course of business. Follow ing a request from Charter, Scientific-Atlanta sent docu ments to Charter statingfalselythat it had increased production costs. It raised the price for set top boxes for the rest of 2000 by $20 per box. As for Motorola, in a written contract Charter agreed to purchase from Mo torola a specific number of set top boxes and pay liqui dated damages of $20 for each unit it did not take. The contract was made with the expectation Charter would fail to purchase all the units and pay Motorola the liquidated damages. To return the additional money from the set top box sales, Scientific-Atlanta and Motorola signed contracts with Charter to purchase advertising time for a price higher than fair value. The new set top box agreements were backdated to make it appear that they were negoti ated a month before the advertising agreements. The backdating was important to convey the impression that the negotiations were unconnected, a point Arthur Ander sen considered necessary for separate treatment of the transactions. Charter recorded the advertising payments to inflate revenue and operating cash flow by approxi mately $17 million. The inflated number was shown on
financial statements filed with the Securities and Ex change Commission (SEC) and reported to the public. Respondents had no role in preparing or disseminating Charters financial statements. And their own financial statements booked the transactions as a wash, under generally accepted accounting principles. It is alleged respondents knew or were in reckless disregard of Char ters intention to use the transactions to inflate its reve nues and knew the resulting financial statements issued by Charter would be relied upon by research analysts and investors. Petitioner filed a securities fraud class action on behalf of purchasers of Charter stock alleging that, by participat ing in the transactions, respondents violated 10(b) of the Securities Exchange Act of 1934 and SEC Rule 10b5. The District Court granted respondents motion to dis miss for failure to state a claim on which relief can be granted. The United States Court of Appeals for the Eighth Circuit affirmed. In re Charter Communications, Inc., Securities Litigation, 443 F. 3d 987 (2006). In its view the allegations did not show that respondents made misstatements relied upon by the public or that they violated a duty to disclose; and on this premise it found no violation of 10(b) by respondents. Id., at 992. At most, the court observed, respondents had aided and abetted Charters misstatement of its financial results; but, it noted, there is no private right of action for aiding and abetting a 10(b) violation. See Central Bank of Denver, N. A. v. First Interstate Bank of Denver, N. A., 511 U. S. 164, 191 (1994). The court also affirmed the District Courts denial of petitioners motion to amend the com plaint, as the revised pleading would not change the courts conclusion on the merits. 443 F. 3d, at 993. Decisions of the Courts of Appeals are in conflict re specting when, if ever, an injured investor may rely upon 10(b) to recover from a party that neither makes a public

misstatement nor violates a duty to disclose but does participate in a scheme to violate 10(b). Compare Simp son v. AOL Time Warner Inc., 452 F. 3d 1040 (CA9 2006), with Regents of Univ. of Cal. v. Credit Suisse First Boston (USA), Inc., 482 F. 3d 372 (CA5 2007). We granted certio rari. 549 U. S. ___ (2007). II Section 10(b) of the Securities Exchange Act makes it unlawful for any person, directly or indirectly, by the use of any means or instrumentality of interstate commerce or of the mails, or of any facility of any na tional securities exchange. [t]o use or employ, in connection with the purchase or sale of any security. any manipulative or deceptive device or contriv ance in contravention of such rules and regulations as the Commission may prescribe as necessary or appro priate in the public interest or for the protection of in vestors. 15 U. S. C. 78j. The SEC, pursuant to this section, promulgated Rule 10b 5, which makes it unlawful (a) To employ any device, scheme, or artifice to de fraud, (b) To make any untrue statement of a material fact or to omit to state a material fact necessary in or der to make the statements made, in the light of the circumstances under which they were made, not mis leading, or (c) To engage in any act, practice, or course of busi ness which operates or would operate as a fraud or deceit upon any person, in connection with the purchase or sale of any secu rity. 17 CFR 240.10b5. Rule 10b5 encompasses only conduct already prohibited by 10(b). United States v. OHagan, 521 U. S. 642, 651
STONERIDGE INVESTMENT PARTNERS, LLC v. SCIENTIFIC-ATLANTA, INC. Opinion of the Court
(1997). Though the text of the Securities Exchange Act does not provide for a private cause of action for 10(b) violations, the Court has found a right of action implied in the words of the statute and its implementing regulation. Superintendent of Ins. of N. Y. v. Bankers Life & Casualty Co., 404 U. S. 6, 13, n. 9 (1971). In a typical 10(b) private action a plaintiff must prove (1) a material misrepresenta tion or omission by the defendant; (2) scienter; (3) a con nection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation. See Dura Pharmaceuticals, Inc. v. Broudo, 544 U. S. 336, 341342 (2005). In Central Bank, the Court determined that 10(b) liability did not extend to aiders and abettors. The Court found the scope of 10(b) to be delimited by the text, which makes no mention of aiding and abetting liability. 511 U. S., at 177. The Court doubted the implied 10(b) action should extend to aiders and abettors when none of the express causes of action in the securities Acts included that liability. Id., at 180. It added the following: Were we to allow the aiding and abetting action pro posed in this case, the defendant could be liable with out any showing that the plaintiff relied upon the aider and abettors statements or actions. See also Chiarella [v. United States, 445 U. S. 222, 228 (1980)]. Allowing plaintiffs to circumvent the reliance re quirement would disregard the careful limits on 10b5 recovery mandated by our earlier cases. Ibid. The decision in Central Bank led to calls for Congress to create an express cause of action for aiding and abetting within the Securities Exchange Act. Then-SEC Chairman Arthur Levitt, testifying before the Senate Securities Subcommittee, cited Central Bank and recommended that aiding and abetting liability in private claims be estab

fact rely upon respondents own deceptive conduct. Liability is appropriate, petitioner contends, because respondents engaged in conduct with the purpose and effect of creating a false appearance of material fact to further a scheme to misrepresent Charters revenue. The argument is that the financial statement Charter released to the public was a natural and expected consequence of respondents deceptive acts; had respondents not assisted Charter, Charters auditor would not have been fooled, and the financial statement would have been a more accurate reflection of Charters financial condition. That causal link is sufficient, petitioner argues, to apply Basics presumption of reliance to respondents acts. See, e.g., Simpson, 452 F. 3d, at 10511052; In re Parmalat Securi ties Litigation, 376 F. Supp. 2d 472, 509 (SDNY 2005). In effect petitioner contends that in an efficient market investors rely not only upon the public statements relating to a security but also upon the transactions those state ments reflect. Were this concept of reliance to be adopted, the implied cause of action would reach the whole market place in which the issuing company does business; and there is no authority for this rule. As stated above, reliance is tied to causation, leading to the inquiry whether respondents acts were immediate or remote to the injury. In considering petitioners argu ments, we note 10(b) provides that the deceptive act must be in connection with the purchase or sale of any secu rity. 15 U. S. C. 78j(b). Though this phrase in part defines the statutes coverage rather than causation (and so we do not evaluate the in connection with require ment of 10(b) in this case), the emphasis on a purchase or sale of securities does provide some insight into the decep tive acts that concerned the enacting Congress. See Black, Securities Commentary: The Second Circuits Approach to the In Connection With Requirement of Rule 10b5, 53 Brooklyn L. Rev. 539, 541 (1987) ([W]hile the in connec
tion with and causation requirements are analytically distinct, they are related to each other, and discussion of the first requirement may merge with discussion of the second). In all events we conclude respondents deceptive acts, which were not disclosed to the investing public, are too remote to satisfy the requirement of reliance. It was Charter, not respondents, that misled its auditor and filed fraudulent financial statements; nothing respondents did made it necessary or inevitable for Charter to record the transactions as it did. The petitioner invokes the private cause of action under 10(b) and seeks to apply it beyond the securities mar ketsthe realm of financing businessto purchase and supply contractsthe realm of ordinary business opera tions. The latter realm is governed, for the most part, by state law. It is true that if business operations are used, as alleged here, to affect securities markets, the SEC enforcement power may reach the culpable actors. It is true as well that a dynamic, free economy presupposes a high degree of integrity in all of its parts, an integrity that must be underwritten by rules enforceable in fair, inde pendent, accessible courts. Were the implied cause of action to be extended to the practices described here, however, there would be a risk that the federal power would be used to invite litigation beyond the immediate sphere of securities litigation and in areas already gov erned by functioning and effective state-law guarantees. Our precedents counsel against this extension. See Ma rine Bank v. Weaver, 455 U. S. 551, 556 (1982) (Congress, in enacting the securities laws, did not intend to provide a broad federal remedy for all fraud); Santa Fe, 430 U. S., at 479480 (There may well be a need for uniform federal fiduciary standards. But those standards should not be supplied by judicial extension of 10(b) and Rule 10b5 to cover the corporate universe (quoting Cary, Federal ism and Corporate Law: Reflections Upon Delaware, 83

Yale L. J. 663, 700 (1974))). Though 10(b) is not limited to preserving the integrity of the securities markets, Bankers Life, 404 U. S., at 12, it does not reach all com mercial transactions that are fraudulent and affect the price of a security in some attenuated way. These considerations answer as well the argument that if this were a common-law action for fraud there could be a finding of reliance. Even if the assumption is correct, it is not controlling. Section 10(b) does not incorporate com mon-law fraud into federal law. See, e.g., SEC v. Zand ford, 535 U. S. 813, 820 (2002) ([Section 10(b)] must not be construed so broadly as to convert every common-law fraud that happens to involve securities into a violation); Central Bank, 511 U. S., at 184 (Even assuming. a deeply rooted background of aiding and abetting tort liability, it does not follow that Congress intended to apply that kind of liability to the private causes of action in the securities Acts); see also Dura, 544 U. S., at 341. Just as 10(b) is surely badly strained when construed to provide a cause of action. to the world at large, Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 733, n. 5 (1975), it should not be interpreted to provide a private cause of action against the entire marketplace in which the issuing company operates. Petitioners theory, moreover, would put an unsupport able interpretation on Congress specific response to Cen tral Bank in 104 of the PSLRA. Congress amended the securities laws to provide for limited coverage of aiders and abettors. Aiding and abetting liability is authorized in actions brought by the SEC but not by private parties. See 15 U. S. C. 78t(e). Petitioners view of primary liabil ity makes any aider and abettor liable under 10(b) if he or she committed a deceptive act in the process of provid ing assistance. Reply Brief for Petitioner 6, n. 2; Tr. of Oral Arg. 24. Were we to adopt this construction of 10(b), it would revive in substance the implied cause of action
against all aiders and abettors except those who commit ted no deceptive act in the process of facilitating the fraud; and we would undermine Congress determination that this class of defendants should be pursued by the SEC and not by private litigants. See Alexander v. Sandoval, 532 U. S. 275, 290 (2001) (The express provision of one method of enforcing a substantive rule suggests that Congress intended to preclude others); FDA v. Brown & Williamson Tobacco Corp., 529 U. S. 120, 143 (2000) (At the time a statute is enacted, it may have a range of plau sible meanings. Over time, however, subsequent acts can shape or focus those meanings); see also Seatrain Ship building Corp. v. Shell Oil Co., 444 U. S. 572, 596 (1980) ([W]hile the views of subsequent Congresses cannot override the unmistakable intent of the enacting one, such views are entitled to significant weight, and particularly so when the precise intent of the enacting Congress is obscure (citations omitted)). This is not a case in which Congress has enacted a regulatory statute and then has accepted, over a long period of time, broad judicial authority to define substan tive standards of conduct and liability. Cf. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U. S. ___, ___ (2007) (slip op., at 1920). And in accord with the nature of the cause of action at issue here, we give weight to Congress amendment to the Act restoring aiding and abetting liability in certain cases but not others. The amendment, in our view, supports the conclusion that there is no liability. The practical consequences of an expansion, which the Court has considered appropriate to examine in circum stances like these, see Virginia Bankshares, Inc. v. Sandberg, 501 U. S. 1083, 11041105 (1991); Blue Chip, 421 U. S., at 737, provide a further reason to reject peti tioners approach. In Blue Chip, the Court noted that extensive discovery and the potential for uncertainty and

p. 45 (1995) (recognizing the 10(b) implied cause of action, and indicating the PSLRA was intended to have Congress. reassert its authority in this area); id., at 26 (indicating the pleading standards covered 10(b) actions). Congress thus ratified the implied right of action after the Court moved away from a broad willingness to imply private rights of action. See Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U. S. 353, 381382, and n. 66 (1982); cf. Borak, supra, at 433. It is appropriate for us to assume that when 78u4 was enacted, Congress accepted the 10(b) private cause of action as then defined but chose to extend it no further. IV Secondary actors are subject to criminal penalties, see, e.g., 15 U. S. C. 78ff, and civil enforcement by the SEC, see, e.g., 78t(e). The enforcement power is not toothless. Since September 30, 2002, SEC enforcement actions have collected over $10 billion in disgorgement and penalties, much of it for distribution to injured investors. See SEC, 2007 Performance and Accountability Report, p. 26, http://www.sec.gov/about/secpar2007.shtml (as visited Jan. 2, 2008, and available in Clerk of Courts case file). And in this case both parties agree that criminal penalties are a strong deterrent. See Brief for Respondents 48; Reply Brief for Petitioner 17. In addition some state securities laws permit state authorities to seek fines and restitution from aiders and abettors. See, e.g., Del. Code Ann., Tit. 6, 7325 (2005). All secondary actors, further more, are not necessarily immune from private suit. The securities statutes provide an express private right of action against accountants and underwriters in certain circumstances, see 15 U. S. C. 77k, and the implied right of action in 10(b) continues to cover secondary actors who commit primary violations. Central Bank, supra, at 191. Here respondents were acting in concert with Charter in
the ordinary course as suppliers and, as matters then evolved in the not so ordinary course, as customers. Un conventional as the arrangement was, it took place in the marketplace for goods and services, not in the investment sphere. Charter was free to do as it chose in preparing its books, conferring with its auditor, and preparing and then issuing its financial statements. In these circumstances the investors cannot be said to have relied upon any of respondents deceptive acts in the decision to purchase or sell securities; and as the requisite reliance cannot be shown, respondents have no liability to petitioner under the implied right of action. This conclusion is consistent with the narrow dimensions we must give to a right of action Congress did not authorize when it first enacted the statute and did not expand when it revisited the law. The judgment of the Court of Appeals is affirmed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. JUSTICE BREYER took no part in the consideration or decision of this case.

Cite as: 552 U. S. ____ (2008) STEVENS, J., dissenting
JUSTICE STEVENS, with whom JUSTICE SOUTER and JUSTICE GINSBURG join, dissenting. Charter Communications, Inc., inflated its revenues by $17 million in order to cover up a $15 to $20 million ex pected cash flow shortfall. It could not have done so ab sent the knowingly fraudulent actions of ScientificAtlanta, Inc., and Motorola, Inc. Investors relied on Char ters revenue statements in deciding whether to invest in Charter and in doing so relied on respondents fraud, which was itself a deceptive device prohibited by 10(b) of the Securities Exchange Act of 1934. 15 U. S. C. 78j(b). This is enough to satisfy the requirements of 10(b) and enough to distinguish this case from Central Bank of Denver, N. A. v. First Interstate Bank of Denver, N. A., 511 U. S. 164 (1994). The Court seems to assume that respondents alleged conduct could subject them to liability in an enforcement proceeding initiated by the Government, ante, at 15, but nevertheless concludes that they are not subject to liabil ity in a private action brought by injured investors be cause they are, at most, guilty of aiding and abetting a violation of 10(b), rather than an actual violation of the statute. While that conclusion results in an affirmance of the judgment of the Court of Appeals, it rests on a rejec
SCIENTIFIC-ATLANTA, INC. STEVENS, J., dissenting
tion of that courts reasoning. Furthermore, while the Court frequently refers to petitioners attempt to expand the implied cause of action,1a conclusion that begs the question of the contours of that cause of actionit is to days decision that results in a significant departure from Central Bank. The Courts conclusion that no violation of 10(b) giving rise to a private right of action has been alleged in this case rests on two faulty premises: (1) the Courts overly broad reading of Central Bank, and (2) the view that reliance requires a kind of super-causationa view con trary to both the Securities and Exchange Commissions (SEC) position in a recent Ninth Circuit case2 and our holding in Basic Inc. v. Levinson, 485 U. S. 224 (1988). These two points merit separate discussion. I The Court of Appeals incorrectly based its decision on the view that [a] device or contrivance is not deceptive, within the meaning of 10(b), absent some misstatement or a failure to disclose by one who has a duty to disclose. In re Charter Communications, Inc., Securities Litigation, 443 F. 3d 987, 992 (CA8 2006). The Court correctly ex plains why the statute covers nonverbal as well as verbal deceptive conduct. Ante, at 7. The allegations in this casethat respondents produced documents falsely claim
ante, at 10 ([w]ere the implied cause of action to be extended to the practices described here. ); ante, at 12 ([t]he practical conse quences of an expansion); ante, at 14 (Concerns with the judicial creation of a private cause of action caution against its expansion. The decision to extend the cause of action is for the Congress, not for us). 2 See Brief for SEC as Amicus Curiae in Simpson v. AOL Time War ner Inc., No. 0455665 (CA9), p. 21 (The reliance requirement is satisfied where a plaintiff relies on a material deception flowing from a defendants deceptive act, even though the conduct of other participants in the fraudulent scheme may have been a subsequent link in the causal chain leading to the plaintiffs securities transaction).

4 Because the kind of sham transactions alleged in this complaint are unquestionably isolated departures from the ordinary course of busi ness in the American marketplace, it is hyperbolic for the Court to conclude that petitioners concept of reliance would authorize actions against the entire marketplace in which the issuing company oper
what it views as a strict division between the realm of financing business and the ordinary business operations. Ante, at 10. But petitioners position does not merge the two: A corporation engaging in a business transaction with a partner who transmits false information to the market is only liable where the corporation itself violates 10(b). Such a rule does not invade the province of ordinary business transactions. The majority states that [s]ection 10(b) does not incor porate common-law fraud into federal law, citing SEC v. Zandford, 535 U. S. 813 (2002). Ante, at 11. Of course, not every common-law fraud action that happens to touch upon securities is an action under 10(b), but the Courts opinion in Zandford did not purport to jettison all refer ence to common-law fraud doctrines from 10(b) cases. In fact, our prior cases explained that to the extent that the antifraud provisions of the securities laws are not coexten sive with common-law doctrines of fraud, it is because common-law fraud doctrines might be too restrictive. Herman & MacLean v. Huddleston, 459 U. S. 375, (1983). Indeed, an important purpose of the federal securities statutes was to rectify perceived deficiencies in the available common-law protections by establishing higher standards of conduct in the securities industry. Id., at 389. I, thus, see no reason to abandon common-law approaches to causation in 10(b) cases. Finally, the Court relies on the course of action Con gress adopted after our decision in Central Bank to argue that siding with petitioner on reliance would run contrary to congressional intent. Senate hearings on Central Bank were held within one month of our decision.5 Less than one year later, Senators Dodd and Domenici introduced S. 240, which became the Private Securities Litigation
ates. Ante, at 11. 5 See S. Rep. No. 10498, p. 2 (1995) (hereinafter S. Rep.).
Reform Act of 1995 (PSLRA), 109 Stat. 737.6 Congress stopped short of undoing Central Bank entirely, instead adopting a compromise which restored the authority of the SEC to enforce aiding and abetting liability.7 A private right of action based on aiding and abetting violations of 10(b) was not, however, included in the PSLRA,8 despite support from Senator Dodd and members of the Senate Subcommittee on Securities.9 This compromise surely provides no support for extending Central Bank in order to immunize an undefined class of actual violators of 10(b) from liability in private litigation. Indeed, as Members of Congressincluding those who rejected restoring a pri vate cause of action against aiders and abettorsmade clear, private litigation under 10(b) continues to play a vital role in protecting the integrity of our securities mar kets.10 That Congress chose not to restore the aiding and

result of a high level of investor confidence in the integrity and effi ciency of our markets. The SEC enforcement program and the avail ability of private rights of action together provide a means for de frauded investors to recover damages and a powerful deterrent against violations of the securities laws); see also Bateman Eichler, Hill Richards, Inc. v. Berner, 472 U. S. 299, 310 (1985) (Moreover, we repeatedly have emphasized that implied private actions provide a most effective weapon in the enforcement of the securities laws and are a necessary supplement to Commission action ); Brief for Former SEC Commissioners as Amici Curiae 4 ([L]iability [of the kind at issue here] neither results in undue liability exposure for non-issuers, nor an undue burden upon capital formation. Holding liable wrongdoers who actively engage in fraudulent conduct that lacks a legitimate business purpose does not hinder, but rather enhances, the integrity of our markets and our economy. We believe that the integrity of our securi ties markets is their strength. Investors, both domestic and foreign, trust that fraud is not tolerated in our nations securities markets and that strong remedies exist to deter and protect against fraud and to recompense investors when it occurs).
verse the decision of the Court of Appeals. III While I would reverse for the reasons stated above, I must also comment on the importance of the private cause of action that Congress implicitly authorized when it enacted the Securities Exchange Act of 1934. A theme that underlies the Courts analysis is its mistaken hostil ity towards the 10(b) private cause of action.11 Ante, at 13. The Courts current view of implied causes of action is that they are merely a relic of our prior heady days. Correctional Services Corp. v. Malesko, 534 U. S. 61, 75 (2001) (SCALIA, J., concurring). Those heady days per sisted for two hundred years. During the first two centuries of this Nations history much of our law was developed by judges in the commonlaw tradition. A basic principle animating our jurispru dence was enshrined in state constitution provisions guar anteeing, in substance, that every wrong shall have a remedy.12 Fashioning appropriate remedies for the viola

11 The Court does concede that Congress has now ratified the private cause of action in the PSLRA. See ante, at 15. 12 Today, the guarantee of a remedy for every injury appears in nearly three-quarters of state constitutions. Ala. Const., Art. I, 13; Ark. Const., Art. II, 13; Colo. Const., Art. II, 6; Conn. Const., Art. I, 10; Del. Const., Art. I, 9; Fla. Const., Art. I, 21; Idaho Const., Art. I, 18; Ill. Const., Art. I, 12; Ind. Const., Art. I, 12; Kan. Const., Bill of Rights, 18; Ky. Const., 14; La. Const., Art. I, 22; Me. Const., Art. I, 19; Md. Const., Declaration of Rights, Art. 19; Mass. Const., pt. I, Art. 11; Minn. Const., Art. 1, 8; Miss. Const., Art. III, 24; Mo. Const., Art. I, 14; Mont. Const., Art. II, 16; Neb. Const., Art. I, 13; N. H. Const., pt. I, Art. 14; N. C. Const., Art. I, 18; N. D. Const., Art. I, 9; Ohio Const., Art. I, 16; Okla. Const., Art. II, 6; Ore. Const., Art. I, 10; Pa. Const., Art. I, 11; R. I. Const., Art. I, 5; S. C. Const., Art. I, 9; S. D. Const., Art. VI, 20; Tenn. Const., Art. I, 17; Tex. Const., Art. I, 13; Utah Const., Art. I, 11; Vt. Const., ch. I, Art. 4; W. Va. Const., Art. III, 17; Wis. Const., Art. I, 9; Wyo. Const., Art. I, 8; see also Phillips, The Constitutional Right to a Remedy, 78 N. Y. U. L. Rev. 1309, 1310, n. 6 (2003) (hereinafter Phillips).
tion of rules of law designed to protect a class of citizens was the routine business of judges. See Marbury v. Madi son, 1 Cranch 137, 166 (1803). While it is true that in the early days state law was the source of most of those rules, throughout our historyuntil 1975the same practice prevailed in federal courts with regard to federal statutes that left questions of remedy open for judges to answer. In Texas & Pacific R. Co. v. Rigsby, 241 U. S. 33, 39 (1916), this Court stated the following: A disregard of the command of the statute is a wrong ful act, and where it results in damage to one of the class for whose especial benefit the statute was en acted, the right to recover the damages from the party in default is implied, according to a doctrine of the common law expressed in 1 Com. Dig., tit. Action upon Statute (F), in these words: So, in every case, where a statute enacts, or prohibits a thing for the benefit of a person, he shall have a remedy upon the same statute for the thing enacted for his advantage, or for the rec ompense of a wrong done to him contrary to the said
The concept of a remedy for every wrong most clearly emerged from Sir Edward Cokes scholarship on Magna Carta. See 1 Second Part of the Institutes of the Laws of England (1797). At the time of the ratifi cation of the United States Constitution, Delaware, Massachusetts, Maryland, New Hampshire, and North Carolina had all adopted constitutional provisions reflecting the provision in Cokes scholarship. Del. Declaration of Rights and Fundamental Rules 12 (1776), re printed in 2 W. Swindler, Sources and Documents of United States Constitutions 198 (1973) (hereinafter Swindler); Mass. Const., pt. I, Art. XI (1780), reprinted in 3 Federal and State Constitutions, Colonial Charters, and Other Organic Laws 1891 (F. Thorpe ed. 1909) (reprinted 1993) (hereinafter Thorpe); Md. Const., Declaration of Rights, Art. XVII (1776), in id., at 1688; N. H. Const., Art. XIV (1784), in 4 id., at 2455; N. C. Const., Declaration of Rights, Art. XIII (1776), in 5 id., at 2787, 2788; see also Phillips 13231324. Pennsylvanias Constitution of 1790 contains a guarantee. Pa. Const., Art. IX, 11, in 5 Thorpe 3101. Connecticuts 1818 Constitution, Art. I, 12, contained such a provision. Reprinted in Swindler 145.

19461974 as the expansion era in implied causes of action under the securities laws). When damage ac tions for violation of 10(b) and Rule 10b5 reached the Supreme Court, the existence of an implied cause of action was not deemed worthy of extended discus sion. Superintendent of Insurance v. Bankers Life & Casualty Co., 404 U.S. 6. (1971). Leist, 638 F. 2d, at 296297 (footnote omitted). In light of the history of court-created remedies and specifically the history of implied causes of action under 10(b), the Court is simply wrong when it states that Congress did not impliedly authorize this private cause of action when it first enacted the statute. Ante, at 16. Courts near in time to the enactment of the securities laws recognized that the principle in Rigsby applied to the securities laws.13 Congress enacted 10(b) with the under standing that federal courts respected the principle that every wrong would have a remedy. Todays decision sim ply cuts back further on Congress intended remedy. I respectfully dissent.
13 See, e.g., Slavin v. Germantown Fire Ins. Co., 174 F. 2d 799 (CA3 1949); Baird v. Franklin, 141 F. 2d 238, 244245 (CA2) (The fact that the statute provides no machinery or procedure by which the individual right of action can proceed is immaterial. It is well established that members of a class for whose protection a statutory duty is created may sue for injuries resulting from its breach and that the common law will supply a remedy if the statute gives none), cert. denied, 323 U. S. 737 (1944); Kardon v. National Gypsum Co., 69 F. Supp. 512, 514 (ED Pa. 1946) ([T]he right to recover damages arising by reason of violation of a statute. is so fundamental and so deeply ingrained in the law that where it is not expressly denied the intention to withhold it should appear very clearly and plainly).