ADS Technologies USBAV-708 User Manual
  1. Manuals
  2. Brands
  3. ADS Technologies Manuals
  4. Software
  5. USBAV-708 INSTANT VIDEOMPX
  6. User manual

ADS Technologies USBAV-708 User Manual

Ads technologies usbav-708 portablecapturedevice: user guide
Hide thumbs Also See for USBAV-708:
Table of Contents

Advertisement

Quick Links

Download this manual See also: Operating Manual
Revision 1.0

Advertisement

Table of Contents
loading
Need help?

Need help?

Do you have a question about the USBAV-708 and is the answer not in the manual?

Questions and answers

Related Manuals for ADS Technologies USBAV-708

Summary of Contents for ADS Technologies USBAV-708

  • Page 1 Revision 1.0...

  • Page 2: Table Of Contents

    Contents Index Introduction Welcome to the guide ...5 ® The origins of DivX video ...5 What is DivX? ...7 Why use DivX? ...8 Quick start guide Introducing VirtualDub ...11 Your first DivX ...12 Your first Multipass ...17 Notes on audio ...19 Forward General concepts ...24 Frames ...24...
  • Page 3 Interlacing What is interlacing? ... 88 Why de-interlace? ... 89 Maintaining interlacing ... 89 Source interlace ... 90 Video buffer verifier What is the video buffer verifier? ... 93 The video buffer verifier explained ... 93 Video buffer verifier ... 94 Profiles What is a profile? ...
  • Page 4 Introduction Guide...

  • Page 5: Introduction

    Introduction Welcome to the guide Welcome to the official guide to DivX With this guide we shall be demystifying the black-art of DivX encoding - leading you step by step through everything from creating your very first DivX video to exploiting the many powerful advanced features that DivX has to offer.
  • Page 6 By September 1999 Gej had achieved his goal of using a high performance codec in combination with the AVI container, and DivX ;-) (the emoticon a jibe at a now failed protected DVD rental system launched by Circuit City) was unleashed upon the Internet.

  • Page 7: What Is Divx

    What is DivX? Firstly, let’s address what DivX is not. DivX is not a stand-alone encoding application; that is when you install DivX you are not installing a program that will allow you choose a file you wish to process and then proceed do the conversion automatically.

  • Page 8: Why Use Divx

    Why use DivX? There are many codecs available these days, so why choose DivX to export your video? ►Quality DivX is a professional grade video codec. The Pro version available to consumers for less than $20 is exactly the same program licensed for use in commercial production environments.
  • Page 9 ►Internet distribution It has not always been easy to share your videos with friends and colleagues via the Internet. Historically, video transferred over the net has been small, jumpy, blocky and suffering from poor color definition. DivX breaks down the barriers to transferring high quality video via the Internet.

  • Page 10: Quick Start Guide

    Quick Start Guide Guide...

  • Page 11: Introducing Virtualdub

    Quick Start Guide Introducing VirtualDub When working with video it is likely that you will require a variety of tools to achieve the best possible results. There are some processes for which dedicated-purpose tools can be invaluable in the world of digital video. Most notable of these tools by far is VirtualDub.

  • Page 12: Your First Divx

    Your first DivX To begin we will use VirtualDub to open an MPEG 1 video file and convert it to DivX format. For the purposes of simplicity our video will not include an audio track at this point. Load VirtualDub by double-clicking on the VirtualDub program icon.
  • Page 13 From the Audio menu set No Audio. This instructs VirtualDub that even if the source file contains audio data we do not want to process it or copy it into our output file. For the present time leave the video mode set to Full Processing Mode. Now we want to configure DivX as the video compressor.
  • Page 14 Congratulations! If you have made it this far then this will be the first time you have seen the DivX encoder. At this point we simply want to configure the encoder with the default settings using the Restore Defaults, button but you might want to take this opportunity to explore the encoder configuration dialogue before doing so.
  • Page 15 The VirtualDub Status window should pop up and allow you to monitor the encoding progress. Here amongst other things you can see the progress bar, how much video and audio data has been encoded so far and a projection of the final file size. The projection should be considered a very rough estimate because DivX may choose to vary the bitrate throughout the...
  • Page 16 Once your encoding has completed both the feedback and status windows will close automatically. You should then be able to play your first DivX movie using The DivX Player (bundled free with DivX) or another media player application on your system. If you would like The DivX Player to open your file by default as opposed to any other player you can rename the AVI file after encoding so that the file extension reads “.divx“...

  • Page 17: Your First Multipass

    Your first Multipass While 1-pass mode is fast and particularly suited to certain situations, Multipass encoding offers more consistent quality by allowing the encoder to first analyze the entire video before encoding it. Load VirtualDub and configure it as you did in Your first DivX steps 1 through 5 but this time set the Variable bitrate mode to “Multipass, 1st pass”.
  • Page 18 Because we have selected Update log file the encoder will record an analysis of this nth pass with respect to the decisions it makes based upon the previous analysis from the log file. In this way we can feed the analysis log from one nth pass into successive nth passes and actually refine the encoding process towards the optimal quality level.

  • Page 19: Notes On Audio

    Notes on audio DivX is a video codec. As such it does not handle audio in any way and we will not cover the very broad topic of audio in detail. However, you will almost certainly want to include audio alongside DivX video and so we will briefly cover a few important points.
  • Page 20 Radium MP3 The Radium MP3 codec is a modified* version of the Fraunhofer IIS MPEG Layer-3 Codec (professional). It is easy to install and simply replaces the Fraunhofer IIS MPEG Layer-3 Codec (advanced) that ships with Windows. The Radium MP3 codec is available from many websites, enter the keywords “Radium MP3 download”...
  • Page 21 Now consider the corresponding correctly interleaved file: AVI Container Audio Video Audio Video Audio Video Here audio chunks are interleaved within the file so as they fall closely alongside the related sequence of video. As playback progresses seeking within the file is minimized, and in fact with appropriate buffering as provided by most devices seeking will not even take place.
  • Page 22 VirtualDub contains two modes for processing audio, these are “Direct stream copy” and “Full processing mode”. In direct stream copy mode no processing (other than interleaving) is performed on the audio stream. If it were desirable when editing an existing AVI file that only the video stream should be manipulated the audio mode would be configured as Direct stream copy in conjunction with “Source audio”.

  • Page 23: Forward

    Forward Guide...

  • Page 24: General Concepts

    Forward General concepts In order to understand all of the encoder features that will be covered in the next section it is first necessary to cover the core concepts behind the MPEG 4 Video Standard upon which DivX 5.1 is based. Lets begin with the way videos are composed.
  • Page 25 When the encoder finds a matching area for a block in the previous frame its position is recorded by use of a vector. A vector is simply a numerical representation of direction and magnitude. For example the vector (4, -7) might mean right 4 units, down 7 units.

  • Page 26: Intra-Frames And Predicted-Frames

    Intra-frames and Predicted-frames The DivX encoder makes use of three frame types during encoding. These are: Intra-frames, Predicted-frames and Bi-directional-frames. More commonly these are known as I, P and B frames. B-frames are discussed later in the guide as an advanced topic. An intra-frame is one in which all of the macroblocks are stored as images rather than as motion vectors.

  • Page 27: Quantizers

    Quantizers DivX uses a technique called quantization to control the accuracy of the image data it stores. Quantizers are similar to the denominator in an expression such as: Data Result Quantizer … and the co-efficient in the inverse quantization: Data Quantizer By maintaining some fixed possible to reduce the value of...
  • Page 28 It soon becomes clear that higher quantizers introduce larger errors due to rounding before inverse quantization; thus when encoding lower frame quantizers lead to higher quality images. Plotting results from the tables with respect to the original series illustrates lower quantizers recreating a more accurate signal. Although lower quantizers give higher quality images they also produce larger file sizes.

  • Page 29: A Key To The Guide

    A key to the guide From this point forward the guide will describe the features available in the DivX encoder. Throughout the rest of the guide an easy to follow iconic key will be used. Features available only in the DivX Pro encoder will be marked The following color-coded sections will appear where appropriate to a feature: This indicates a feature is not compatible with the DivX Certified Program.
  • Page 30 This indicates there are CLI parameters associated with the feature. Additionally, at the beginning of each feature description you will see the Quick Guide bar. QUICK GUIDE Here the Quick Guide bar indicates that there are performance tips and usage notes associated with the feature.

  • Page 31: Bitrate Mode

    Bitrate mode Guide...

  • Page 32: What Is The Bitrate

    Bitrate mode What is the bitrate? In a computer the minimum unit of storage is one bit. A bit can represent two values, 0 or 1, and hence computers use the binary number system (base 2) as opposed to the denary system (base 10, 0-9) that humans use. When several bits are combined they can be used to store more complex data.

  • Page 33: Bitrate Calculation

    Bitrate calculation 60 minute video with 128 Kbps MP3 audio for 700MB CD-R 1. Calculate how many seconds long the video is The number of minutes The number of seconds in one minute 3,600 seconds duration 2. Calculate how many bits in total there are available The number of megabytes 1,048,576 The number of bytes in one megabyte...
  • Page 34 The bitrate calculated is the average bitrate for the video. DivX may choose to vary the actual bitrate throughout the video as it encodes, a technique known as variable bitrate encoding. Since not all sequences in a video are equally complex in terms of image and motion, it is impossible to maintain a constant bitrate throughout the file without introducing substantial quality differences from frame-to-frame.

  • Page 35: 1-Pass

    1-Pass In 1-Pass mode the encoder will output a working DivX video stream as it receives the source video. This is both the easiest and fastest method of encoding DivX video while targeting an average bitrate, and is particularly suited to video capture applications where it is impossible to recreate the source video identically more than once.
  • Page 36 1-Pass mode is particularly useful when you are capturing from a live source and you desire control over the bitrate/file size. When capturing from a live source and attempting to maintain a consistent quality throughout the video you should use 1-Pass Quality Based mode. When working from static sources (for example re-processing existing stored video) you will achieve more consistent quality by using Multipass mode.

  • Page 37: Multipass

    Multipass In Multipass mode you are required to run the video through the encoder two or more times. On every successive pass the video must be identical to that which you used in your first pass. As discussed in “Bitrate mode—What is the bitrate?”, DivX uses a rate control algorithm to vary the bitrate throughout the video in order to achieve a consistent quality while targeting the specified average bitrate.

  • Page 38: Multipass Log File

    Multipass Log file Log header ##map version <Value> Value is the version number describing the format of the log file. nframes <Value> Value is the total number of frames listed in the log file. Frame numbering begins from zero. timescale <Value> Value is some arbitrary number representing the number of intervals in one second of video.
  • Page 39 type <Frame_type> Frame_type is the type of each frame and can be one of ‘I’,’P’ or ‘B’, representing intra-frame, predicted frame or bi-directional frame respectively. total_bits <Value> Value is the total number of bits used by the encoder to encode the frame.

  • Page 40: Multipass, Nth Pass

    Because DivX Pro includes the EKG tool for graphically displaying and manipulating the contents of the log file it is rarely necessary to edit it manually. Multipass, nth pass Once first pass has been completed the same video must be encoded one or more times using Multipass, nth pass mode.
  • Page 41 Moving the slider left of center, or lowering the modulation value, biases bandwidth in favor of high-motion frames and consequently away from low-motion frames. Moving the slider right of center, or raising the modulation value, biases bandwidth in favor of low-motion frames and consequently away from high-motion frames.
  • Page 42 Fastest Performance/Quality mode is unavailable during Multipass encoding because it does not perform the motion estimation necessary for the rate- control to work effectively. If you intend to use either Slow or Slowest Performance/Quality modes you can substantially decrease your encoding time by encoding all but the final pass using Standard Performance/Quality.
  • Page 43 Bitrate mode is one of: -bv1 <bitrate> 1-Pass -b1q <quantizer> 1-Pass, quality-based -bvn1 <bitrate> Multipass, 1st pass -bvnn <bitrate> Multipass, nth pass Bitrate is an integer value between 0 and 16,000 specifying the number of kilobits per second. Quantizer is floating point value between 1.0 and 31.0 specifying the fixed quantizer to be used.

  • Page 44: Mv Re-Use

    MV re-use During multi-pass encoding it is necessary to pass the same source video through the encoder two or more times. One of the more computationally intensive parts of encoding is motion estimation, where the encoder tries to find vectors predicting the movement of each block from frame-to-frame via the motion search.

  • Page 45: 1-Pass, Quality-Based

    1-Pass, Quality-based In 1-Pass quality-based mode the encoder will output a working DivX video stream as it receives the source video, similar to in 1-Pass mode except that you specify the quality rather than the target bitrate. This mode is only accessible when profiles have been disabled from the Select Profile Wizard.
  • Page 46 1-Pass, Quality-based mode is not explicitly unsupported by DivX Certified devices, however the Video Buffer Verifier is disabled in this mode and there is no guarantee that the video stream produced will not exceed the capabilities of the certified device. When 1-Pass, Quality-based mode is enabled any Video Buffer Verifier CLI parameters are disregarded.

  • Page 47: Performance/Quality

    Performance/ Quality Guide...

  • Page 48: Fastest

    Performance/Quality Fastest The fastest performance/quality mode causes the encoder to perform no motion search when encoding. Because no motion estimation is done all blocks will either be intra-blocks or predicted with a null motion vector. In this respect the encoded video will resemble an MJPEG sequence.
  • Page 49 Performance/Quality: -pq <mode> Mode is one of: Fastest Standard Slow Slowest Guide...

  • Page 50: Standard

    Standard The standard performance/quality mode enables the motion search algorithm and is functionally equivalent to Slowest mode in previous versions of DivX 5, but contains algorithm enhancements that improve video quality. Standard mode respects the Maximum Keyframe Interval and Scene Change Threshold settings (described later).

  • Page 51: The Rate-Distortion Algorithm

    The Rate-Distortion algorithm New to DivX 5.1 is the Rate-Distortion algorithm, enabled to varying degrees when the Performance/Quality setting is either Slow or Slowest. The rate- distortion algorithm allows the outcome of various decisions made by the encoder to be evaluated intelligently with respect to bit spend against quality gain where previously simple algorithms would govern the process.
  • Page 52 Here points above the rate-distortion curve evaluate as good decisions, those below the curve as bad decisions. The best decisions fall towards the upper-left of the plot (higher quality and lower bit spend). It follows that the encoder must derive this rate-distortion curve from somewhere. The curve is approximated by some function inside the encoder accepting a value that changes the shape of the curve.

  • Page 53: Slow

    Slow Slow performance/quality mode enables the performance-optimized Rate-Distortion algorithm, designed to dramatically improve video quality at low bitrates while avoiding the performance hit incurred by the fully quality-orientated rate distortion algorithm. The Rate-Distortion algorithm attempts to best balance bit spend against quality gain by intelligently evaluating all of the possible outcomes of various decisions made by the encoder.

  • Page 54: Guide

    Slowest Slowest performance/quality mode enables the quality-optimized Rate-Distortion algorithm, designed to dramatically improve video quality at low bitrates. Slowest mode will yield the highest possible quality available but at the expense of the longest encoding duration. The Rate-Distortion algorithm attempts to best balance bit spend against quality gain by intelligently evaluating all of the possible outcomes of various decisions made by the encoder.

  • Page 55: Psychovisual Enhancement

    Psychovisual Enhancement Guide...

  • Page 56: What Is Psychovisual Enhancement

    Psychovisual Enhancement What is psychovisual enhancement? DivX is what is known as a lossy codec—that is after encoding and decoding a video the result will not be identical to the source, the video will have been degraded to some extent. In other words it will have lost detail. DivX, like other lossy codecs you may be familiar with such as MP3 or Vorbis, uses encoding techniques based upon human perception to intelligently lose detail where it is least likely to be noticed before degrading the remainder of the...

  • Page 57: The Psychovisual Enhancement System

    The human visual system is far less sensitive to high frequencies in an image than it is to low frequencies. One perceptual technique DivX uses during lossy compression is the reduction in accuracy of higher frequency co-efficients, saving bits while causing the least perceivable quality degradation. It is in fact these DCT co-efficients that are quantized by DivX when image data is encoded (see Forward—Quantizers).

  • Page 58: Experiencing Pve

    Because both psychovisual enhancement methods mask artifacts in textured areas where they are least visible the video quality appears to improve when psychovisual enhancements are enabled. You can visualize the concept by imagining two characters having a conversation while standing in front of a tree— the psychovisual enhancement process might enhance details in the characters faces at the expense of causing some artifacting amongst the leaves of the tree where it would be least perceivable.

  • Page 59: Fast Psychovisual Enhancement

    Fast psychovisual enhancement Fast psychovisual enhancement mode attempts to manipulate the DCT co-efficients selectively so that the resulting noise in the decoded image is positioned where it will be least visible, for example in areas of strong texture. When encoding a multipass video without MV re-use enabled you can save time by enabling psychovisual enhancements from the second-last pass onwards only.

  • Page 60: Slow Psychovisual Enhancement

    Slow psychovisual enhancement Slow psychovisual enhancement mode analyses each block and the blocks surrounding it in turn to ensure any enhancement does not introduce significant blocking and ringing artifacts, thus slow mode is less likely to introduce artifacts than fast mode.
  • Page 61 Source pre- processing Guide...

  • Page 62: Source Pre-Processing

    Source pre-processing What is pre-processing? In the modern computer age digital video can be acquired from a wide and varied array of sources. DV Camcorders, DVD video discs, TV capture cards and even web-cameras have all become so accessible at the consumer level that for DivX to consistently produce high quality output it is necessary to provide a means of correcting video issues attributable to particular sources.

  • Page 63: Spatial Processing

    Spatial filtering Pre-processing consists of two filter types, the first is a spatial filter. The spatial filter is concerned with each pixel in the frame and the pixels surrounding it. Noise manifests itself as high frequency changes between the colors of adjacent pixels.

  • Page 64: Source Pre-Processing

    Source pre-processing Source pre-processing applies spatial and temporal filtering prior to encoding in order to reduce noise in the source video. Four presets are defined allowing control of the filter intensity. These are light, normal, strong and extreme. Both light and normal presets are very carefully balanced and neither should cause visible degradation to the video.

  • Page 65: Crop And Resize

    Crop and Resize Guide...

  • Page 66: Cropping And Resizing Via The Encoder

    Crop and Resize Cropping and resizing via the encoder Cropping and resizing are two of the most basic filters than can be applied to video, and yet not all video applications provide this functionality. Both techniques alter the dimensions of the video area, cropping by discarding areas of the video and resizing by scaling the video along its horizontal and/or vertical axis.
  • Page 67 This extra bit spend means throughout the video less bits are available to encode the actual picture and thus video quality suffers. By cropping borders no bits are wasted outside the actual picture area and thus overall quality is improved. Resizing is used in a variety of situations, the simplest case being where there is a desire to reduce the video dimensions.

  • Page 68: Crop

    Crop Crop allows you to specify the number of pixels from each edge of the video that will be discarded before encoding. The cropping area is fixed throughout the video. Crop top Crop bottom Cropping in some applications, such as VirtualDub, can require colorspace conversions that reduce encoding performance and minimally degrade the video.

  • Page 69: Resize

    Resize Resizing scales the video dimensions on one or both axis. You can select the type of sampling that is performed in the source image to calculate the color of each pixel in the resized image. Different sampling options create slightly different effects in the resized image.
  • Page 70 The encoder resize filters are very highly optimized. Consider using them in place of external resize filters provided with video editing applications. Resizing in certain applications, such as VirtualDub, can require colorspace conversions that reduce encoding performance and minimally degrade the video.

  • Page 71: Mpeg4 Tools

    MPEG4 Tools Guide...

  • Page 72: Bi-Directional Encoding

    Bi-directional encoding What is bi-directional encoding? As discussed in Forward—General concepts, there are three different frame types available for the DivX encoder to select from. These are intra-frames, predicted-frames and bi-directional-frames, or I, P and B frames respectively. Recall that in an I-frame all blocks are intra-blocks and are encoded as the image filling the block.

  • Page 73: The Ordering Of Frames

    As described earlier in the guide, predicting blocks consumes far less bits than encoding them intra-blocks. Because B-frames can be both forwards and backwards predicted in general B-frames will have a lower proportion of intra- blocks than any other type of frame. Since B-frames themselves are never referenced by any other frame it is possible for the encoder to use a higher quantizer than is normal for their coding since it need not be concerned with impacting upon the quality of other frames...
  • Page 74 Now consider the re-ordered sequence of frames: Frame 3 (P) Frame 2 (B) Frame 1 (I) Frame 5 (P) Frame 4 (B) Although Frame 2 still makes reference to both frame 1 and frame 3, frame 3 is decoded first. It is now clear that in the re-ordered sequence all frames can be decoded, as every frame makes reference only to others that have been previously decoded (all arrows point left).

  • Page 75: Bi-Directional Encoding

    Bi-directional encoding Bi-directional encoding allows the encoder to use the bi-directional frame type, supporting both forwards and backwards motion prediction. B-frames offer the best compression ratio of all frame types and are invaluable to producing high quality video at low bitrates. Bi-directional encoding can be toggled from the second step of the Select Profile Wizard dialogue under MPEG4 Tools.
  • Page 76 Bi-directional encoding: Enabled only when CLI parameter is present. Guide...

  • Page 77: Quarter Pixel

    Quarter Pixel What is quarter-pixel? In a digitized video each frame is composed of a 2- dimensional array of pixels, the small colored squares that when viewed collectively from a distance make up the picture. Recall from Forward-Macroblocks and motion that during the motion search DivX attempts to locate a matching image for every block in the current frame from a similarly sized area in the reference frame.

  • Page 78: Why Use Quarter-Pixel

    Why use quarter-pixel? Motion vectors accurate to quarter-pixel resolution allow each motion- compensated frame to be reconstructed more accurately than those accurate to half-pixel resolution. The result tends to be that video appears sharper and of slightly higher quality. Consider that the eye is most sensitive to motion where objects in a scene are moving very slowly.

  • Page 79: Quarter-Pixel

    Quarter-pixel Quarter-pixel increases the resolution of the motion search to one quarter of one pixel, double that of the default half-pixel resolution. Quarter-pixel resolution motion vectors reduce the error in the motion-compensated frame but increase the search time, leading to longer encoding durations. Quarter-pixel can be enabled from the second step of the Select Profile Wizard dialogue under MPEG4 Tools only when Disable profiles has...
  • Page 80 If you are not creating content strictly for your personal use be aware of the decoding implications of quarter-pixel motion accuracy. Slower computers may perform very poorly when attempting to decode quarter-pixel video— some may fail to decode in real-time even when the quality slider is at its minimum setting.

  • Page 81: What Is Global Motion Compensation

    Global motion compensation What is global motion compensation? Global motion compensation reduces the bit spend for motion estimation by deriving the motion of some blocks from a global motion present in the frame, as opposed to explicitly specifying a unique vector for each.

  • Page 82: Global Motion Compensation

    Global motion compensation Global motion compensation reduces the bit spend for motion estimation by deriving the motion vectors of some blocks during decoding from a common global motion present in the scene, as opposed to specifying a unique vector for each. Global motion compensation can be enabled from the second step of the Select Profile Wizard dialogue under MPEG4 Tools only when Disable profiles has been...

  • Page 83: Advanced

    Advanced Guide...

  • Page 84: Scene-Change Threshold

    Advanced Scene-change threshold QUICK GUIDE As described in Forward—Predicted-frames and Intra-frames, each frame in a video can be one of three types: intra, predicted or bi-directional. It follows that there must be some logic controlling frame type selection—the encoder is not human and can’t actually recognize where one scene changes into the next.
  • Page 85 Note that what the encoder views to be a scene-change does not necessarily correspond with what humans actually perceive to be a scene-change—no matter what value you configure the scene-change threshold to be there is no guarantee that the encoder will key-frame on an actual scene change, just as there is no guarantee that the encoder will not key-frame elsewhere.

  • Page 86: Maximum Key-Frame Interval

    Maximum key-frame interval The maximum key-frame interval defines the maximum number of consecutive frames that may be a predicted type. If maximum key-frame interval frames have been encoded since the last intra-frame, including of the current frame, the current frame will be forced intra. A fixed limit on the maximum interval between key- frames is essential in ensuring that seek and error recovery is prompt throughout the encoded video –...
  • Page 87 Interlacing Guide...

  • Page 88: What Is Interlacing

    Interlacing What is interlacing? Interlacing is a method adopted from the old days of analogue television that allowed the frame rate of a video to be doubled by broadcasting only half a frame at a time. This was achieved by dividing each frame into two fields, each occupying alternating horizontal lines.

  • Page 89: Why De-Interlace

    Why de-interlace? During playback of an interlaced video half of one frame is continually being merged with the previous picture. If the encoder were to treat each frame as progressive (a complete picture that is not interlaced) then inevitably artifacting would occur when half of a new frame was merged with the remaining half of the previous frame, particularly where the new frame differs significantly from the last.

  • Page 90: Source Interlace

    Source interlace Source interlace specifies the interlacing format of the source video and how interlacing should be handled by the encoder. Encode as progressive The encoder will assume that the source video is progressive (not interlaced) and no de- interlacing will be performed prior to encoding. Encode as interlaced The encoder will assume that the source video is interlaced and will encode each field in...
  • Page 91 Source interlace: -d <mode> Mode is one of: Encode as progressive (parameter may be omited) Encode as interlaced De-interlace all frames Guide...
  • Page 92 Video Buffer Verifier Guide...

  • Page 93: Video Buffer Verifier

    Video buffer verifier What is the video buffer verifier? The encoding profiles available in DivX 5.1 create video that is guaranteed to play back on DivX Certified devices. To facilitate this it is necessary to control aspects of the video stream produced by the encoder in such a way that the capabilities of the target device are not exceeded.

  • Page 94: Video Buffer Verifier

    Video buffer verifier The video buffer verifier prevents the encoder exceeding the ability of a target system to read and buffer the video stream as to prevent pauses in playback. Each DivX Certified Profile has specific VBV parameters associated with it that ensure playback performance on DivX Certified devices.
  • Page 95 Video buffer verifier: -vbv <bitrate>, <buffer size>, <initial occupancy> Bitrate is the maximum sustainable read bitrate in bits per second of the target system. Buffer size is the size in bits of the buffer associated with the video stream in the target system.

  • Page 96: Profiles

    Profiles Guide...

  • Page 97: What Is A Profile

    Profiles What is a profile? As DivX grew over the years into a professional-grade standard for low-bitrate video, hardware manufactures realized its potential in the consumer market and developed devices claiming to be DivX compatible. In truth these players would typically support some, but not all, versions of DivX, and then only limited features from these versions.

  • Page 98: Profiles

    Profiles Profiles enforce that the encoder creates a video stream compatible with DivX Certified devices. By selecting the profile of the DivX Certified device from the Select Profile Wizard prior to encoding the video stream will always play correctly on the device. Any given profile may limit the available encoder options based upon the minimum requirements defined by the certification program for that profile.
  • Page 99 Profiles: -profile <number> Number is one of: Disable profiles Handheld profile Portable profile Home Theatre profile High-definition profile Guide...
  • Page 100 DivX Certified™ Program Guide...

  • Page 101: Divx Certified™ Program

    DivX Certified™ Program What is the DivX Certified Program? The certification program was developed by DivX Networks to ensure the ability of certified devices to play all versions of DivX video without problems. DivX Certified devices come in various forms, each has a suitable profile associated with it that defines its abilities.

  • Page 102: What's In A Profile

    What’s in a profile? Each profile is associated with a set of minimum features a device must support in order to achieve certification. The following table lists profile-specific features. 176 x 144 Maximum resolution 15 fps Macroblocks per 1485 second Maximum average 128 kbps bitrate...

  • Page 103: Divx Certified Program Requirements

    DivX Certified Program requirements Aside from profile-specific features, there are requirements that all DivX Certified devices must meet. The following table lists the generic requirements of certified devices. Feature All DivX 3.11 movies on 1 CD, anything under 1 mbps average bitrate All DivX 4 content DivX 5 content with no GMC and no QPel DivX video created for Video on Demand...
  • Page 104 E.K.G. Guide...

  • Page 105: Electrokompressiongraph

    Electrokompressiongraph What is the EKG? We saw in Bitrate mode—Multipass that after each pass of a Multipass encoding a log file is written containing an analysis of the video encoded during the pass, used in turn to optimize the rate-control strategy for any successive pass. Normally the rate-control will use the information from the log file in an unbiased fashion, lending equal weight to all frames and basing frame quantizers upon the bitrate and frame complexity.

  • Page 106: How The Ekg Works

    How the EKG works As discussed in Forward-General concepts-Quantizers, DivX uses quantizers to control the picture quality in each frame. Lower quantizers equate to higher quality and bit-spend, while higher quantizers equate to lower quality and bit- spend. The aim of the EKG application is to control the modulation value associated with each frame in the multipass log file.

  • Page 107: Using The Ekg

    Using the EKG Using the EKG it is possible to display graphically the contents of the multipass log file after only the 1st pass, however in order to preview the encoded video as you manipulate the modulation control at least one nth pass must have been performed.
  • Page 108 Click the Edit button on the toolbar to enter editing mode. Clicking on the graph will now operate the modulation control system. Click on any frame to select it. The bar will turn purple to indicate that it has been selected and the modulation grab-bar will appear.

  • Page 109: Divx Decoder

    DivX Decoder Guide...

  • Page 110: About The Decoder

    DivX Decoder About the decoder The DivX decoder is capable of playing all DivX 3, 4 and 5 series video, applying special post-processing effects that enhance video quality, and making best use of the advanced features provided by your graphics hardware to improve playback performance.

  • Page 111: De-Blocking

    De-blocking The most visible artifact occurring in low- bitrate DivX video is called blocking, and appears as small 8x8 squares in the video. The artifact is caused by the DCT algorithm used by DivX to encode the image, and becomes strongest where video is encoded at very low bitrates due to high quantization.

  • Page 112: Post-Processing Levels

    Post-processing levels The manual de-blocking and de-ringing controls do not control the strength of the post-processing, but rather where and how it is applied. Each pixel in a DivX video is composed of three color channels—one luminance channel and two chrominance channels. The eye is most sensitive to changes in the luminance channel and so de-blocking gives preference to this.

  • Page 113: Quality Settings

    Quality settings The quality settings page allows configuration of decoder options relating to the rendering of DivX video. ►Smooth playback When a DivX video contains bi-directional frames smooth playback makes use of a buffer so the decoder need only decode one frame for every frame it displays.

  • Page 114: Overlay Extended

    ►Overlay Extended When enabled the decoder will attempt to use the hardware overlay instead of the software overlay. The hardware overlay is a feature of some graphics cards that increases both rendering performance and quality when video is scaled. When overlay extended is enabled only one video file may be played concurrently.

  • Page 115: Acknowledgements

    Acknowledgements Guide...

  • Page 116: Thanks And Credits

    Acknowledgements Thanks and credits Permission to illustrate this guide with images taken from Killer Bean 2 was very kindly given by Jeff Lew. See his work online and read about his upcoming DVD, Learning 3D Character Animation: http://www.jefflew.com Thanks to the entire DivX Advanced Research Centre (DARC) team for their assistance in writing this guide: Eugene Kuznetsov ...

Table of Contents