Vc1 Codec
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Vc1 Codec as PDF for free.
More details
- Words: 5,269
- Pages: 15
VC-1 for Digital Rapids Stream
Microsoft VC-1 Encoder Settings Microsoft VC-1 Encoder Settings.................................................................................................................. 1 General ...................................................................................................................................................... 2 Rate Control .............................................................................................................................................. 3 Timecode................................................................................................................................................... 4 GOP Settings............................................................................................................................................. 4 Closed Captioning ..................................................................................................................................... 6 Advanced Settings..................................................................................................................................... 6 The Microsoft VC-1 encoder is used to create video elementary streams. It can be used in conjunction with the group profile to create an asf file that includes both video and audio or a live stream. It can also be used to create a Bluray compatible VC-1 file that can be used in a Bluray authoring program.
Microsoft VC-1 Encoder Settings
Page 1 of 15
VC-1 for Digital Rapids Stream
General Display Size vs Encode Size The Display frame size will only be available when the Profile selected is Advanced. This feature is not available for Main or Simple profile. For most encodes, the Display frame size settings should match the height and width of the Encode frame size. Specifying a different value for display resolution than for encoded resolution can help to enable some encoding scenarios. For example, video may be scaled and encoded at a low resolution but displayed at full resolution. This can reduce network bandwidth usage when live streaming content.
Frame Rate Set the frame rate by choosing an option from the list, or specifying Custom and entering a frame rate.
Never drop frames This setting is only supported for 1-pass CBR encoding. Enabling this setting ensures that the target frame rate is maintained, even in cases where there are insufficient bits to be encoded, by inserting an explicit skipped frame flag into the bit stream. The skipped frame flag is an indication that the current frame is a visual duplicate of the previous frame. When disabled, the encoder extends the duration of the previous frame to compensate for a dropped frame. For IPTV operation you must enable this setting, so that frame-based metadata such as closed captioning can be applied to a frame.
VC-1 Profile VC-1 supports 3 profiles, and each profile supports specific features, bit rates, and resolutions. Once you have set a profile, the level will be set automatically by the encoder, based on your other settings. You should select a profile based on the requirements of you decoder, or playback device. Profile Simple
Main
Advanced
Level
Max Bit Rate
Representative Resolutions by Frame Rate
Low
96 Kbps
176 × 144 @ 15 Hz (QCIF)
Medium
384 Kbps
240 × 176 @ 30 Hz 352 × 288 @ 15 Hz (CIF)
Low
2 Mbps
320 × 240 @ 24 Hz (QVGA)
Medium
10 Mbps
720 × 480 @ 30 Hz (480p) 720 × 576 @ 25 Hz (576p)
High
20 Mbps
1920 × 1080 @ 30 Hz (1080p)
L0
2 Mbps
352 × 288 @ 30 Hz (CIF)
L1
10 Mbps
720 × 480 @ 30 Hz (NTSC-SD) 720 × 576 @ 25 Hz (PAL-SD)
Microsoft VC-1 Encoder Settings
Page 2 of 15
VC-1 for Digital Rapids Stream
L2
20 Mbps
720 × 480 @ 60 Hz (480p) 1280 × 720 @ 30 Hz (720p)
L3
45 Mbps
1920 × 1080 @ 24 Hz (1080p) 1920 × 1080 @ 30 Hz (1080i) 1280 × 720 @ 60 Hz (720p)
L4
135 Mbps
1920 × 1080 @ 60 Hz (1080p) 2048 × 1536 @ 24 Hz
.
Field/Frame Mode This setting is only supported for Advanced Profile. Use this to indicate whether the source video is progressive, interlaced with top field first, or interlaced with bottom field first. Note that if you choose Progressive, this will NOT deinterlace the video. If you need to deinterlace the video either use the Digital Rapids hardware deinterlacer or the software deinterlacer plugin.
Complexity Level There are 6 complexity settings, ranging from Fastest (best performance) to Extreme (best quality). The highest setting you can use in real time will be determined by your system’s CPU speed and number of CPU cores.
Rate Control Rate Control Mode There are 5 available modes: 1-pass CBR: Normally used for streaming video 1-pass VBR: Useful for creating files for later processing, but has no rate control, so file size will not be predictable 2-pass CBR: 1-pass CBR with Look ahead rate control enabled may yield superior results than 2pass CBR 2-pass VBR peak constrained: Use for file-based encodes 2-pass VBR unconstrained: Use for file-based encodes where the playback device can handle a wide range of bitrates
Bitrate The value entered for bitrate will be proportional to file resulting file size. The range is 1kbps to 135Mbps. This setting is not used when Rate Control Mode is set to 1-pass VBR.
Peak Bitrate This setting is only used for 2-pass VBR peak Constrained. The Peak bitrate setting determines how many bps are allocated to the frames of the video that are hardest to encode.
Microsoft VC-1 Encoder Settings
Page 3 of 15
VC-1 for Digital Rapids Stream
VBV Buffer Size (in ms) Not used with 1-pass VBR or 2-pass unconstrained VBR modes. Enter a value for the desired VBV buffer size, in milliseconds (ms). For streaming content from Windows Media Services with Advanced Fast Start and Fast Cache modes, use a buffer size that corresponds to a duration of 8000 milliseconds. Lower buffer sizes/durations are useful when attempting to encode for low latency connections. The allowable VBV range depends on the Profile. Simple Profile: corresponds to buffer sizes of 1 to 157,696 bytes Main Profile: corresponds to buffer sizes of 1 to 5,001,216 bytes Advanced profile: corresponds to buffer sizes of 1 to 33,792,000 bytes To calculate the VBV buffer size in bytes, use the following equation: VBVBuffer (bytes) = Bit rate in kbps x buffer duration in seconds x (1000 bits/kbits / 8 bits/byte) For a 500 kbps encode with a 8 second buffer duration, the VBV Buffer size would be = 500 x 8 x 125 = 500,000 bytes.
Look ahead rate control This setting is only supported for 1-pass CBR. Look ahead rate control mode optimizes the tradeoff between video quality and bit usage in scenarios that require the use of short buffer and GOP sizes, such as live broadcast over IP networks. It works by applying greater compression to B-frames, thus freeing up more bits for higher quality I-frames.
Timecode If you would like to embed timecode in the VC-1 file, then you can enable this setting. This will enable the Timecode settings button. Click this button to select whether you want to embed timecode from the source video, or from a user-specified starting timecode.
GOP Settings GOP Size: Max key frame distance Enter a value for the maximum duration between key frames in seconds. The encoder may insert additional keyframes at a shorter interval, depending on whether or not a scene change has been detected. Note that a higher maximum key frame distance value will potentially yield a better compression, while a lower value will allow you to stop and re-start the video as well as to scrub the video more smoothly. If you are using B-frames the duration must be greater than or equal to the number of B frames + 1 divided by the number of frames per second.
Microsoft VC-1 Encoder Settings
Page 4 of 15
VC-1 for Digital Rapids Stream
Fixed GOP Duration If you require a fixed GOP duration, that is you do not want the encoder to insert additional keyframes based on scene detection, then you can enable this setting. When Fixed GOP Duration is enabled, the Adaptive GOP and Look Ahead settings will be disabled.
Closed GOP A group of pictures (GOP) can be either closed or open. A closed GOP does not contain frames that depend on adjacent GOPs. Closed GOPs are mainly used for chapter points on optical discs or for files encoded for VOD servers; they are not required for playback in a Windows Media Player.
Adaptive GOP This setting cannot be enabled at the same time as “Look Ahead Rate Control” or “Fixed GOP Duration”. When you enable this setting the encoder resets the count for the maximum key frame distance at each key frame. When this setting is disabled the encoder counts the maximum key frame distance starting from the first key frame and inserts an additional keyframe at a regular interval, regardless as to whether or not additional key frames have been inserted by the encoder due to a scene detection. For example, assume the Maximum key frame distance is 8 frames, with the Look Ahead setting enabled and an I-frame inserted at the 5th frame due to a scene change. The following would be the GOP structure depending on whether Adaptive GOP were enabled or disabled: Frames: 1-2-3-4-5-6-7-8-1-2-3-4-5-6-7-8-1-... Enabled: I-B-P-B-I-B-P-B-P-B-P-B-I-B-P-B-P-... Disabled: I-B-P-B-I-B-P-B-I-B-P-B-P-B-P-B-I-...
Look Ahead This setting cannot be enabled at the same time as “Look Ahead Rate Control” or “Fixed GOP Duration”. It can only be used with 1-pass encoding modes. Look Ahead allows the encoder to insert I-frames and B-frames based on based on content analysis, specifically scene change detection, fade detection, and flash detection. Adding I-frames and B-frames based on content analysis will optimize the compression. This optimization offers some of the B frame quality improvements that would otherwise require 2-pass encoding.
Max consecutive B Frames This setting is not available when using Simple Profile. Enter a value for the number of B-frames between key frames. The valid range is 0 to 7, the recommended value is 1.
Microsoft VC-1 Encoder Settings
Page 5 of 15
VC-1 for Digital Rapids Stream
Closed Captioning The close captioning that is embedded into the VC-1 data is the type of captioning that is used by hardware playback devices such as set top boxes. It is not the close captioning that can be played back using software players (such as the Windows Media Player).
Mode Indicate if the source for the closed captions will be from line 21 of the video input or from an SCC file.
SCC file (field 1) / (field 2) If you selected the SCC mode, then you can use this field to navigate to the SCC file.
Advanced Settings When you click on this button you will see additional settings that may be use to control the encoder.
[Quality] Quantization Setting For 1-pass VBR, this is the fixed quantization value which controls the quality of the encoding. Lower values will provide better visual quality, but will use more bits per frame. Generally values below 8 yield good results and values above 8 may show distracting artifacts. The valid range is 1 to 31. This setting is not used for 2-pass VBR modes. For 1-pass or 2-pass CBR, this represents the maximum frame quantization value. If there are not enough bits for the next frame to be at or below the maximum frame quantization value, then that frame will not be encoded, and those bits will be allocated to the subsequent frame. Thus, a lower fixed quantization value may result in dropped frames. In general, you should choose the highest value that provides the minimum acceptable visual quality in order to reduce the number of dropped frames.
[B Frames] B Frame delta quantization setting The B-frame delta QP specifies the amount of the increase in QP for B-frames relative to the anchor frame QP on a per-macroblock basis. The QP value is incremented in whole steps, with a valid range of 1 to 30. A higher QP value means a higher compression ratio. Increasing B-frame delta QP can sometimes result in better video quality because this can free up bits to better compress the key frames from which the B-frames are temporally predicted. This method is only supported when the Number of B frames is greater than 0. Not supported for Simple profile. By default, the encoder will use dynamic B frame delta QP settings.
[Quantization] Adaptive quantization setting Adaptive quantization is an encoding method that attempts to balance the bit distribution within a picture between smooth areas having fine details and textured areas having coarse details. Unlike differential quantization encoding, which uses different quantizer levels for macro blocks within the image, adaptive dead zone methods use the same quantizer level for the entire image but vary the size of the quantizer dead zone depending on the level of textures in the image. The dead zone is created during the quantization step of image compression. It represents the bin in which all AC coefficients that quantize to zero are stored. AC coefficient values close to zero commonly
Microsoft VC-1 Encoder Settings
Page 6 of 15
VC-1 for Digital Rapids Stream represent noise and subtle picture detail. Increasing the dead zone therefore ensures more low-frequency image detail is lost during quantization, which can in turn reduce the data size of compressed frames. The adaptive dead zone method dynamically increases the size of the dead zone in macro blocks containing textured areas. In the context of encoding with rate control, this often translates to lower QP and higher quality in smooth areas due to more bits being available overall. Increasing the adaptive dead zone strength controls how textured areas are mapped to larger deadzones. The recommended strength is the lowest adaptive dead-zone strength. Using higher values can result in image detail being eroded too readily from textured areas. Adaptive dead zone methods can be used with quantization option, quantization strength for P frames and quantization strength for B frames for a combined approach to perceptual optimization. While usually effective on film sources with natural noise (such as film grain), these methods are not meant for use as generic encoding enhancements and should be used with caution. Higher values mean stronger quantization. The following table lists the possible values: Value
Description
0
Off (Default)
1-15
Conservative range
16-20
Strong range
[Quantization] Differential quantization option Differential quantization (DQuant) is a form of perceptual optimization where the amount of compression is varied on different parts of the frame. DQuant enables macro blocks containing smooth or dark areas to be encoded at a quantizer level lower than the rest of the image, yielding more accurate encoding of details in those macro blocks. DQuant can improve video quality in smooth areas containing very fine detail or gradients, or very dark uniform areas because those areas are prone to blocking artifacts at high quantizer levels. The drawback to using DQuant is that using lower QPs for certain macro blocks can use up too many bits for the entire frame, resulting in a higher general QP for the rest of the macro blocks in the frame. In other words, improved quality in targeted areas might result in reduced quality in the rest of the image. Using more than 2 levels of quantizers can also add an additional overhead to compressed sample sizes due to the necessity to signal different quantizer levels for each macroblock. DQuant applied only to I and P frames usually produces the best results. Note that DQuant in Main profile doesn't actually apply to I-frames, but the I+P and I+P+ B settings parameters are used in the normal fashion. Use a setting of 0 for Simple profile or 1-pass VBR. Recommended settings for DQuant: 0 for moderate to low bitrates (anything under 15 Mbps for HD). 2 for high-bit-rate encoding.
Microsoft VC-1 Encoder Settings
Page 7 of 15
VC-1 for Digital Rapids Stream The following table lists the possible values: Value
Description
0
DQuant perceptual optimization will not be used. (Default)
1
Only I-frames will be optimized.
2
I-frames and P-frames will be optimized, but B-frames will not.
3
I, P, and B-frames will be optimized.
[Quantization] Differential quantization strength for P frames Specifies the strength of the DQuant perceptual optimization that will be used for P frames. The stronger the optimization, the more CPU-intensive the encoding will be. The default is 0 (off). Other possible values are 1 and 2. Higher values correspond to stronger optimization. Specify 0 for Simple profile or 1pass VBR. Must be non-zero when the differential quantization setting is 2 or 3. Otherwise, this value must be zero. Recommendation: 0 should be used for most content. Try setting 1 if you notice blocking in smooth regions after initial encoding and setting 2 if you still see blocking after trying setting 1.
[Quantization] Differential quantization strength for B frames Specify the strength of the Dquant perceptual optimization that will be used for B frames. The stronger the optimization, the more CPU-intensive the encoding will be. The default is 0 (off). Other possible values range from 1 to 4. Higher values correspond to stronger optimization. Specify 0 for Simple profile or 1-pass VBR. Must be non-zero when the differential quantization option is 3. Otherwise, this value must be zero. Recommendation: 0
[Filters] Enable denoise filter The denoise filter is a noise reduction filter. It can improve the quality of noisy video sources, such as film containing visible grain or video that contains noise as a result of low-light conditions. This filter should be disabled if the source video does not contain visible noise artifacts. De-noising is generally performed during the preprocessing phase in Digital Rapids encoding hardware, however, this filter can be useful when preprocessing is not an option. When using this filter you cannot preview the noise reduction separately from the encoder, as you can with the Digital Rapids hardware noise reduction filter.
[Filters] Enable In-loop filter The in-loop filter is not supported for the Simple profile. The in-loop filter reduces blocking artifacts during encoding to improve the quality of P and B frames. It is also used when decoding, which means it can reduce performance during playback. Although the in-loop filter can reduce image detail in individual frames, the overall quality of the video improves. The biggest downside to using in-loop filtering is the additional decoding performance cost, which can be a problem for low-power playback device, such as cell phones. The In-Loop Filter will typically increase CPU requirements for a given encode by about 15%.
Microsoft VC-1 Encoder Settings
Page 8 of 15
VC-1 for Digital Rapids Stream
[Filters] Enable Median filter You should disable this filter for source video that does not contain visible noise artifacts as the median filter can introduce trailing artifacts behind moving objects. The median filter improves motion estimation processing by factoring out noise artifacts. This can improve the quality of very noisy video and reduce the size of the compressed data. Note that this filter is not the same as median blur filters found in many video editing and post-processing applications.
[Filters] Enable overlap smoothing The overlap smoothing filter helps to reduce blocking artifacts by smoothing the borders between adjacent macroblocks. This tends to make the image appear softer, but can improve the appearance of low bit rate video that contains many blocking artifacts.
[Filters] Enable noise edge removal filter The noise edge removal filter detects noise in frame edges and removes it. The noise edge detection only works for frame edges that contain three lines of noise, or less. The filter copies lines adjacent to the noisy lines to complete the frame. A noisy frame edge is usually caused by the vertical blanking interval (VBI) data from a frame of broadcast television being visible. The VBI is the first 21 scan lines of a broadcast frame. When a television signal is recorded by a capture card, the VBI is usually removed from the frame. The noisy edge detection and correction filter can only correct an edge that has 3 or fewer lines of noise.
[Color] Apply Color Formats This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Matrix Coefficient This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Primaries This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Transfer Character This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Motion Estimation] Delta Motion Vector Range Index This setting is for interlaced content only, and most useful for unpredictable motion. Specify 0 except for Advanced profile and interlaced encoding.
Microsoft VC-1 Encoder Settings
Page 9 of 15
VC-1 for Digital Rapids Stream Motion estimation settings control how the codec searches for motion in the frame. These settings can have a dramatic effect on quality and an even more dramatic effect on encoding time. The following table lists the possible values: Value
Description
0
Off. (Default)
1
Improves coding efficiency for highly spread-out horizontal delta motion vector distributions.
2
Improves coding efficiency for highly spread-out vertical delta motion vector distributions.
3
Improves coding efficiency for highly spread-out horizontal and vertical delta motion vector distributions.
[Motion Estimation] Motion Search Level Motion search level controls whether and how color is used in motion searches. Including chroma in motion estimation can significantly improve the quality of encoded video when chroma changes happen where luma changes do not. For example, motion graphics, cel animation, and screen recordings can be significantly improved with this setting. Motion search with luma and true chroma will yield the best quality, but at the highest performance cost. The two adaptive modes and the nearest-integer chroma mode provide reasonable compromises between quality and performance. Adaptive modes apply chroma search to the 50 percent of the blocks in the frame that are predicted to have the most benefit. This provides most of the value of chroma search with only half the encoding performance reduction The default depends on the complexity level. The following table lists the possible values: Value
Description
0
Luma only. The VC-1 encoder searches for motion in luminance values only. Provides fastest performance (encoding speed).
1
Luma with nearest integer chroma. Provides a compromise between quality and performance.
2
Luma with true chroma. Provides the best quality with the lowest performance.
3
Macroblock adaptive with nearest integer chroma. Provides a compromise between quality and performance.
4
Macroblock adaptive with true chroma. Provides a compromise between quality and performance. (recommended setting)
[Motion Estimation] Motion Search Method This setting controls the size of the area the codec will search for an element of a frame that may have moved since a previous frame. Larger search ranges can better detect fast motion, but require more processing time. Processing time roughly doubles with each search range increase. Setting the search
Microsoft VC-1 Encoder Settings
Page 10 of 15
VC-1 for Digital Rapids Stream range too high can also lead to false positives, so it's important to set the motion search window to a range adequate for the video. The following table lists the possible values: Value
Description
0
SAD. Provides the best performance.
1
Hadamard. Provides the best quality.
2
Macroblock-adaptive. This option configures the codec to make decisions about which method to use on each macroblock. This can potentially reduce overall computation required for encoding by performing the computationally-intensive Hadamard transform only when appropriate. (recommended setting)
[Motion Estimation] Macro Block Mode Cost Method Used by the codec to determine which macroblock mode to use. The default is 0 when encoder complexity is less than 2. Otherwise, the default is 1. The following table lists the possible values: Value
Description
0
SAD/Hadamard. This option configures the codec to account for only distortion when computing cost.
1
RD cost. This option configures the codec to account for both rate and distortion when computing cost.
[Motion Estimation] Motion Vector Cost Method Used by the codec to estimate the amount of processing required for motion vector coding. The codec uses the cost to determine which features will be used in encoding. The following table lists the possible values: Value
Description
0
Static method. This option uses the same motion vector cost estimate for all macroblocks.
1
Dynamic method. This option varies the motion vector cost between macroblocks to achieve optimal visual quality.
Recommended: Setting 1 if DQuant option is not used. Otherwise, use setting 0.
[Motion Estimation] Motion Search Range The range used in motion searches. Specify 0 when using Simple profile. The following table lists the possible values: Value
Description
0
+63.75/-64.0 H, +31.75/-32.0 V (Default)
1
+127.75/-128.0 H, +63.75/-64.0 V
Microsoft VC-1 Encoder Settings
Page 11 of 15
VC-1 for Digital Rapids Stream
2
+511.75/-512.0 H, +127.75/-128.0 V
3
+1023.75/-1024.0 H, +255.75/-256.0 V
4
Macroblock adaptive. The VC-1 encoder selects the most efficient mode. The maximum range never exceeds ±1024.0 H, ±256.0 V.
Recommendation: setting 4, except when using Simple profile (when you must use 0)
[Advanced] Letter Box Present This feature enables the codec to determine the location of letterbox matte bars at the top and bottom of each frame. When detected, the letterboxing is excluded from motion search operations, saving some bits and speeding up encoding. Letterbox detection is dynamic and should correctly detect changes in video mattes. However, there may be cases where detection does not immediately find a change, particularly when video frames contain mostly black regions.
[Advanced] Key pop reduction Key pop, a type of compression artifact, occurs when the encoder uses different algorithms to compress key frames and predicted frames. Using different algorithms can lead to differences in fine details, texture, or film grain. After compression, key frames may appear to stand out or otherwise appear different. The patterned nature of frame compression can make this problem even worse by creating a rhythmic pulsing in the video during playback. Key pop may also be called key frame pulsing. Key pop reduction is most useful for shorter GOP encodes, like those targeted for optical discs. Key pop reduction can also cause some softness in the video, and therefore may not be appropriate for all content, particularly content that uses long GOPS where they are not needed. Recommendation: Off unless key frame popping is visible.
[Advanced] Number of threads Each instance of the VC-1 encoder can use up to four threads to distribute encoding tasks. Each thread runs on a separate processor. Each thread processes a different horizontal portion of the frame. Motion search operations only happen within each horizontal portion. Because most motion in video is typically horizontal, such as panning, motion estimation usually works well. However, if the content contains a lot of vertical motion, some slight loss of coding efficiency may occur. Valid values are 1, 2 and 4. Note that specifying higher values will not raise an error. The encoder will simply use four threads. Recommendation: One thread for pictures less than 128 lines in height, two threads for pictures between 128 and 256 lines, and four threads for larger pictures.
[Advanced] Affinity Mask The affinity mask is used with the number of threads setting to specify which processors to use. Each binary digit (bit) in the affinity mask represents an individual processor. Setting a bit to one makes the VC1 encoder use the corresponding processor to run an encoding thread. For example, to make the VC-1 encoder use processors zero, one, two, and three, the low-order byte of the affinity mask would be 00001111 (decimal value of 15 and a hexadecimal value of 0xF).
Microsoft VC-1 Encoder Settings
Page 12 of 15
VC-1 for Digital Rapids Stream If the encoding computer has 8 processors, then processors four, five, six and seven would use an affinity mask of 11110000 (decimal value of 240 and a hexadecimal value of 0xF0).
[Advanced] CPU Usage This is only supported for 1-pass CBR encoding. This setting specifies the percentage of available processing to use for encoding. The range is 1 – 100, and the default is 80. This method can dynamically vary the complexity of the encoding to ensure a fixed encoding time, regardless of the complexity of the input video stream. Setting a lower target CPU usage value implies lower complexity and, therefore, faster encoding, while making more system resources available for other components that the application controls. Setting a higher value implies potentially higher complexity and more utilization of CPU resources by the encoder itself. Recommendation: Use the highest value that leaves sufficient resources for other processes on the computer.
[Advanced] Pixel width Use this setting along with the Pixel Height setting to specify the pixel aspect ratio. The pixel aspect ratio is defined as pixel width / pixel height. For a square pixel this ratio is 1:1. The valid range for this setting is 1 to 256. For display on a computer VGA monitor or an HDTV monitor, use an aspect ratio of 1:1. For display on an NTSC monitor, for D1/DV use 10:11 and for D1/DV anamorphic widescreen use 40:33. For display on a PAL monitor, for CCIR-601 use 16:15, for D1/DV use 59:54, and for D1/DV anamorphic widescreen use 118:81. This is not to be confused with the display aspect ratio, which defines the ratio between the width and height of the video display.
[Advanced] Pixel Height Use this setting along with the Pixel Height setting to specify the pixel aspect ratio. The pixel aspect ratio is defined as pixel width / pixel height. For a square pixel this ratio is 1:1. The valid range for this setting is 1 to 256. For display on a computer VGA monitor or an HDTV monitor, use an aspect ratio of 1:1. For display on an NTSC monitor, for D1/DV use 10:11 and for D1/DV anamorphic widescreen use 40:33. For display on a PAL monitor, for CCIR-601 use 16:15, for D1/DV use 59:54, and for D1/DV anamorphic widescreen use 118:81. This is not to be confused with the display aspect ratio, which defines the ratio between the width and height of the video display.
[Advanced] Pixel Index This setting is used for HD DVD and Blu-ray disc encoding. This value is completely independent of the Pixel Width and Pixel Height settings. A value of 1 will set 1:1 aspect ratio for Blu-Ray and a value of 5 will set 4:3.
Microsoft VC-1 Encoder Settings
Page 13 of 15
VC-1 for Digital Rapids Stream
Value
Description
1
(PAR=1:1(square))
2
(PAR=12:11)
3
(PAR=10:11)
4
(PAR=16:11)
5
(PAR=40:33)
6
(PAR=24:11).
7
(PAR=20:11).
8
(PAR=32:11)
9
(PAR=80:33)
10
(PAR=18:11).
11
(PAR=15:11)
12
(PAR=64:33)
13
(PAR=160:99)
14
(PAR=SMPTE Reserved (illegal)).
15
(PAR=custom)
[Advanced] Video Type This setting is used to specify the method used to encode progressive or interlaced source video. Value
Description
0
Progressive video. (Default for progressive and for Simple or Main profiles)
1
The VC-1 encoder encodes all frames as interlaced frames. This method is suitable for content that does not contain fast motion. Best when the content contains random motion.
2
The VC-1 encoder encodes all frames as pairs of interlaced fields. This is usually the most efficient method, especially if the content contains fast motion, such as panning. (Default for interlaced)
3
The VC-1 encoder automatically determines the most efficient interlaced encoding method (frame or field). This method is suitable for content that does not contain fast motion.
4
The VC-1 encoder automatically determines the most efficient encoding method (frame, field, or progressive). This is the best method for content that contains a mixture of frame and field types.
Microsoft VC-1 Encoder Settings
Page 14 of 15
VC-1 for Digital Rapids Stream
[Advanced] Output Type RAW: The encoder generates output for a container format, such as ASF. ES: The encoder generates output for an elementary stream with an entry point start code inserted for each GOP. Sequence start codes are inserted as needed. ES_SH: The encoder generates output for an elementary stream with both entry point and sequence start codes inserted for each GOP. For BluRay choose this setting.
Microsoft VC-1 Encoder Settings
Page 15 of 15
Microsoft VC-1 Encoder Settings Microsoft VC-1 Encoder Settings.................................................................................................................. 1 General ...................................................................................................................................................... 2 Rate Control .............................................................................................................................................. 3 Timecode................................................................................................................................................... 4 GOP Settings............................................................................................................................................. 4 Closed Captioning ..................................................................................................................................... 6 Advanced Settings..................................................................................................................................... 6 The Microsoft VC-1 encoder is used to create video elementary streams. It can be used in conjunction with the group profile to create an asf file that includes both video and audio or a live stream. It can also be used to create a Bluray compatible VC-1 file that can be used in a Bluray authoring program.
Microsoft VC-1 Encoder Settings
Page 1 of 15
VC-1 for Digital Rapids Stream
General Display Size vs Encode Size The Display frame size will only be available when the Profile selected is Advanced. This feature is not available for Main or Simple profile. For most encodes, the Display frame size settings should match the height and width of the Encode frame size. Specifying a different value for display resolution than for encoded resolution can help to enable some encoding scenarios. For example, video may be scaled and encoded at a low resolution but displayed at full resolution. This can reduce network bandwidth usage when live streaming content.
Frame Rate Set the frame rate by choosing an option from the list, or specifying Custom and entering a frame rate.
Never drop frames This setting is only supported for 1-pass CBR encoding. Enabling this setting ensures that the target frame rate is maintained, even in cases where there are insufficient bits to be encoded, by inserting an explicit skipped frame flag into the bit stream. The skipped frame flag is an indication that the current frame is a visual duplicate of the previous frame. When disabled, the encoder extends the duration of the previous frame to compensate for a dropped frame. For IPTV operation you must enable this setting, so that frame-based metadata such as closed captioning can be applied to a frame.
VC-1 Profile VC-1 supports 3 profiles, and each profile supports specific features, bit rates, and resolutions. Once you have set a profile, the level will be set automatically by the encoder, based on your other settings. You should select a profile based on the requirements of you decoder, or playback device. Profile Simple
Main
Advanced
Level
Max Bit Rate
Representative Resolutions by Frame Rate
Low
96 Kbps
176 × 144 @ 15 Hz (QCIF)
Medium
384 Kbps
240 × 176 @ 30 Hz 352 × 288 @ 15 Hz (CIF)
Low
2 Mbps
320 × 240 @ 24 Hz (QVGA)
Medium
10 Mbps
720 × 480 @ 30 Hz (480p) 720 × 576 @ 25 Hz (576p)
High
20 Mbps
1920 × 1080 @ 30 Hz (1080p)
L0
2 Mbps
352 × 288 @ 30 Hz (CIF)
L1
10 Mbps
720 × 480 @ 30 Hz (NTSC-SD) 720 × 576 @ 25 Hz (PAL-SD)
Microsoft VC-1 Encoder Settings
Page 2 of 15
VC-1 for Digital Rapids Stream
L2
20 Mbps
720 × 480 @ 60 Hz (480p) 1280 × 720 @ 30 Hz (720p)
L3
45 Mbps
1920 × 1080 @ 24 Hz (1080p) 1920 × 1080 @ 30 Hz (1080i) 1280 × 720 @ 60 Hz (720p)
L4
135 Mbps
1920 × 1080 @ 60 Hz (1080p) 2048 × 1536 @ 24 Hz
.
Field/Frame Mode This setting is only supported for Advanced Profile. Use this to indicate whether the source video is progressive, interlaced with top field first, or interlaced with bottom field first. Note that if you choose Progressive, this will NOT deinterlace the video. If you need to deinterlace the video either use the Digital Rapids hardware deinterlacer or the software deinterlacer plugin.
Complexity Level There are 6 complexity settings, ranging from Fastest (best performance) to Extreme (best quality). The highest setting you can use in real time will be determined by your system’s CPU speed and number of CPU cores.
Rate Control Rate Control Mode There are 5 available modes: 1-pass CBR: Normally used for streaming video 1-pass VBR: Useful for creating files for later processing, but has no rate control, so file size will not be predictable 2-pass CBR: 1-pass CBR with Look ahead rate control enabled may yield superior results than 2pass CBR 2-pass VBR peak constrained: Use for file-based encodes 2-pass VBR unconstrained: Use for file-based encodes where the playback device can handle a wide range of bitrates
Bitrate The value entered for bitrate will be proportional to file resulting file size. The range is 1kbps to 135Mbps. This setting is not used when Rate Control Mode is set to 1-pass VBR.
Peak Bitrate This setting is only used for 2-pass VBR peak Constrained. The Peak bitrate setting determines how many bps are allocated to the frames of the video that are hardest to encode.
Microsoft VC-1 Encoder Settings
Page 3 of 15
VC-1 for Digital Rapids Stream
VBV Buffer Size (in ms) Not used with 1-pass VBR or 2-pass unconstrained VBR modes. Enter a value for the desired VBV buffer size, in milliseconds (ms). For streaming content from Windows Media Services with Advanced Fast Start and Fast Cache modes, use a buffer size that corresponds to a duration of 8000 milliseconds. Lower buffer sizes/durations are useful when attempting to encode for low latency connections. The allowable VBV range depends on the Profile. Simple Profile: corresponds to buffer sizes of 1 to 157,696 bytes Main Profile: corresponds to buffer sizes of 1 to 5,001,216 bytes Advanced profile: corresponds to buffer sizes of 1 to 33,792,000 bytes To calculate the VBV buffer size in bytes, use the following equation: VBVBuffer (bytes) = Bit rate in kbps x buffer duration in seconds x (1000 bits/kbits / 8 bits/byte) For a 500 kbps encode with a 8 second buffer duration, the VBV Buffer size would be = 500 x 8 x 125 = 500,000 bytes.
Look ahead rate control This setting is only supported for 1-pass CBR. Look ahead rate control mode optimizes the tradeoff between video quality and bit usage in scenarios that require the use of short buffer and GOP sizes, such as live broadcast over IP networks. It works by applying greater compression to B-frames, thus freeing up more bits for higher quality I-frames.
Timecode If you would like to embed timecode in the VC-1 file, then you can enable this setting. This will enable the Timecode settings button. Click this button to select whether you want to embed timecode from the source video, or from a user-specified starting timecode.
GOP Settings GOP Size: Max key frame distance Enter a value for the maximum duration between key frames in seconds. The encoder may insert additional keyframes at a shorter interval, depending on whether or not a scene change has been detected. Note that a higher maximum key frame distance value will potentially yield a better compression, while a lower value will allow you to stop and re-start the video as well as to scrub the video more smoothly. If you are using B-frames the duration must be greater than or equal to the number of B frames + 1 divided by the number of frames per second.
Microsoft VC-1 Encoder Settings
Page 4 of 15
VC-1 for Digital Rapids Stream
Fixed GOP Duration If you require a fixed GOP duration, that is you do not want the encoder to insert additional keyframes based on scene detection, then you can enable this setting. When Fixed GOP Duration is enabled, the Adaptive GOP and Look Ahead settings will be disabled.
Closed GOP A group of pictures (GOP) can be either closed or open. A closed GOP does not contain frames that depend on adjacent GOPs. Closed GOPs are mainly used for chapter points on optical discs or for files encoded for VOD servers; they are not required for playback in a Windows Media Player.
Adaptive GOP This setting cannot be enabled at the same time as “Look Ahead Rate Control” or “Fixed GOP Duration”. When you enable this setting the encoder resets the count for the maximum key frame distance at each key frame. When this setting is disabled the encoder counts the maximum key frame distance starting from the first key frame and inserts an additional keyframe at a regular interval, regardless as to whether or not additional key frames have been inserted by the encoder due to a scene detection. For example, assume the Maximum key frame distance is 8 frames, with the Look Ahead setting enabled and an I-frame inserted at the 5th frame due to a scene change. The following would be the GOP structure depending on whether Adaptive GOP were enabled or disabled: Frames: 1-2-3-4-5-6-7-8-1-2-3-4-5-6-7-8-1-... Enabled: I-B-P-B-I-B-P-B-P-B-P-B-I-B-P-B-P-... Disabled: I-B-P-B-I-B-P-B-I-B-P-B-P-B-P-B-I-...
Look Ahead This setting cannot be enabled at the same time as “Look Ahead Rate Control” or “Fixed GOP Duration”. It can only be used with 1-pass encoding modes. Look Ahead allows the encoder to insert I-frames and B-frames based on based on content analysis, specifically scene change detection, fade detection, and flash detection. Adding I-frames and B-frames based on content analysis will optimize the compression. This optimization offers some of the B frame quality improvements that would otherwise require 2-pass encoding.
Max consecutive B Frames This setting is not available when using Simple Profile. Enter a value for the number of B-frames between key frames. The valid range is 0 to 7, the recommended value is 1.
Microsoft VC-1 Encoder Settings
Page 5 of 15
VC-1 for Digital Rapids Stream
Closed Captioning The close captioning that is embedded into the VC-1 data is the type of captioning that is used by hardware playback devices such as set top boxes. It is not the close captioning that can be played back using software players (such as the Windows Media Player).
Mode Indicate if the source for the closed captions will be from line 21 of the video input or from an SCC file.
SCC file (field 1) / (field 2) If you selected the SCC mode, then you can use this field to navigate to the SCC file.
Advanced Settings When you click on this button you will see additional settings that may be use to control the encoder.
[Quality] Quantization Setting For 1-pass VBR, this is the fixed quantization value which controls the quality of the encoding. Lower values will provide better visual quality, but will use more bits per frame. Generally values below 8 yield good results and values above 8 may show distracting artifacts. The valid range is 1 to 31. This setting is not used for 2-pass VBR modes. For 1-pass or 2-pass CBR, this represents the maximum frame quantization value. If there are not enough bits for the next frame to be at or below the maximum frame quantization value, then that frame will not be encoded, and those bits will be allocated to the subsequent frame. Thus, a lower fixed quantization value may result in dropped frames. In general, you should choose the highest value that provides the minimum acceptable visual quality in order to reduce the number of dropped frames.
[B Frames] B Frame delta quantization setting The B-frame delta QP specifies the amount of the increase in QP for B-frames relative to the anchor frame QP on a per-macroblock basis. The QP value is incremented in whole steps, with a valid range of 1 to 30. A higher QP value means a higher compression ratio. Increasing B-frame delta QP can sometimes result in better video quality because this can free up bits to better compress the key frames from which the B-frames are temporally predicted. This method is only supported when the Number of B frames is greater than 0. Not supported for Simple profile. By default, the encoder will use dynamic B frame delta QP settings.
[Quantization] Adaptive quantization setting Adaptive quantization is an encoding method that attempts to balance the bit distribution within a picture between smooth areas having fine details and textured areas having coarse details. Unlike differential quantization encoding, which uses different quantizer levels for macro blocks within the image, adaptive dead zone methods use the same quantizer level for the entire image but vary the size of the quantizer dead zone depending on the level of textures in the image. The dead zone is created during the quantization step of image compression. It represents the bin in which all AC coefficients that quantize to zero are stored. AC coefficient values close to zero commonly
Microsoft VC-1 Encoder Settings
Page 6 of 15
VC-1 for Digital Rapids Stream represent noise and subtle picture detail. Increasing the dead zone therefore ensures more low-frequency image detail is lost during quantization, which can in turn reduce the data size of compressed frames. The adaptive dead zone method dynamically increases the size of the dead zone in macro blocks containing textured areas. In the context of encoding with rate control, this often translates to lower QP and higher quality in smooth areas due to more bits being available overall. Increasing the adaptive dead zone strength controls how textured areas are mapped to larger deadzones. The recommended strength is the lowest adaptive dead-zone strength. Using higher values can result in image detail being eroded too readily from textured areas. Adaptive dead zone methods can be used with quantization option, quantization strength for P frames and quantization strength for B frames for a combined approach to perceptual optimization. While usually effective on film sources with natural noise (such as film grain), these methods are not meant for use as generic encoding enhancements and should be used with caution. Higher values mean stronger quantization. The following table lists the possible values: Value
Description
0
Off (Default)
1-15
Conservative range
16-20
Strong range
[Quantization] Differential quantization option Differential quantization (DQuant) is a form of perceptual optimization where the amount of compression is varied on different parts of the frame. DQuant enables macro blocks containing smooth or dark areas to be encoded at a quantizer level lower than the rest of the image, yielding more accurate encoding of details in those macro blocks. DQuant can improve video quality in smooth areas containing very fine detail or gradients, or very dark uniform areas because those areas are prone to blocking artifacts at high quantizer levels. The drawback to using DQuant is that using lower QPs for certain macro blocks can use up too many bits for the entire frame, resulting in a higher general QP for the rest of the macro blocks in the frame. In other words, improved quality in targeted areas might result in reduced quality in the rest of the image. Using more than 2 levels of quantizers can also add an additional overhead to compressed sample sizes due to the necessity to signal different quantizer levels for each macroblock. DQuant applied only to I and P frames usually produces the best results. Note that DQuant in Main profile doesn't actually apply to I-frames, but the I+P and I+P+ B settings parameters are used in the normal fashion. Use a setting of 0 for Simple profile or 1-pass VBR. Recommended settings for DQuant: 0 for moderate to low bitrates (anything under 15 Mbps for HD). 2 for high-bit-rate encoding.
Microsoft VC-1 Encoder Settings
Page 7 of 15
VC-1 for Digital Rapids Stream The following table lists the possible values: Value
Description
0
DQuant perceptual optimization will not be used. (Default)
1
Only I-frames will be optimized.
2
I-frames and P-frames will be optimized, but B-frames will not.
3
I, P, and B-frames will be optimized.
[Quantization] Differential quantization strength for P frames Specifies the strength of the DQuant perceptual optimization that will be used for P frames. The stronger the optimization, the more CPU-intensive the encoding will be. The default is 0 (off). Other possible values are 1 and 2. Higher values correspond to stronger optimization. Specify 0 for Simple profile or 1pass VBR. Must be non-zero when the differential quantization setting is 2 or 3. Otherwise, this value must be zero. Recommendation: 0 should be used for most content. Try setting 1 if you notice blocking in smooth regions after initial encoding and setting 2 if you still see blocking after trying setting 1.
[Quantization] Differential quantization strength for B frames Specify the strength of the Dquant perceptual optimization that will be used for B frames. The stronger the optimization, the more CPU-intensive the encoding will be. The default is 0 (off). Other possible values range from 1 to 4. Higher values correspond to stronger optimization. Specify 0 for Simple profile or 1-pass VBR. Must be non-zero when the differential quantization option is 3. Otherwise, this value must be zero. Recommendation: 0
[Filters] Enable denoise filter The denoise filter is a noise reduction filter. It can improve the quality of noisy video sources, such as film containing visible grain or video that contains noise as a result of low-light conditions. This filter should be disabled if the source video does not contain visible noise artifacts. De-noising is generally performed during the preprocessing phase in Digital Rapids encoding hardware, however, this filter can be useful when preprocessing is not an option. When using this filter you cannot preview the noise reduction separately from the encoder, as you can with the Digital Rapids hardware noise reduction filter.
[Filters] Enable In-loop filter The in-loop filter is not supported for the Simple profile. The in-loop filter reduces blocking artifacts during encoding to improve the quality of P and B frames. It is also used when decoding, which means it can reduce performance during playback. Although the in-loop filter can reduce image detail in individual frames, the overall quality of the video improves. The biggest downside to using in-loop filtering is the additional decoding performance cost, which can be a problem for low-power playback device, such as cell phones. The In-Loop Filter will typically increase CPU requirements for a given encode by about 15%.
Microsoft VC-1 Encoder Settings
Page 8 of 15
VC-1 for Digital Rapids Stream
[Filters] Enable Median filter You should disable this filter for source video that does not contain visible noise artifacts as the median filter can introduce trailing artifacts behind moving objects. The median filter improves motion estimation processing by factoring out noise artifacts. This can improve the quality of very noisy video and reduce the size of the compressed data. Note that this filter is not the same as median blur filters found in many video editing and post-processing applications.
[Filters] Enable overlap smoothing The overlap smoothing filter helps to reduce blocking artifacts by smoothing the borders between adjacent macroblocks. This tends to make the image appear softer, but can improve the appearance of low bit rate video that contains many blocking artifacts.
[Filters] Enable noise edge removal filter The noise edge removal filter detects noise in frame edges and removes it. The noise edge detection only works for frame edges that contain three lines of noise, or less. The filter copies lines adjacent to the noisy lines to complete the frame. A noisy frame edge is usually caused by the vertical blanking interval (VBI) data from a frame of broadcast television being visible. The VBI is the first 21 scan lines of a broadcast frame. When a television signal is recorded by a capture card, the VBI is usually removed from the frame. The noisy edge detection and correction filter can only correct an edge that has 3 or fewer lines of noise.
[Color] Apply Color Formats This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Matrix Coefficient This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Primaries This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Color] Color Transfer Character This setting is for HD DVD encoding only. Refer to the HD DVD specification for details. Ensure this off for Blu-ray encoding.
[Motion Estimation] Delta Motion Vector Range Index This setting is for interlaced content only, and most useful for unpredictable motion. Specify 0 except for Advanced profile and interlaced encoding.
Microsoft VC-1 Encoder Settings
Page 9 of 15
VC-1 for Digital Rapids Stream Motion estimation settings control how the codec searches for motion in the frame. These settings can have a dramatic effect on quality and an even more dramatic effect on encoding time. The following table lists the possible values: Value
Description
0
Off. (Default)
1
Improves coding efficiency for highly spread-out horizontal delta motion vector distributions.
2
Improves coding efficiency for highly spread-out vertical delta motion vector distributions.
3
Improves coding efficiency for highly spread-out horizontal and vertical delta motion vector distributions.
[Motion Estimation] Motion Search Level Motion search level controls whether and how color is used in motion searches. Including chroma in motion estimation can significantly improve the quality of encoded video when chroma changes happen where luma changes do not. For example, motion graphics, cel animation, and screen recordings can be significantly improved with this setting. Motion search with luma and true chroma will yield the best quality, but at the highest performance cost. The two adaptive modes and the nearest-integer chroma mode provide reasonable compromises between quality and performance. Adaptive modes apply chroma search to the 50 percent of the blocks in the frame that are predicted to have the most benefit. This provides most of the value of chroma search with only half the encoding performance reduction The default depends on the complexity level. The following table lists the possible values: Value
Description
0
Luma only. The VC-1 encoder searches for motion in luminance values only. Provides fastest performance (encoding speed).
1
Luma with nearest integer chroma. Provides a compromise between quality and performance.
2
Luma with true chroma. Provides the best quality with the lowest performance.
3
Macroblock adaptive with nearest integer chroma. Provides a compromise between quality and performance.
4
Macroblock adaptive with true chroma. Provides a compromise between quality and performance. (recommended setting)
[Motion Estimation] Motion Search Method This setting controls the size of the area the codec will search for an element of a frame that may have moved since a previous frame. Larger search ranges can better detect fast motion, but require more processing time. Processing time roughly doubles with each search range increase. Setting the search
Microsoft VC-1 Encoder Settings
Page 10 of 15
VC-1 for Digital Rapids Stream range too high can also lead to false positives, so it's important to set the motion search window to a range adequate for the video. The following table lists the possible values: Value
Description
0
SAD. Provides the best performance.
1
Hadamard. Provides the best quality.
2
Macroblock-adaptive. This option configures the codec to make decisions about which method to use on each macroblock. This can potentially reduce overall computation required for encoding by performing the computationally-intensive Hadamard transform only when appropriate. (recommended setting)
[Motion Estimation] Macro Block Mode Cost Method Used by the codec to determine which macroblock mode to use. The default is 0 when encoder complexity is less than 2. Otherwise, the default is 1. The following table lists the possible values: Value
Description
0
SAD/Hadamard. This option configures the codec to account for only distortion when computing cost.
1
RD cost. This option configures the codec to account for both rate and distortion when computing cost.
[Motion Estimation] Motion Vector Cost Method Used by the codec to estimate the amount of processing required for motion vector coding. The codec uses the cost to determine which features will be used in encoding. The following table lists the possible values: Value
Description
0
Static method. This option uses the same motion vector cost estimate for all macroblocks.
1
Dynamic method. This option varies the motion vector cost between macroblocks to achieve optimal visual quality.
Recommended: Setting 1 if DQuant option is not used. Otherwise, use setting 0.
[Motion Estimation] Motion Search Range The range used in motion searches. Specify 0 when using Simple profile. The following table lists the possible values: Value
Description
0
+63.75/-64.0 H, +31.75/-32.0 V (Default)
1
+127.75/-128.0 H, +63.75/-64.0 V
Microsoft VC-1 Encoder Settings
Page 11 of 15
VC-1 for Digital Rapids Stream
2
+511.75/-512.0 H, +127.75/-128.0 V
3
+1023.75/-1024.0 H, +255.75/-256.0 V
4
Macroblock adaptive. The VC-1 encoder selects the most efficient mode. The maximum range never exceeds ±1024.0 H, ±256.0 V.
Recommendation: setting 4, except when using Simple profile (when you must use 0)
[Advanced] Letter Box Present This feature enables the codec to determine the location of letterbox matte bars at the top and bottom of each frame. When detected, the letterboxing is excluded from motion search operations, saving some bits and speeding up encoding. Letterbox detection is dynamic and should correctly detect changes in video mattes. However, there may be cases where detection does not immediately find a change, particularly when video frames contain mostly black regions.
[Advanced] Key pop reduction Key pop, a type of compression artifact, occurs when the encoder uses different algorithms to compress key frames and predicted frames. Using different algorithms can lead to differences in fine details, texture, or film grain. After compression, key frames may appear to stand out or otherwise appear different. The patterned nature of frame compression can make this problem even worse by creating a rhythmic pulsing in the video during playback. Key pop may also be called key frame pulsing. Key pop reduction is most useful for shorter GOP encodes, like those targeted for optical discs. Key pop reduction can also cause some softness in the video, and therefore may not be appropriate for all content, particularly content that uses long GOPS where they are not needed. Recommendation: Off unless key frame popping is visible.
[Advanced] Number of threads Each instance of the VC-1 encoder can use up to four threads to distribute encoding tasks. Each thread runs on a separate processor. Each thread processes a different horizontal portion of the frame. Motion search operations only happen within each horizontal portion. Because most motion in video is typically horizontal, such as panning, motion estimation usually works well. However, if the content contains a lot of vertical motion, some slight loss of coding efficiency may occur. Valid values are 1, 2 and 4. Note that specifying higher values will not raise an error. The encoder will simply use four threads. Recommendation: One thread for pictures less than 128 lines in height, two threads for pictures between 128 and 256 lines, and four threads for larger pictures.
[Advanced] Affinity Mask The affinity mask is used with the number of threads setting to specify which processors to use. Each binary digit (bit) in the affinity mask represents an individual processor. Setting a bit to one makes the VC1 encoder use the corresponding processor to run an encoding thread. For example, to make the VC-1 encoder use processors zero, one, two, and three, the low-order byte of the affinity mask would be 00001111 (decimal value of 15 and a hexadecimal value of 0xF).
Microsoft VC-1 Encoder Settings
Page 12 of 15
VC-1 for Digital Rapids Stream If the encoding computer has 8 processors, then processors four, five, six and seven would use an affinity mask of 11110000 (decimal value of 240 and a hexadecimal value of 0xF0).
[Advanced] CPU Usage This is only supported for 1-pass CBR encoding. This setting specifies the percentage of available processing to use for encoding. The range is 1 – 100, and the default is 80. This method can dynamically vary the complexity of the encoding to ensure a fixed encoding time, regardless of the complexity of the input video stream. Setting a lower target CPU usage value implies lower complexity and, therefore, faster encoding, while making more system resources available for other components that the application controls. Setting a higher value implies potentially higher complexity and more utilization of CPU resources by the encoder itself. Recommendation: Use the highest value that leaves sufficient resources for other processes on the computer.
[Advanced] Pixel width Use this setting along with the Pixel Height setting to specify the pixel aspect ratio. The pixel aspect ratio is defined as pixel width / pixel height. For a square pixel this ratio is 1:1. The valid range for this setting is 1 to 256. For display on a computer VGA monitor or an HDTV monitor, use an aspect ratio of 1:1. For display on an NTSC monitor, for D1/DV use 10:11 and for D1/DV anamorphic widescreen use 40:33. For display on a PAL monitor, for CCIR-601 use 16:15, for D1/DV use 59:54, and for D1/DV anamorphic widescreen use 118:81. This is not to be confused with the display aspect ratio, which defines the ratio between the width and height of the video display.
[Advanced] Pixel Height Use this setting along with the Pixel Height setting to specify the pixel aspect ratio. The pixel aspect ratio is defined as pixel width / pixel height. For a square pixel this ratio is 1:1. The valid range for this setting is 1 to 256. For display on a computer VGA monitor or an HDTV monitor, use an aspect ratio of 1:1. For display on an NTSC monitor, for D1/DV use 10:11 and for D1/DV anamorphic widescreen use 40:33. For display on a PAL monitor, for CCIR-601 use 16:15, for D1/DV use 59:54, and for D1/DV anamorphic widescreen use 118:81. This is not to be confused with the display aspect ratio, which defines the ratio between the width and height of the video display.
[Advanced] Pixel Index This setting is used for HD DVD and Blu-ray disc encoding. This value is completely independent of the Pixel Width and Pixel Height settings. A value of 1 will set 1:1 aspect ratio for Blu-Ray and a value of 5 will set 4:3.
Microsoft VC-1 Encoder Settings
Page 13 of 15
VC-1 for Digital Rapids Stream
Value
Description
1
(PAR=1:1(square))
2
(PAR=12:11)
3
(PAR=10:11)
4
(PAR=16:11)
5
(PAR=40:33)
6
(PAR=24:11).
7
(PAR=20:11).
8
(PAR=32:11)
9
(PAR=80:33)
10
(PAR=18:11).
11
(PAR=15:11)
12
(PAR=64:33)
13
(PAR=160:99)
14
(PAR=SMPTE Reserved (illegal)).
15
(PAR=custom)
[Advanced] Video Type This setting is used to specify the method used to encode progressive or interlaced source video. Value
Description
0
Progressive video. (Default for progressive and for Simple or Main profiles)
1
The VC-1 encoder encodes all frames as interlaced frames. This method is suitable for content that does not contain fast motion. Best when the content contains random motion.
2
The VC-1 encoder encodes all frames as pairs of interlaced fields. This is usually the most efficient method, especially if the content contains fast motion, such as panning. (Default for interlaced)
3
The VC-1 encoder automatically determines the most efficient interlaced encoding method (frame or field). This method is suitable for content that does not contain fast motion.
4
The VC-1 encoder automatically determines the most efficient encoding method (frame, field, or progressive). This is the best method for content that contains a mixture of frame and field types.
Microsoft VC-1 Encoder Settings
Page 14 of 15
VC-1 for Digital Rapids Stream
[Advanced] Output Type RAW: The encoder generates output for a container format, such as ASF. ES: The encoder generates output for an elementary stream with an entry point start code inserted for each GOP. Sequence start codes are inserted as needed. ES_SH: The encoder generates output for an elementary stream with both entry point and sequence start codes inserted for each GOP. For BluRay choose this setting.
Microsoft VC-1 Encoder Settings
Page 15 of 15
Related Documents
Vc1 Codec
June 2020 6
Vc1-+eaie.pdf
June 2020 4
Codec Content.docx
November 2019 9
Nanda Codec
November 2019 21
Voip Codec
May 2020 0