ccesp2

User’s GuideVersion 1.10

Custom Technology Corporation2-12-4 Hirakawa-cho, Chiyoda-ku, Tokyo102-0093 Japan

e-mail: [email protected] site: http://www.cinemacraft.com/web site: http://www.ctech.co.jp/

Cinema Craft Encoder SP2 User’s GuideVersion 1.10

Edition2008 January Revised Video trimming section.2006 July First edition

All rights reserved. No part of this publication may be reproduced, electronicallyor mechanically, in any form or by any means, including copying, recording or byany information storage and retrieval system, without permission in writing fromCustom Technology Corporation.

Information contained in this publication is believed to be accurate and reliable.However, Custom Technology Corporation shall not be responsible for any in-fringement of patent rights or other rights of a third party arising out of use ofthis information.

Cinema Craft is a registered trademark of Custom Technology Corporation.

Microsoft, Windows and Windows NT are registered trademarks of MicrosoftCorporation. Adobe and Premiere are registered trademarks of Adobe SystemsIncorporated. Intel, Pentium and MMX are registered trademarks of Intel Corpo-ration. Celeron is a trademark of Intel Corporation. AMD and AMD Athlon areregistered trademarks of Advanced Micro Devices, Inc. QuickTime is a trademarkof Apple Computer, Inc. HASP is a registered trademark of Aladdin KnowledgeSystems Ltd. Company names and product names contained in this publicationare trademarks or registered trademarks of respective companies. The symbolsTMand R©are not indicated in this publication.

The specifications on this software and information contained in this manual maybe changed without prior notice.

Contents

1 Getting started 51.1 How to run Cinema Craft Encoder SP2 . . . . . . . . 51.2 Main window . . . . . . . . . . . . . . . . . . . . . . . 71.3 Command line options . . . . . . . . . . . . . . . . . . 121.4 Precautions on input files . . . . . . . . . . . . . . . . 13

2 Input file setup 152.1 Combining movie files for one MPEG output . . . . . 162.2 Select external audio file . . . . . . . . . . . . . . . . . 172.3 Video trimming . . . . . . . . . . . . . . . . . . . . . . 17

3 Basic encoder setting 213.1 Title . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213.2 Timecode . . . . . . . . . . . . . . . . . . . . . . . . . 213.3 Output files . . . . . . . . . . . . . . . . . . . . . . . . 23

3.3.1 Video file . . . . . . . . . . . . . . . . . . . . . 243.3.2 Video information file . . . . . . . . . . . . . . 243.3.3 Pulldown information file . . . . . . . . . . . . 243.3.4 Audio file . . . . . . . . . . . . . . . . . . . . . 25

3.4 Audio setting . . . . . . . . . . . . . . . . . . . . . . . 253.4.1 Bitrate . . . . . . . . . . . . . . . . . . . . . . . 253.4.2 Add CRC . . . . . . . . . . . . . . . . . . . . . 253.4.3 File size . . . . . . . . . . . . . . . . . . . . . . 253.4.4 Mode . . . . . . . . . . . . . . . . . . . . . . . 253.4.5 Sampling frequency . . . . . . . . . . . . . . . 263.4.6 Audio run out . . . . . . . . . . . . . . . . . . 26

i

ii CONTENTS

3.5 Encoding mode setting . . . . . . . . . . . . . . . . . . 263.5.1 Output type . . . . . . . . . . . . . . . . . . . 273.5.2 Encoding method . . . . . . . . . . . . . . . . . 293.5.3 Output stream type . . . . . . . . . . . . . . . 303.5.4 V/C (VBR/CBR) : Fluctuation of the bit allo-

cation . . . . . . . . . . . . . . . . . . . . . . . 313.5.5 Pass . . . . . . . . . . . . . . . . . . . . . . . . 31

3.6 Video information file . . . . . . . . . . . . . . . . . . 313.6.1 Creating video information file . . . . . . . . . 323.6.2 Renewing the video information . . . . . . . . 333.6.3 VBR bit allocation . . . . . . . . . . . . . . . . 34

3.7 Frame size . . . . . . . . . . . . . . . . . . . . . . . . . 343.7.1 Area setting . . . . . . . . . . . . . . . . . . . . 34

3.8 Frame rate . . . . . . . . . . . . . . . . . . . . . . . . 383.9 File size . . . . . . . . . . . . . . . . . . . . . . . . . . 383.10 Bitrate and Q.factor . . . . . . . . . . . . . . . . . . . 393.11 Pulldown setting . . . . . . . . . . . . . . . . . . . . . 40

3.11.1 Pulldown . . . . . . . . . . . . . . . . . . . . . 413.11.2 Inverse 3:2 pulldown . . . . . . . . . . . . . . . 423.11.3 Auto 2-pass detection . . . . . . . . . . . . . . 44

3.12 Aspect ratio . . . . . . . . . . . . . . . . . . . . . . . . 443.13 Deinterlacing . . . . . . . . . . . . . . . . . . . . . . . 44

4 Advanced video setting 454.1 Setting GOP configuration . . . . . . . . . . . . . . . . 45

4.1.1 Picture type . . . . . . . . . . . . . . . . . . . . 454.1.2 GOP sequence . . . . . . . . . . . . . . . . . . 474.1.3 GOP header . . . . . . . . . . . . . . . . . . . 484.1.4 SEQ header . . . . . . . . . . . . . . . . . . . . 48

4.2 Add sequence end code . . . . . . . . . . . . . . . . . 484.3 Encode the last frame as an I frame . . . . . . . . . . 484.4 DVD Video multiangle . . . . . . . . . . . . . . . . . . 484.5 Close all GOPs . . . . . . . . . . . . . . . . . . . . . . 494.6 Disable auto I frame insertion . . . . . . . . . . . . . . 494.7 Equalize each GOP’s bit length . . . . . . . . . . . . . 504.8 Disable adaptive Q-matrix switching . . . . . . . . . . 504.9 Panscan . . . . . . . . . . . . . . . . . . . . . . . . . . 504.10 Output top field first stream . . . . . . . . . . . . . . . 51

CONTENTS iii

4.11 Offset line . . . . . . . . . . . . . . . . . . . . . . . . . 514.12 Luminance level . . . . . . . . . . . . . . . . . . . . . . 514.13 Quantization scale type . . . . . . . . . . . . . . . . . 524.14 Quantization matrices setting . . . . . . . . . . . . . . 534.15 Packet size . . . . . . . . . . . . . . . . . . . . . . . . 55

5 VBR bit allocation 575.1 Bitrate / Quantization scale graph . . . . . . . . . . . 58

5.1.1 Bitrate graph . . . . . . . . . . . . . . . . . . . 585.1.2 Quantization scale graph . . . . . . . . . . . . 595.1.3 Operation of graph . . . . . . . . . . . . . . . . 60

5.2 Bitrate settings . . . . . . . . . . . . . . . . . . . . . . 635.2.1 Global bitrate settings . . . . . . . . . . . . . . 635.2.2 Local bitrate settings . . . . . . . . . . . . . . 645.2.3 Keep the previous bit allocation . . . . . . . . 655.2.4 Apply the changes . . . . . . . . . . . . . . . . 655.2.5 Cancel the changes . . . . . . . . . . . . . . . . 655.2.6 Update the graph . . . . . . . . . . . . . . . . 66

5.3 GOP information . . . . . . . . . . . . . . . . . . . . . 66

6 Picture settings 696.1 Entry list . . . . . . . . . . . . . . . . . . . . . . . . . 69

6.1.1 Adding an entry to the list . . . . . . . . . . . 696.1.2 Operation buttons for the list . . . . . . . . . . 726.1.3 Saving/Loading the entry list . . . . . . . . . . 72

6.2 Simple setting . . . . . . . . . . . . . . . . . . . . . . . 726.2.1 Material types . . . . . . . . . . . . . . . . . . 726.2.2 Complexity of material . . . . . . . . . . . . . 736.2.3 Parameter sets of Simple setting . . . . . . . . 74

6.3 Detailed setting . . . . . . . . . . . . . . . . . . . . . . 746.3.1 Caption . . . . . . . . . . . . . . . . . . . . . . 756.3.2 Low and High filter . . . . . . . . . . . . . . . 756.3.3 Vertical filter . . . . . . . . . . . . . . . . . . . 776.3.4 Dithered quantization . . . . . . . . . . . . . . 776.3.5 Quantizer characteristics . . . . . . . . . . . . . 786.3.6 Letterbox . . . . . . . . . . . . . . . . . . . . . 786.3.7 Intrablock DC precision . . . . . . . . . . . . . 796.3.8 Block scan order . . . . . . . . . . . . . . . . . 79

iv CONTENTS

6.3.9 Fades on a static scene . . . . . . . . . . . . . . 796.3.10 Black screen . . . . . . . . . . . . . . . . . . . . 806.3.11 Progressive frame . . . . . . . . . . . . . . . . . 806.3.12 Monochrome . . . . . . . . . . . . . . . . . . . 806.3.13 Save/Load a parameter set . . . . . . . . . . . 80

7 Inverse 3:2 pulldown setting 817.1 Overview of the setting . . . . . . . . . . . . . . . . . 81

7.1.1 How to find the pulldown phase . . . . . . . . . 817.1.2 Pulldown list . . . . . . . . . . . . . . . . . . . 83

7.2 Modifying the pulldown list . . . . . . . . . . . . . . . 837.2.1 Explanation of the pulldown list . . . . . . . . 837.2.2 Operation buttons for the list . . . . . . . . . . 857.2.3 Viewer windows . . . . . . . . . . . . . . . . . 867.2.4 Current point and its phase . . . . . . . . . . . 86

7.3 Phase calculation . . . . . . . . . . . . . . . . . . . . . 897.4 Threshould setting . . . . . . . . . . . . . . . . . . . . 90

7.4.1 Pulldown information file . . . . . . . . . . . . 907.4.2 Progressive frame pairing . . . . . . . . . . . . 917.4.3 Repeat field tolerance . . . . . . . . . . . . . . 917.4.4 Initial setting . . . . . . . . . . . . . . . . . . . 927.4.5 Apply . . . . . . . . . . . . . . . . . . . . . . . 927.4.6 Detail . . . . . . . . . . . . . . . . . . . . . . . 92

7.5 Saving/Loading the pulldown list . . . . . . . . . . . . 927.6 Effects of chapter points . . . . . . . . . . . . . . . . . 93

7.6.1 Improper frame for a chapter point . . . . . . . 93

8 Chapter setting 978.1 What is “Chapter” ? . . . . . . . . . . . . . . . . . . . 988.2 Creating chapter list . . . . . . . . . . . . . . . . . . . 988.3 Creating chapter list file . . . . . . . . . . . . . . . . . 1008.4 Loading and saving chapter list . . . . . . . . . . . . . 100

9 Optional settings 1019.1 Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

9.1.1 Reading method for QuickTime movies . . . . 1019.2 Output . . . . . . . . . . . . . . . . . . . . . . . . . . 1029.3 Setting extensions . . . . . . . . . . . . . . . . . . . . 103

CONTENTS v

9.4 Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . 104

10 Template settings 10710.1 Applying template . . . . . . . . . . . . . . . . . . . . 10810.2 Operation at template setting window . . . . . . . . . 10810.3 Operation at encoder setting window . . . . . . . . . . 11010.4 Editing template . . . . . . . . . . . . . . . . . . . . . 11210.5 Deleting template . . . . . . . . . . . . . . . . . . . . . 11210.6 Editing template list . . . . . . . . . . . . . . . . . . . 112

11 Encoder control list 11511.1 Creating and saving encoder control list . . . . . . . . 11511.2 Loading encoder control list . . . . . . . . . . . . . . . 11611.3 Editing encoder control list . . . . . . . . . . . . . . . 116

11.3.1 Editing ECL . . . . . . . . . . . . . . . . . . . 11611.4 Encode . . . . . . . . . . . . . . . . . . . . . . . . . . 118

12 Adobe Premiere plug-in 11912.1 The basic operation . . . . . . . . . . . . . . . . . . . 11912.2 Encoder settings on plug-in . . . . . . . . . . . . . . . 12612.3 Encoder settings for plug-in . . . . . . . . . . . . . . . 127

13 Removing Cinema Craft Encoder SP2 129

14 Troubleshooting 13114.1 VBR bitrate . . . . . . . . . . . . . . . . . . . . . . . . 13114.2 VBV buffer overflow . . . . . . . . . . . . . . . . . . . 13214.3 Mux buffer overflow . . . . . . . . . . . . . . . . . . . 132

vi CONTENTS

MPEGLA Statement:

Use of this product for encoding video information for the purpose of pro-ducing prerecorded video programs products for commercial sale or rentalincluding by way of examples and without limitation, digital video disk anddigital video tapes, or for the purpose of storing encoded video programsfor distribution by a video server is expressly prohibited without a licenseunder applicable patents marked on this product, or on the container, userdocumentation or specification sheet for this product.

Introduction

Cinema Craft Encoder SP2 is a program for converting movie filesto an MPEG-2 (or MPEG-1) files at high-speed with high picturequality. This manual explains the functions and manner of use ofCinema Craft Encoder SP.

For details on the basic operation and the terminology of the op-erating system and on the detailed technology of MPEG, refer to therespective manuals and guides.

Operation environment

The following hardware and software are required to use Cinema CraftEncoder SP2.

• A computer with a processor that supports SSE or Enhanced3D Now !

• 512MB of RAM• Microsoft Windows Vista or Windows XP• QuickTime 7 for encoding QuickTime files• Adobe Premiere Pro/Pro 1.5/Pro 2.0/Pro CS3 to use plug-in

version

1

2 CONTENTS

Specifications

Compression method

System ISO/IEC 13818-1 | ITU-T Rec. H.222.0(Program Stream only)ISO/IEC 11172-1

Video ISO/IEC 13818-2 | ITU-T Rec. H.262 (MP@ML)ISO/IEC 11172-2

Audio ISO/IEC 11172-3 (Layer 2)

Output video format

Encoding method Constant Bitrate (CBR) (One pass)Constant Bitrate (CBR) (Multipass)Variable Bitrate (VBR) (One pass)Variable Bitrate (VBR) (Multipass)

Bitrate maximum 15 Mbits/secFrame rate 23.976/24/25/29.97/30 (frames/sec)Aspect ratio SAR 1:1, DAR 4:3, DAR 16:9 or DAR 2.21:1GOP configuration I, P frame interval (M) : 1 ∼ 3

I frame interval (N) : max. 15 (a multiple of M)

Output audio format

Bitrate 1 64/96/112/128/160/192/224/256/320/384 (kbits/sec)

Sampling frequency 32/44.1/48 (kHz)Quantization bits 16 bitsMode Stereo, Joint Stereo, Dual channel and Monaural

Input file format

The following descriptions are for the standalone version.

1In the case of one channel, the bitrate is half of each.

CONTENTS 3

Video

AVI files AVI 1.0 (including reference file)AVI 2.0OpenDMLframeserver (Avisynth, VirtualDub)

QuickTime files mov filesDV (dif) files

Still-image (sequence) Bitmap filesTarga filesTiff filesPNG filesJPEG files

Audio

Wave files uncompressed ones onlyAIFF files uncompressed ones only

Chapter 1

Getting started

1.1 How to run Cinema Craft EncoderSP2

This section focuses on standalone version. See page 119 for AdobePremiere plug-in version.

Figure 1.1: Cinema Craft Encoder SP2 Startup Screen

The basic steps for encoding files with Cinema Craft Encoder SP2

5

6 CHAPTER 1. GETTING STARTED

are as follows (explanations for each parameter are described in thefollowing chapter).

� Start Cinema Craft Encoder SP2. To start the program, do oneof the following:

• Double-click the icon of Cinema Craft Encoder SP2.

• From Programs of the Start menu, select Cinema CraftEncoder SP2.

• Start Cinema Craft Encoder SP2 from command prompt.See 1.3 Command line options on page 12.

� Select files you would like to encode (AVI, QuickTime, etc) andload them in the main window of Cinema Craft Encoder SP2.To load the files, do one of the following:

• Select a file (or files) from Explorer, and drag it (them)onto the main window of Cinema Craft Encoder SP2.

• Select Add in the Edit menu or shortcut menu.

• Click the button in the toolbar.

Figure 1.2: Shortcut menu

� Open Encoder setting window and set the parameters forencoding. To open Encoder setting window, do one of thefollowing:

• Double-click a file in the list of the main window.

1.2. MAIN WINDOW 7

• Drag a file onto the main window while holding down theSHIFT key. Then the file is loaded in the list and Encodersetting window appears at once.

• Choose a file in the list of the main window, and selectSetting in the Edit menu or shortcut menu.

• Click the button in the toolbar.

� Start encoding. To start encoding, do one of the following:

• Click Encode now button on the Encoder setting win-dow. Only the file currently setting parameters will beencoded.

• Select a file (or files) in the list of the main window, andthen select Encode Selected Items in the Edit menu orshortcut menu. Only the file(s) selected in the list will beencoded.

• Select Encode in the File menu, or click button in thetoolbar, or press F5 key. Then all of the files loaded in themain window will be encoded.

When encoding starts, the window as shown in Fig. 1.3 ap-pears. Also, an icon as shown in Fig. 1.4 is displayed in thenotification area. If you click this icon, the screen of CinemaCraft Encoder SP2 can be erased from a desktop. Click thisicon again to re-display the screen of Cinema Craft EncoderSP2. If a mouse cursor is carried on this icon during encoding,a progress percentage is shown.

� When encoding ends, the window closes.

� To end Cinema Craft Encoder SP2, click the close button atthe upper right of the main window, or select Exit in the Filemenu.

1.2 Main window

When you start Cinema Craft Encoder SP2, the main window (Fig. 1.5)will appear. On the main window, the list called Encoder Control List


Figure 1.3: Window during encoding

Figure 1.4: Icon in the notification area

Figure 1.5: Main window

1.2. MAIN WINDOW 9

is displayed. An entry in the list holds parameter settings for one out-put. You can encode the entries one by one, or batch-encode all ofthe entries in the list. (See also how to start encoding in Section 1.1described above.)

The columns of the list show the information listed below. Andyou can change the order of the columns by dragging one.

Title Shows the name of the entry. Initially,the source file name is displayed.

Files Shows how many files are input for an output.Frames Shows how many frames are to be encoded.Duration Shows the duration to be encoded.Video Encoding mode and bitrateAudio Encoding mode and bitrate (for MPEG Audio)

� To see more detailed information, select Property in the short-cut menu (Fig. 1.2).

� About The operations of Encoder Control List, see Chapter 11(page 115).

� The order of columns of the list are changeable by dragging acolumn.

File menu

Figure 1.6: File menu

File menu is used to open and save the project file called an ECLfile, to start encoding, and to close the application.


Edit menu

Figure 1.7: Edit menu

Edit menu is the same as shortcut menu. Please read Chapter 11(page 115) about operation of the encoder control list that uses thismenu.

Setting opens the Encoder setting window. Append opens thesubwindow used when combining two or more input files. Propertyopens a window that display the detailed information on input andoutput files.

View menu

Figure 1.8: View menu

Selection of Toolbar displays a tool bar on the upper part of anencoder control list. Selection of Status Bar displays a status bar onthe lower part of the list. The status bar displays currently selectedtemplate name.

1.2. MAIN WINDOW 11

Selection of Grid shows the grid on an encoder control list, a chap-ter list, a pull down list, and the entry list of the picture settingswindow.

In Duration menu, the display unit of Duration of an encodercontrol list can be changed into Milliseconds or a Timecode.

Option menu

Figure 1.9: Option menu

Option menu is used to setup a setting about reading of an inputfile, an output folder, extensions, and others settings. For details,please see Chapter 9 (page 101).

Template menu

Figure 1.10: Template menu

A click of Setting will display a template selection screen. Fordetails, please see Chapter 10 (page 107).

Help menu

Figure 1.11: Help menu

The version number of a program can be checked.


1.3 Command line options

You can also start Cinema Craft Encoder SP2 by typing cct2.exeor cct2.com from command prompt, and the following command lineoptions are provided. (cct2.com outputs the encoding informationon the console window.)

-auto Starts a program and encoding immediately.-batch Starts a program and encoding immediately

and quits a program when finished.-ecl filename Loads specified ecl file1.-t template name Specifies a template.filename Specifies an input file.

USAGE:

Case 1 It starts a program and encoding f:\a.avi and f:\b.avi 2

with a template “NTSC DVD”3.

cct2 -t "NTSC DVD" f:\a.avi f:\b.avi -auto

Case 2 It starts a program and encoding with the direction of f:\c.ecl 4,and quits a program when finished.

cct2 -ecl f:\c.ecl -batch

Case 3 It outputs verbose log to log.txt in addition to the act ofcase 2.

cct2.exe -ecl f:\c.ecl -batch -verbose > log.txt

1About an ECL (Encoder Control List) file, see page 115.2If an input movie is located in a current folder, the full path description is

not necessary.3Templates should be created in advance, see page 107.4ECL file should be created in advance.

1.4. PRECAUTIONS ON INPUT FILES 13

1.4 Precautions on input files

Note that Cinema Craft Encoder SP2 cannot read AVI or Quick-Time files without appropriate codec5 for them. Also, QuickTime 7is required to read a QuickTime movie.

5Codec is an abbreviation for Coder-Decoder.

Chapter 2

Input file setup

An input file setting is needed when one of the following applies.

• Combine two or more files and output one MPEG file.

• Input an audio from another file.

• Trim the source video.

To perform input file setting, select the target item in the list ofthe main window, and then select Append in the shortcut menu toshow the subwindow displayed in Fig. 2.1.

Figure 2.1: Sub window to add extra files

15

16 CHAPTER 2. INPUT FILE SETUP

2.1 Combining movie files for one MPEG

output

Select input files in the same way as operation in a main window.When all the files are registered into a subwindow, click OK but-ton and return to a main window. You can see the number of filesregistered at the Files column of the list of main window.

In Fig. 2.2, four files are registered into the subwindow. Whenreturned to the main window, the value of the column of Files is 4.

Figure 2.2: Four files are added in the subwindow

Figure 2.3: After returned to main window

You can combine input files only when frame size andframe rate are the same.

2.2. SELECT EXTERNAL AUDIO FILE 17

2.2 Select external audio file

Video and an audio can be inputted from separate files. The inputaudio file should be either uncompressed Wave file or uncompressedAIFF file.

To specify an external audio file, click a target item in the list ofsubwindow, and then select Audio in the shortcut menu (Fig. 2.4)to show Audio window. In the Audio window, select Use externalaudio file, and then specify an audio file name. You can also selectan audio file from common dialog box by clicking right-hand side ofthe text box.

Figure 2.4: Shortcut menu of subwindow

Figure 2.5: Audio input window

To synchronize audio with video, select Sync with video checkbox.

When a synchronization is chosen as video, an in point is specifiedand it is a way of a video file. When this option selected, video andaudio will be in sync even when video is trimmed.

2.3 Video trimming

For each of the specified input file, video trimming can be performedby frame number or timecode.


To trim video, open Video trimming window shown in Fig. 2.6by double-clicking an item in the list of subwindow, or select an itemin the list of subwindow and select Trimming in the shortcut menu.

Figure 2.6: Video trimming window

When you open the Video trimming window, the viewer windowalso appears to help search points. On the title bar of the viewerwindow shown is the frame number. The frame number 0 indicatesthe first frame. Use the upper slider or directly change the framenumber or timecode of Current fields to find a trim point, and thenclick In or Out to set the point.

Frame numbers for input file and output file are displayed sepa-rately. Next to the frame number for input file, time and duration ofthe frame are displayed. While each frame duration is the same forAVI file, QuickTime file may contain frames with different durations.

Note that In point is inclusive and Out point is exclusive. Forexample, if you set 0 as In point and 300 as Out point, CinemaCraft Encoder SP2 encodes 0th frame through 299th frame.

2.3. VIDEO TRIMMING 19

� Using the slider, you can only move to the last frame. If youset that frame as an Out point, the last frame will be missing.So please make sure to specify the (last + 1) frame as an Outpoint in this case.

Chapter setting

In the Video trimming window, you can also specify chapter points.The pictures at chapter points are encoded as I pictures.

Movie playback setting

To enable movie playback on Video trimming window, select En-able movie play if possible in the Option menu of subwindow(Fig. 2.7).

Figure 2.7: Option menu of subwindow

� Cinema Craft Encoder SP2 does not playback still-image se-quence in the current version (1.00.00.01).

Operation buttons

Use the following buttons to operate movie play.


(QuickTime only) Play backwards.Play forwards.Step backward. Shows the previous frame.Step forwards. Shows the next frame.(QuickTime only) Double speed play backwards.Double speed play forwards.Stop playback.Return to top.Go to end.

Changing playback speed

During movie playback, you can change the playback speed by movingthe Speed slider. Move the slider to the left to slow down, and tothe right to speed up. You can specify the top speed by changingMax speed value. The mimimum speed will be inverse number ofthe maximum speed.

Chapter 3

Basic encoder setting

This chapter explains the parameteres of Cinema Craft Encoder SP2.To set parameters, open Encoder setting window. To open thewindow, please refer to section 1.1 (page 5).

3.1 Title

Specify a title that is shown in the list of the main window. Appro-priate title will make the Encoder Control List (the list of the mainwindow) easier to follow. This setting does not affect encoding.

3.2 Timecode

Since a timecode is not generally included in both AVI and QuickTimefiles, Cinema Craft Encoder SP2 generates a timecode internally andembeds it in MPEG stream.

Timecode should be tied to a frame number at Timecode settingin the Encoder setting window shown in Fig. 3.1.

Initially, the first frame (frame number 0) is tied to the timecode01:00:00:00. Although you can change this setting, you should avoidusing the timecode 00:00:00:00 since this value has special meaning toCinema Craft Encoder SP2 especially when you perform Pulldownor Inverse 3:2 Pulldown.

21

22 CHAPTER 3. BASIC ENCODER SETTING

Figure 3.1: Encoder setting window

3.3. OUTPUT FILES 23

If you need a dropped frame timecode, select DF check box.

3.3 Output files

Specify output files and their types here at Output files field. Thereare 4 types of output files listed below.

• Video file• Video information file• Pulldown information file• Audio file

To set or change the name of output files, click ... button to showa dialog box and specify the file name there, or directly type the filename in the text box.

Figure 3.2: Output files setting

� You can specify the default output folder. How to change thedefault folder for output files is descibed at 9.2 Specifying afolder for output files on page 101.

� You can change the default extensions of the output files. Howto change an extension is described at 9.3 Setting extensionson page 103. However, concerning audio files, only the extensionof MPEG-1 audio can be changed.


3.3.1 Video file

Specify an output video file name. When Multipass VBR is se-lected in encoding mode, you can save the video output into two filesalternately. One is for an odd number of times of encoding (1,3,5,...passes), another is for an even number of times (2,4,6,...). If youspecify different names for two video files, you can at least keep theencoding result of the previous pass even if you cancel encoding. Thefile name for 2,4,6,...passes should be specified in the text box of(Second Pass).

3.3.2 Video information file

Specify a video information file name. Video information file (VAFfile) is used when multipass encoding and VBR bit allocation setting.

This file will be automatically created when multipass encoding, oryou can create it before VBR bit allocation setting. If you need thevideo information file in advance, select 1-pass CBR or 1st passof VBR (1-pass VBR also will do.) in encoding mode, and setthe parameters as similarly as possible to ones for multipass encodingthat will be done later.

� See page 32 Create the video information file for moredetail.

3.3.3 Pulldown information file

Specify a pulldown information file name. Pulldown information file(i32 file) holds the information for inverse 3:2 pulldown.

It will be used when you carry out Auto 2-pass detection (seepage 44 Auto 2-pass detection). You also need this file to applythe changes of Threshold setting of Inverse 3:2 pulldown setting(see page 42 Inverse 3:2 pulldown).

The pulldown information file (i32 file) will be created automati-cally when you carry out encoding with Inverse 3:2 pulldown op-tion.

3.4. AUDIO SETTING 25

3.3.4 Audio file

Specify an output audio file name. You can select either MPEG-1(Layer2) audio, Wave, or AIFF file as the output audio format. Ifyou select Wave or AIFF file, the output audio stream will not becompressed.

3.4 Audio setting

3.4.1 Bitrate

Set the bitrate for an MPEG Audio stream. The initial value is 256kilobits/sec. (half for monaural).

3.4.2 Add CRC

Select whether to add CRC data to an MPEG Audio stream. WhenSVCD or VCD is selected in encoding mode, you cannot change thissetting.

3.4.3 File size

The estimated file size for the output audio is shown here. When yououtput System Stream, please look at the File size of Video setting(see also page 38 3.9 File size).

3.4.4 Mode

Select the channel mode for an MPEG Audio stream.

• Monaural

The left and right channels are mixed and encoded as monaural.

• Joint stereo

Encoded in joint stereo format. Joint stereo is a special stereoformat for MPEG Audio. In this mode, the differential informa-tion on sound volume between the left and right is encoded as abalance signal. Encoding quality is slightly better than stereo.


• Stereo

Encoded in stereo.

• Dual channel

Encoded in dual channel mode.

3.4.5 Sampling frequency

Select a sampling frequency for the output audio stream. Since itaffects audio quality, it is recommended not to change it from originalrate except when necessary.

3.4.6 Audio run out

The length of output audio stream can be equal to or longer than thatof video stream. If the value of Audio run out is positive, audiostream will be longer by the specified frame length.

This setting is intended to be used to output Wave or AIFF audiofile that is to be fed to Dolby Digital Encoder, which may requireextra audio to fill its access unit.

3.5 Encoding mode setting

The encoding mode is set by selecting option buttons located in upperright of Encoder setting window. Each mode is described below.

Figure 3.3: Video encoding mode

3.5. ENCODING MODE SETTING 27

3.5.1 Output type

Select one from the output types listed below.

• MPEG-2• MPEG-2 for DVD

• MPEG-1• SVCD

• VCD

MPEG-2

Outputs MPEG-2 video stream. Image quality is better than that ofMPEG-1, but MPEG-2 decoder or software DVD player is needed toplay it back. It is especially appropriate for interlaced sources.

MPEG-2 for DVD

Outputs DVD compatible stream. If the resolution of the source file(frame size and frame rate) does not match with DVD standard, itwill be automatically converted.

When used as Adobe Premiere plug-in, make sure to specify audiosampling frequency to 48,000 Hz on Adobe Premiere side.

� If output frame is greater than input frame, picture will belocated at top left corner and extra area will be black-filled.Centering or resizing can be done on Area setting window.

� In DVD mode, when you select VBR, 9,800 kbps is alwayswritten as maximum bitrate in the sequence header of outputMPEG files whatever value you may set. It is for more flexiblebit allocation to achieve higher quality. However, some author-ing softwares seem to calculate the bitrate for a DVD using this9,800 kbps as video bitrate. And when multiplexing with PCMaudio, they consider the total bitrate out of DVD acceptablerange 10.08 Mbps. If your authoring software has this problem,please select MPEG-2 mode. Then the maximum bitrate inthe sequence header will be the value you specified. Note that


image

Figure 3.4: Frame size extension and picture location

actual bitrate does not exceed your maximum bitrate settingeven if 9,800 kbps is written in the sequence header.

MPEG-1

Outputs MPEG-1 video stream. Windows media player can playback MPEG-1 files without additional DLLs, but image quality isnot as good as that of MPEG-2. MPEG-1 does not support interlaceencoding.

SVCD

Outputs Super Video CD compatible stream. If the resolution of thesource file does not match with SVCD standard, it will be automati-cally converted.

When used as Adobe Premiere plug-in, make sure to specify audiosampling frequency to 44,100 Hz on Adobe Premiere side.

� If output frame is greater than input frame, picture will belocated at top left corner and extra area will be black-filled.Centering or resizing can be done on Area setting window.(See Fig. 3.4.)

VCD

Outputs Video CD compatible stream. If the resolution of the sourcefile does not match with VCD standard, it will be automatically con-

3.5. ENCODING MODE SETTING 29

verted.When used as Adobe Premiere plug-in, make sure to specify audio

sampling frequency to 44,100 Hz on Adobe Premiere side.

� If output frame is greater than input frame, picture will belocated at top left corner and extra area will be black-filled.Centering or resizing can be done on Area setting window.(See Fig. 3.4.)

3.5.2 Encoding method

Select one from the encoding methods listed below.

• 1-pass VBR• 1-pass CBR

• Multipass VBR• Multipass CBR

• 1st pass of VBR

1-pass VBR

Outputs VBR (variable bitrate) stream encoded in a single pass. Thismode allows you to specify Q.factor and minimum and maximumbitrates, however, average bitrate cannot be specified. The bitrate ofeach GOP will be varied, and quantization scale will be kept constant.However, certain movements of quantization scale may occur to keepthe bitrate limits.

File size in 1-pass VBR mode shows you possible largest size.

� About Q.factor, see 3.10 Bitrate and Q.factor describedlater.

1-pass CBR

Outputs CBR (constant bitrate) stream encoded in a single pass. Thebitrate will be kept constant throughout the stream.


Multipass VBR

Outputs VBR (variable bitrate) stream encoded in multiple passes.The bitrate will be varied to keep constant quality of the image .It can be selected only when you select MPEG-2 or MPEG-2 forDVD.

� See also 3.6 Video information file on page 31.

Multipass CBR

Outputs CBR stream encoded in multiple passes. It is provided tocreate the stream for multiangle DVD.

� See also 3.6 Video information file on page 31.

1st pass of VBR

It is provided to create the video infomation file for Multipass VBR.This mode allows you to specify minimum and average and maximumbitrates.

3.5.3 Output stream type

Select between the two output stream types listed below.

• Elementary Stream (ES)• System Stream (SS)

Elementary Stream (ES)

Outputs a video file and an audio file separately without multiplexing.It cannot be selected when you select SVCD or VCD.

System Stream (SS)

Outputs a single file with multiplexed video and audio. If MPEG-2is selected, the output is a program stream. It cannot be selectedwhen you select one of VBR modes.

3.6. VIDEO INFORMATION FILE 31

3.5.4 V/C (VBR/CBR) : Fluctuation of the bitallocation

In VBR streams, Cinema Craft Encoder SP2 performs bit alloca-tion balancing the quality of image on its own valuation basis. Thevaluation basis can be changed by V/C value between 0 and 100.The initial value is 30. As V/C becomes smaller, a stream becomesmore like VBR keeping the quality of the image with heavily up anddown in bitrate. As V/C becomes larger, a stream becomes even likeCBR keeping the bitrate constant but without stability of the imagequality.

� It is not applied to CBR stream.

� When the distortion of complicated part is more noticeable thanthat of flat part, smaller V/C value may improve image qualityof complicated part. On the other hand, when the distortionof flat part is more noticeable, larger V/C value may bring thebetter result. Although you can also adjust the setting for acertain part of the footage one by one at Local bitrate settingin VBR bit allocation (see page 57), V/C parameter canchange the bit allocation through the footage more effectivelyat a time.

3.5.5 Pass

Specify a number of encoding passes. You can set the number up to99.

� See also page 32 Create a video inforamtion file.

3.6 Video information file

The video information file (VAF file) is necessary for multipass en-coding. The information in this file is used for the calculation de-scribed below. And when you open VBR bit allocation window(see page 57), the video information file should be created in advance.


Multipass VBR

Generally at least two passes are required to create a VBR streamfit the average bitrate to the specified one. On the first pass, thecomplexity of images is scanned and recorded to the video informationfile. Then on the next pass, encoding will be carried out calculatingthe bit allocation for each frame using the video information gatheredduring the first pass. In multipass encoding, this information will bemodified again and again to achieve the higher quality of image.

Multipass CBR

In CBR mode, the more passes will result more constant and stablerbit allocation for GOPs, and also the bit allocation for each picturein a GOP will have been modified to achieve the higher quality ofimage.

3.6.1 Creating video information file

When you carry out encoding without existing video information file,a single pass for creating the file will be added first automatically. Sothe number of passes actually done will be one more than specified.When creating the file automatically, 1-pass CBR will be done forMultipass CBR, and 1st pass of VBR for Multipass VBR.

When you carry out encoding with Create new option, an existingvideo information file will be discarded and new one will be createdin the same way.

� If you do not recreate a video information file after changingsome parameters and try to encode using the existing videoinformation file, you may encounter an error. (For more detail,see Renewing the video information described below.) Toavoid the error, delete the exising video information file or selectCreate new when encoding.

When you create an video information file before multipass encod-ing, select 1-pass CBR or 1st pass of VBR (1-pass VBR alsowill do.) in encoding mode, and set the parameters as similarly aspossible to ones for multipass encoding that will be done later.

3.6. VIDEO INFORMATION FILE 33

� VBR bit allocation (see page 57) is the only setting thatrequires you to create a video inforamtion file in advance.

3.6.2 Renewing the video information

Parameters recorded in a video information file are ones that have agreat impact on encoding. And if the information is different from theactual parameter setting, it cannot fulfill its function. So a videoinformation file must be renewed after changing the param-eters listed below. To recreate a video information file, delete theexising video information file or select Create new when encoding.

input files to be combined mapping between a frame and a timecode In points and Out points chapter points frame size and frame rate pulldown setting inverse 3:2 pulldown setting letterbox (on picture settings) intrablock DC precision (on picture settings) block scan order (on picture settings) progressive frame (on picture settings) black screen / monochrome (on picture settings) GOP configuration (N, M, GOP header interval, sequence header

interval) quantization scale type quantization matrices setting

In addition to the list above, a video information file should berecreated when you made a substantial change in bitrate setting. Thechange of bitrate setting may not cause an error, however a renewedinformation file will bring the better results in limited number of en-coding passes. When an average bitrate is increased twice or more or


decreased less than half, it is better to recreate the video informationfile.

3.6.3 VBR bit allocation

In Multipass VBR mode, you can specify average, minimum andmaxmum bitrates throughout the footage. And moreover, you canspecify different bitrate in a certain part of the footage using VBRbit allocation window. To open the window, you need a video in-formation file created in advance. About the setting on VBR bitallocation window, see page 57.

3.7 Frame size

Specify a frame size of output video file. It must be a multiple of 16for both width and height. When the value is not a multiple of 16, itwill be changed automatically.

When MPEG-2 for DVD is selected, the width must be either720 or 704. If output frame is greater than input frame, picturewill be located at top left corner and extra area will be black-filled.Centering or resizing can be done on Area setting window.

3.7.1 Area setting

By specifying the region of interest for both input and output frames,you can do cropping, blanking, zooming, and locating settings. Also,you can select interpolation method and mirroring axis.

Setting information

In this field is displayed current area setting information.Cropping Shows cropping area of input frameBlanking Shows blanking area of output frameZooming Shows the rate of enlargement of output frameOffset line Offset line can be changed here.

(For details, see Offset line on page 51)Field order Indicates whether field order is reversed or not

3.7. FRAME SIZE 35

Figure 3.5: Area setting window


� The sum of the value of top cropping and offset line is the actuallines that cut out from upper part.

� Please make sure to select Output top field first stream inVideo setting window if and only if the field order is reversed.(See Output top field first stream on page 51.)

Source file

trimmingarea

(x,y)

h

w

Figure 3.6: ROI of source file

Frame size Displays frame size of source fileRegion of interest(x,y) Enter top left point of source ROI(w,h) Enter width and height of source ROI

The picture of input file is cut out from (x,y) with size (w,h).

� If cropping is not necessary, set (x,y) = (0,0) and (w,h) beingoriginal frame size.

Destination file

Frame size Displays frame size of output fileRegion of interest(x,y) Enter top left point of destination ROI(w,h) Enter width and height of destination ROI

3.7. FRAME SIZE 37

The picture of specified size (w,h) is displayed in the specifiedposition (x,y). Press Move to center button to move picture tocenter.

imagearea

(x,y)

h

w

Figure 3.7: ROI of destination file

� If (w,h) is smaller than output frame size, remaining area willbe black-filled.

� If (w,h) of input file and (w,h) of output file differ, picture willbe resized to fit output ROI.

Interpolation methods

Cinema Craft Encoder SP2 provides four interpolation methods forresizing picture:

• Nearest neighbor

• Linear interpolation

• Cubic interpolation• Lanczos interpolation

The nearest neighbor algorithm is the fastest, while other methodsyield higher quality results, but are slower.

Progressive frame

If Progressive frame is selected, resizing is performed per frame,otherwise, per field.


� This option does not affect progressive frame flag.

� To extract one field from interlaced source, select this option to-gether with Nearest neibor, and then halve the output height.

Mirroring

Selecting Horizontal axis mirrors the source image about a horizon-tal axis. And selecting Vertical axis mirrors the source image abouta vertical axis.

3.8 Frame rate

Select the frame rate of output video file. If input and output framerates differ, frame insertion or frame skip will occur. To avoid this,clear Rate conv option. However, it will change duration of outputvideo file, and will be out of sync with audio unless Sync with videois selected.

� If both MPEG-2 for DVD and 23.976 are selected, pull-down is automatically selected to convert frame rate to 29.97fps.

3.9 File size

An estimated file size for the video is shown. When you output SystemStream, it will be calculated including audio.

Since Bitrate and File size are linked together, if you change thefile size, the average bitrate will be calculated automatically.

The file size can be displayed in one of the units listed below:

• B (bytes) Shows file size in bytes. Since bitrate can be setonly in kbps (internally, multiple of 400 bits), some errors areinevitable.

• kB (kilobytes) Shows file size in kilobites (103 = 1, 000 bytes).

3.10. BITRATE AND Q.FACTOR 39

• KiB (kibibytes) Shows file size in kibibytes (210 = 1, 024bytes).

• MB (megabytes) Shows file size in megabytes (106 = 1, 000, 000bytes).

• MiB (mebibytes) Shows file size in mebibyte (220 = 1, 048, 576bytes).

� When 1-pass VBR is selected in encoding mode, the possiblelargest size will be shown.

3.10 Bitrate and Q.factor

Each encoding mode requires the parameters for bitrate setting asfollows.

Encoding mode Avg Min Max Q.MPEG-2/DVD (1-pass VBR) -

MPEG-2/DVD (1-pass CBR) - - -MPEG-2/DVD (Multipass CBR) - - -MPEG-2/DVD (Multipass VBR) -MPEG-2/DVD (1st pass of VBR) -MPEG-1 - - -SVCD - - -VCD - - - -

Avg, Min, Max, Q. indicate average bitrate, minimum bitrate,maximum bitrate, and Q.factor, respectively.

Bitrate

Bitrate indicates bits per second. The higher the bitrate, the betterthe quality but larger the file size. (1 kilo bits is regarded as 1,000bits.)

For example, if the average bitrate is 6,000 kbits/sec and the movielength is 2 hours, the file size of encoded video elementary stream is:

2 × 60 × 60 × 6000 × 1000÷ 8 = 5, 400, 000, 000 bytes


(To convert from bits to bytes, it is divided by 8.)

� An upper bitrate limit for SVCD is 2460 kbps.

� A bitrate for VCD is always locked at 1150 kbps.

Q.factor

Q.factor is a parameter unique to Cinema Craft Encoder, and can beset only when 1-pass VBR mode. Cinema Craft Encoder calculatesquantization scales for each picture type (I,B,P) using Q.factor.When you set Q.factor, refer to the following table.

Q.factor Description1 ∼ 40 achieves a higher quality of image

but a lower compression rate40 ∼ 80 Standard80 ∼ 120 achieves a higher compression rate

but a lower quality of image120 ∼ disregards a quality of image

3.11 Pulldown setting

When creating a NTSC video from a film, it is necessary to convertframe rate from 24 fps to 30fps (60 fields/sec). 3:2 pulldown is acommon method for this convertion. And the 3:2 pulldowned sourcewill be encoded efficiently and achieve better result applying Inverse3:2 pulldown.

� When you apply 3:2 pulldown, do not set 00:00:00:00 as a time-code for the first frame.

� When you apply 3:2 pulldown, Output top field first streamshould be selected. If the source is bottom field first, you shouldalso confirm that Offset line is 1 or odd number.

� When you output DVD video stream with this option, GOP se-quence should be M=3, N/M=4. To specify GOP sequence,

3.11. PULLDOWN SETTING 41

click Advanced button in the lower right of the encoder set-ting window. About Advanced video setting window, seepage 45.

3.11.1 Pulldown

Select this option when you convert film source at 24 (23.976) fpsinto NTSC video at 30 (29.97) fps. When it is selected, Inverse3:2 pulldown will be also applied automatically. This setting willbe allowed when output is MPEG-2 format file and 23.976 or 24 isselected as Frame rate.

2:3 or 3:2

In 3:2 pulldown method, 2 fields are added to fill 10 fields (5 frames)of video that make 4 frames of film.

You can select the pattern of where to add fields between 2:3 and3:2. (See Fig. 3.8)

24 fps

10 fields

30 fps

1 2 3 4

1 2 3 4 5

copy copy

1 2 3 4

1 2 3 4 5

copy copy

Figure 3.8: 2:3 and 3:2 pattern

� If the total number of encoded frames is a multiple of 4, the playback time of the source and the output will be same. However,if it cannot be divided by 4, the gap of the play back timewill be caused between the source and the output. And how aremainder when devided by 4 will make a number of frames isdifferent between 2:3 and 3:2. (See the table below.) So it willbe better to alternate 2:3 and 3:2 patterns if you encode thesource devided into multiple streams.


Number of frames left Number of frames to be encodedwhen devided by 4 2:3 3:2

1 1 12 2 33 4 4

� In plug-in version, frame rate setting of Premiere should be23.98(24) fps.

3.11.2 Inverse 3:2 pulldown

If the source is 3:2 pulldowned NTSC footage, you can expect im-provement in the quality by applying Inverse 3:2 pulldown.

The 3:2 pulldowned footage has only 48/60 = 4/5 of informationcompared to those not 3:2 pulldowned. Inverse 3:2 pulldown de-tects and avoids encoding redundant copied fields, then encoding ef-ficiency will be gained and the image quality will be improved withmore allocatable bits.

� If the source is letter boxed, you can except upper and lowerblack areas from 3:2 pulldown detection to make the detectionmore precise. The setting for a letterbox source is on Picturesetting window. See page 78 for more detail.

� When you select Inverse 3:2 pulldown, auto 3:2 pulldowndetection will be done when encoding. If you would rather setpulldown patterns manually or would like to confirm the resultof auto detection to modify, click Settings button to open thewindow for a manual setting. See page 81 Inverse 3:2 pull-down for more detail.

� 3:2 (2:3) pulldowned source is only footage to apply this option.Other patterns of pulldown such as 2:3:3:2, 2:3:3:2:2 (PAL �

NTSC), 2:2 (film � PAL) are not supported.

� When you apply inverse 3:2 pulldown, Output top field firststream should be selected. If the source is bottom field first,you should also confirm that Offset line is 1 or odd number.

3.11. PULLDOWN SETTING 43

� When you output DVD video stream with this option, GOP se-quence should be M=3, N/M=4. To specify GOP sequence,click Advanced button in the lower right of the encoder set-ting window. About Advanced video setting window, seepage 45.

� When you apply Inverse 3:2 pulldown against 720× 486 sourcefile, see also 12.2 Encoder settings for plug-in program /Parameters / Inverse 3:2 pulldown on page 126.

Cautions when applying inverse 3:2 pulldown

The error of pulldown detection causes jerky movement of de-coded images. If this problem happens everywhere on the de-coded stream, the material does not seem to suit for pulldowndetection. In that case, clear the Inverse 3:2 pulldown checkbox and encode again.

Do not apply pre-filters to the material before 3:2 pulldowndetection, because it makes the precision of 3:2 pulldown detec-tion lower. However, there is no problem to use Cinema CraftEncoder SP2’s internal filters because those filters are appliedafter 3:2 pulldown detection.

If the original images are compressed, deterioration caused bythe compression affects the precision of pulldown detection.Therefore do not perform heavy compression over the footage.It is recommended to avoid using codec whose compression ra-tio is higher than DV codec, and also to avoid recompressionduring video editing.

When multipass encoding, DO NOT FORGET to recre-ate the video information file if you change some set-tings in inverse 3:2 pulldown.

Chapter setting has an impact on inverse 3:2 pulldown opera-tion. At a frame pointed as a chapter point, the first field ofthe frame is forced to be on the top field even disturbing theproper pulldown phase. If you see the pulldown phase out oforder, confirm and modify the location of the chapter point to


keep the pulldown phase. However, this rarely if ever effects onthe quality of image. See page 93 for more detail.

3.11.3 Auto 2-pass detection

Auto 2-pass detection will allow you precise Inverse 3:2 pull-down without editing the pulldown list manually.

When you carry out encoding with Inverse 3:2 pulldown andAuto 2-pass detection option, auto pulldown detection will bedone and a pulldown information file will be created gathering theinformation of each line in the source. Then the pulldown list will bereviewed using the pulldown information file. If any improvement canbe expected, Cinema Craft Encoder SP2 will modify the pulldown listand add one more pass automatically to renew VAF file.

� If you apply the pulldown list that is manually set, do not selectAuto 2-pass detection, or the pulldown list you set will bediscarded.

3.12 Aspect ratio

Aspect ratio is the ratio between the width and height of the screen.The MPEG-2 playback device (e.g., DVD player) refers to this setting,and modifies the ratio when outputting the data to a TV monitor. Itdoes not mean frame size conversion.

DAR (Display Aspect Ratio) works on aspect ratio of the frame.Available DAR values are 4:3, 16:9, 2.21:1. 4:3 is for a standard TVmonitor. For wide screen materials, 16:9 will be appropriate.

SAR (Sample Aspect Ratio) works on aspect ratio of the pixel. IfSAR 1:1 is selected, the ratio of frame width and height of the outputwill be the same ratio as source.

3.13 Deinterlacing

If you intend to playback an interlaced video on a non-interlaceddisplay (e.g., a computer monitor), artifacts will be seen as jaggededges. Deinterlacing is useful to reduce such jaggies.

Chapter 4

Advanced video setting

This section describes the advanced video settings. The setting win-dow is displayed by clicking Advanced video setting button inEncoder setting window (Fig. 3.1).

4.1 Setting GOP configuration

In MPEG, encoding, playing, editing is based on GOP that consistsof a group of pictures.

4.1.1 Picture type

MPEG uses three types of pictures for spatial and temporal coding,that is, I, P and B pictures.

I picture (intra-coded picture)

An I picture holds all the picture information on one picture withinitself. It is not necessary to refer to another picture to decode, butcompression efficiency is not good in comparison with other typesof pictures. Therefore, when the bitrate is the same, the picturequality of a stream of I pictures is lower than that of other types ofstreams. To edit encoded streams, however, it is more convenient touse numerous I pictures.

45

46 CHAPTER 4. ADVANCED VIDEO SETTING

Figure 4.1: Advanced video setting window

4.1. SETTING GOP CONFIGURATION 47

P picture (predictive-coded picture)

A P picture consists of motion vectors referring to previous (justbefore) I picture or P picture and differential information between apicture comprised of these motion vectors and an original picture. Todecode a P picture, referred pictures are required, but compressioncan be done more efficiently than I pictures. However, in a sequencewith continual P pictures, a deterioration in the picture quality of Ppictures may occurs with distance from the referred I picture whendecoding. That results from error accumulation in the gap of IDCT1

calculation method between the encoder and a decoder.

B picture (bidirectionally predictive-coded picture)

A B picture consists of motion vectors referring to previous (justbefore) I picture or P picture and/or future (just after) I picture or Ppicture, and differential information between a picture comprised ofthese motion vectors and an original picture. Therefore, to decode a Bpicture, referred two pictures are required, but compression efficiencyis even better than that of P pictures. Since a B picture is not referredby other pictures, continual B pictures do not cause accumulationof errors unlike the case of P pictures. However, the sequence ofB pictures means the distance from a referred picture, then causedecrease of the motion compensation effect.

4.1.2 GOP sequence

You can change the GOP sequence modifying M and N/M. Maffects the number of B picture, and N/M does the number of Ppicture. You can see the GOP sequence you set above the settingboxes for M and N/M. The initial setting is M=3, N/M=5.

� Generally, M=3, N/M=5 is used for NTSC DVD, M=3, N/M=4for PAL DVD.

� When you create a stream for DVD with applying inverse 3:2pulldown, M=3, N/M=4 should be set.

1Inverse Discrete Cosine Transform


4.1.3 GOP header

Specify the number of frames to which a GOP header is added. Thisvalue stands for the GOP length. 1 ∼ 15 can be set. Since the GOPlength must be a multiple of N, the value can be limited to 1 × Nin some cases. For example, supposing M=3, N/M=5, the valuebecomes N=15, then no other value than 1 × N will be allowed.

4.1.4 SEQ header

Specify how many GOPs between the sequence headers. The initialsetting is 1.

4.2 Add sequence end code

To add the sequence end code, select Add sequence end code checkbox. You should not clear this check box except when necessary.

4.3 Encode the last frame as an I frame

This option will achieve a clear image of the last picture of resultedDVD when you pause the DVD player at the end of the stream.

4.4 DVD Video multiangle

If DVD Video multiangle is selected, the encoder outputs a videoElementary Stream for multiangle DVD. Multipass CBR is recom-mended as encoding mode.

It is equivalent to selecting the following three options at the sametime.

• Close all GOPs

• Disable auto I frame insertion• Equalize each GOP’s bit length

4.5. CLOSE ALL GOPS 49

4.5 Close all GOPs

If Close all GOPs is selected, all GOPs are closed. In closed GOP,individual frames in a GOP do not refer the frames outside the GOP,and B pictures can be correctly decoded even in random access mode.

� Even if Close all GOPs is not selected, the GOP at the scenechange point is automatically closed unless Restrict auto Iframe insertion is selected.

� If Close all GOPs is selected, image quality will be slightlydegraded. Therefore, do not select this except when necessary.

4.6 Disable auto I frame insertion

Cinema Craft Encoder automatically detects scene change and en-codes the first frame of a new scene as I frame. This function isimportant for improving image quality. However, if you are to cre-ate streams for multiangle DVD, auto I frame insertion may cause aproblem since every I frame should be the same position in an eachangle. In this case, select Disable auto I frame insertion to avoidauto I frame insertion.

� If I frame is inserted, the length of GOP will be changed. If youwould like to keep the constant GOP length, select Disableauto I frame insertion check box.

� If Disable auto I frame insertion is selected, image qualitywill be slightly degraded. Therefore, do not select this exceptwhen necessary.

� Even if Disable auto I frame insertion is selected, I framesas chapter points will be put in the location you set. Whenyou create a stream for multiangle DVD, you must be carefulto match chapter points in each stream.


4.7 Equalize each GOP’s bit length

Although slight bitrate movement in CBR is allowed to improve imagequality, Equalize each GOP’s bit length option forces the bitamount of each GOP to be equal.

� If Equalize each GOP’s bit length is selected, image qualitywill be slightly degraded. Therefore, do not select this exceptwhen necessary.

4.8 Disable adaptive Q-matrix switching

Cinema Craft Encoder SP2 (later than 2.67) adjusts quantization ma-trix automatically to improve the quality of image when MultipassVBR is selected. However, some DVD player does not support suchMPEG streams as encoded with adaptive quantization matrix, andmay cause the block noise when playback. When you encounter theproblem, you can avoid it with this option.

What is adaptive Q-matrix switching ?

There are some cases that Cinema Craft Encoder SP2 cannot keepthe minimum bitrate. It is considered to be caused by that the lowestquantization scale cannot raise the bitrate under a single quantizationmatrix. To solve the problem, Cinema Craft Encoder SP2 automati-cally creates other matrices cutting the value of each element in thematrices by half, quarter and so on, and applies another one to pro-duce a much lower distortion for much higher bitrate. This switchingof quantization matrix can occur at every GOP.

Thus, adaptive Q-matrix switching can improve the quality of im-age, especially with lower bitrate sources or CG works or the edge offading out.

4.9 Panscan

Panscan is one of formats for displaying wide-screen movies on 4:3TV. In panscan format, the image is stripped off at the left and right

4.10. OUTPUT TOP FIELD FIRST STREAM 51

side of the screen. (In another format called letterbox, black barsare added at the top and bottom of the screen to fill 4:3 TV screeninstead of cutting off the image.)

If you prefer to create DVD-Video displayed in panscan format,select Panscan. This setting works on MPEG-2 output.

4.10 Output top field first stream

Specify the field order of output stream. If it is selected, an outputstream is top field first. If not, an output is bottom field first. Thissetting works on MPEG-2 output.

It is important to set correct field order, because incorrect fieldorder causes stuttered motion. To set field order correctly, you haveto care about Offset line setting.

• If the field order is same between the source and the output,Offset line should be 0 or even number.

• If the field order is different between the source and the output,Offset line should be 1 or odd number.

4.11 Offset line

This parameter specifies offset line from which encoding starts. Inanother words, it implies how many lines should be cropped from thetop. For example, if you set 0, no line is cropped and encoding startsfrom the top. When the output is MPEG-2 video, it affects the fieldorder. If this setting is incorrect, the movement of an encoded videowill be jerky. Please be careful about both field orders of output andsource material. (See also Output top field first stream describedabove.)

4.12 Luminance level

Cinema Craft Encoder uses a YUY2 pixel format internally. If thesource format is RGB, the color space conversion into YCbCr will beperformed using this luminance range.


In ITU-R BT. 601-52, the range of luminance is settled from 16 to235. If this setting is not suitable, you can select “0 to 255”.

When “16 to 235” is specified

RD = 219R + 16 × 256GD = 219G + 16 × 256BD = 219B + 16 × 256

Y =77RD + 150GD + 29BD

216

CR =131RD − 110GD − 21BD

216+ 128

CB =−44RD − 87GD + 131BD

216+ 128

When “0 to 255” is specified

Y =77R + 150G + 29B

28

CR =131R− 110G− 21B

28+ 128

CB =−44R − 87G + 131B

28+ 128

In any case, decimals in the result of the division are omitted.

4.13 Quantization scale type

When you output a MPEG-2 file, you can select a type for the quan-tization scale used when quantizing DCT coefficients between Linearand Nonlinear. Generally, MPEG-2 outputs use Nonlinear.

� MPEG-1 outputs always use Linear.

2Standards and specifications that can be used to broadcast or interchangedigitally coded television signals.

4.14. QUANTIZATION MATRICES SETTING 53

4.14 Quantization matrices setting

Quantize matrices setting window will appear when you clickQuantization matrices button. You can choose quantization ma-trices for encoding or create the new one on the window,.

Figure 4.2: Quantize matrices setting window

Preset

Several presets of quantization matrices are provided in advance. Youcan list them by clicking � next to preset box and select one.

Preset matrices and their characteristics are the following :

• StandardCinema Craft Encoder SP2 sets this one as a standard matrix.It is the matrix that was used for the evaluation of the MPEG-2international standard algorithm. For natural images, it will beappropriate.

• MPEG standardThis one is a default matrix under the MPEG-2 internationalstandard. If Standard does not work as fine as you expected,it may help you to improve the quality of image with higherbitrate.


• Smooth (CG etc.)It will be appropriate for smooth and less noise sources like CG.

• Very low bitrateIt is intended to be used when encoding NTSC or PAL con-formed source with the bitrate of 4 Mbps or less.

• Ultra low bitrateIt is intended to be used when encoding NTSC or PAL con-formed source with the bitrate of 2 Mbps or less. Note that youcannot always lower the bitrate as you would like. How far youcan lower the bitrate depends on your source.

Create quantization matrices

On Quantize matrices setting window, you will see two matrices,one for Intra block and another for Non-intra block. You can setthese matrices separately. To modify a matrix, type the value directlyon the matrix.

� Allowable range of the value is : 1 to 127 3.

� (1, 1) element in intra block must be 8 under MPEG standard.

And the buttons below a matrix will help you to modify the value.

• button copies (i, j) element and put them in (j, i) when i < j.

• button copies (i, j) element and put them in (j, i) when i > j.

• Transpose button interchanges rows and columns in a matrix.It means the interchange of (i, j) and (j, i) elements.

When there is no difference between (i, j) and (j, i) elements, thesethree buttons appear to be grayed out and do not work. In otherwords, you can see if these is some difference between the elementsor not by the status of the buttons.

3Cinema Craft Encoder SP2 limits the value to 127 in spite of MPEG standard255.

4.15. PACKET SIZE 55

To save the matrix you created, click Save button and type the newname of the matrix. Then you can see it in the list of the presets.

To delete a matrix setting, select one from the list of presets andclick Delete button.

4.15 Packet size

The default value of packet size of program streams is 2048 bytes.You can change this value.

Chapter 5

VBR bit allocation

When Multipass VBR mode, the parameters such as V/C (seepage 31.), Avg/Min/Max bitrates are allowed as you can see themon Encoder setting window. In addition to them, you can changethe bitrate setting partly in GOPs on VBR bit allocation window(Fig. 5.1).

To open the window, click VBR bit allocation button on En-coder setting window. Then a fluctuation graph of bitrate or quan-tization scale will appear on VBR bit allocation window. And theviewer window will also appear at the same time. You can see howand where the bits are allocated on the graph and the viewer, thenmodify the bit allocation or change the picture quality settings callingPicture settings window.

� The video information file (VAF file) should be created in ad-vance to open VBR bit allocation window. And MultipassVBR should be selected in encoding mode.

� The viewer shows the first I picture of the GOP currently re-ferred.

57

58 CHAPTER 5. VBR BIT ALLOCATION

Figure 5.1: VBR bit allocation window

5.1 Bitrate / Quantization scale graph

VBR bit allocation window will show the calculated plan of bitallocation that will be done at the next encoding. You can see thefluctuation of bitrate or quantization scale on the graph. The X-axisindicates time of source, and the Y-axis indicates bitrate or quanti-zation scale.

5.1.1 Bitrate graph

To show the bitrate fluctuation graph, select Bitrate option button(radio button) at the right side of the graph. Then the graph such asFig. 5.2 will appear.

Three colored vertical short lines on the graph indicate white as Ipicture, green as P picture, and red as B picture. Yellow bar indicatesthe average bitrate of the GOP. Blue band in the back means thebit allocatable band between maximum and minimum bitrate. Thedouble vertical line is a current time indicator that shows the timecodepoint currently referred. And pink vertical line indicates that the

5.1. BITRATE / QUANTIZATION SCALE GRAPH 59

GOP is closed.

Figure 5.2: Bitrate graph

5.1.2 Quantization scale graph

Quantization scale affects distortion level. The smaller quantizationscale becomes, the smaller distortion will occur. It means that thequality of the image will be well kept. On the other hand, the largerquantization scale will cause the larger distortion and lose the qualityof image.

To show the fluctuation graph of quantization scale , select Q.scaleoption button (radio button) at the right side of the graph. Then thegraph such as Fig. 5.3 will appear.

As well as bitrate graph, three colored vertical short lines on thegraph indicate white as I picture, green as P picture, and red as Bpicture. Yellow bar indicates the average quantization scale of theGOP. The double vertical line is a current time indicator that showsthe timecode point currently referred. And pink vertical line indicatesthat the GOP is closed.

And you will see the background of the graph is painted in green,gray or red. These colors implies the quality of the encoded image.Green implies that you can expect a good result. Gray representsthe acceptable quality. Red means the distortion will be more visible.The implication is helpful to foresee the quality of results and modifythe settings to avoid the poor quality. Against red painted area, youcan raise the minimum bitrate or apply filters strongly for improvingthe quality.


Figure 5.3: Quantization scale graph

5.1.3 Operation of graph

Switching of graphs

To switch graphs, select Bitrate or Q.scale option button (radiobutton) at the right side of the graph.

Changing the upper and lower limit of the graph

To change the upper limit, click − + button in upper left of thegraph. The value of the upper limit is displayed above the buttons,and the unit of bitrate graph is Mbps. − + button in lower leftof the graph changes the lower limit. The value of the lower limit isdisplayed below the buttons. When you change the upper and lowerlimit of the graph, the graph will be stretched or squashed vertically.You can give an appearance you like.

Graph reset

It gives the fixed values for the limits of the graph. In bitrate graph,#1 gives 40.0 for the upper and 0.0 for the lower. #2 gives 10.0

for the upper and 0.0 for the lower. In quantization scale graph, #1gives 30.0 for the upper and 0.0 for the lower. #2 gives 15.0 for theupper and 0.0 for the lower.

5.1. BITRATE / QUANTIZATION SCALE GRAPH 61

Scrolling vertically

To scroll the graph vertically, click ⇑ ⇓ buttons above graph resetbuttons.

Scrolling horizontally

The graph will move horizontally by GOPs, following the current timeindicator (the double vertical line). You can see the current timecodeat Pos in GOP information section located in the lower right ofthe window. And timecodes of the edges of the graph are displayedbelow the left and right edges of the graph.

To move the current time indicator, there are several ways listedbelow.

• Click the location on the graph you would like to go, then thecurrent time indicator will point there.

• Drag the slider just below the graph.

• To go to the top of the footage, click button below the leftedge of the graph. To go to the I picture (white short line) inlast GOP of the footage, click button below the right edge ofthe graph.

• When you click the buttons at the left of the slider, you canmove forward or backward by the time scale of the button youpress. The time scales you can select are 10 min / 1 min / 10sec / GOP. And negative values move backward, and positivevalues move forward. Then the current time indicator will pointthe I picture of the destination GOP.

• When you type the destination timecode at the boxes of Set inGOP information in lower right of the window then clickbutton, you can put the current time indicator at the I pictureof the destination GOP. Instead of clicking button, draggingthe timecode onto the graph will bring the same result. You canalso move to the just previous referred timecode in the boxes ofPrev by dragging or clicking button. And furthermore, you


can move to the timecodes in In and Out boxes in local bitratesettings section as well.

• You can set a frequently referred timecode as a label. Then youcan move there by just selecting the label.

Creating a label

Put a current time indicator on a timecode you would like toset a label. Type the name of a label in the box below the sliderand click ⇒ button at the right of the box. Then the label willbe put in the label box next to ⇒ button. You can see the listof labels by clicking � in the right.

Deleting a label

When you delete a currently selected label, click Del buttonnext to the label box. To select a label, click � next to thelabel box to open the list and select one. When you delete allof the labels in the list, click Del all button.

Moving to a timecode pointed by a label

Click � next to the label box to open the list, and select one.

Viewer window

You can see the image of source at the I picture of currently referredGOP in the viewer window. And a source file name and a framenumber (0 indicates the first frame of a source file) are displayed onthe title bar of the window. When the multiple files are input for anoutput, timecode of the source files will be combined then becomeconsecutive. However, note that a frame number on the title bar ofthe viewer window is not combined one, but shows a frame numberin each source file.

5.2. BITRATE SETTINGS 63

� The image in the viewer is not a result of applying the modifiedbitrate setting, but just an image of source to specify the timepoint for the parameter setting.

� The image in the viewer comes from source files, however thetimecode of the graph comes from MPEG file. In some cases, itmay make you confuse.

When you apply inverse 3:2 pulldown, amount of frames willdiffer between source files and MPEG file. It may cause a gapof ±1 field between viewer image and the actual time point inthe MPEG file when the referred time point is on some pulldownphases.

When you apply frame rate conversion (e.g., 3:2 pulldown),timecodes will differ between source files and MPEG file. Insuch case, when you call Picture setting window, you will seethat the timecode is different from one in VBR bit allocationwindow. However, you will also see that the frame number ofthe viewers are same in both setting windows.

5.2 Bitrate settings

You can modify the bitrate setting throughout the footage or partly,seeing the fluctuation of the graph.

5.2.1 Global bitrate settings

You can modify the bitrate setting and V/C setting as you have doneon Encoder setting window. The modified setting will be appliedthrough the footage.

Avg/Min/Max

You can modify the average, minimum, maximum bitrate setting. Ifyou change the setting, the graph will be updated automatically.

� About the update of the graph, see Update the graph de-scribed later.


V/C (VBR/CBR)

You can change a fluctuation of the bit allocation. See page 31 V/Cfor more detail.If you change the setting, the graph will be updated automatically.

File size

An estimated file size for the video is shown.

5.2.2 Local bitrate settings

You can change the bitrate setting in a closed section. And you caneven keep the bit allocation used in the previous encoding in a closedsection, discarding newly planned one.

In/Out points of a closed section

Specify In and Out points of a closed section you would like to changethe bitrate setting. With Exclude out point option, the local bi-trate will be applied from In to the just previous GOP of Out you set.Without the option, it will be applied from In to the GOP includingOut point.

To specify In and Out points, you can type the timecodes to theboxes directly. Or clicking ⇒ button will put the timecode pointedby the current time indicator into the box. Moreover, dragging thecurrent time indicator or timecodes in GOP information will putthe timecode in the box.

Min/Max

Specify minimum and maximum bitrate for the closed section. Youcannot set local bitrate values beyond global bitrate settings. How-ever, local minimum bitrate is allowed to rise above global averagebitrate.

If you change global bitrates after setting local bitrates, you willsee that the blue band for local bitrate settings is not updated andstill out of global bitrates. In this case, actual local bitrate values willbe matched to global bitrate settings automatically.

5.2. BITRATE SETTINGS 65

5.2.3 Keep the previous bit allocation

To keep the bit allocation used in the previous encoding, specify Inand Out points then click Lock button. Once you locked the section,any changes you made later will not be applied in the section. So,when you got a good result partly in the previous encoding, you canavoid loosing the quality of the parts.

In the locked section, the blue band in the back, which indicatesthe bit allocatable band between maximum and minimum bitrate,will disappear.

� When you open VBR bit allocation window, the graph willshow you the newly planned bit allocation for the next encoding,which is calculated from the video information file (VAF file).Once you locked a section, the planned bit allocation on thegraph will be discarded and replaced with the bit allocationused in the previous encoding.

� Even if you lock the bit allocation, the slightest modificationmay occur to keep VBV (Video Buffer Verifier) model of MPEG-2.

5.2.4 Apply the changes

To apply the local bitrate settings and update the graph, click Setbutton.

� About the update of the graph, see Update the graph de-scribed later.

5.2.5 Cancel the changes

To cancel the local bitrate settings, specify In and Out points thenclick Reset button.

� When you cancel the changes, the bit allocation in a sectionwill be back to the initially planned one. That is the one yousaw when you just opened the window. Note that it is not thecancellation of a just previous change you made.


� When you point the top of the footage as In and the last Ipicture as Out, the cancellation will be applied through thefootage then the graph will be back to the one you saw whenyou just opened the window.

� When you make the changes on VBR bit allocation window,the changes will be written to the video information file imme-diately. If you need to keep the video information in a stepof changing the settings, you should save the VAF file as an-other name. Then you can go back to the step when you wouldlike, not to the initial one by Reset button. To refer to an-other named VAF file, close VBR bit allocation window togo back to Encoder setting window, then change the videoinformation file name in Output files section, and open VBRbit allocation window again.

5.2.6 Update the graph

When you make the changes, planning calculation for the renewed bitallocation will start. During the calculation, the indicator next to thetimecode box of the right edge of the graph will turn to [Planning]from [Clean]. When finished calculation, the graph will be updatedand the indicator will turn back to [Clean]. Note that the graph hasnot renewed until [Planning] turns to [Clean].

� If you make the changes during [Planning], just calculating bitallocation will be discarded and the calculation for the new onewill start.

5.3 GOP information

Pos

Pos shows the timecode pointed by the current time indicator.In VBR bit allocation window, timecode comes from MPEG file’s,and the image in the viewer comes from source files.When you apply inverse 3:2 pulldown, amount of frames will differbetween source files and MPEG file. It may cause a gap of ±1 field

5.3. GOP INFORMATION 67

between viewer image and the actual time point in the MPEG filewhen the referred time point is on some pulldown phases.When you apply frame rate conversion (e.g., 3:2 pulldown), timecodeswill differ between source files and MPEG file. In this case, whenyou call Picture setting window, you will see that the timecode isdifferent from one in VBR bit allocation window. However, youwill also see that the frame number of the viewers are same in bothsetting windows.

Set

When you specify the destination timecode in Set box then clickbutton, you can jump to the I picture of the distination GOP. Insteadof button, dragging the timecode onto the graph will bring thesame result.

Prev

Prev shows the timecode just previously pointed by the current timeindicator. You can move to the timecode by dragging or clickingbutton.

BR

It shows average bitrate in a GOP pointed by the current time in-dicator. Numbers in parentheses shows the minimum and maximumbitrate of the stream.

Q.

It shows average Q.scale of each picture type (I,P,B) in a GOP pointedby the current time indicator.

Chapter 6

Picture settings

Picture settings window (Fig. 6.1) will appear when you click Pic-ture Quality button in the lower left of Encoder setting window(Fig. 3.1). When you open the window, the viewer window will alsoappear.

You can set the filters or the parameters for improving the qual-ity on the window. Pre-processing such as filters applied by othersoftwares may cause a loss of precision in pulldown detection wheninverse 3:2 pulldown. On the other hand, the filters here will be ap-plied after pulldown detection and does not affect the precision of thedetection.

6.1 Entry list

Parameter sets based on timecode should be put into the entry list.

6.1.1 Adding an entry to the list

To add an entry to the list, do the following :

� Specify the starting timecode for a parameter set at TC frombox. The first entry of the list is TC in that has no entrynumber. The first timecode is assigned to TC in. When youadd the later timecode, the entry will get an entry number 1. If

69

70 CHAPTER 6. PICTURE SETTINGS

Figure 6.1: Picture settings window

6.1. ENTRY LIST 71

you select check box for to below TC in, the timecode of thenext entry will be added at the same time.

� To put the TC from timecode into to box, click ⇓ buttonin the right. To put the to timecode into TC from box,click ⇑ button. Dragging the timecode will bring the sameresult.

� Set the parameters. The explanation of each parameter is de-scribed later on page 72. The parameter setting you made willbe applied from the TC from timecode to the timecode justbefore the next entry.

� Click Set button below the timecode boxes to put the parameterset into the list as an entry. Then you can see the entry numberin # column of the list, the timecode in the TC, and the captionin the Description column.

� There is another way to build the entry list: add all of thetimecodes for the entries first, then set the parameters for eachentry later. To set the parameters later, select the entry in thelist then set the parameters. When completed the setting, clickApply button in the lower right of the window or Set buttonbelow the timecode boxes to update the entry list.

� You can see the image of source of the currently referred time-code in the viewer window. Note that the image in the vieweris not a result of applying the parameter set, but just an imageof source to specify the time point for the setting.

� On the title bar of the viewer, a source file name and a framenumber (0 indicates the first frame of a source file) are displayed.When the multiple files are input for an output, timecode of thesource files will be combined then become consecutive. However,note that a frame number on the title bar of the viewer windowis not combined one, but shows a frame number in each sourcefile.


6.1.2 Operation buttons for the list

Chapters

It does not give an list operation, but opens the chapter list window.

Del

It deletes the currently selected entry from the list. You can alsodelete the entry by Delete key.When you delete an entry, the duration included in the entry willbe combined to the just previous entry and the parameter set of theprevious one will be applied there.Note that you cannot delete the entry TC in.

Reset

It deletes all of the entries except TC in.

6.1.3 Saving/Loading the entry list

You can save the entry list to a file in text format or in CSV format.To save the list, click List Save button. To load the file you saved,click List Load button.

6.2 Simple setting

There are two ways of setting filters. In Simple setting, a properparameter set is automatically chosen by specifying the type and thecomplexity of source material. In Detailed setting, you can selectthe filters and their values individually. And Simple setting is linkedto Detailed setting, so you can see the contents of a parameter setof Simple setting in Detailed setting section.

6.2.1 Material types

Select the material type from the following three alternatives.

6.2. SIMPLE SETTING 73

Figure 6.2: Simple setting

• Natural picture

It is appropriate for live-action images.

• Computer Graphics

It is appropriate for smooth images like computer graphics.

• Animation

It is appropriate for high contrast images such as cell animation.

6.2.2 Complexity of material

Specify the complexity of material in 5 levels. When it seems plainor flat, move the slider to the left. When it seems more complicated,move the slider to the right.

The examples for choosing a level of the complexity are listed below.Flat:

• Simple flat picture• Horizontal motion

Complicated:

• Complicated picture

• Complicated motion

– Zooming

– Surface of the water, flickering flame, stirring leaves in thewind

– Crossfading between pictures with different motions


6.2.3 Parameter sets of Simple setting

The tables below show the parameter sets of Simple setting.

NaturalParameter 1 2 3 4 5Low - 32 32 24 16High - - - - -Vertical filter - - - 6 8Dithered Q. -Y 8 8 - - -Dithered Q. -C 4 4 - - -Quantizer characteristics 24 20 16 12 8

C.G.Parameter 1 2 3 4 5Low - 32 32 24 16High - - - - -Vertical filter - - - 16 16Dithered Q. -Y 18 18 12 - -Dithered Q. -C 9 9 6 - -Quantizer characteristics 5 4 3 2 1

AnimationParameter 1 2 3 4 5Low 32 32 32 24 16High - - - - -Vertical filter - - - 6 8Dithered Q. -Y 18 18 - - -Dithered Q. -C 9 9 - - -Quantizer characteristics 40 32 24 16 8

6.3 Detailed setting

In Detailed setting, you can select the filters you need and set theirvalues individually.

6.3. DETAILED SETTING 75

Figure 6.3: Detailed setting

6.3.1 Caption

Specify the name of the parameter set. The caption you set will bedisplayed on Description column of the enrty list when you clickSet button. It is not necessary, but will help you follow the list witheasy name.

6.3.2 Low and High filter

You can set these filters seeing their frequency characteristics on filtercharacteristics chart

Low

It is a low pass filter to cut out the higher frequency components atspecific frequencies. It tends to make the image blur or smooth.


Figure 6.4: Low and High filter

Cut-off Frequency Move the slider in the left hand.Allowed range : 4 - 47

Slope Move the slider in the right hand.Allowed range : 0 - 200The smaller the value, the flatterattenuation slope becomes.The bigger the value, the steeperthe slope becomes.You can see the slope on the chart.

� The default value of cut-off frequency is 4.5MHz. It will be ableto cut the mosquito noise of almost all the sources with thefrequency. If you set the lower value than 4.5MHz, you shouldcarefully check the result not to cause a problem.

High

It is a high boost filter to increase gain at specific frequencies. It tendsto make the image greasy emphasising details like edges or lines.


Turnover frequency Move the slider in the left hand.Allowed range : 4 - 47

Slope Move the slider in the right hand.Allowed range : 0 - 200The smaller the value, the flatterthe slope becomes.The bigger the value, the steeperthe slope becomes.You can see the slope on the chart.

Filter Characteristics Chart

It shows the frequency characteristics of High and Low filters you set,gain (dB) on the vertical axis and frequency (Hz) on the horizontalaxis. When you put the mouse pointer on the chart, you can see thepointed value in the right. When you set High and Low filters, becareful about the slope because too steep slope may cause overshootor ringing.

6.3.3 Vertical filter

It detects the texture domain and reduces speckle noise, keeping thesharpness of the edges. The value you can set is from 1 to 64. Thebigger the value, the larger domain will be detected as a texture.

� If your source is interlaced, specify the value 16 or less. If youapply bigger value, it may cause a jerky motion.

� If your source is progressive or 3:2 pulldowned, the value isrecommended not to exceed 32.

6.3.4 Dithered quantization

It adds a small noise (dither) before quantization to reduce the quan-tization error and to distribute the remaining error more effectively.


It can help you remove contouring artifact1 in fading in/out or in agradation of the color. You can set the values for Y (luminance) andC (chrominance) signals separately.

� The addition of dithers will cause a loss of the quality ofimage where there is no contouring artifact. You shouldset the value moderately not to exceed 40.

6.3.5 Quantizer characteristics

This is a parameter for the balance of bit allocation between simpleand complicated part. The range you can set is 0 to 100.

As the value becomes closer to 0, a higher bit amount is allocatedto complicated part of the image. As the value becomes closer to 100,a higher bit amount is allocated to flat part of the image. When thevalue is close to 0, the mosquito noise at the edges (noise causing hazypart along the edges, looking like flying mosquitoes) is less noticeable,but the contouring noise (noise that looks like contour line patterns,which appear in flat and wide areas, such as a dark background) ismore noticeable. The opposite occurs when the value is closer to 100.

The optimal setting depends on the footage, but roughly speaking,16 to 40 is recommended. If the bitrate is relatively high, greatervalue may bring better result.

6.3.6 Letterbox

Letterbox is a format to show an entire wide image of movie on 4:3TV screen, adding black bars at the top and bottom of the screen.If your source is letterboxed, you can improve the precision of 3:2pulldown detection, specifying these black areas to except them fromthe pulldown detection. Or when the noise is on the black areas, youcan remove it and paint them black.

1noise that looks like contour line patterns, which appear in flat and wideareas, such as a dark background


Fill UsageNo Does not paint the area.Black Paints the specified area black.White Paints the area white to confirm the proper size of

the area before applying Black.

� The preview of filling is not displayed. So you need to outputa few second duration to see if the white filling areas coversthe black areas of the source without sticking out, then applyBlack.

6.3.7 Intrablock DC precision

Specify the bit precision of the DC coefficient of intra-blocks. 10is recommended for DVD-Video. This setting works on MPEG-2output.

6.3.8 Block scan order

Specify how to scan 8 × 8 DCT blocks when coding the quantizeddata. This setting works on MPEG-2 output. Generally, Alternatescan is used for the interlaced source, Zigzag is for the progressivesource. If you select Auto, block scan order is automatically selecteddepending on Progressive frame setting (page 80 6.3.11 Progres-sive frame). When Progressive frame is specified, zigzag scan willbe applied on the frame. When Progressive frame is not specified,alternate scan will be applied.

� If MPEG-1 is selected, Zigzag scan is always performed sinceMPEG-1 does not support alternate scan.

6.3.9 Fades on a static scene

It may improve the quality of the image when you get poor encodedresult of the fading in/out on a static scene.


6.3.10 Black screen

Paints the screen black. It can help you to cut the noise on the blackscreen.

6.3.11 Progressive frame

Select when the footage is progressive. This setting works on MPEG-2 output. If you apply Inverse 3:2 pulldown, this setting does notwork.

� Usually, Cinema Craft Encoder SP2 tries motion detection orsampling color information both in fields and frames, then selectone that brings better compression efficiency. When you selectProgressive frame, such operation will be done only in frames.

� If your source is 2:2 pulldowned (film � PAL), it may bring abetter result.

� If you apply this option to interlaced source, it may cause declinein quality.

6.3.12 Monochrome

It discards the chrominance information and outputs black and whiteimage.

6.3.13 Save/Load a parameter set

You can save a parameter set to a file in text format or in CSVformat. To save the list, click Settings Save button. To load the fileyou saved, click Settings Load button.

Chapter 7

Inverse 3:2 pulldownsetting

When you carry out encoding with Inv 3:2 pulldown option, thepulldown pattern will be automatically detected. You can see andmodify the auto detected result or build the pulldown pattern listmanually on Inverse 3:2 pulldown setting window (Fig. 7.3). Or ifyou know the pulldown pattern in advance, you can load the pulldownpattern list from a text file.

To open Inverse 3:2 pulldown setting window, click Settingbutton in the right of Inv 3:2 pulldown option.

7.1 Overview of the setting

The pulldown phases corresponding to 10 fields that fill 4 film frames(A,B,C,D) are named as A1/A2, B1/B2/B3, C1/C2, D1/D2/D3.(See Fig. 7.1) What you should do is to find a proper phase for afield, then build the list of pulldown pattern.

7.1.1 How to find the pulldown phase

To find a pulldown phase of a field, you should first find a combinationof two fields that makes an original frame of film. When you put two

81

82 CHAPTER 7. INVERSE 3:2 PULLDOWN SETTING

24 fps

10 fields

30 fps

A B C D

1 2 3 4 5

A1 A2 B1 B2 B3 C1 C2 D1 D2 D3

Figure 7.1: Pulldown phases of fields

fields that compose a film frame on another, you can see a progressivepicture without jaggies of interlace artifact.

Figure 7.2: Combinations of fields

When you open Inverse 3:2 pulldown setting window, twoviewer windows will also appear. You can see two combination pic-tures of temporally neighbored fields at a time, one is a picture com-bined current and previous field, another is combined current andnext fields. Here just look at Fig. 7.2. Field B3 is a repeat field ofB1, and D3 is also a copy of D1. When you refer B2 field, you can seethe same progressive pictures (combinations of B1/B2 and B2/B3)on the viewers. When you refer D2, the same progressive pictureswill appear as well. So it can be said that there is a rule that thesame progressive pictures appear when referring B2 or D2. Moreover,if you refer the bottom field when the same pictures appear, you can

7.2. MODIFYING THE PULLDOWN LIST 83

tell the pulldown phase of the field B2. If the top field, it is D2.During the single pulldown pattern, you can see the same progressivepictures every 5 fields.

7.1.2 Pulldown list

When you specify the pulldown phase of the field, put it into thepulldown list. After putting several entries, you can trim the list byReduce redundancy button. Reduce redundancy button willsearch the entries that are supposed to be under a single pulldownpattern and bundle them into one entry.

Information of the pulldown list is recorded in ECL file. When en-coding, the inverse 3:2 pulldown is carried out following the pulldownlist.

7.2 Modifying the pulldown list

You can see Pulldown list in the left side of Inverse 3:2 pulldownsetting window. Once you carried out encoding with Inv 3:2 pull-down check, the result of the auto pulldown detection will be listedup. If you have not carried out encoding yet, only an entry thatindicates auto detection is in the list.

� On Adobe Premiere plug-in program, the result of the detec-tion is not displayed automatically when you open Inverse 3:2pulldown setting window. To list the result, you need to loadthe pulldown list manually from the file. The pulldown list isautomatically saved as a text file (.txt) with same name andin the same folder as a video output stream. Click Load buttonto read the pulldown list file.

7.2.1 Explanation of the pulldown list

The following table is an explanation of the list.


Figure 7.3: Inverse 3:2 pulldown setting window


Column Description# Entry numberTC Timecode/Field number

(1: top field / 2: bottom field)Field# Number of fields in the entryPhase Pulldown phase

A1–D3 pulldown phase of the fieldProg progressiveVideo interlaced (not pulldowned part)Auto auto detection

In the list of the auto detected result, you willsee Auto entry in the last. And the last entrywill point the field next to the existing last TC.

The fields in an entry keep a single pulldown pattern. For example,look at Fig. 7.3. Two fields in the first entry are progressive, next6080 fields in the second entry keep a pulldown pattern that startsfrom B1 phase.

7.2.2 Operation buttons for the list

You can modify the list using the buttons below the pulldown list.

Buttons DescriptionUndo Reverses the last change.Redo Cancels the last undo.Cache Puts the content of the list into the memory

and keeps it until closing the window.Recall Loads the list information from the memory.Swap Puts the content of the list into the memory

and loads the list information from the memory.Reduce redundancy If there are multiple entries under a single

pulldown pattern, bundles them into one entry.Discard changes Removes the list, and reloads the initial list.Delete all Removes the list.


� To delete an entry, select an entry and press Delete key. Whenyou delete an entry, the duration included in the entry will becombined to the just previous entry.

� Chapters button does not give an list operation, but opens thechapter list window.

7.2.3 Viewer windows

When you open Inverse 3:2 pulldown setting window, the twoviewer windows will also appear. You can see two combination pic-tures of temporally neighboring fields at a time, one is a picture com-bined current and previous field, the other is combined current andnext fields. On the title bar of a viewer window, combined two fields,each of which is indicated by a frame number (frame number 0 indi-cates the first frame of a source file) and a field order and a source filename, are shown. When the multiple files are input for an output,timecode of the source files will be combined then become consec-utive. However, note that a frame number on the title bar of theviewer window is not combined one, but shows a frame number ineach source file.

For example, when 5B / 6T : test.avi is shown on the title bar,the picture on the viewer is a combination of 5B and 6T fields oftest.avi. 5B indicates the bottom field of the frame number 5, and6T does the top field of the frame number 6.

When you refer to the first or last field of the footage, the previousor next field to make a combination picture does not exist. In thosecases, a black screen will be combined instead of the nonexistent field.If a black screen is combined, the title bar of the viewer shows -1 asa frame number and (null) as a source file name such as -1B / 0T: (null) / test.avi.

7.2.4 Current point and its phase

In this section of the window, you can look into a field and specify apulldown phase of it. The field you refer is combined to the previousand next fields and the combination pictures are displayed on the twoviewer window.


Figure 7.4: Current point and its phase

Search

When some entries already in the list, you can look for an entry thatmeet a condition. If you would like an entry whose duration is lessthan or equal to a certain number of fields, type a number of fields inthe upper box of the Search section and click < = . Similarly, for anentry whose duration is greater than or equal to a certain number offields, type a number of fields in the lower box of the Search sectionand click > = . If you would like an entry that is not pulldownedbut interlaced part of the footage, click Video button.

When you click a button for search, the current point will move tothe first field of the searched entry. You can select the direction ofsearch by the option buttons next to ⇑ ⇓.

Specifying the current point

The currently referred timecode should be put in the box in the upperpart of the section. The field order should be selected by the optionbuttons next to the timecode box. 1 indicates top field, 2 does bottomfield.

< button moves the current point to the previous field, and >button moves to the next. button moves to the first field of thecurrently referred entry.

There are other ways to insert a time point into the timecode box.When you click another entry on the pulldown list, the first field of theentry will be pointed as a current point and its timecode will be putin the box. Or when you drag a timecode from Phase calculationsection of the window, the timecode will be put in the box.


The field referred here is combined to the previous and next fieldsand the combination pictures are displayed on the two viewer window.

Specifying the pulldown phase

The pulldown phase should be specified by selecting an option but-ton. The possible patterns or pulldown phases are arranged with thebuttons below the timecode box. Click one among them.

� When you move the current point that is under a pulldownphase sequence of the entry, you can see the phase will alsomove. For example, if you are in the entry that starts from B1phase, when you move forward the current point field by field,a corresponding phase button will be selected in order such asB1,B2,B3,C1,C2... Until you specify another phase at a point,the phase sequence of currently selected entry on the list will bekept. And if you move to a field that belongs to another entryby typing a timecode directly in the box, the selected entry onthe list is not automatically moved to an entry that includes thecurrently referred field. In that case, when you move forwardthe current point by > , the movement of the phase buttonswill kept a phase sequence of the currently selected entry. Becareful which entry is currently selected.

When you build the pulldown list manually or confirm the resultof the auto detection, you should find a field that shows the sameprogressive combination pictures on two viewers. Then, if the field isthe top field, you can tell it is in D2 phase. If the bottom field, it isin B2 phase.

� If the viewers keep showing the same progressive pictures for awhile, it is supposed to be in Progressive pattern, not a partof the pulldown phase sequence. In that case, you should add aProgressive entry.

� If you cannot see the same progressive pictures on viewers re-currently, it is supposed to be in not pulldowned part. In thatcase, you should add a Interlaced entry.

7.3. PHASE CALCULATION 89

� If you cannot specify the pattern or phase in a part, add anAutomatic detection entry.

Adding and deleting an entry

When you specify the pulldown phase or pattern for a field, add anentry to the pulldown list by clicking Set button. To delete an entry,select an entry in the list and click Delete button.

7.3 Phase calculation

It calculates a phase of a given field from a base point, supposing itis under a single pulldown sequence.

Figure 7.5: Phase calculation

At first, a base point should be put in the top timecode box inPhase calculation section. You should specify the timecode andits field order, and its phase. To specify the phase, click - - buttonin the right several times until the caption of the button shows thephase you would like. Secondly, put the fields that you would like toknow a phase in 3 boxes from the bottom. You can caltulates threepoints at a time.

If you specify a base point, a phase of a given point will be auto-matically calculated. Without a base point in the top box, when youselect a phase at one of given fields by clicking the phase button, thefield will become a base point and phases of other fields are calculated.

� The second box from the top shows the currently referred field.

� A number next to the phase buttons shows how many fieldsapart from a base point.


Operation buttons for the calculator

⇒ Inserts the currently referred timecodeinto the box.

< Moves the current point to the previous field> Moves the current point to the next field.- - , A1 - D3 Specifies a pulldown phase.⇑ Put the result of calculation into the box of

Current point and its phase section.Searches for an entry including the field in the box.

� You can drag and insert a timecode in the box of Currentpoint and its phase into a box in Phase calculation. Andyou can also drag and insert a timecode from a box in Phasecalculation into a box in Current point and its phase.However, the information of a field order and a pulldown phasewill not be copied.

� When you use < > button at the top box, the phase buttonwill show a same phase in another field order. That is, if yousee A1 on the button then click > , the field order changesto the bottom field and you will see C1 on the button. If B1phase, you will see D1 . Note that the phase button of thetop box does not follow the pulldown sequence in order.

7.4 Threshould setting

You can change the threshold values for rebuilding the pulldown list.If you carry out Auto 2-pass detection, you should set them beforeencoding. After the auto detection, you can rebuild the pulldown listby changing the threshold values.

7.4.1 Pulldown information file

When you carry out encoding with selected Pulldown info file checkbox, Cinema Craft Encoder SP2 will look into every interlaced field

7.4. THRESHOULD SETTING 91

Figure 7.6: Threshold setting

and analyze the correlation with the fields listed below and record theinformation to the pulldown information file. The extension of thefile is .i32.

• 2 fields apart• 1 field apart

If you can see close correlation with a field apart from 2, it will besupposed an interlaced frame. If close correlation with a next field,it will be supposed a progressive frame. Two parameters describedlater specify at which threshold correlation will be considered close.

7.4.2 Progressive frame pairing

It specifies the threshold correlation between fields that can make aprogressive frame. The default value is 80 (%). The larger the valuebecome, the closer correlation is required and the more frames will beencoded as interlaced. The smaller the value, the more frames will beencoded as progressive. Note that an incorrect combination of fieldsmay be allowed to be encoded as a progressive frame with too smallvalue. On the other hand, too large value may cause a decline in thecoding efficiency. The value between 70 and 90 will be recommended.

7.4.3 Repeat field tolerance

A repeated field of 3:2 pulldown should be basically the same as anoriginal field that is 2 fields apart. However, it tends to become differ-ent because of filtering or the compression when editing or recording.This value specifies the difference threshold to which extent the differ-ences can be ingnored between a repeated field and an original field.The value between 4 and 16 is recommended.


7.4.4 Initial setting

Sets the values back to the initial ones. The default values are Pro-gressive frame pairing = 80 and Repeat field tolerance = 6.

7.4.5 Apply

It rebuilds the pulldown list applying the threshould values. It justchanges the list information. To apply the changes to VAF file, youneed to carry out encoding.

� To carry out this button, i32 file is required.

7.4.6 Detail

It outputs the inverse 3:2 pulldown information to a text file. Thecontent of the inforamtion is listed below. It will be useful to modifya chapter point.

• Information of every field recorded in i32 file• The result of auto 3:2 pulldown detection• Combination of fields that will be encoded as a frame

7.5 Saving/Loading the pulldown list

Saving the pulldown list

You can save the pulldown list to a file in text or CSV format. Tosave a list, click Save button, and type a file name.

Loading the pulldown list

You can load a pulldown list saved to a file using Load button. Whenyou know the pulldown pattern in advance, it will be efficient thatyou make a pulldown list in text or CSV format and load it.

In a pulldown list file, a timecode, a field number and a pulldownpattern should be in a line. The formats are the following :

7.6. EFFECTS OF CHAPTER POINTS 93

text format CSV format01:00:00:00.0 Prog 01,00,00,00,0,Prog01:00:00:01.0 B1 01,00,00,01,0,B1

� Be careful that the notation of field number differs from thaton the setting window. In a pulldown list file, 0 indicates thetop field and 1 indicates the bottom field.

� About the description of the pulldown pattern, see Explana-tion of the pulldown list on page 83.

� If a pulldown phase does not match a field number in an entry,a pulldown list cannot be loaded correctly.

� If you load a pulldown list when some entries are already on thelist, the entries in a file are added leaving the existing entries.

7.6 Effects of chapter points

Cinema Craft Encoder SP2 gives priority to putting a chapter pointat the location specified over keeping a proper pulldown sequence.Even if the source keeps a simple pulldown sequence through thefootage, Cinema Craft Encoder SP2 will break it and replace it withvideo or progressive sequence to put a chapter point. (However, ifthe interval between chapter points becomes less than 6 frames, youwill not always put a chapter point there.) If the pulldown list isunexpectedly untidy with so many entries, it may be caused by thechapter points put in the improper locations. When you encounterthe case, it will be better to check the chapter points. If you needsome modifications, you can call chapter list window by chaptersbutton.

7.6.1 Improper frame for a chapter point

Chapter points on bottom field first frame

If a chapter point you set is on B3/C1 or C2/D1 frame (see Fig. 7.7),Cinema Craft Encoder SP2 cannot perform inverse 3:2 pulldown atthat point. And if a chapter point you set is on D2/D3 frame, Cinema


Figure 7.7: Combinations of fields

Craft Encoder SP2 does not perform inverse 3:2 pulldown at thatpoint in order not to make the frame bottom field first. In both cases,B3/C1 and C2/D1 frames are encoded as interlaced video. Thereforeit is recommended that you set chapter points only on A orB frames.

Chapter points on C or D frame

When the source is 24p, you should specify chapter points using thetimebase of 24 fps even if you would like to apply 3:2 pulldown whileencoding. Looking at Fig. 7.7, you will see that the C and D frameswill be encoded as bottom field first. Since chapter points must be ontop field first frames, Cinema Craft Encoder SP2 rearranges the pull-down pattern if you set a chapter point on C or D so that the targetframe will be encoded as A or B. If you don’t like this rearrangement,avoid setting a chapter point on C or D frame.

� If you apply 3:2 pulldown provided by Cinema Craft EncoderSP2, it’s better setting chapters only on the frames that meetthe conditions below.

7.6. EFFECTS OF CHAPTER POINTS 95

Pattern Reminder of frame number divided by 42:3 0 or 13:2 0 or 3

� To see a pulldown list resulted from 3:2 pulldown by CinemaCraft Encoder SP2, once carry out encoding, then select 29.97and click Setting button next to Inv 3:2 pulldown check boxto open Inverse 3:2 pulldown setting window. Note thatyou cannot change the pulldown list resulted from 3:2 pulldownby Cinema Craft Encoder SP2 to apply when encoding.

A frame made of split fields

In Fig. 7.7, video frames B3/C1 and C2/D1 combine the fields thatis not proper pair for a progressive frame, and shows not a clearimage but a confusing picture mixing two images. You should avoidsetting a chapter point on such frame made of split fields.

� You can easily confirm a frame whether it is made of split fieldsor not when you refer a chapter point on any viewer window.

Chapter 8

Chapter setting

To set chapter points, click Chapters button in Encoder setting(Fig. 3.1 on page 22) window to show Chapter list window (Fig. 8.1).You can also open Chapter list window from Picture settingswindow or Inverse 3:2 pulldown setting window.

Figure 8.1: Chapter list window

97

98 CHAPTER 8. CHAPTER SETTING

8.1 What is “Chapter” ?

DVD players offer random playback using an index called “chapter”.In MPEG, since a playback starts only at I frame, a chapter point

should be encoded as I frame. And a sequence header and a GOPheader will be inserted and a closed GOP (each frame in the GOPdoes not refer to the data outside the GOP) will be made up to playback at a chapter point smoothly.

� This setting is to insert an I frame for a chapter at any locationyou would like. To use it as a DVD chapter, you need to set itin your authoring software.

� When you apply 3:2 pulldown of Cinema Craft Encoder SP2and select timecode for the base unit, you should set a chapterpoint by original timecode used in the source (23.976 fps or 24fps). And a chapter point should be set 6 frames apart fromthe previous one. If closer than 6 frames to another, it canhappen that Cinema Craft Encoder SP2 refuses the location ofthe point.

� About the cautions for chapter setting when applying inverse3:2 pulldown, see page 93.

8.2 Creating chapter list

When you open Chapter list window, the viewer window will alsoopen to show the source image of the specific time point. On thetitle bar of the viewer window, the file name of the source is shown.When you select timecode as a base unit for Frame search, a framenumber of a reference point (frame number 0 indicates the first frame)is shown as well.

� The base unit of the time is timecode only.

� There are not a slider for the viewer and a playback buttonhere. To search a source image of a time point, click up anddown arrow buttons next to the box for the timecode or type aspecific timecode in the box.

8.2. CREATING CHAPTER LIST 99

� When the multiple files are input for an output, Chapter listwindow will show you combined timecode of them. However,note that a frame number on the title bar of the viewer windowis not combined one, but shows a frame number in each sourcefile.

Adding a chapter point

To search a chapter point with the viewer, click up and down arrowbuttons next to the box for the timecode or type a specific timecodein the box. When you find the time point of the chapter, type achapter name in the text box on the right of the timecode box andclick Register button. Then the time point is set as a chapter pointand added to the chapter list. To apply changes you have made, clickApply button. OK button will also apply changes and then closethe window.

� A chapter name you enter here will not be embedded in outputMPEG, however, it will help you follow a chapter list.

� A chapter name is not essential item for chapter setting.

Modifying a chapter name

Select a chapter point in the list, then modify the name in the textbox and click Set button.

To apply changes you have made, click Apply button. OK buttonwill also apply changes and then close the window.

Deleting a chapter point

Select a chapter point in the list, then modify the name in the textbox and click Delete button.

To apply changes you have made, click Apply button. OK buttonwill also apply changes and then close the window.

100 CHAPTER 8. CHAPTER SETTING

8.3 Creating chapter list file

When you know the timecodes for chapter points in advance, it willbe efficient that you make a chapter list in a text or CSV format andload it.

In a chapter list file, a timecode and a chapter name (not essential)should be in a line. The formats are the following:

text format CSV format01:00:00:00 Chap1 01,00,00,00,Chap101:00:01:00 Chap2 01,00,01,00,Chap201:00:02:00 Chap3 01,00,02,00,Chap3

8.4 Loading and saving chapter list

You can export a chapter list to a file in a text or CSV format. Toexport a list, click Save button and specify a file name. To load achapter list file, click Load button.

� If you load a chapter list when some entries are already on thelist, the points in a file are added leaving the existing points.

Chapter 9

Optional settings

In this chapter, we describe parameters that can be set from optionmenu.

9.1 Input

In Input page, you can set parameters that relates to file input.

File decoding

Cinema Craft Encoder SP2 tries first to decode a movie file in YUY2format. If it works, it can skip color space conversion into YCbCr,which causes loss of image quality, (But all of the movie files cannotbe decoded in YUY2 format.) By clear Try decoding in YUY2format of the page shown inFig. 9.1, you can avoid this action. How-ever, it is not necessary to change it unless you find some problemsdecoding in YUY2 format.

9.1.1 Reading method for QuickTime movies

Since the frame rate of QuickTime file may change in the middle ofthe stream, some frames may be used more than once, and some maynot be used at all. To avoid this, select Read frames at constant

101

102 CHAPTER 9. OPTIONAL SETTINGS

Figure 9.1: Input setting tab

frame rate check box. Note that some QuickTime files may not beproperly read with this option.

9.2 Output

Select the default output destination of a stream from the following:

• Same folder as input fileThe same folder as an input file

• Current folderThe current folder selected when Cinema Craft Encoder SP2starts

• Folder specified belowThe specified folder

� This setting will be applied to the source files loaded later.When you load a source and open Encoder setting window,

9.3. SETTING EXTENSIONS 103

Figure 9.2: Output setting tab

you will see the folder specified here in the textbox of Outputfiles section.

9.3 Setting extensions

You can change extensions for the following output files.

• Encoder control list• MPEG-2 Video elementary stream• MPEG-1 Video elementary stream• MPEG-2 Program stream• MPEG-1 System stream• MPEG-1 Audio elementary stream• Video CD• Super Video CD• Video information file


Figure 9.3: Extensions setting tab

9.4 Miscellaneous

Log file

To outpu a log file, enter a log file name and its log level. The log willbe appended to the specified file every time you carry out encoding.

How to see the log file.

Log level 1 includes the following information:

• Version number of the program• Input file names• Output file names• Codec information• Encoding start time• Encoding end time• Performance

9.4. MISCELLANEOUS 105

Figure 9.4: Misc setting tab

• Source duration• Elapsed time• A breakdown of time elapsed

In addition, log level 3 outputs the following information:

• Encoded frame numbers and field numbers at every GOP en-coded.

• cur: Average bitrate of current GOP.• avg: Average bitrate so far.• I: Q scale of I picture in a GOP.• P: Average Q scale of P pictures in a GOP.• B: Average Q scale of B pictures in a GOP.• (closed/open): Whether the GOP is closed or open.


Action when encoding is finished

Cinema Craft Encoder SP2 can shutdown the system or play soundwhen finished encoding. However, this setting will not work whenyou selected Encode Selected Items.

Optimization

Generally, SSE2 instruction set will improve the processing speed.However, some of the processors will lose the performance with SSE2.If you encounter the case, remove the check of this option.

Chapter 10

Template settings

You can keep frequently used parameter sets as templates up to 36.And you can export them to CSV files.

Figure 10.1: 6 templates are registered

107

108 CHAPTER 10. TEMPLATE SETTINGS

10.1 Applying template

To set up a template, select Setting in Template menu to showTemplates window. When an input file is registered to main window,the template selected here is applied.

Use the last used parameter set

A parameter set last used are applied.

Use the selected template

You can select a template from the template list. In Fig. 10.1, sixtemplates are registered.

� The currently selected template is shown in the status bar ofthe main window.

10.2 Operation at template setting win-

dow

Initially, the templates listed below are registered. You can create anew template modifying one of them.

Standard The initially provided settingNTSC DVD Standard + frame rate (29.97)NTSC DVD w IVTC Standard + Inverse 3:2 pulldown

+ GOP (M=3, N/M=4)NTSC DVD multiangle Standard + DVD Video multianglePAL DVD Standard + frame rate (25)

+ GOP (M=3, N/M=4)PAL DVD multiangle PAL DVD + DVD Video multiangle

� Select Setting in Template menu of main window. In Tem-plates window, select Use the selected template, and thendouble-click an item in the list to show Template setting win-dow (Fig. 10.2).

10.2. OPERATION AT TEMPLATE SETTING WINDOW 109

Figure 10.2: Template setting window


� Change parameters.

� Save the template.

When adding to a template list

Click OK button and return to Templates window. The tem-plate is registered as the name that you entered at Title textbox. Keep in mind that the existing template with the samename will be overwritten.

When saving to a file

Click Save button to show a common dialog box, then enter afile name and a template name, and then click Save button tosave it to the file.

10.3 Operation at encoder setting window

� Open Encoder setting window and set parameters.

� Click Save button to show Save parameters window (Fig. 10.3).

When adding to a template list

Select Save to template, and <Create new template...>,and then click OK button. When Template name settingwindow appears (Fig. 10.4), enter template name, and then clickOK button to close the window.

When saving to a file

Click Save button to show a common dialog box, then enter afile name and a template name, and then click Save button tosave it to the file. In this case, the parameter set is not addedto the template list.

10.3. OPERATION AT ENCODER SETTING WINDOW 111

Figure 10.3: Save parameters window

Figure 10.4: Template name setting window


10.4 Editing template

To edit a template, follow the instructions below:

� Select Setting in Template menu in the main window. Selectand double-click an item in the list you want to edit and showTemplate setting window.

� Modify parameters.

� Click OK button to save the setting.

� To edit a template exported in a file, click Load button in theTemplate setting window to load the template.

10.5 Deleting template

To delete a template, follow the instructions below:

� Click the template from Template menu in Select and double-click an item in the list you want to edit and show Templatesetting window.

� Select an item you want to delete in the list, and then pressDelete key.

� If you delete all of the templates, initially provided 6 templateswill be automatically add to the list when you re-open the win-dow.

10.6 Editing template list

To edit the template list, follow the instructions below:

� Select Setting in Template menu in the main window andopen Template setting window. Select an item in the list youwould like to change the position, and then click buttonsin the upper right of the list to move up and down.

10.6. EDITING TEMPLATE LIST 113

� Click OK button to save the setting.

� To add a template to the list from file, click Load button inTemplates window.

Chapter 11

Encoder control list

Encoder control list is the list of encoding setting informations for theinput files loaded in the main window. You can save it as a projectfile called ECL file.

� An extension of ECL file is initially set as .ecl, however, youcan change it. How to change an extension is described at 9.3Setting extensions on page 103.

� See page 126 to save ECL file from plug-in.

11.1 Creating and saving encoder control

list

Encoder control list is created by the following procedure.

� Load the files to be encoded into the main window, and set theparameters for them. About the basic operation, see Chapter1 Getting started on page 5.

� After setting all of files to be batch encoded, select Save asfrom File menu of the main window or click button.

115

116 CHAPTER 11. ENCODER CONTROL LIST

11.2 Loading encoder control list

To load he saved ECL file, do one of the following:

• Drag an ECL file on the main window of Cinema Craft EncoderSP2 from Explorer.

• Select Open in File menu of Cinema Craft Encoder SP2.• Click the button in the toolbar of the main window.

11.3 Editing encoder control list

To load the saved ECL file, do one of the following:

• Drag ECL file(s) onto the main window from Windows Explorer.

• Select Project from Open of the File menu.

• Click button in the top of the window.

11.3.1 Editing ECL

Shortcut menu

When right-clicked on the list of main window,a menu as shown in the left figure will appear.(It’s the same as Edit menu.)

• Encode Selected ItemsEncodes the selected items in the list.

� To encode all of the items, select Encode in File menu,or click button in the toolbar of the main window.

11.3. EDITING ENCODER CONTROL LIST 117

• AddAdds an input file to the end of the list.

• SettingOpens Encoder setting window to modify parameters for theselected item.

• AppendOpens a subwindow. In the subwindow, you can input mul-tiple files to combine and produce one MPEG file. Also, youcan trim each movie file by double-clicking an item or selectingTrimming in the shortcut menu to show Video trimmingwindow.

� For details, please see Chapter 2 Input file setup.

• UpMoves up the position of the selected file in the encoder controllist . This command cannot work if the selected file is already atthe top. The same operation can be done by clicking buttonin the toorbar in the main window.

• DownMoves down the position of the selected file in the encoder con-trol list. This command cannot work if the selected file is al-ready at the bottom. The same operation can be done by click-ing button in the toorbar in the main window.

• CopyCopies and adds the selected file in the encoder control list.The same operation can be done by clicking button in thetoorbar in the main window.

• DeleteDeletes the selected file(s) in the encoder control list. The sameoperation can be done by clicking button in the toorbar inthe main window.

• PropertyShows the detailed information of input files and output files ofthe selected item.

118 CHAPTER 11. ENCODER CONTROL LIST

� When you select two or more files in the encoder control list,only Encode Selected Items, Add and Delete commandsare available.

11.4 Encode

To encode all of the items in the list, select Encode in File menu,or click button in the toolbar.

To save a pulldown list to an ECL file, re-save the encoder controllist after finished encoding.

Chapter 12

Adobe Premiere plug-in

This chapter explains Cinema Craft Encoder SP2 for Adobe PremierePro. To setup Cinema Craft Encoder SP2 for Adobe Premiere, selectthe check box for Adobe Premiere and specify the install destina-tion on Install option dialog of setup program.

12.1 The basic operation

To use Cinema Craft Encoder SP2 plug-in for Adobe Premiere Pro,follow the instructions below.

� From File menu, select Export, then select Movie... to openExport Movie dialog box. (Fig. 12.1 on page 120)

� Click Settings... button and open Movie Export Settingsdialog box (Fig. 12.3 on page 122).

� Click each item in the list of Export Movie Settings windowto check parameters. After checking the following contents, clickOK button and return to Movie Export window.

General

• File TypeSelect General from the left menu, and select Cinema

119

120 CHAPTER 12. ADOBE PREMIERE PLUG-IN

Figure 12.1: Movie Export menu

Figure 12.2: Export Movie window

12.1. THE BASIC OPERATION 121

Craft MPEG from the File Type list.

• Compile SettingsClick Compile Settings button to show Encoder Con-trol window (Fig. 12.4). In this window, specify input/outputECL file settings and template selection. Also, if Onlyshow the Encoder setting dialog when the Shift keyis down. is selected, Cinema Craft Encoder SP2 will startencoding without showing Encoder setting window.

• RangeSelect encoding range.

• ExportTo output audio, select Export Audio option.

Video

Select Video from the left menu, and then check video set-tings. The parameters that set here will be passed to CinemaCraft Encoder SP2. Since Cinema Craft Encoder SP2 does notperform frame rate conversion, set proper frame rate here. In-correct Pixel Aspect Ratio setting may cause resizing andblanking.

� To perform 3:2 pulldown (converts film source to NTSCvideo) on Cinema Craft Encoder SP2 by selecting Pull-down, frame rate must be 23.976 for both on Adobe Pre-miere and Cinema Craft Encoder SP2.

� When using 720 × 486 source for inverse 3:2 pulldown,frame size must be set to 720 × 486, otherwise, inverse3:2 pulldown fails. Also, please see Cautions on inverse3:2 pulldown on page 126.

Keyframe and Rendering

Select Keyframe and Rendering from the left menu, andthen check keyframe and rendering settings.


Figure 12.3: Movie Export Settings window

Figure 12.4: Encoder Control window


Figure 12.5: Movie Export Settings window (Video)

� Cinema Craft Encoder SP2 ignores field order setting inRendering Options.

� If you select Deinterlace Video Footage here, don’t se-lect Deinterlacing on Cinema Craft Encoder SP2, other-wise, it decreases image quality.

� If you select Add keyframes at Markers, marked frameswill be encoded as I frames. However, to perform 3:2 pull-down, you must specify I frames at Chapter list on Cin-ema Craft Encoder SP2.

Audio

Select Audio from the left menu, and then check audio settings.The parameters that set here will be passed to Cinema CraftEncoder SP2. Please make sure that settings here match thoseon Cinema Craft Encoder SP2.

� Since Cinema Craft Encoder SP2 does not perform sam-pling frequency conversion, set proper sampling frequencyhere. If sampling frequency here does not match that onCinema Craft Encoder SP2, encoding will fail.


Figure 12.6: Movie Export Settings window (Keyframe and Render-ing)

Figure 12.7: Movie Export Settings window (Audio)


� Click OK button and return to Export Movie window.

� Click Save button on Export Movie window (Fig. 12.2 onpage 120), then Encoder setting will open.

Figure 12.8: Plug-in Encoder setting window

� Set parameters on Encoder setting window, and then clickOK or Encode Now button to start encoding.

� When OK button is pressed, Cinema Craft Encoder SP2starts encoding after closing Encoder setting window.

� When Encode Now button is pressed, Cinema Craft En-coder SP2 starts encoding without closing Encoder set-ting window.


� When Cancel button is pressed, Cinema Craft EncoderSP2 closes the window without encoding.

12.2 Encoder settings on plug-in

Most of the settings are same as that on standalon version. However,please be careful about the following things.

Parameters to be set on Adobe Premiere

Sampling frequency and frame rate should be set on Adobe Premiere.

Parameters to be set on both sides

Audio mode (stereo/monaural), frame size, and deinterlacing settingexists on both side. Please be careful not to be converted twice. Fordetails, refer to the previous section.

Cautions on inverse 3:2 pulldown

When performing inverse 3:2 pulldown on 720× 486 source, cautionsmust be made with the Preset Project selection and Output framesize setting on Adobe Premiere. With any other preset and framesize, pulldown detection will fail.

Premiere Preset Output frame size

Pro/Pro1.5 Video for Windows 720 (480)×486720 × 486

Pro2.0 Custom Desktop 720 (480)×486720 × 486

Saving ECL

You can save encoder setting parameters to an ECL file if you selectCreate ECL file and specify filename at Encoder Control window(Fig. 12.4).

Cinema Craft Encoder SP2 saves ECL file when

12.3. ENCODER SETTINGS FOR PLUG-IN 127

• OK button on Encoder setting window is clicked.

• finished encoding.

� When Encode Now button is clicked, Cinema Craft EncoderSP2 does not save parameters to ECL file.

� If you would like to save parameters without encoding, clickOK button, and then cancel encoding.

12.3 Encoder settings for plug-in

Some operations differ from that of standalone program.

Any viewer windows of Cinema Craft Encoder SP2 are not avail-able in plug-in program. You should confirm the timecodesusing the monitor of Premiere before opening the window ofCinema Craft Encoder SP2.

Plug-in program does not use ECL file, so the project includingchapter list, pulldown list and entry list of picture setting willnot be saved in a project file. You need to save those list as atext or CSV file manually and load them when setting.

Note that only the pulldown list will be automatically saved asa text file with same name and in the same folder as a videooutput stream when encoding. When you open Inverse 3:2pulldown setting window to edit a pulldown list, you need toload this file manually.

� You can load just previous list data, selecting ones at templateselecting window.

Chapter 13

Removing Cinema CraftEncoder SP2

To remove Cinema Craft Encoder SP2 from the system, do the fol-lowing:

� Open the control panel.

� Start Add/Remove Programs applet.

� Select “Cinema Craft Encoder SP2” from the list box in theInstall/Uninstall page, and click Add/Remove... button.

129

Chapter 14

Troubleshooting

14.1 VBR bitrate

When encoded in VBR mode, actual bitrate may be less than thatspecified.

The cause considered

Case 1 The number of encoding passes was not enough to correctbit allocation.

Case 2 The pictures of the movie was so simple that given bitratewas not necessary to achive the best quality.

Solution

Case 1• Increase the encoding pass.

Case 2• Clear Disable adaptive Q-matrix switching option in the

Video settings window.• Turn off all of the filter settings.• Change GOP configuration and increase I pictures. However,

too many I frames may affect picture quality.

131

132 CHAPTER 14. TROUBLESHOOTING

14.2 VBV buffer overflow

Under certain circumstances, VBV overflow may occur during encod-ing. VBV is a hypothetical decoder that is conceptually connectedto the output of the encoder. Encoder stops encoding when the VBVcauses buffer overflow.

Solution

• Change the GOP configuration (Ex. M = 3, N/M = 5 � M =3, N/M = 4).

• Increase or decrease video bitrate.• Change filter settings.• Edit source movie to eliminate problematic frames.

14.3 Mux buffer overflow

Under certain circumstances, Mux overflow may occur during multi-plexing video and audio. When it happens, encoding stops.

Solution

• Increase or decrease audio bitrate.• Change filter settings.• Edit source movie to eliminate problematic frames.• Output video and audio separately as elementary streams, and

then multiplex them by using external multiplexer to producesystem stream.

ccesp2

Documents

pulldown setting

pulldown list

pulldown information

chapter list

audio setting

simple setting

pulldown phase

setting extensions