Rec. ITU-R BT.1620 1

RECOMMENDATION ITU-R BT.1620

Data structure for DV-based audio, data and compressed video at a data rate of 100 Mbit/s (Question ITU-R 12/6) (2003)

The ITU Radiocommunication Assembly, considering a) that applications within professional television production and post-production have been identified where DV-based video compression can offer operational and economic advantages when compared to serial digital interface (SDI)-based operations; b) that three data rates have been proposed within the same compression family to serve different applications (25 Mbit/s, 50 Mbit/s and 100 Mbit/s); c) that the sampling rasters for each of the three applications are different; d) that for the international exchange of high-definition programme material the ITU-R recommends the application of Recommendation ITU-R BT.709; e) that audio, auxiliary data and metadata elements form an integral part of these applications; f) that these elements are multiplexed into a single data stream for transport and further processing; g) that the compression quality and functional characteristics must be identical and reproducible in complex production chains; h) that for this purpose all details of parameters used for coding and multiplexing must be defined, recommends 1 that for applications in professional television production and post-production using DV-based compression at 100 Mbit/s, the parameters given in the SMPTE 370M-2002 − Data Structure for DV-Based Audio, Data and Compressed Video at 100 Mb/s − 1080/60i, 1080/50i and 720/60p, be used. NOTE 1 – SMPTE 370M-2002 includes a normative reference to SMPTE 296M − 1280 × 720 Progressive Image Sample Structure − Analog and Digital Representation and Analog Interface. The following formats listed in Table 1 of SMPTE 296M shall not be considered part of this Recommendation.

Table 1 item System nomenclature Frame rate 3 1280 × 720/50 50 6 1280 × 720/25 25 7 1280 × 720/24 24 8 1280 × 720/23.98 24/1.001

2 Rec. ITU-R BT.1620

Summary of SMPTE Standard 370M-2002

This Standard defines the DV-based data structure at a 100 Mbit/s sampling rate for the interface of digital audio, sub-code data and compressed video with the following parameters:

− 1080/60i system: 1920 × 1080 image sampling structure, 59.94 Hz field rate, interlace format.

− 1080/50i system: 1920 × 1080 image sampling structure, 50 Hz field rate, interlace format.

− 720/60p system: 1280 × 720 image sampling structure, 59.94 Hz frame rate, progressive format.

The Standard does not define the DV-based data structure for the interface of digital audio, sub-code data and compressed video with the following parameters:

− 720/50p system: 1280 × 720 image sampling structure, 50 Hz frame rate, progressive format.

NOTE 1 − SMPTE Standards 370M-2002 and 296M-2001 are given in Annexes 1 and 2. SMPTE Standards 370M-2002 and 296M-2001 refer to Versions 2002 and 2001 respectively only, which are the versions approved by Administrations of Member States of the ITU in application of Resolution ITU-R 1-3 on 03-05- 03. By agreement between ITU and SMPTE, these Versions were provided and authorized for use by SMPTE and accepted by ITU-R for inclusion in this Recommendation. Any subsequent versions of SMPTE Standard 370M and 296M which have not been accepted and approved by Radiocommunication Study Group 6 are not part of this Recommendation. For subsequent versions of SMPTE documents, the reader should consult the SMPTE website: http://www.smpte.org/.

Rec. ITU-R BT.1620 3 Annex 1

SMPTE 370M-2002

SMPTE STANDARD  ' !    !#  (!  $   ))*+ ).)+2)().)+3)(4)+2)

Page 1 of 62 pages

Table of contents                           !"     "#    $$    !     %  & " % 

1 Scope  '                   !#  "  (    (     ))*+,'     -    !#        "     &# "    . /01(    (  "#    ).)+2)().)+3)( 4)+2),  '""#         5  ).)+2)%  M   57)).) " "  (37,7#01 (    , $    5))*+  ).)+3)%  M   57)).) " "  (3)#01 (    , $    5))*+  4)+2)%  M   5.)4) " "  (37,7#01  ( "  , $    5))*+     ( 2)#01%        ).)+2) 4)+2)%  8  (  3)#01%    %  ).)+3)%  ,'  ).)#%      ).)+2) ).)+3)%  (( 4)#%    %  4)+2)%  , 

$% " '))%'0&:$&';:<   *:': =$'>& !'&?&: & @ && 373A,0  ,(A = ( ;)2)4 D 7()) B7C42#))

4 Rec. ITU-R BT.1620

SMPTE 370M-2002

2 Normative references  '"  (   "     (      ,     E   (     "             "   "    % %"          ,  &#77(  '  <  '#$ ? %  !"  !   +*='&*#777'(  < N'  $ $  *='&2)*#777('N3+2)0"#! =   %  N!"      #=      *='&4*#77.('N7)).) "  " = !"      *   =    *='&72*#774('N.)4)= "  "   N " !"      "   

3 Data processing  3.1 General

" (  (         " !#   ,   %     !<           " "     , !       "           , !           '  ,   / " !#   ,<" ,     %      ,  3.1.1 Video encoding parameter  '   "     %      %*='& 4* *='&72*,

3.1.2 Audio encoding parameter  ' "   ./01( 2# -  1 %&,  3.1.3 Subcode data  '         *='&*,  3.1.4 Frame structure  ).)+2) ).)+3)%  (  (   (        ,'        # "  , 4)+2)%  (                    ).)+2) %  , $-  %(         %  ).)+2)%  ,  &                           ).)# %  (    4)+2)%  ,'  (   ).)+2) 4)+2) %     ,

Page 2 of 62 pages

Rec. ITU-R BT.1620 5

SMPTE 370M-2002



 /"+ ?$+  " !$' A" " G  1   " 4* $   " <  72*

.+)  .    

  !<     !<

  &  "

  *   Figure 1 – Data processing block diagram  

 3.2 Data structure  '          "    " ,'       !< ,  & !<  )!<-   2)01%   !<-   3)#01 %  ,  & !<-      (  (>H (  (     "!</  %8   0   5!</    5!</  >H 5!</    57!</   53!</     "  (   !< /     #%  !  44 %    , !<   %     ) 47,<"      !<- , 

Page 3 of 62 pages

6 Rec. ITU-R BT.1620

SMPTE 370M-2002

!  

<      '   <  

!<-  !<- )() !<- () !<- #() !<- )( !<- #(

!<-   !<  

   !< -  0       >H   I 

!</ 0)() $)() $() )() () () )() )() () () ()

%    !</  !<   )########################47    !</ ! !     J) 2)01%   J 3)01%    Figure 2 – Data structure 

Page 4 of 62 pages

Rec. ITU-R BT.1620 7

SMPTE 370M-2002

!</ 0)( $)( $( )( ( (

 )( )( ( ( ( ( 3( 2( 4( .( 7( )( ( ( ( (

( 3( 2( 4( .( 7( )( ( ( ( ( 3( 2( 4( .( 7(

( )( ( ( ( ( 3( 2( 4( .( 7( )( ( ( ( (

( 3( 2( 4( .( 7( 3)( 3( 3( 3( 3( 33( 32( 34( 3.( 37(

( 2)( 2( 2( 2( 2( 23( 22( 24( 2.( 27( 4)( 4( 4( 4( 4(

3( 43( 42( 44( 4.( 47( .)( .( .( .( .( .3( .2( .4( ..( .7(

2( 7)( 7( 7( 7( 7( 73( 72( 74( 7.( 77( ))( )( )( )( )(

4( )3( )2( )4( ).( )7( )( ( ( ( ( 3( 2( 4( .( 7(

 .( )( ( ( ( ( 3( 2( 4( .( 7( )( ( ( ( (

 !</       5!<    J)(((  0)(  5!</     $)( $( 5!</    )( ( 5!</>H   )( .( 5!</    )( ( 5!</    Figure 3 – Data structure of a DIF sequence  

3.3 Header section  3.3.1 ID  '!  !</    (" ( % B!)(!(!C, '  !   !</, 

Page 5 of 62 pages

8 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 1 – ID data of a DIF block   %     )  !) ! ! *  $' !- ! 4  $' !- ! 2 $') !- ! 3  !-) !    <$ !    <= !     !  ?     ! )    '!  "5  $'5   %B C  !-5 !<-   B  C  <$(<=5 $    !</B 3C  :'&M<=   *='&*  ! 5 !</  B 2C   5   %   5         !            Table 2 – Section type    %    % $' $' $') ) ) ) 0   ) )    )  ) >H )      ) )   )     )       

Page 6 of 62 pages

Rec. ITU-R BT.1620 9

SMPTE 370M-2002

 Table 3 – DIF sequence number for the 60-Hz system

!<-     !<-    !- !- !- !-) ) ) ) ) ) ) ) )   ) )  )  ) )    )  ) )  )  )  3 )   ) 2 )    4  ) ) ) .  ) )  7  )  )     )        ) )      )        )           

Table 4 – DIF sequence number for the 50 Hz system

!<-     !<-    !- !- !- !-) ) ) ) ) ) ) ) )   ) )  )  ) )    )  ) )  )  )  3 )   ) 2 )    4  ) ) ) .  ) )  7  )  ) )  )      ) )      )        )             Table 5 – DIF channel number  <$ <= !<    )  )5      5  ) ) 5     ) 5  

Page 7 of 62 pages

10 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 6 – DIF block number  !</    !</   ! 4 ! 2 ! 3 !  !  !  !  ! ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )   ) ) ) ) ) )  )  ) ) ) ) ) )           ) ) ) )   )   ) ) ) )       5 5 5 5 5 5 5 5 5              3.3.2 Data  '  B % C !</     4, %  4    % . 47   ,   Table 7 - Data (payload) in the header section  %        3 2 4 . ####### 47 *  !<  '< '< '<  ------  )      ------        ------        ------        ------   =' = = =  ------   =' = = =  ------ ?   =') =) =) =)  ------    !<5!<-  "  )J)!<-   !< B2)#01%  C  J!<-   !< B3)#01%  C  ='( =( =(  =           /   ! B=' J ))(=J))(=J))(=J))C(  "     !  " $, "   /(        ,  '<5'   " "  '<5'   " " !</  '<5'   " ">H !</  '<5'   " " !</  )J   J  ,  5      !       ,

Page 8 of 62 pages

Rec. ITU-R BT.1620 11

SMPTE 370M-2002

3.4 Subcode section

3.4.1 ID  '!  !</       ,,,'  %  )),

3.4.2 Data

'     B % C    !< /          "  , '      2; ( .% "(      7%  !</,;  !<

-    ) ,& ;   ; !-  % ( <<(   ;   % 3% ,  %    )  3)3 47 7%  $)( !  

!

).7 2 4  3   3) ; ) ;  ;  ;  ;  ; 3

%    )  3)3 47 7%  $( !  

!

).7 2 4  3   3) ; 2 ; 4 ; . ; 7 ; ) ; 

; ; !) ! << ;  .%     Figure 4 – Data in the subcode section

Page 9 of 62 pages

12 Rec. ITU-R BT.1620

SMPTE 370M-2002

3.4.2.1 SSYB ID

' .  ; !B!)(!C,  <!(  !B=(=(=)C(B='( ='(=')C( ;   B%(%(%(%)C,   Table 8 – SSYB ID  ;    ;    ;      ) 2  3 4     !) ! !) ! !) ! 4 <  <  <   2 =     ='   3 =     ='    =)      =')     %  %  %   %  %  %   %  %  % )   %)   %)   %) :'&M J  %   <5'          !< ,   J     !<    )J   !<     '    !<    !<-   )(((( 2)01%     !<-   )(((((3 3)01%      '  !<    !<-   3(2(4(.(7 2)01%     !<-   2(4(.(7()( 3)01%         (     ,

3.4.2.2 SSYB data  & ;   %    /3%  " 3,' 7  /   B=$)%  " 1 C,' )  / "  ;    !< ,     ; ; << ;  !) !

3% 

= /

=$) =$ =$ =$ =$   Figure 5 – Pack in SSYB 

Page 10 of 62 pages

Rec. ITU-R BT.1620 13

SMPTE 370M-2002

Table 9 – Pack header table  >==&            )))) ))) ))) )) ))) )) )) ) N  ?:A&   >!: !&:  )))) :>$& :>$&    >!: !&:  ))) :>$& :>$&  $: ':? $: ':?   )))    '*&  )) $:!&     ;  ))) @:>=    ))         : <:     Table 10 – Mapping of packets in SSYB data  '    '  ;     !<   !<  )                '$ '$  @   3 '$   2     4     .     7 '$ '$ ) @    '$   :'& '$J'  /,  @J  %"   /,  J!         , '$  @       , '     ?$' %   3.4.2.2.1 Time code pack (TC)

'   "    /,'        /       ,

Page 11 of 62 pages

14 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 11 – Mapping of time code pack 2)01%   * ?  =$) ) ) )  ) )   '&  =$ $< !< > '<*& <*& '&  =$ =$ > '&$: ! &$: ! '&  =$ @<) > '* >'& * >'& '&  =$ @< @< > '0:> 0:>  3)01%   * ?  =$) ) ) )  ) )   '&  =$ $<   > '<*& <*& '&  =$ @<) > '&$: ! &$: ! '&  =$ @< > '* >'& * >'& '&  =$ =$ @< > '0:> 0:>  :'&M!  " +*='&*,  $<5$    )J % 1   J% 1   !<5!   "  )J       J!      =$5   /  %    )J&  J:  @<5  %"   "   5  %   3.4.2.2.2 Binary group pack (BG)  '   " %"   /,  %"       %"   /      ,

Table 12 – Mapping of binary group pack   * ?  =$) ) ) )  )  ) ) =$  ;@:>=  ;@:>= =$  ;@:>=  ;@:>= =$  ;@:>=2  ;@:>=3 =$  ;@:>=.  ;@:>=4 

Page 12 of 62 pages

Rec. ITU-R BT.1620 15

SMPTE 370M-2002

3.5 VAUX section  3.5.1 ID  '!  !</ >H     ,,,'  %  )),

3.5.2 Data

'  B % C !</ >H " 2,'"   >H / "    !<- ,  '  3 /( 3% "(    %  >H!</ % ,  

      %   <<,  '  (   3 / !<- ,>H / !</ -   %  ) ,'     /  ,  '   " >H / >H!</,>H  /BC   >H   /B$C     ,'  ">H / !</ 

!<-              <<, 

>H      (  : <: /( <<(    ,   Byte position number ).....33.22.44.47

)( ! )     3 2 4 . 7 )    

( ! 3 2 4 . 7 )     3 2 4 . 7

( ! )     3 2 4 . 7 )    

= /  = / = /   

=$) =$ =$ =$ =$   Figure 6 – Data in the VAUX section  

Page 13 of 62 pages

16 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 13 – Mapping of VAUX pack in a DIF sequence  = /    = /  &!<-  :!<-  7 )  )  $

 &!<- 5  !<-   )(((2(. 2)#01%    !<-   )(((2(.() 3)#01%     :!<- 5  !<-   ((3(4(7 2)#01%    !<-   ((3(4(7( 3)#01%    3.5.2.1 VAUX source pack (VS)

'   " >H  /,   Table 14 – Mapping of VAUX source pack  * ?  =$) )   ) ) ) ) ) =$         =$         =$   3)+2) ';=& =$ )          3)+2)5  )J2)01%    J3)01%  

';=&5"  %  < 2)01%    )))J).)+2)#))*+  B  ).)C  ))J).)+2)#))*+  B  )3C  )))J4)+2)#))*+    :  J   < 3)01%    )))J).)+3)#))*+    :  J    5      !       ,  3.5.2.2 VAUX source control pack

' 3 ">H   /, 

Page 14 of 62 pages

Rec. ITU-R BT.1620 17

SMPTE 370M-2002

Table 15 – Mapping of VAUX source control pack  * ?  =$) )   ) ) ) )  =$ $@*       =$   ) )  != =$ << < <$    ) ) =$           $@*5$%"   "  %    ))J$%   :  J   !=5! %    ))J257  :  J   <<5< + " <  ).)%  B 2C <<       (      "  B 2C   )J:%       J      , <  4)%  B 4C <<              (           "    ,   )J:%     ,   J       ,  <5<  + " <  ).)%  B 2C   <    "  B 2C   )J<    J< , <  4)%  B 4C   <     "   ,   )J  ,   J  ,   Table 16 – FF/FS for the 1080-line system  << < :     <       B(- C,  ) <       B(- C, )  <   , ) ) <   , 

Page 15 of 62 pages

18 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 17 – FF/FS for the 720 line system  << < :               B(- C,  )          B(- C, )      , ) )     ,    <$5<  " " <  ).)%   <$                  ,   )J          J!          <  4)%     <$                    ,   )J           J!            5      !       ,  3.6 Audio section  3.6.1 ID  '!  !</       ,,,'  %  ),

3.6.2 Data

'  B % C !</   " 4,'  !</     3%    % B>HC 4%      %    ,2,, ,2,,,

 %    ) 4. 47

!    %     Figure 7 – Data in the audio section

3.6.2.1 Audio encoding

3.6.2.1.1 Source coding

&     "      ./01(   2#  -  1 , ' %     "    ,            /,

Page 16 of 62 pages

Rec. ITU-R BT.1620 19

SMPTE 370M-2002

3.6.2.1.2 Emphasis

' "         #  3)+3µ,<    "   "(        ,

3.6.2.1.3 Audio error code

     ( .)))    "                , '     "            %        ,

A   .)))(    .)),

3.6.2.1.4 Relative audio-video timing

).)%  M    "     -       3)    1         ,  4)%  M    "     -       3)    1          ,

3.6.2.1.5 Audio frame processing  '      ,&    2) 2))    2)#01%   7)    3)#01%          (  (   % ,<  2)#01%  (           # -  5  2))(2)(2)(2)(2) ,

:           2)      2)#01 %     7      3)#01 %  ,'           %  ,  3.6.2.2 Audio shuffling

' 2#            % , '   %    * (     %   ? ( " .,     !<-  !</  

 ,' %   !BJ)(((,,,,,C  #         % !  ,  '      "    % "- 5  2)01%  M   !<   5J)5 $0($0  J5 $0($0  J5 $03($02  J5 $04($0.  !<-   5B 'B+CKB CC 3  $0($0($03($04   B 'B+CKB CC 3K3  $0($0($02($0.   !</  5B CK 'BB 3C+3C 

Page 17 of 62 pages

20 Rec. ITU-R BT.1620

SMPTE 370M-2002

 %    5 .K 'B+3C    "  %  7K 'B+3C    "  %     J) 27  3)01%  M   !<   5J)5 $0($0  J5 $0($0   J5 $03($02  J5 $04($0.  !<-   5B 'B+CKB CC 2  $0($0($03($04 B 'B+CKB CC 2K2  $0($0($02($0.   !</  5B CK 'BB 3C+.C   %    5 .K 'B+3C    "  %  7K 'B+3C    "  %     J) 7    * 2  ? 3)7. 423)

> ? 3)7. 423)

.  .    Figure 8 – Conversion of audio sample to audio data bytes    3.6.2.3 Audio auxiliary data (AAUX)

>H        " 4 7,'>H /    >H /     B>H % C,'"  >H /  3%   " 7(  >H / "  ,= /   ) . " 7,'       /  ,   ' .   "   >H  /, >H    /BC   >H      /B$C        ,   

Page 18 of 62 pages

Rec. ITU-R BT.1620 21

SMPTE 370M-2002

 %    ) 4. 47 !    %  

3% 

)(   /  ) (   /   (   /   (   /   (   /   3(   /  3 2(   /  2 4(   /  4 .(   /  .

= / = /   

=$) =$ =$ =$ =$   Figure 9 – Arrangement of AAUX packs in audio auxiliary data  

Table 18 – Mapping of AAUX pack in a DIF sequence    /     &!<-  :!<-  = /   )    $   &!<- 5  !<-   )(((2(. 2)#01%   !<-   )(((2(.() 3)#01%     :!<- 5 !<-   ((3(4(7 2)#01%   !<-   ((3(4(7( 3)#01%     3.6.2.3.1 AAUX source pack (AS)  '>H  /"  ' 7, 

Page 19 of 62 pages

22 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 19 – Mapping of AAUX source pack

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source control pack (ASC)

'>H   /"  ' ),

Page 20 of 62 pages

Rec. ITU-R BT.1620 23

SMPTE 370M-2002

Table 20 – Mapping of AAUX source control pack   * ?  =$) )  )  ) ) )  =$ $@*     &<$ <! <!     =$ &$ &$ & & ' & ! ' & ! =$ !< =&&! =$          $@*5$%"   "  %     ))J$%    :  J   &<$5&     "   ))J&     )J&     :  J    &<$     /,  &$'5 "     )J "     J    "       "  (&$')      /-   3 2!<-     ,  &$& !5 "    )J "    J    "      " (&$& !)  "    / -  3 2!<-     ,  <!&'5< "  "     )J< "   J< "  ' <!& '    %      "     B &$ ' J ) C, <!&'    "  (    "          ""   ,<!&')    "  (     "    ,  <!&& !5< "  "    )J< "   J< "  ' <!& & !    %      "    B &$ & ! J ) C,  <!&& !    " (    "          ""   ,<!&& !)    " (    "    ,  !<5!   "   )J      J<      =&&!5 'B C

Page 21 of 62 pages

24 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 21 – SPEED code definition  $   ' * ?  2)#01%   3)#01%   ))))))) )+)BJ)C )+))BJ)C )))))) +) +)) 5 5 5 )))) ))+) ))+))BJC 5 5   ))) )+)BJC   5     )      !   !    5      !       ,  3.7 Video section

3.7.1 ID

'!  !</      ,,,'  %  )),  3.7.2 Data  !    B % C    !< /        44 %           (  ,    %      ,'44 %         /,  3.7.2.1 DIF block and compressed macro block  $  !</     /$*((E(/    2)#01%      3)#01%  ,  '  "       !< /       /   5  2)01%  M   BJ)8M8KKCN  BJ)8M8KKCN  B/J)8/M48/KKCN  B J)8 M38 KKCN   JBKK KC )8  JBKK K2C )8  JBKK K.C )8  JBKK K)C )8  JBKK KC )8 ! -JB3 K3/C 38 ! J 'BB3 K3/K243C+3C8   ! -(! J$*( ((/  B! -KC(! J$*(((/

Page 22 of 62 pages

Rec. ITU-R BT.1620 25

SMPTE 370M-2002

 B! -KC(! J$*(((/  B! -KC(! J$*(()(/  B! -KC(! J$*(((/ O O O O     ! -5!</    ! 5!<-     5!/  ( 5      /  /5*  /    /  3)01%  M   BJ)8M8KKCN  B/J)8/M48/KKCN  BJ)8M8KKCN   JBKKC 8  JBKK2C 8  JBKK.C 8  JBKK)C 8  JBKKC 8 ! -JB3K33/C 38 ! J 'BB3K33/C+3C8   ! -(! J$*( ((/  B! -KC(! J$*(((/  B! -KC(! J$*(((/  B! -KC(! J$*(()(/  B! -KC(! J$*(((/ O O O  B/J)8/M48/KKCN ! -J3/8 ! J8   ! -()! J$*)(()(/  B! -KC()! J$*)(((/  B! -KC()! J$*)(((/  B! -KC()! J$*)(((/  B! -KC()! J$*)(((/ O     ! -5!</    ! 5!<-     5!/  5      /  /5*  /    /

Page 23 of 62 pages

26 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 22 – Video DIF blocks and compressed macro blocks for the 60-Hz system

!<  !<-  $   !</        / )() $*)((() () $*)(2(() () $*)(.(() ) () $*)()()() ) () $*)((() 5 5 5 5 5 5 5 7 () $*)(((2 )( $*(2(() ( $*()(() ( $*((() ) ( $*(()()  ( $*(.(() 5 5 5 5 5 5 5 7 ( $*(4((2 5 5 5 5 )( $*((() ( $*(.(() ( $*()(() ) ( $*(()()  ( $*(2(() 5 5 5 5 5 5 5 7 ( $*(3((2 

Page 24 of 62 pages

Rec. ITU-R BT.1620 27

SMPTE 370M-2002

Table 23 – Video DIF blocks and compressed macro blocks for the 50-Hz system

!<  !<-  $   !</        / )() $*)((() () $*)(2(() () $*)(.(() ) () $*)()()() () $*)((() 5 5 ) 5 5 5 5 5 ) () $*)(((2 )() $*)(()() () $*)((()  5 5 () $*)(((2 )( $*(2(() ( $*()(() ( $*((() ) ( $*(()() ( $*(.(() 5 5  5 5 5 5 5 ) ( $*(4((2 )( ዉ  5 5 ( ዉ 5 5 5 5 )( $*((() ( $*(4(() ( $*(7(() ) ( $*(()() ( $*(3(() 5 5  5 5 5 5 5 ) ( $*(((2 )( ዉ  5 5 ( ዉ 

Page 25 of 62 pages

28 Rec. ITU-R BT.1620

SMPTE 370M-2002

4 Video compression

'          "    ).)+2) %  (  ).)+3) %  (   4)+2)%  ,  4.1 Video structure  4.1.1 Video sampling structure  ' "  %*='&4*  ).)#%  ( *='&72*   4)#%  ,'    B;C   # " B$($ C  ,    )#   .     %   "  B    "/" C,  4.1.1.1 Video frame pixel structure  ).)+2)%  M  ' " " ;"  7'    1 %  "  8   'J,))+B4,3)2C   7)   72)  # "       

" ),' " "     $ $ "        " "      ;" ,&          B#3). 3)4C% " *   " ,  ).)+3)%  M  ' " " ;"  7'    1 %  "  8   'J+B4,3)2C   7)   72)  # "       

" ,' " "     $ $ "        " "      ;" ,&          B#3). 3)4C% " *   " ,  4)+2)%  M  ' " " ;"  2)'    1 %  "  8   'J,))+B4,3)2C   .)   2)  # "       

" ,' " "     $ $ "        " "      ;" ,&          B#3). 3)4C% " *   " ,  4.1.1.2 Video frame line structure   ).)%  M

3)    ;( $(  $  "             , '            , 

Page 26 of 62 pages

Rec. ITU-R BT.1620 29

SMPTE 370M-2002

4)%  M

4) ;($ $ "          ,'          ,  4.1.1.3 Horizontal resampling   ).)+2)%  M

7) 1 % ;"     .),'72) 1 % $  $ "     2),'  "            -  .    ,B  ,C  ).)+3)%  M

7) 1 % ;"     ),'72) 1 % $  $ "     4),'  "            -  .    ,B  ,C  4)+2)%  M

'.) 1 % ;"     72),'2) 1 % $  $  "        .) , '    "               -  .    ,B  ,C   Table 24 – Construction of input video  ).)+2)%   ).)+3)%   4)+2)%   ; 4,3+,))*01 4,3*01 4,3+,))*01  " - % $($  4,3+,))*01 4,3*01 4,3+,))*01 ; )) 23) '     $($  )) .3 '          ; 7) .)  $($  72) 2) '       3 43) '        ).) 4) < 32) '     2 43 <3.  G  1  &   %-  1 )  ;($ $ ,    )7 '   "  "  57) ; G  1.44  -  1 "  /52 $($  " " %53 G  1.74  

Page 27 of 62 pages

30 Rec. ITU-R BT.1620

SMPTE 370M-2002

 Input Luminance (Y) T

line 21 First active line in a field line 584

line 22

Output line 585 3 Luminance (Y) 2 T

line 21 First active line in a field line 584

line 22

line 585

Input Color difference (CR, CB) 2T

line 21 First active line in a field line 584

line 22

Output line 585 Color difference (CR, CB) 3T

line 21 First active line in a field line 584

line 22

line 585

T = 1.001/74.25µs

First pixel in active period 

Figure 10 – Sampling structure for the 1080/60i system

Page 28 of 62 pages

Rec. ITU-R BT.1620 31

SMPTE 370M-2002

 Input Luminance (Y) T

line 21 First active line in a field line 584

line 22

Output line 585 4 Luminance (Y) 3 T

line 21 First active line in a field line 584

line 22

line 585

Input Color difference (CR, CB) 2T

line 21 First active line in a field line 584

line 22

Output line 585 8 Color difference (CR, CB) 3 T

line 21 First active line in a field line 584

line 22

line 585

T = 1/74.25µs

First pixel in active period 

Figure 11 – Sampling structure for the 1080/50i system



Page 29 of 62 pages

32 Rec. ITU-R BT.1620

SMPTE 370M-2002

 Input Luminance (Y) T

First active line line 26 in a frame line 27

line 28

Output line 29 4 Luminance (Y) 3 T

First active line line 26 in a frame line 27

line 28

line 29

Input Color difference (CR, CB) 2T

First active line line 26 in a frame line 27

line 28

Output line 29 8 Color difference (CR, CB) 3 T

First active line line 26 in a frame line 27

line 28

line 29

T = 1.001/74.25µs

First pixel in active period

Figure 12 – Sampling structure for the 720/60p system   4.1.2 DCT block 

';($( $     !$'/ "    ).)%  ( "   4)#%  ,!$'/       "     "       "   1       , '         1                 %             , <   ).)# %  (%J)(((2   1 ( %J((3(4    ,  !$'/ "    

Page 30 of 62 pages

Rec. ITU-R BT.1620 33

SMPTE 370M-2002

).)+2)%  M  ' "   1 !$'/  " 3,'  1  "     3!$'/     ,=    ))!$'/,   ;53  !$'/2) 1 !$'/J2))!$'/

$53  !$'/.) 1 !$'/J).))!$'/ $53  !$'/.) 1 !$'/J).))!$'/  ).)+3)%  M

' "   1 !$'/  " 2,'  1  "     3!$'/     ,=    .2))!$'/,   ;53  !$'/.) 1 !$'/J))!$'/

$53  !$'/7) 1 !$'/J3)!$'/ $53  !$'/7) 1 !$'/J3)!$'/  4)+2)%  M

' "   1 !$'/  " 4,'  1  "     7)!$'/     ,=    2))!$'/,   ;57)  !$'/) 1 !$'/J).))!$'/

$57)  !$'/2) 1 !$'/J3))!$'/ $57)  !$'/2) 1 !$'/J3))!$'/   4.1.3 Macro block  &   / " !$'/,<" .  ).)#%   " 7  4)# %        / !$'/,  4.1.3.1 Arrangement of macro block  ).)+2)%  M  *  / "       ,   5 ""  /  =    3))  / " ),  &   /      /  !$'/;  1 %

  % E  (   % E  !$'/$    % E  !$'/$   ' 8   (24    /.) 1   /J32)  /,  &       /     1 % E   !$' / ;(   1 % E  

!$'/$   1 % E  !$'/$  ' 8   (    /) 1   /J)  /,

Page 31 of 62 pages

34 Rec. ITU-R BT.1620

SMPTE 370M-2002

  5 ""  /    ")  /  ) 4   ")  /  . 3  " " ),  )   /  2  "         /  )  1    /  2  % " )8   (2)    /7) 1   /J3))  /  ).)+3)%  M  *  / "       ,   5 ""  /  =    2)43  / " ,  &   /      /  !$'/;  1 %

  % E  (   % E  !$'/$    % E  !$'/$   ' 8   (24    /7) 1   /J2))  /,  &       /     1 % E   !$' / ;(   1 % E  

!$'/$   1 % E  !$'/$  ' 8   (    /3 1   /J3  /,   5 ""  /  *  /        "  ,'"      /)     / " ,'       "/8   (    522    /7) 1   /J37)  / "  5    /3 1   /J3  /  4)+2)%  M  =    4))  / " 8   (3    /2) 1   /J4))  /  4.1.3.2 Divided blocks  ).)+2)%  M  *  /      %/ " ,&   %/0    / 1 %   /  %,  0  %/0    / 5  !/5J)50 ( J50 (K

Page 32 of 62 pages

Rec. ITU-R BT.1620 35

SMPTE 370M-2002

J50 K( J50 K(K    ( J)(((,,,(7  J)((((      (            /, &   /    )     /3 1   /,  ).)+3)%  M  *  /        %/ " ,&   %/0   1 % E    /,  0  %/0    / 5   !/5J)50 ( J50 (K J50 K( J50 K(K   ( J)(((,,,( J)((((      (            /, &   /          /3 1   /,  4)+2)%  M  *  /      %/ <" 3,&   %/ 0   / 1 %   /  %,  0  %/0    / 5  !/5J)50 ( J50 (K J50 K3( J50 K3(K    ( J)(((,,,(   J)((((      (      /,& / 3     /) 1   /,

Page 33 of 62 pages

36 Rec. ITU-R BT.1620

SMPTE 370M-2002



Left x Right Top 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 Field 1

0,1 1,1 2,1 3,1 4,1 5,1 6,1 7,1 Field 2

0,2 1,2 2,2 3,2 4,2 5,2 6,2 7,2 Field 1

y 0,3 1,3 2,3 3,3 4,3 5,3 6,3 7,3 Field 2

0,4 1,4 2,4 3,4 4,4 5,4 6,4 7,4 Field 1

0,5 1,5 2,5 3,5 4,5 5,5 6,5 7,5 Field 2

0,6 1,6 2,6 3,6 4,6 5,6 6,6 7,6 Field 1

Bottom 0,7 1,7 2,7 3,7 4,7 5,7 6,7 7,7 Field 2

Pixel x = 6 y = 7   Figure 13 – DCT block and the pixel coordinates for the 1080-line system



Left x Right Top 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0

0,1 1,1 2,1 3,1 4,1 5,1 6,1 7,1

0,2 1,2 2,2 3,2 4,2 5,2 6,2 7,2

y 0,3 1,3 2,3 3,3 4,3 5,3 6,3 7,3

0,4 1,4 2,4 3,4 4,4 5,4 6,4 7,4

0,5 1,5 2,5 3,5 4,5 5,5 6,5 7,5

0,6 1,6 2,6 3,6 4,6 5,6 6,6 7,6

Bottom 0,7 1,7 2,7 3,7 4,7 5,7 6,7 7,7

Pixel x = 6 y = 7  Figure 14 – DCT block and the pixel coordinates for the 720-line system

Page 34 of 62 pages

Rec. ITU-R BT.1620 37

SMPTE 370M-2002

 Luminance DCT block Left 160 DCT blocks Right Top

Color difference DCT block 8x8 pixels Left 80 DCT blocks Right Top    Figure 15 – DCT block arrangement for the 1080/60i system     Luminance DCT block Left 180 DCT blocks Right Top

Color difference DCT block 8x8 pixels Left 90 DCT blocks Right Top    Figure 16 – DCT block arrangement for the 1080/50i system     Luminance DCT block Left 120 DCT blocks Right Top

Color difference DCT block 8x8 pixels Left 60 DCT blocks Right Top    Figure 17 – DCT block arrangement for the 720/60p system  

Page 35 of 62 pages

38 Rec. ITU-R BT.1620

SMPTE 370M-2002

  CB CR DCT4 DCT4 DCT5 DCT6 DCT7 ' Y DCT5 DCT0 DCT1 CB CR DCT6 Y DCT2 DCT3 DCT7 DCT0 DCT1 DCT2 DCT3   Left Right Left Right     / 

Figure 18 – Macro block and DCT blocks for the 1080-line system         CB CR DCT4 ' Y DCT5 DCT0 DCT1 DCT6 DCT2 DCT3 DCT7   Left Right

Figure 19 – Macro block and DCT blocks for the 720-line system

Page 36 of 62 pages

Rec. ITU-R BT.1620 39

SMPTE 370M-2002

Step1: Arranging macro blocks

10 10 10 10 10 10 10 10 macro blocks

A0 A1 A2 A3 A4 A5 A6 A7 4 macro blocks

60 macro blocks

A8 A9 A10 A11 A12 A13 A14 A15 3 macro blocks A16 1 macro block

Bottom macro blocks

Step2: Rearranging macro blocks 90 macro blocks

A0 4 macro blocks A1 4 macro blocks A2 4 macro blocks A3 4 macro blocks A4 4 macro blocks A5 4 macro blocks A6 4 macro blocks A7 4 macro blocks A8 3 macro blocks A9 3 macro blocks 60 A10 3 macro blocks macro A11 3 macro blocks A12 3 macro blocks blocks A13 3 macro blocks A14 3 macro blocks A15 3 macro blocks B16 4 macro blocks

10 macro blocks

Rearranging A16 to B16 A16 B16 0 1 2 3 38 39 0 1 2 8 9 10 11 12 18 19 20 21 22 28 29 30 31 32 38 39

Figure 20 – Arrangement of macro blocks for the 1080/60i system

Page 37 of 62 pages

40 Rec. ITU-R BT.1620

SMPTE 370M-2002

Step1: Arranging macro blocks

90 macro blocks

A0 1 macro block

66 macro blocks

A1 1 macro block Bottom macro blocks 45 macro blocks

Step2: Rearranging macro blocks

main unit 90 macro blocks

66 macro blocks

edge unit 90 macro blocks 45 macro blocks

A0 A1 1 macro block    Figure 21 – Arrangement of macro blocks for the 1080/50i system

Page 38 of 62 pages

Rec. ITU-R BT.1620 41

SMPTE 370M-2002

     60 macro blocks

45 macro blocks

    Figure 22 – Arrangement of macro blocks for the 720/60p system    

Page 39 of 62 pages

42 Rec. ITU-R BT.1620

SMPTE 370M-2002

7)  /

0)() 0)( 0)( 0)(7   / 0() 0( 0( 0(7   / 0() 0( 0( 0(7   /

2)   /

037() 037( 037( 037(7   /

777 7  /

3  / 3  /

0)() 0)( 0)(. 0)( 0)( 0)(7 0() 0( 0(. 0( 0( 0(7 ) )     / /

03.() 03.( 03.(. 03.( 03.( 03.(7

divided block h=0 divided block h=1

3  / 3  /

0() 0( 0(. 0( 0( 0(7 0() 0( 0(. 0( 0( 0(7 ) )     / /

037() 037( 037(. 037( 037( 037(7

divided block h=2 divided block h=3    Figure 23 – Divided blocks for the 1080/60i system

Page 40 of 62 pages

Rec. ITU-R BT.1620 43

SMPTE 370M-2002

main unit 7)  /

0)() 0)( 0)( 0)(7   / 0() 0( 0( 0(7   / 0() 0( 0( 0(7   /

22   /

023() 023( 023( 023(7   /

777 7  /

3  / 3  /

0)() 0)( 0)(. 0)( 0)( 0)(7 0() 0( 0(. 0( 0( 0(7       / /

02() 02( 02(. 02( 02( 02(7

divided block h=0 divided block h=1

3  / 3  /

0() 0( 0(. 0( 0( 0(7 0() 0( 0(. 0( 0( 0(7       / /

023() 023( 023(. 023( 023( 023(7

divided block h=2 divided block h=3    Figure 24 – Divided blocks for the 1080/50i system

Page 41 of 62 pages

44 Rec. ITU-R BT.1620

SMPTE 370M-2002

2)  / 2)  /

0)() 0)( 0)( 0)(7    03() 03( 03( 03(7    / / 0() 0( 0( 0(7  02() 02( 02( 02(7 

3 3     / /

0() 0( 0( 0(7  0.7() 0.7( 0.7( 0.7(7 

2 22 2 2 22 2     / /    

)  / )  /

0)() 0)( 0)(. 0)( 0)( 0)(7 0() 0( 0(. 0( 0( 0(7 3 3     / /

0() 0( 0(. 0( 0( 0(7

divided block h=0 divided block h=1

)  / )  /

03() 03( 03(. 03( 03( 03(7 02() 02( 02(. 02( 02( 02(7 3 3     / /

0.7() 0.7( 0.7(. 0.7( 0.7( 0.7(7

divided block h=2 divided block h=3    Figure 25 – Divided blocks for the 720/60p system

Page 42 of 62 pages

Rec. ITU-R BT.1620 45

SMPTE 370M-2002

4.1.4 Super block  &   / 4  /,  ).)+2)%  M  ' "    / /" 2,' /   3)  /,   )    /3 1   /J3)  /,  ).)+3)%  M  ' "    / /" .,' /   33  /,       /3 1   /J33  /,  ' "    3  /,       /3 1   /J3  /,  4)+2)%  M  ' "    / /" ),' /   3)  /,   )    /3 1   /J3)  /,  4.1.5 Definition of super block number, macro block number and value of the pixel    /  M'  /    ((E" 2(.( ),  ((E 5 / J)(G( 5        / J)(G(7 2)#01%    J)(G( 3)#01%   E5  1      / EJ)(G(  *  /  M'  /    *((E(/,'% /   /      /" 4  ).)+2)%  (" 7  ).)+3)%  (  "   4)+2)%  ,'   " "    /(        " /,  *((E(/ ((E5   /   /5   /     //J)(G(2  = M'   =((E(/(B(%C,'    ((E( /(B(%C,'%  !$'/     /" . 7,'  " "  !$'/(  !$'     " ,'%  %      !$'/  ,,,   =((E(/(B(%C   ((E(/5   /   5 !$'/     / B(%C5    !$'/J)(G(4%J)(G(4

Page 43 of 62 pages

46 Rec. ITU-R BT.1620

SMPTE 370M-2002

? E " ) ' ) ()() ()( ()( ()( ()(     (() (( (( (( (( /  (() (( (( (( ((

 (() (( (( (( ((

 (() (( (( (( ((

3 (3() (3( (3( (3( (3(

2 (2() (2( (2( (2( (2(  4 (4() (4( (4( (4( (4(

. (.() (.( (.( (.( (.(

7 (7() (7( (7( (7( (7(     /J7 7  / EJ     Figure 26 – Super blocks and macro blocks in a divided block for the 1080/60i system         /((EBJ)(,,,((J)(,,,(7(EJ)(,,,(C

)     3 2 4 .

7 )     3 2 4

. 7 )     3 2

/    Figure 27 – Macro block order in a super block for the 1080/60i system

Page 44 of 62 pages

Rec. ITU-R BT.1620 47

SMPTE 370M-2002

/

? E " ) ' ) ()() ()( ()( ()( ()(     (() (( (( (( (( /  (() (( (( (( ((

 (() (( (( (( ((

 (() (( (( (( ((

3 (3() (3( (3( (3( (3(

2 (2() (2( (2( (2( (2(

 4 (4() (4( (4( (4( (4( . (.() (.( (.( (.( (.(

7 (7() (7( (7( (7( (7(

) ()() ()( ()( ()( ()(     /J) 7  / EJ "  )(() )(( )(( )(( )((    Figure 28 – Super blocks and macro blocks for the 1080/50i system       /((EBJ)(,,,((J)(,,,()(EJ)(,,,(C

)     3 2 4 .

7 )     3 2 4 / . 7 )     3 2

  /)((EBEJ)(,,,(C

)   ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 2    Figure 29 – Macro block order in a super block for the 1080/50i system

Page 45 of 62 pages

48 Rec. ITU-R BT.1620

SMPTE 370M-2002

? E " ) ' ) ()() ()( ()( ()( ()(

 (() (( (( (( ((

 (() (( (( (( ((

 (() (( (( (( ((

 (() (( (( (( ((

3 (3() (3( (3( (3( (3(

2 (2() (2( (2( (2( (2(  4 (4() (4( (4( (4( (4(

. (.() (.( (.( (.( (.(

7 (7() (7( (7( (7( (7(  

  /J7 EJ

  /J4  /   Figure 30 – Super blocks and macro blocks in a divided block for the 720/60p system

   /((EBJ)(,,,((J)(,,,(7(EJ)(,,,(C

)     3 2 4 . 7 )     3 2 4 . 7 )    /  3 2 )     3 2 4 . 7 )     3 2 4 . 7 )     3 2   Figure 31 – Macro block order in a super block for the 720/60p system

Page 46 of 62 pages

Rec. ITU-R BT.1620 49

SMPTE 370M-2002

4.1.6 Definition of video segment and compressed macro block  "     /            ,  2)#01%  M  *( ((/  JBKC )(J *((-(/ JBK2C )(-J *(( (/ JBK.C )( J *(((/ JBK)C )(J) *(( (/ JBKC )( J   5 /J)(G( 5        /J)(G(7 /5   /     //J)(G(2  3)#01%  M  /  *( ((/  JBKC (J *((-(/ JBK2C (-J *(( (/ JBK.C ( J *(((/ JBK)C (J) *(( (/ JBKC ( J   5 /J)(G( 5        /J)(G() /5   /     //J)(G(2  "    *( ((/ J)( J(J) *((-(/ J)(J(-J *(( (/ J)(J( J *(((/ J)(J(J *(( (/ J)(J( J   /5   /     //J)(G(2

 &   "                 ((/     * ( ((/8 * ((-(/8*(( (/8*(((/8 *(( (/,  '  #          -   %    * ( ((/  * (( (/, '      "            .3#%      ,                    /, &       /    44 %      $*,& "       #      $((/  $*( ((/8$*((-(/8$*(( (/8$*(((/8 $*(( (/ 5  $*( ((/5  ' /                        / * ( ((/  %        /*((-(/8 *(( (/8 *(((/8 *(( (/,

Page 47 of 62 pages

50 Rec. ITU-R BT.1620

SMPTE 370M-2002

$*((-(/5  ' /                        / * ((-(/  %        /*( ((/8 *(( (/8 *(((/8 *(( (/,  $*(( (/5  ' /                        / * (( (/  %        /*( ((/8 *((-(/8 *(((/8 *(( (/,  $*(((/5  ' /                        / * (((/  %        /*( ((/8 *((-(/8 *(( (/8 *(( (/,  $*(( (/5  ' /                        / * (( (/  %        /*( ((/8 *((-(/8 *(( (/8 *(((/,  4.2 DCT processing  <    "   1                 !$' /   ).)# %  ,&"  "  1       !$'/ 4)#%  ,  '!$'    2 !$'/   ((E(/(B(%C 2     ((E(/(B (C  5  =((E(/(B(%C     $((E(/(B (C     ,  <  J) J)(   !$ ,      $ ,  4.2.1 DCT mode  <  ).)#%  (     !$'       -  %        , '      .#.# #!$'    .#.##!$' , ' .#.# # !$'          ,'.#.## !$'          ",  <  4)#%  ( .#.# #!$'   ,  ' !$'    !$'/   /,  " ( .#.##!$'  (     E  !$'/   "   # "!$'/       ,  '"!$' "   "        !$' ( .#.# #!$'  .#.##!$' ,  !$'5 44 $((E(/(B (CJ$BC$B C55B=((E(/(B(%C$:B4B%KC+2C$:B4 BKC+2CC %J)J)

Page 48 of 62 pages

Rec. ITU-R BT.1620 51

SMPTE 370M-2002

 !$'5 44 =((E(/(B(%CJ55B$BC$B C$((E(/(B (C$:B4B%KC+2C$:B4 BKC+2CC J) J)   5 $B CJ),3+   J) $B CJ),3  J 4 $BCJ),3+  J) $BCJ),3 J 4    DCT blocks Rearranged DCT blocks

: Field 1 pixels : Field 2 pixels    Figure 32 – Rearrangement of pixels in the 8-8-field-DCT mode   4.2.2 Weighting  '!$' $((E(/(B (C " %-  1  ,!  -  1        "    "  "   ).)+2)%  ("    ).)+3)%  ( " 3  4)+2)%  ,  4.2.3 Output order  <" 2      "  , 

Page 49 of 62 pages

52 Rec. ITU-R BT.1620

SMPTE 370M-2002

?   $   horizontal horizontal

)324 )324 ) . 2 4 . . 7   ) . 2 4 3 2 2    2 4 . . 7 .  3  2 4 3 3 2 .  7  4 . 7 7 )  3 .  4 3 2 4 )  7 72  . . 7 )   2 7  3 3 4 )  . 7 74 Vertical  . 7 )    . ) Vertical  2 2 )  . .2 7) 3 7 .     7. ) 3 2 .  . .2 4474)7 2   3 2 . 7. )72 2   7 7 7747 4  3 . 7 ))2 4  7 72 74))72  Figure 33 – Quantizer matrix for the 1080/60i system   ?   $   horizontal horizontal

)324 )324 ) . 2 4 . . 7   ) . 2 4 3 2 2    2 4 . . 7 .  3  2 4 3 3 2 .  7  4 . 7 7 )  3 .  4 3 2 4 )  7 72  . . 7 )   2 7  3 3 4 )  . 7 74 Vertical  . 7 )    . ) Vertical  2 2 )  . .2 7) 3 7 .     7. ) 3 2 .  . .2 4474)7 2   3 2 . 7. )72 2   7 7 7747 4  3 . 7 ))2 4  7 72 74))72   Figure 34 – Quantizer matrix for the 1080/50i system   ?   $   horizontal horizontal

)324 )324 ) . 2 4 . . 7   ) .  2 2 2 . . ..  2 4 . . 7 .  2.   2 2 2 . 42 .2 .  4 . 7 7 )  2. 72  2 2 . . .) . .7  . . 7 )  2 7 7.  2 2 . .) . 2..27 Vertical  . 7 )  2 .2 72 ) Vertical  2 . .) . 2.7.)2 3 7 .  2 .2 .. 72). 3 . 42 . 2.437. 2   2. 7 72 72. 2 . .2 ..2.7.2 4  2. 72 7. )).2 4 .. .77)2.27   Figure 35 – Quantizer matrix for the 720/60p system

Page 50 of 62 pages

Rec. ITU-R BT.1620 53

SMPTE 370M-2002

 horizontal

)324 )   2 4 3 2 . 7   3 .  4 4 )    7  . 2     )  7 3   3 3 Vertical   )   ) 2 3 33 3    7 4 3 32 2 2  3 . . 3 34 2) 2 4 2 4 7 3) 3. 37 2 2    Figure 36 – Output order of a weighted DCT block

4.3 Quantization  4.3.1 Introduction  A" !$'  %-  1             "       /   7 ,

4.3.2 Bit assignment for quantization  A" !$'      5  !$   B7 C5.423)       B#33 33C  $   B C5)7.423)  " K     B#)4 )4C  4.3.3 Quantization step  '-  1  BG# C                 / "    ""  ,G#  % -  1    BG :C           3, G :       %   /, '         %!$'/,  !        ,<  ( $ % G# , "   -  1 $          7(          ,      ( $  " % " G#   "   "       /  "  -  1$ 7 ,

Page 51 of 62 pages

54 Rec. ITU-R BT.1620

SMPTE 370M-2002

 Table 25 – Quantization step

 $    )        .    .    2     .   3 3 )   G  1  2 2       4 4    BG :C . .    7 2  2  ) . 2 4   ) ) .)     ..    . 72   . 32   3 3 )   

4.4 Variable length coding (VLC)     "  "        "    -  1 $       "  ,:    $   !$'/    "   "     " 2, "      5    " 5 '   $ -  1 ) B J)(…(2C     5    E      $ -  1 ) B J)(…(33C   B ( C5 '   "    ,  ' 2 "   " B ( C,  ("    "  ,A    1 ( "    %           "    , <    %    (       B ( C     ( B #()C  B)( C,    "    4,'    *    "     ?  ,'*   -      ?    E   ," OP  ,  A -  1$ "     1 (J)  A -  1$   1 (J  A          " -  1    1     !$' /(  "  % " &: B/C ))  %       ,

Page 52 of 62 pages

Rec. ITU-R BT.1620 55

SMPTE 370M-2002

Table 26 – Length of codewords

    "  )324.7)324.7) 33 ) 3322444...... 77777 7 3 3    3 4 4. ..7 ) ) )         34.7 7 )        2.7) )     2 . 7   3  4 7 ) 2  4 7  4  .   .  .   7  .  )  .    7   7   7   7 3 

 2      :'&   "   ,  '" &: J,

Page 53 of 62 pages

56 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 27 – Codewords of variable length coding

B ( C $  ?"  B ( C$  ?"  B ( C $  ?"  )  )) K   ))))) 4  ))))) )  )) K   )))) .  )))) &: ))    )))) 7))))   )   ))) )  ))) )  ))) K 3  )))) 4  )))) )  )) 2  ))) .  )))   )))   ))  3 )))   ))   ))  4 )) 3K 7K K ) 3 ))   ))))  4 )))) ) 2 )  3 )))  . )))   ))))  . )))  7 )))   ))) ) . ))  ) )) 2K ) 4 ))) ) 7 )))   ))) ) . )) ) ) ))  3 )) 3  )))) )  ))  2 )) 2  ))) )  )  4 )   ))) 3  ))))) 2 ) )))))   ))   )))) 4 ) )))) 4K   )))  3 ))))   %    ) 7 ))  2 ))) )K ))  ) ) ))  7 )))) J2 2 )  )  ) ))) 2 ) )) 4  )))))   ))) )  )))) .  )))) ) ) )))  )  )))))) 7  ))))  ) ))

)  ))) 2  )))))  %  3K )    ))))   ))))  J 33   )))  2 )))) K   )))   ))) ) 33   3 ))   )))) .K  2 ))))   )))  4 )))  ) )))) )  )))  ) )))  )  ))  ) ))) )  ))) 3 ) )) ) 3 )) ) 2 )) ) 4 ) :'& B()C5) 3     )(   3K2 K. K K K )J, B)(C5 4 2 3     )(  . 4K2 2K 3K2 K. K K K )J, " ,&:   &/, 

Page 54 of 62 pages

Rec. ITU-R BT.1620 57

SMPTE 370M-2002

4.5 Arrangement of a compressed macro block    "       /,&     /  44%  ,' "       /  " 4,  'B      /C  '             /    5(( (),' .  ',  G :B-  1   CMG : -  1        /,$  G :   7,  !$M  !$B  !$'/     /(J)(…(4C  !$ ( !$' (      !$'/,  * ?  !$5.423) ))    . )5!$     )5!$'     J))J.#.# #!$'   J.#.##!$'   J 4         !         )5     $M  $ "      " $   "  ((/,'  

;)(;(;(;($)($($ )( $     #   ( ;)(;(;(;($)( $ .)   $ ) $  2  " 4,!$    "  $  !$'/!$'/  ((E(/(  "    ""   #        /$*((E(/," 4(    "    "  *     (  ?    " ,'  ($                "   ,   %    3,,,,, 3,,,,,  3 ,,,,,  3 ,,,,,  3 ,,,,, 3 33,,,,, 2 23 ,,,,, 4 4,,,,, 47 *  ' ! ! ! ! ! ! ! !  $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ G )     3 2 4

: ?

;) ; ; ; $) $ $ ) $  )%  )%  )%  )%  )%  )%  .%  .%    Figure 37 – Arrangement of a compressed macro block

Page 55 of 62 pages

58 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table 28 – Definition of STA   '        /      ) &   &      $   %  ) ) ) )      ) )  ) '%       )  ) ) '%  '%   )   ) '%$    )    &        )  ) '%      ) )    '%  '%     ) '%$       &                '%5       /      /        %  , '% 5       /      /         , '%$5'    / (        , '% '   %  "-        /  )J) J)  "  "  , '%5'   %  "-        /   "  ,  :'&   < 'J)(         /,'  , < 'J(      ,   Table 29 – Codewords of the QNO   G     G :  - - - -)  ) ) ) ) )   ) ) )    ) )  )   ) )      )  ) )   )  )  3   )   ) 2  )    4   ) ) ) .    ) )  7   )  ) )    )      ) )    )      )      3

Page 56 of 62 pages

Rec. ITU-R BT.1620 59

SMPTE 370M-2002

4.6 Arrangement of a video segment     (        -  1 $     , <"  .   "   "  $ ((/    #    ,'       /,% <((E(/(         !$'/!$'/  (( E(/(,  - (  ((E(/((   "   5!$ (!$'   (     (  $        !$' /    ((E(/(, $   $  ((E(/(     "     " 2      &: ,'*   -       ?    E    ,  ' "      "   "       5  = 5'   ((E(/(    #   8  =  5 '         ((E(/(                     /8  = 5'     ((E(/(           "  ,   "   "    "     BJ)8M8KKCN B2)01%  CJ)8 BJ)CJ8 J8  BJ)8M8KKCN BMCN  JBKC 8 JBK2C 8 JBK.C 8 JBK)C 8 JBKC 8 J8-J8 J8J)8 J8 O N  JJJJJ8 J)8-J8 J8J8 J8 O  B/J)8/M48/KKCN  J 8%J8  J)8  +S  -       S+ +S     "  $((/% ,S+ +S S+  BEJ)8EM38EKKCN *%J)8  +S*%  -       S+  +S       /*((%(/% ,S+   BJ)8M.8KKCN    J  B ((%(/((<((%(/(C8   *%J B*%(  C8  O   B%JJCN%J-8J8O  B%JJ-CN%J 8J8O

Page 57 of 62 pages

60 Rec. ITU-R BT.1620

SMPTE 370M-2002

 B%JJ CN%J8J8O  B%JJCN%J 8J8O  B%JJ CN%J8J 8O O +S S+  BEJ)8EM38EKKCN   BJ)8M.8KKCN   *%J  B*%(<((%(/(C8  O  J B(*%C8   B%JJCN%J-8J8O  B%JJ-CN%J 8J8O  B%JJ CN%J8J8O  B%JJCN%J 8J8O  B%JJ CN%J8J 8O O +S S+  BEJ)8EM38EKKCN   BJ)8M.8KKCN  J  B(<((%(/(C8 O   B%JJCN%J-8J8O  B%JJ-CN%J 8J8O  B%JJ CN%J8J8O  B%JJCN%J 8J8O  B%JJ CN%J8J 8O O O O O      B )(  )CN+S!   )  *    %    ),S+ +S'  ) "   * ,S+   JB  "T C8+S "T        ,S+   B  C8 O  B ( CN +S$  *    ?  ,S+  JB "T C8+S$ "T      ,S+ +S   ,S+   B C8 O  '  "             /" , '  (                     /(        %   %    ,     "            /          (   #    "          ,'       %     #      5  *   ?   ))))))))))))) 

Page 58 of 62 pages

Rec. ITU-R BT.1620 61

SMPTE 370M-2002

'  7  !$  (      !$'             &:  " 7,  A     /(      "(            (       /    ,   %    $   ,GGG, ,GGG, ,GGG,  ,GGG, ,GGG, 3 ,GGG, 2 ,GGG 4 ,GG, 47  /    ' 

$*( ((/ <( ((/() <( ((/( <( ((/( <( ((/( <( ((/( <( ((/(3 <( ((/(2 <( ((/(4 G

:

 '   $*((-(/ <((-(/() <((-(/( <((-(/( <((-(/( <((-(/( <((-(/(3 <((-(/(2 <((-(/(4 G

:   '   $*(( (/ <(( (/() <(( (/( <(( (/( <(( (/( <(( (/( <(( (/(3 <(( (/(2 <(( (/(4 G

:   '   $*(((/ <(((/() <(((/( <(((/( <(((/( <(((/( <(((/(3 <(((/(2 <(((/(4 G

:   '   $*(( (/ <(( (/() <(( (/( <(( (/( <(( (/( <(( (/( <(( (/(3 <(( (/(2 <(( (/(4 G

:  $ ;) ; ; ; $) $ ) $  )%  )%  )%  )%  )%  )%  .%  .%    Figure 38 – Arrangement of a video segment after the bit rate reduction

DC error

MSB  ) DCI ) ) m0 ) ) c1 ) ) c0 8bits ) ) AC )  )  EOB LSB ) )

Figure 39 – Video error code

Page 59 of 62 pages

62 Rec. ITU-R BT.1620

SMPTE 370M-2002

Annex A B C Relationship between compression format and other documents

<" ,        B   C     " !#   ,  

!    B*C

!' !' *  

$   B%*C !<  

  " B*C 4*  " 72*

 &   " +=  "  *   "

Figure A.1 – Block diagram of D-xx recorder

Page 60 of 62 pages

Rec. ITU-R BT.1620 63

SMPTE 370M-2002

Annex B B C Digital filter for sampling-rate conversion

2)

3) 3)

) )

)

)  B C  B

) 2

)

0 ~  ~  ~  ~   < - %BJ4,3*01C

Figure B.1 – Template for insertion loss frequency characteristic

),)3

) ),

#),)3  B C  B

< - %BJ4,3*01C ) ~  ~ 

Figure B.2 – Pass band ripple tolerance

Page 61 of 62 pages

64 Rec. ITU-R BT.1620

SMPTE 370M-2002

Table B.1 – Parameter of digital filter

  ; ),)3 ),3 ), ),3 ),33 ).)+2) $ ($ ),)3 ),3 ),24 ),3 ),43 ; ),)3 ),3 ),43 ),3) ),2) ).)+3) $ ($ ),)3 ),3 ),.43 ),3 ),) ; ),)3 ),3 ),43 ),3) ),2) 4)+2) $ ($ ),)3 ),3 ),.43 ),3 ),) 

Annex C B C Compression specification

'           &$2.#0!  3#2) 3)#3)%  ,  Annex D B C Abbreviations and acronyms  >H     %  =     ! =  ! =   ! ='  ' /  !      %   >H  / $  >H   / $@*  $%"   "  %   $*  $    / !   !</   !$'  !     !<  !"     !<  !   " !-  !<-    !<  !<-  " &<$  &     " &:   &/ ?<  ?/  " G :  G  1    G>  G  1         $'    % *=   " - % ;    %/ '        / ';=&  "  % %   %/   '<  '   " " >H    %  ?$    " "   >H  / $  >H   /

Annex E B C Bibliography  &$2.#B774C( "N0 # !" $   "%  >"2(3 * " '  $   >B33#2)(23#3)(3#2) 3)#3)%  C N= 5!<  33#2) 23#3)%   N= 50!<  3#2) 3)#3)%  

Page 62 of 62 pages

Rec. ITU-R BT.1620 65 Annex 2

SMPTE 296M-2001 Revision of SMPTE STANDARD ANSI/SMPTE 296M-1997 for Television — 1280 × 720 Progressive Image Sample Structure — Analog and Digital Representation and Analog Interface

Page 1 of 14 pages

Contents 1 Scope

1 Scope 1.1 This standard defines a family of progressive 2 Normative references image sample systems for the representation of stationary 3 General or moving two-dimensional images sampled tempo- 4 Timing rally at a constant frame rate and having an image 5 System colorimetry format of 1280 pixels by 720 lines and an aspect 6 Raster structure ratio of 16:9 as given in table 1. All systems in the 7 Digital representation table have the common characteristic that all the 8 Digital timing reference sequences (SAV, EAV) samples gathered within a single temporal unit, a frame, 9 shall be spatially contiguous and provide a complete 10 Bit-parallel interface description of that frame (4.2) This standard specifies: 11 Analog sync 12 Analog interface – R′G′B′ color encoding; Annex A Production aperture – R′G′B′ analog and digital representation; Annex B Pre- and post-filtering characteristics – Y′P′BP′R color encoding, analog representation, and Annex C Bibliography analog interface; and – Y′C′BC′R color encoding and digital representation.

Table 1 – Image sampling systems Luma or R′G′B′ Active Luma or R′G′B′ Luma sample Total samples per lines per sampling periods per lines System active line frame Frame frequency total line per nomenclature (S/AL) (AL/F) rate, Hz (fs), MHz (S/TL) frame 1 1280 × 720/60 1280 720 60 74.25 1650 750 2 1280 × 720/59.94 1280 720 60/1.001 74.25/1.001 1650 750 3 1280 × 720/50 1280 720 50 74.25 1980 750 4 1280 × 720/30 1280 720 30 74.25 3300 750 5 1280 × 720/29.97 1280 720 30/1.001 74.25/1.001 3300 750 6 1280 × 720/25 1280 720 25 74.25 3960 750 7 1280 × 720/24 1280 720 24 74.25 4125 750 8 1280 × 720/23.98 1280 720 24/1.001 74.25/1.001 4125 750 NOTE – For systems 4 through 8, analog video interface is not preferred. See clause 12.

Copyright © 2001 by THE SOCIETY OF MOTION PICTURE AND TELEVISION ENGINEERS Approved 595 W. Hartsdale Ave., White Plains, NY 10607 January 4, 2001 (914) 761-1100

66 Rec. ITU-R BT.1620

SMPTE 296M-2001

Designers should be aware that serial digital inter- ITU-R BT.709-4 (09/00), Parameter Values for the faces for formats other than Y′C′BC′R have not been HDTV Standards for Production and International defined. Programme Exchange

A bit-parallel digital interface is incorporated by refer- 3 General ence in clause 10. 3.1 The specification of a system claiming NOTE – Throughout this standard, references to signals compliance with this standard shall state: represented by a single primed letter (e.g., R′, G′, and B′) are equivalent to the nomenclature in earlier documents of – which of the systems of table 1 are imple- the form ER′, EG′, and EB′, which in turn refer to signals to which the transfer characteristics in 5.4 have been applied. mented; Such signals are commonly described as being gamma corrected. – which of the analog R′G′B′ or Y′P′BP′R and/or which of the digital R′G′B′ or Y′C′BC′R interfaces are imple- 1.2 This standard specifies multiple system formats mented; and (table 1). It is not necessary for an implementa- tion to support all formats to be compliant with – whether the digital representation employs eight bits this standard. However, an implementation must or 10 bits per sample in its uniformly quantized (linear) state which of the system formats are supported. PCM coding.

2 Normative references 3.2 Digital codeword values in this standard are expressed as decimal values in the 10-bit repre- The following standards contain provisions which, sentation. An eight-bit system shall either round through reference in this text, constitute provisions of or truncate to the most significant eight bits as this standard. At the time of publication, the editions specified in 7.10. indicated were valid. All standards are subject to revision, and parties to agreements based on this 4 Timing standard are encouraged to investigate the possibility of applying the most recent edition of the standards 4.1 Timing shall be based on a reference clock of listed below. the sampling frequency indicated in table 1, which shall be maintained to a tolerance of ± 10 ppm. SMPTE 274M-1998, Television — 1920 × 1080 Scan- ning and Analog and Parallel Digital Interfaces for 4.2 A frame shall comprise the indicated total Multiple Picture Rates lines per frame, each line of equal duration as determined by the sampling frequency (fs) and the SMPTE 291M-1998, Television — Ancillary Data samples per total line (S/TL). Samples may be Packet and Space Formatting obtained in an optoelectronic conversion process sequentially, simultaneously, or via a combination SMPTE RP 160-1997, Three-Channel Parallel Analog of both, provided all samples in the frame are Component High-Definition Video lnterface contiguous in the image and obtained within the same temporal frame period. The samples within SMPTE RP 177-1993 (R1997), Derivation of Basic each line shall be uniformly delivered to and col- Television Color Equations lected from the interface in a spatially left-to-right sequence; lines in a frame shall be uniformly CIE Publication 15.2 (1986), Colorimetry, Second delivered to and collected from the interface in Edition a spatially top-to-bottom sequence. Lines are numbered in time sequence according to the raster IEC 60169-8 (1978-01), Radio Frequency Connectors, structure described in clause 6. Part 8: R.F. Coaxial Connectors with Inner Diameter of Outer Conductor 6.5 mm (0.256 in) with Bayonet 4.3 Timing instants in each line shall be defined with Lock — Characteristic Impedance 50 Ohms (Type respect to a horizontal datum denoted by 0H which BNC) plus amendments IEC 60169-8-am1 (1996-03) is established by horizontal synchronizing (sync) and IEC 60169-8-am2 (1997-11) information in clauses 8 and 11. Each line shall be

Page 2 of 14 pages

Rec. ITU-R BT.1620 67

SMPTE 296M-2001

divided into a number of reference clock intervals, NOTE – Because the Y′ component is computed from of equal duration, as specified by the column S/TL nonlinear R′G′B′ primary components rather than from the in table 1. The time between any two adjacent sam- linear tristimulus RGB values, it does not represent the true ple instants is called the reference clock interval T. luminance value of the signal, but only an approximation. T = 1/fs. To distinguish it from luminance, the term luma is used for the Y′ signal. For more information, see e.g. Poynton, A Technical Introduction to Digital Video. 5 System colorimetry 5.6 Color-difference component signals P′B and 5.1 Equipment shall be designed in accordance P′R, having the same excursion as the Y′ com- with the colorimetric analysis and optoelectronic ponent, shall be computed as follows: transfer function defined in this clause. This cor- responds to ITU-R BT.709. 0.5 P ′ = ( B ′ – Y ′ ) B 1 – 0.0722 5.2 Picture information shall be linearly repre- sented by red, green, and blue tristimulus values 0.5 ( ) P ′R = R ′ – Y ′ (RGB), lying in the range 0 (reference black) to 1 1 – 0.2126 (reference white), whose colorimetric attributes are based upon reference primaries with the follow- P′B and P′R are filtered and may be coded as C′B and ing chromaticity coordinates, in conformance C′R components for digital transmission. Example with ITU-R BT.709, and whose white reference filter templates are given in figure B.2. conforms to CIE D65 as defined by CIE 15.2: 6 Raster structure CIE x CIE y Red primary 0.640 0.330 6.1 For details of vertical timing, see figures 1 Green primary 0.300 0.600 and 2. Blue primary 0.150 0.060 Reference white 0.3127 0.3290 6.2 In a system according to this standard, each frame shall comprise 750 lines including: 5.3 From the red, green, and blue tristimulus values, three nonlinear primary components, – Vertical blanking: lines 1 through 25 inclusive R′, G′, and B′, shall be computed according to (including vertical sync, lines 1 through 5 inclusive) the optoelectronic transfer function of ITU-R and lines 746 through 750 inclusive; and BT.709, where L denotes a tristimulus value and V′ denotes a nonlinear primary signal: – Picture: 720 lines, lines 26 through 745 inclusive.

 0 ≤ L< 0.018 6.3 Ancillary signals, as distinct from ancillary ′= 4.5L, V  0.45 data, may be conveyed during vertical blanking, 1.099L – 0.099, 0.018 ≤ L ≤ 1.0 lines 7 through 25 inclusive. The portion within each of these lines that may be used for ancillary 5.4 To ensure the proper interchange of picture data is defined in 9.3. Ancillary signals shall not information between analog and digital repre- convey picture information although they may be sentations, signal levels shall be completely employed to convey other related or unrelated contained in the range specified between refer- signals, coded similarly to picture information. ence black and reference white specified in 7.6 Further specification of ancillary signals is out- and 12.4, except for overshoots and under- side the scope of this standard. shoots due to processing. 6.4 During time intervals not otherwise used, the 5.5 ′ The Y component shall be computed as a R′, G′, B′ or Y′, P′B, C′B, P′R, and C′R components weighted sum of nonlinear R′G′B′ primary compo- shall have a blanking level corresponding to zero. nents, using coefficients calculated from the reference primaries according to the method of SMPTE RP 177: 6.5 The production aperture defines a region 1280 samples by 720 lines. The horizontal ex- Y′ = 0.2126R′ + 0.7152G′ + 0.0722B′ tent of the production aperture shall have the

Page 3 of 14 pages

68 Rec. ITU-R BT.1620

SMPTE 296M-2001

0 FRAME

27

Picture Picture Picture

Figure 1 – Vertical timing (analog representation)

FRAME

Picture Picture Picture

Figure 2 – Vertical timing (digital representation)

7 Digital representation 50% point of its leading transition at reference luma sample 0 and the 50% point of its trailing 7.1 Digital representation shall employ either transition at luma sample 1279. The production ′ ′ ′ ′ ′ ′ aperture defines the maximum extent of picture R G B or Y C BC R components, as defined in information. For further information, consult annex clause 5 or clause 6, uniformly sampled. A. NOTE – Each component is prepared as an individual channel. Combinations of channels may be presented to an 6.6 The aspect ratio of the image represented by appropriate interface for signal interchange. For example, the production aperture shall be 16:9. The sample the Y′ channel and the multiplexed C′B/C′R channel data aspect ratio is 1:1 (square pixels). together comprise a source format for the serial interface specified in SMPTE 292M.

6.7 The center of the picture shall be located at 7.2 The digital signals described here are the center of the production aperture, midway assumed to have been filtered to reduce or pre- between samples 639 and 640, and midway vent aliasing upon sampling. For information between lines 385 and 386. regarding filtering, consult annex B.

7.3 The characteristics of the digital signals are 6.8 Each edge of the picture width, measured based on the assumption that the location of any at the 50% amplitude point, shall lie within six required sin (x)/x correction is at the point where reference clock intervals of the production aperture. the signal is converted to an analog format.

Page 4 of 14 pages

Rec. ITU-R BT.1620 69

SMPTE 296M-2001

7.4 R′G′B′ signals and the Y′ signal of the 7.6 Digital R′, G′, B′, and Y′ components shall Y′C′BC′R interface shall be sampled ortho- be computed as follows: gonally, line- and picture-repetitive, at the sam- n-8 pling frequency, fs. The period of the sampling L′D = Floor (219DL′+ 16D + 0.5); D = 2 clock shall be denoted T. R′G′B′ samples shall be cosited with each other. where L′ is the component value in abstract terms 7.5 A luma sampling number in a line is denoted in from zero to unity, n takes the value 8 or 10 corre- this standard by a number from 0 through one sponding to the number of bits to be represented, and less than the total number of samples in a line. L′D is the resulting digital code. The unary function Luma sample number zero shall correspond to the Floor yields the largest integer not greater than its first active video sample. The luma sample num- argument. bering is shown in figure 3. Note that the distance between 0H and the start of SAV is 256 samples. NOTE – This scaling places the extrema of R′, G′, B′, and Y′ components at codewords 64 and 940 in a 10-bit repre- NOTE – The active video digital representation is 1280 clock sentation or codewords 16 and 235 in an eight-bit repre- periods (0-1279) in length. sentation.

0 H

Analog Waveform (Y’R’G’B’)

256T A.T Durations in B.T References 4T 40T 40T 4T 1280T Clock Periods C.T (T)

Luma Sample p abcde f ghi jklmno p abcd Numbering Duration in 4T BT4T 1280T ref. Clock periods Digital ANCILLARY DATA or Data EAV SAV VIDEO DATA (EAV) BLANKING CODEWORDS Stream

NOTES 1 Horizontal axis not to scale. 2 0H is the analog horizontal timing reference point, and in the analog domain, is regarded as the start of the line. 3 A line of digital video extends from the first word of EAV to the last word of video data.

Figure 3 – Analog and digital timing relationships

Page 5 of 14 pages

70 Rec. ITU-R BT.1620

SMPTE 296M-2001

7.7 Digital C′B and C′R components of the Y′C′BC′R handling of ancillary data between eight-bit and 10-bit inter- set shall be computed as follows: faces in annex D.

n-8 7.11 For Y′, R′, G′, and B′ signals, undershoot C′D = Floor (224DC′ + 128D + 0.5); D = 2 and overshoot in video processing may be accommodated by the use of codewords 4 where C′ is the component value in abstract terms through 63 and codewords 941 through 1019 from – 0.5 to + 0.5, and C′ is the resulting digital D in a 10-bit system, or codewords 1 through 15 code. The unary function Floor yields the largest and codewords 236 through 254 in an eight-bit integer not greater than its argument. system. ′ ′ NOTE – This scaling places the extrema of C B and C R at ′ ′ codewords 64 and 960 in a 10-bit representation or code- For C B and C R signals, undershoot and overshoot words 16 and 240 in an eight-bit representation. in video processing may be accommodated by the use of codewords 4 through 63 and codewords 961 7.8 C′B and C′R signals shall be horizontally through 1019 in a 10-bit system, or codewords 1 subsampled by a factor of two with respect to the through 15 and codewords 241 through 254 in an Y′ component. C′B and C′R samples shall be eight-bit system. cosited with even-numbered Y′ samples. The sample number zero of C′B and C′R corresponds 8 Digital timing reference sequences to the first active video 0 sample. For information (SAV, EAV) regarding filtering, consult annex B. 8.1 SAV (start of active video) and EAV (end of The subsampled C′B and C′R signals shall be time- active video) digital synchronizing sequences multiplexed on a sample basis, in the order C′BC′R. shall define synchronization across the digital The first data word of an active line shall be a C′B interface. Figures 2, 3, and 4 show the relation- sample. The multiplexed signal is referred to as ship of the SAV and EAV sequences to digital and C′B/C′R. analog video.

NOTE – Systems 7 and 8 have 2063 C′B sample periods 8.2 An SAV or EAV sequence shall comprise and 2062 C′R sample periods per line. The C′B/C′R multi- four consecutive codewords: a codeword of all plexer must be reset every line at sample number zero. ones, a codeword of all zeros, another code- word of all zeros, and a codeword including F 7.9 Code values having the eight most signifi- (frame), V (vertical), H (horizontal), P3, P2, cant bits all zero or all one — that is, 10-bit codes P1, and P0 (parity) bits. An SAV sequence 0, 1, 2, 3, 1020, 1021, 1022, and 1023 — are shall be identified by having H = 0; EAV shall employed for synchronizing purposes and shall have H = 1 (tables 3 and 4 show details of the be prohibited from video, ancillary signals, and coding). ancillary data. 8.3 When digitized, every line shall include a 7.10 A system having an eight-bit interface shall four-sample EAV sequence commencing 110 address the conversion of 10-bit video data to clocks prior to 0H (for systems 1 and 2); 440 eight bits with an appropriate process that mini- clocks prior to 0H (for system 3); 1760 clocks mizes video artifacts such as quantization noise. prior to 0H (for systems 4 and 5); 2420 clocks Ancillary data in 10-bit format shall be converted prior to 0H (for system 6);and 2585 clocks prior to eight-bit format by truncating the two least to 0H (for systems 7 and 8). When digitized, every significant bits. In both cases, when converting line shall include a four-sample SAV sequence eight-bit data to 10-bit data, the two least signifi- commencing 256 clocks after 0H (for all systems cant bits of the 10-bit word shall be set to 0. [1, 2, 3, 4, 5, 6, 7, and 8]). The EAV sequence immediately preceding the 0H datum of line 1 NOTE – SMPTE is addressing rounding for all eight-bit/10- shall be considered to be the start of the digital bit digital video standards. SMPTE 291M describes the frame as shown in figure 2.

Page 6 of 14 pages

Rec. ITU-R BT.1620 71

SMPTE 296M-2001

10-bit 8-bit

960 240

C’B/C’R

512 128 EAV EAV SAV 64 16

940 235

Y’,R’,G’,B’ EAV EAV SAV

64 16

a - d eh k-n o luma sample numbers a-d e

0H NOTES Notes 1 Figure 3/table 21 Figureshow 3/table numbering 2 show numbering of luma of lumasample sample numbers numbers for each each of the of systems the systems covered covered in this standard. in this standard. 2 0H is the analog2 O Hhorizontal is the analog timinghorizontal reference timing reference point. point. Figure 4 – Digital representation — Horizontal timing details

Table 2 – Values for figures 3 and 4 and table 5 for different systems Luma sample numbering Systems abcde f gh i j k lmnop 1,2 1280 1281 1282 1283 1284 1350 1389 1390 1391 1430 1646 1647 1648 1649 0 1279 3 1280 1281 1282 1283 1284 1680 1719 1720 1721 1760 1976 1977 1978 1979 0 1279 4,5 1280 1281 1282 1283 1284 3000 3039 3040 3041 3080 3296 3297 3298 3299 0 1279 6 1280 1281 1282 1283 1284 3660 3699 3700 3701 3740 3956 3957 3958 3959 0 1279 7,8 1280 1281 1282 1283 1284 3825 3864 3865 3866 3905 4121 4122 4123 4124 0 1279 System Duration in reference clock periods (T)

ABC 1,2 70 362 1650 3 400 692 1980 4,5 1720 2012 3300 6 2380 2672 3960 7,8 2545 2837 4125 NOTE – Figure 3 and table 2 representations show nominal relationship values between the analog and digital representations. See figure 5 and table 5 for tolerance values for the analog sync including rise and fall tolerances.

Page 7 of 14 pages

72 Rec. ITU-R BT.1620

SMPTE 296M-2001

Table 3 – Video timing reference codes Bit 9 876543210 number (MSB) (LSB) Word Value 010231111111111 100000000000 200000000000 3 1 F V H P3P2P1P00 0

Table 4 – Protection bits for SAV and EAV Bit number9876543210 1 F V H P3P2P1P00 0 Function Fixed Fixed Fixed 01000000000 Value 11001110100 of (F/V/H) 21010101100 31011011000

8.4 F/V/H flags is employed, the first four samples after EAV are reserved for other usage. – The EAV and SAV of all lines shall have F = 0. 9.3 The interval between the end of SAV and the – The EAV and SAV of lines 1 through 25 inclusive start of EAV of any line that is outside the vertical and lines 746 through 750 inclusive shall have V = 1. extent of the picture (as defined in clause 6.2), and that is not employed to convey digitized – The EAV and SAV of lines 26 through 745 inclusive ancillary signals, may be employed to convey shall have V = 0. ancillary data packets. – The EAV of line 1 shall be considered the start of the digital frame. NOTE – Currently SMPTE is defining the switching point(s) for all serial digital video interfaces. The reader is cautioned 8.5 A line which in the analog representation to be aware that ancillary data should be placed taking into is permitted to convey ancillary signals may account the switching point. convey digitized ancillary signals. 9.4 Ancillary data packets may be conveyed ′ ′ ′ 9 Ancillary data across each of the three R , G , and B channels, or across each of the three Y′, C′B/C′R channels. 9.1 Ancillary data may optionally be included in the blanking intervals of a digital interface 9.5 In the case of 10-bit representation, intervals according to this standard. not used to convey SAV, video data, EAV, ancil- lary signals, or ancillary data shall convey the 9.2 The interval between the end of EAV and codeword 64 (black) in the R′, G′, B′, Y′ channels, the start of SAV may be employed to convey or 512 in the C′B/C′R channels. They shall be 16 ancillary data packets. Designers should be and 128, respectively, in the case of 8-bit repre- aware that when SMPTE 292M serial interface sentation.

Page 8 of 14 pages

Rec. ITU-R BT.1620 73

SMPTE 296M-2001

9.6 For specifications of the details of ancillary 11 Analog sync data, see SMPTE 291M. 11.1 Details of analog sync timing are shown in 10 Bit-parallel interface figures 1, 3, and 5 and are summarized in table 5.

The electrical and mechanical parameters of the 11.2 A positive zero-crossing of a trilevel sync bit-parallel interface are specified in clauses 10, 11, 12, pulse shall define the 0H datum for each line. A and 13 of SMPTE 274M, which are incorporated negative-going transition precedes this instant by reference. It is anticipated that in future revisions by 40 reference clock intervals, and another of SMPTE 274M that auxiliary component A will be negative-going transition follows this instant by eliminated. 40 reference clock intervals.

+ 300 mV

BLANKING Vertical 0 Sync

- 300 mV

BROAD PULSE d b g

0H

+ 350 mV + 300 mV

P’B,P’R 0

- 300 mV - 350 mV

0H + 700 mV

+ 300 mV

Y’,R’,G’,B’ 0

- 300 mV c a e f

0H Figure 5 – Analog levels and timing

Page 9 of 14 pages

74 Rec. ITU-R BT.1620

SMPTE 296M-2001

Table 5 – Analog sync timing 11.6 Each frame shall commence with five verti- (Systems 1 - 8) cal sync lines, each having a broad pulse. The leading 50% point of a broad pulse shall be 260T Duration Tolerance after the preceding trilevel zero crossing. The (T) (T) trailing 50% point of a broad pulse shall be a See figure 5 40 ± 3 1540T after the preceding trilevel zero crossing.

b See figure 5 1540 – 6 + 0 12 Analog interface c See figure 5 40 ± 3 NOTE – This clause applies to all frame rates in table 1. d See figure 5 260 – 0 + 6 However, direct analog interconnection of slow-rate sys- tems (30 Hz and below) is not preferred, except for synchro- e See figure 5 260 – 0 + 6 nizing signals.

f See figure 5 1540 – 6 + 0 12.1 An analog interface according to this stand- ′ ′ ′ Sync rise and ard may employ either the R G B component set fall time 4 ± 1.5 or the Y′P′BP′R component set. g Total line Table 2 ’C’ ′ ′ ′ ′ – 12 12.2 R G B and Y channels shall have a nomi- Active line 1280 + 0 nal bandwidth of 30 MHz.

12.3 Each component signal shall be conveyed electrically as a voltage on an unbalanced coaxial 11.3 Positive transition of a trilevel sync pulse cable into a pure resistive impedance of 75 Ω. shall be skew-symmetric with a rise time from ± 10% to 90% of 4 1.5 reference clock periods. 12.4 For the Y′, R′, G′, and B′ components, ref- The 50% (midpoint) point of each negative tran- erence black (zero) in the expressions of 5.5 sition shall be coincident with its ideal time shall correspond to a level of 0 Vdc, and refer- ± within a tolerance of 3 reference clock periods. ence white (unity) shall correspond to 700 mV.

11.4 The trilevel sync pulse shall have structure 12.5 P′ B and P′R components are analog ver- and timing according to figures 3 and 5. The sions of the C′B and C′R components of 5.6, in positive peak of the trilevel sync pulse shall have which zero shall correspond to a level of 0 Vdc a level of +300 mV ± 6 mV; its negative peak and reference peak level (value 0.5 of equations shall have a level of –300 mV ± 6 mV. The in 5.6) shall correspond to a level of +350 mV. amplitude difference between positive and nega- tive sync peaks shall be less than 6 mV. 12.6 Trilevel sync according to clause 11 shall be added to each analog component. 11.5 Each line that includes a vertical sync pulse shall maintain blanking level, here denoted zero, 12.7 Each of the electrical signals in an analog except for the interval(s) occupied by sync interface employs a connector that shall conform pulses. During the horizontal blanking interval, to IEC 60169-8, with the exception that the areas not occupied by sync shall be maintained impedance of the connector may be 75 Ω, or to at blanking level, here denoted zero. SMPTE RP 160.

Page 10 of 14 pages

Rec. ITU-R BT.1620 75

SMPTE 296M-2001

Annex A (informative) Production aperture

A.1 Production aperture – Amplitude clipping of analog signals due to the finite dynamic range imposed by the quantization process; A production aperture for the studio digital signal defines an active picture area of 1280 pixels × 720 lines as produced – Use of digital blanking in repeated analog-digital-analog by signal sources such as cameras, telecines, digital video conversions; and tape recorders, and computer-generated pictures conform- ing to this standard. – Tolerance in analog blanking.

A.2 Analog blanking tolerance A.4 Clean aperture

A.2.1 The duration of the maximum active analog video signal A.4.1 The bandwidth limitation of an analog signal (pre- and post- measured at the 50% points is standardized as 1280 clock filtering) can introduce transient ringing effects which intrude periods. However, the analog blanking period may differ from into the active picture area. Also, multiple digital blanking equipment to equipment and the digital blanking may not operations in an analog-digital-analog environment can in- coincide with the analog blanking in actual implementation. crease transient ringing effects. Furthermore, cascaded spatial filtering and/or techniques for handling the horizontal and vertical edges of the picture (associated with complex digital A.2.2 To maximize the active video duration in picture origi- processing in post-production) can introduce transient dis- nation sources, it is desirable to have analog blanking match turbances at the picture boundaries, both horizontally and digital blanking. However, recognizing the need for reason- vertically. It is not possible to impose any bounds on the able tolerance in implementation, analog blanking may be number of cascaded processes which might be encountered wider than digital blanking (see figures 3 and 5). in the practical post-production system. Hence, recognizing the reality of those picture edge transient effects, the defini- A.2.3 To accommodate a practical implementation of analog tion of a system design guideline is introduced in the form blanking within various studio equipment, a tolerance of six of a subjectively artifact-free area, called clean aperture. clock periods is provided at the start and end of active video. Accordingly, the analog tolerance to parameters b and e of A.4.2 The clean aperture defines an area within which table 5 are as follows: picture information is subjectively uncontaminated by all edge transient distortions. In order to minimize the effects Nominal Tolerance on subsequent compression or transmission processes, the Parameter Definition value (T) (T) contaminated area should be confined within 16 pixels and 9 lines of the production aperture edges. b 0H to end of – 6 active video 1540 + 0 A.4.3 The clean aperture of the picture defines a region e 0H to start of – 0 1248 samples in width by 702 lines high, symmetrically active video 260 + 6 located in the production aperture. The clean aperture shall be substantially free from transient effects due to Preferred practice is to provide a full production aperture signal blanking and picture processing. An encroachment of 6 at the output of an analog source prior to first digitization, reserv- samples maximum on each of the left and right edges of ing the tolerance for possible subsequent analog processes. the production aperture is allowed for horizonal blanking errors generated by analog processing. Vertical blanking shall be as specified with zero tolerance. A.2.4 The relationship of the associated analog repre- sentation (inclusive of this tolerance) with the production A.4.4 This yields a minimum clean aperture of 1248 horizontal aperture is shown in figure 5. active pixels by 702 active lines whose quality is guaranteed for final release. The clean aperture lies within the produc- A.3 Transient regions tion aperture as shown in figure A.1.

A.3.1 This standard defines a picture aspect ratio of A.4.5 It is good practice to minimize variations in analog 16:9 with 1280 pixels per active line and 720 active lines blanking and to use techniques in digital processing that per frame. However, digital processing and associated minimize or prevent transients in the allowed contaminated spatial filtering can produce various forms of transient area as well as inside the clean aperture. effects at picture blanking edges and within adjacent active video that should be taken into account to allow practical implementation of the studio standard.

A.3.2 Analog transients. The following factors contribute to these effects:

– Bandwidth limitation of component analog signals (most noticeably, the ringing on color-difference signals);

– Analog filter implementation; Figure A.1 – Clean aperture

Page 11 of 14 pages

76 Rec. ITU-R BT.1620

SMPTE 296M-2001

Annex B (informative) B.2 The passband frequency of the component Y′, R′, G′, Pre- and post-filtering characteristics and B′ signals is nominally 30 MHz.

B.1 Figure B.1 depicts example filter characteristics for pre- and post-filtering of Y′, R′, G′, and B′ component signals. B.3 The value of the amplitude ripple tolerance in the Figure B.2 depicts example filter characteristics for pre- passband is ± 0.05 dB relative to the insertion loss at 100 and post-filtering of P′B and P′R component signals. kHz.

Figure B.1 – Example filter template for Y′ and R′G′B′ components

Page 12 of 14 pages

Rec. ITU-R BT.1620 77

SMPTE 296M-2001

60

50 dB 50

40 dB 40

30

20 Insertion loss loss (dB) Insertion

10 6 dB Relative to the valueat 100 kHz

0 0 0.20 fs 0.25 fs 0.30 fs 0.37 fs Frequency Template for insertion loss

+0.05

0

-0.05 Insertion loss loss (dB) Insertion

NOTE - The value of fs is given in table 1 Relative to the valueat 100 kHz 0 0.20 fs Template for passband ripple

+0.110 +0.075

0

Group Delay (T) -0.075 -0.110 Relative to the valueat 100 kHz

0 0.14 fs 0.20 fs Template for passband group-delay

Figure B.2 – Example filter template for P′B and P′R components

B.4 The insertion loss characteristics of the filters are B.5 The specifications for group-delay in the filters are suf- frequency-scaled from the characteristics of ITU-R ficiently tight to produce good performance while allowing BT.601. Insertion loss is 12 dB or more at half the the practical implementation of the filters. sampling frequency of the Y′, R′, G′, and B′ components, and 6 dB or more at half the sampling frequency of the P′B and P′R components relative to the insertion loss at 100 kHz.

Page 13 of 14 pages

78 Rec. ITU-R BT.1620

SMPTE 296M-2001

Annex C (informative) Bibliography

SMPTE 292M-1998, Television — Bit-Serial Digital Interface ITU-R BT.1361 (1998), Worldwide Unified Colorimetry and for High-Definition Television Systems Related Characteristics of Future Television and Imaging Systems ITU-R BT.601-5 (10/95), Studio Encoding Parameters of Digital Television for Standard 4:3 and Wide-Screen 16:9 Poynton, Charles. A Technical Introduction to Digital Video. Aspect Ratios John Wiley & Sons

Page 14 of 14 pages