rfc2880.txt

来自「RFC 的详细文档！」· 文本 代码 · 共 1,730 行 · 第 1/5 页

   See also: bits 44,45.

3.2 DIS/DTC bit 16 (MR coding)

   All Group 3 fax systems are required to support Modified Huffman (MH)
   1-dimensional coding for bi-level images.  (A bi-level image is one
   with just two pixel states such as black and white, as opposed to a
   grey-scale or colour image.)

   Setting this bit indicates additional support for Modified Read (MR)
   2-dimensional coding for bi-level images.

   Both MH and MR coding are described in ITU T.4 [7].

   See also: bits 31,78,79.

3.3 DIS/DTC bits 17,18 (width)

   All Group 3 fax systems are required to support 215mm paper width.
   These bits can be set to indicate additional support for 255mm and
   303mm paper widths.

   See also: bits 76,77.

3.4 DIS/DTC bits 19,20 (length)

   All Group 3 fax systems are required to support 297mm paper length.
   These bits can be set to indicate additional support for 364mm and
   unlimited paper lengths.

   See also: bits 76,77.

3.5 DIS/DTC bit 31 (MMR coding)

   Setting this bit indicates support for Modified Modified Read (MMR)
   2-dimensional coding for bi-level images, in addition to the required
   support for MH coding.

   MMR coding is described in ITU T.6 [8].

   See also: bits 16,78,79.






McIntyre & Klyne             Informational                      [Page 7]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


3.6 DIS/DTC bit 36 (JBIG multi-level coding)

   If multi-level images are to be handled, support for JPEG coding is
   required (i.e. bit 68 must be set).  Setting bit 36 indicates
   additional support for JBIG lossless coding of multi-level images.

   JBIG coding for multi-level images is described in ITU T.43 [10] and
   T.4 Annex G [7].

   See also: bits 68,69.

3.7 DIS/DTC bit 37 (plane interleave)

   Setting this bit indicates support for plane interleave for JBIG-
   coded multi-level images in addition to stripe interleave, which is
   standard for JBIG multi-level images.

   JBIG coding for multi-level images is described in ITU T.43 [10] and
   T.4 Annex G [7].

   See also: bit 36.

3.8 DIS/DTC bits 41,42,43 (resolution)

   Setting these bits indicates support for resolutions in addition to
   200*100dpi and 200*200dpi, or 8*3.85dpmm and 8*7.7dpmm.  (Or in
   addition to the basic 200*200dpi resolution when using a multi-level
   image mode.)

   Bit 41 indicates support for 8*15.4dpmm bi-level images
   (independently of the settings of bits 44 and 45).

   Bit 42 indicates support for 300*300dpi bi-level images
   (independently of the settings of bits 44 and 45).  Also applies to
   multi-level images or MRC mask if bit 97 is set.

   Bit 43 indicates support for 400*400dpi and/or 16*15.4dpmm bi-level
   images, depending upon the settings of bits 44 and 45.  Also applies
   to multi-level images or MRC mask if bit 97 is set.

   See also: bits 15,44,45,97.










McIntyre & Klyne             Informational                      [Page 8]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


3.9 DIS/DTC bits 44,45 (preferred units)

   These bits are used to indicate the preferred resolution units for
   received images.  Because the exact resolution and x/y pixel density
   measures in dpi or dpmm are slightly different, some image size and
   aspect ratio distortion may occur if the sender and receiver use
   different units.

   Even when sender and recipient have different preferred units, image
   transfer must be accomplished.  For most fax uses, the dpi and dpmm
   measurements are sufficiently close to each other that the difference
   is not noticed.

   The preferred units setting affects the detailed interpretation of
   the following resolutions:

                dpi       dpmm      (dpi equivalent)
                ---       ----
      Base      200*100   8*3.85    204*98
      Bit 15    200*200   8*7.7     204*196
      Bit 43    400*400   16*15.4   408*391

   But terminals are required to accept the inch- and metric-based
   measures given above as equivalent, distorting the image if necessary
   to accommodate the differences.

   See also: bits 15, 43

3.10 DIS/DTC bit 68 (JPEG)

   This bit indicates support for JPEG coding of multi-level images.

   JPEG coding for multi-level images is described in ITU T.81 [12] and
   T.4 Annex E [7].

   See also: bits 15,69,73

3.11 DIS/DTC bit 69 (colour)

   This bit indicates support for colour images, in addition to just
   grey-scale.

   Both grey-scale and colour require multi-level coding.  The
   difference is that grey is single component while colour is multi-
   component.

   See also: bits 36,68,73.




McIntyre & Klyne             Informational                      [Page 9]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


3.12 DIS/DTC bit 71 (bits/sample)

   Standard support for multi-level images uses 8 bits per sample.
   Setting this bit indicates additional support for 12 bits per sample.

   For a grey-scale multi-level image, there is just one sample per
   pixel, giving 256 or 4096 possible pixel values.  For a full colour
   multi-level image there are three samples per pixel, giving 256^3
   (16777216) or 4096^3 (6871946736) possible values.

   When related to a mapped colour image, bit 71 also affects the
   maximum number of entries in the mapping table:  when it is reset, up
   to 4096 different pixel values can be used;  when set, up to 65536
   different values can be used.

   See also: bit 68.

3.13 DIS/DTC bit 73 (no subsampling)

   Standard support for JPEG-coded multi-level images uses 4:1:1
   chrominance subsampling.  That is, for each 4 luminance samples in
   the image data there is a single chrominance sample.

   Setting this bit indicates that JPEG-coded colour images without
   subsampling can also be supported.  This is not applicable to JBIG
   coding (bit 36).

   See also: bits 68,69.

3.14 DIS/DTC bit 74 (custom illuminant)

   Standard support for multi-level images requires use of D50
   illuminant.  Setting this bit indicates that a custom illuminant also
   can be supported for multi-level images (both grey-scale and colour).
   Details of the custom illuminant are contained in the image data.

   Use of a custom illuminant with multi-level images is described in
   ITU T.4 Annex E [7].

   See also: bits 36,68.

3.15 DIS/DTC bit 75 (custom gamut)

   Standard support for a default colour gamut is required for multi-
   level images.  Setting this bit indicates that a custom gamut also
   can be supported for multi-level images (both grey-scale and colour).
   Details of the custom gamut are contained in the image data.




McIntyre & Klyne             Informational                     [Page 10]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


   The default gamut (L*=[0,100], a*=[-85, 85], b* = [-75, 125]), and
   use of a custom gamut with multi-level images is described in ITU
   T.42 [9] and T.4 Annex E [7].

   See also: bits 36,68.

3.16 DIS/DTC bits 76,77 (paper size)

   All Group 3 faxes are required to support A4 paper size.  These bits
   can be set to indicate additional support for North American letter
   and legal paper sizes.

   See also: bits 17,18,19,20.

3.17 DIS/DTC bit 78 (JBIG bi-level coding)

   Setting bit 78 indicates support for JBIG coding of bi-level images
   (using T.85 encoding rules), in addition to the required support for
   MH coding.

   JBIG coding of bi-level images is described in ITU T.85 [14].

   See also: bits 16,31,79.

3.18 DIS/DTC bit 79 (JBIG stripe size)

   Setting bit 79 (along with bit 78) indicates support for the 'LO'
   option with JBIG coded bi-level images.  Basic bi-level JBIG coding
   uses 128 lines per stripe;  the 'LO' option allows other stripe sizes
   to be used.

   The stripe size is used for all stripes except the last, which may
   have fewer lines than the indicated value.

   JBIG coding of bi-level images is described in ITU T.85 [14].

   See also: bits 16,31,78.

3.19 DIS/DTC bit 92,93,94 (MRC maximum functional mode)

   If these bits are all zero, then Mixed Raster Content (MRC) coding is
   not supported.  Otherwise, they represent a number in the range 1-7
   that indicates an MRC maximum functional mode.

   MRC coding of images is described in ITU T.44 [11] and T.4 Annex H
   [17].

   See also: bits 68,95.



McIntyre & Klyne             Informational                     [Page 11]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


3.20 DIS/DTC bit 95 (MRC stripe size)

   Standard support for MRC uses a maximum stripe size of 256 lines.
   When this bit is set, the maximum stripe size is a full page.

   This bit is meaningful only if bits 92-94 indicate an MRC coding
   capability.  MRC coding of images is described in ITU T.44 [11] and
   T.4 Annex H [17].

   See also: bits 92,93,94.

3.21 DIS/DTC bit 97 (resolution)

   Setting this bit indicates that the additional resolutions indicated
   by bits 42 and 43 may be used for multi-level images and any MRC mask
   layer.

   When this bit is set, bit 42 implies 300dpi and bit 43 implies 400dpi
   for multi-level or MRC mask layer images (irrespective of the
   preferred units indicated by bits 44 and 45).

   See also: bits 42,43,68.

3.22 DIS/DTC bit 98 (resolution)

   Setting this bit indicates that the additional resolution 100*100dpi
   may be used for multi-level images.

        NOTE: 100dpi is not used for bi-level images, including
        the MRC mask layer.

   See also: bit 36,68,92-94.

4. Summary of T.30 capability dependencies

   This section contains a number of decision tables that indicate the
   allowable combinations of T.30 DIS/DTC mask bits.

   Within the decision table bodies, the following symbols are use to
   indicate values of T.30 DIS/DTC bits:

      0  = bit set to '0'
      1  = bit set to '1'
      x  = don't care bit value: may be '0' or '1'
      *0 = bit must be '0' ('1' is invalid in given combination)
      *1 = bit must be '1' ('0' is invalid in given combination)
      #  = bits in row combined to form a numeric value




McIntyre & Klyne             Informational                     [Page 12]

RFC 2880           Internet Fax T.30 Feature Mapping         August 2000


4.1 Image coding

   MH coding is required as a minimum for Group 3 fax operation.

4.1.1 Bi-level coding

<------- T.30 bits --------->
15|16|31|36|37|68|69|73|78|79||Description
--+--+--+--+--+--+--+--+--+--++---------------------------------------
 x| 0| 0|  |  |  |  |  | 0| 0||Compression = [MH]
 x| 1| 0|  |  |  |  |  | 0| 0||Compression = [MH,MR]
 x| 0| 1|  |  |  |  |  | 0| 0||Compression = [MH,MMR]
 x| 1| 1|  |  |  |  |  | 0| 0||Compression = [MH,MR,MMR]
 x| 0| 0|  |  |  |  |  | 1| 0||Compression = [MH,T.85]