G I F (tm)

                    Graphics Interchange Format (tm)

                     A standard defining a mechanism

                    for the storage and transmission

                  of raster-based graphics information

                              June 15, 1987

                    (c) CompuServe Incorporated, 1987

                           All rights reserved

           While this document is copyrighted, the information

         contained within is made available for use in computer

         software without royalties, or licensing restrictions.

         GIF and 'Graphics Interchange Format' are trademarks of

                        CompuServe, Incorporated.

                          an H&R Block Company

                       5000 Arlington Centre Blvd.

                          Columbus, Ohio 43220

                             (614) 457-8600

                                                                    Page 2

             Graphics Interchange Format (GIF) Specification

                            Table of Contents

       INTRODUCTION . . . . . . . . . . . . . . . . . page 3

       GENERAL FILE FORMAT  . . . . . . . . . . . . . page 3

       GIF SIGNATURE  . . . . . . . . . . . . . . . . page 4

       SCREEN DESCRIPTOR  . . . . . . . . . . . . . . page 4

       GLOBAL COLOR MAP . . . . . . . . . . . . . . . page 5

       IMAGE DESCRIPTOR . . . . . . . . . . . . . . . page 6

       LOCAL COLOR MAP  . . . . . . . . . . . . . . . page 7

       RASTER DATA  . . . . . . . . . . . . . . . . . page 7

       GIF TERMINATOR . . . . . . . . . . . . . . . . page 8

       GIF EXTENSION BLOCKS . . . . . . . . . . . . . page 8

       APPENDIX A - GLOSSARY  . . . . . . . . . . . . page 9

       APPENDIX B - INTERACTIVE SEQUENCES . . . . . . page 10

       APPENDIX C - IMAGE PACKAGING & COMPRESSION . . page 12

       APPENDIX D - MULTIPLE IMAGE PROCESSING . . . . page 15

Graphics Interchange Format (GIF)                                    Page 3

Specification

INTRODUCTION

       'GIF' (tm) is CompuServe's standard for defining generalized  color

  raster   images.    This   'Graphics  Interchange  Format'  (tm)  allows

  high-quality, high-resolution graphics to be displayed on a  variety  of

  graphics  hardware  and is intended as an exchange and display mechanism

  for graphics images.  The image format described  in  this  document  is

  designed  to  support  current  and  future image technology and will in

  addition serve as a basis for future CompuServe graphics products.

       The main focus  of  this  document  is  to  provide  the  technical

  information  necessary  for  a  programmer to implement GIF encoders and

  decoders.  As such, some assumptions are made as to terminology relavent

  to graphics and programming in general.

       The first section of this document describes the  GIF  data  format

  and its components and applies to all GIF decoders, either as standalone

  programs or as part of  a  communications  package.   Appendix  B  is  a

  section  relavent to decoders that are part of a communications software

  package and describes the protocol requirements for entering and exiting

  GIF mode, and responding to host interrogations.  A glossary in Appendix

  A defines some of the terminology used in  this  document.   Appendix  C

  gives  a  detailed  explanation  of  how  the  graphics  image itself is

  packaged as a series of data bytes.

               Graphics Interchange Format Data Definition

GENERAL FILE FORMAT

       +-----------------------+

       | +-------------------+ |

       | |   GIF Signature   | |

       | +-------------------+ |

       | +-------------------+ |

       | | Screen Descriptor | |

       | +-------------------+ |

       | +-------------------+ |

       | | Global Color Map  | |

       | +-------------------+ |

       . . .               . . .

       | +-------------------+ |    ---+

       | |  Image Descriptor | |       |

       | +-------------------+ |       |

       | +-------------------+ |       |

       | |  Local Color Map  | |       |-   Repeated 1 to n times

       | +-------------------+ |       |

       | +-------------------+ |       |

       | |    Raster Data    | |       |

       | +-------------------+ |    ---+

       . . .               . . .

       |-    GIF Terminator   -|

       +-----------------------+

Graphics Interchange Format (GIF)                                    Page 4

Specification

GIF SIGNATURE

       The following GIF Signature identifies  the  data  following  as  a

  valid GIF image stream.  It consists of the following six characters:

            G I F 8 7 a

       The last three characters '87a' may be viewed as a  version  number

  for  this  particular  GIF  definition  and will be used in general as a

  reference  in  documents  regarding  GIF  that   address   any   version

  dependencies.

SCREEN DESCRIPTOR

       The Screen Descriptor describes the overall parameters for all  GIF

  images  following.  It defines the overall dimensions of the image space

  or logical screen required, the existance of color mapping  information,

  background  screen color, and color depth information.  This information

  is stored in a series of 8-bit bytes as described below.

             bits

        7 6 5 4 3 2 1 0  Byte #

       +---------------+

       |               |  1

       +-Screen Width -+      Raster width in pixels (LSB first)

       |               |  2

       +---------------+

       |               |  3

       +-Screen Height-+      Raster height in pixels (LSB first)

       |               |  4

       +-+-----+-+-----+      M = 1, Global color map follows Descriptor

       |M|  cr |0|pixel|  5   cr+1 = # bits of color resolution

       +-+-----+-+-----+      pixel+1 = # bits/pixel in image

       |   background  |  6   background=Color index of screen background

       +---------------+          (color is defined from the Global color

       |0 0 0 0 0 0 0 0|  7        map or default map if none specified)

       +---------------+

       The logical screen width and height can both  be  larger  than  the

  physical  display.   How  images  larger  than  the physical display are

  handled is implementation dependent and can take advantage  of  hardware

  characteristics  (e.g.   Macintosh scrolling windows).  Otherwise images

  can be clipped to the edges of the display.

       The value of 'pixel' also defines  the  maximum  number  of  colors

  within  an  image.   The  range  of  values  for 'pixel' is 0 to 7 which

  represents 1 to 8 bits.  This translates to a range of 2 (B & W) to  256

  colors.   Bit  3 of word 5 is reserved for future definition and must be

  zero.

Graphics Interchange Format (GIF)                                    Page 5

Specification

GLOBAL COLOR MAP

       The Global Color Map is optional but recommended for  images  where

  accurate color rendition is desired.  The existence of this color map is

  indicated in the 'M' field of byte 5 of the Screen Descriptor.  A  color

  map  can  also  be associated with each image in a GIF file as described

  later.  However this  global  map  will  normally  be  used  because  of

  hardware  restrictions  in equipment available today.  In the individual

  Image Descriptors the 'M' flag will normally be  zero.   If  the  Global

  Color  Map  is  present,  it's definition immediately follows the Screen

  Descriptor.   The  number  of  color  map  entries  following  a  Screen

  Descriptor  is equal to 2**(# bits per pixel), where each entry consists

  of three byte values representing the relative intensities of red, green

  and blue respectively.  The structure of the Color Map block is:

             bits

        7 6 5 4 3 2 1 0  Byte #

       +---------------+

       | red intensity |  1    Red value for color index 0

       +---------------+

       |green intensity|  2    Green value for color index 0

       +---------------+

       | blue intensity|  3    Blue value for color index 0

       +---------------+

       | red intensity |  4    Red value for color index 1

       +---------------+

       |green intensity|  5    Green value for color index 1

       +---------------+

       | blue intensity|  6    Blue value for color index 1

       +---------------+

       :               :       (Continues for remaining colors)

       Each image pixel value received will be displayed according to  its

  closest match with an available color of the display based on this color

  map.  The color components represent a fractional intensity  value  from

  none  (0)  to  full (255).  White would be represented as (255,255,255),

  black as (0,0,0) and medium yellow as (180,180,0).  For display, if  the

  device  supports fewer than 8 bits per color component, the higher order

  bits of each component are used.  In the creation of  a  GIF  color  map

  entry  with  hardware  supporting  fewer  than 8 bits per component, the

  component values for the hardware  should  be  converted  to  the  8-bit

  format with the following calculation:

       <map_value> = <component_value>*255/(2**<nbits> -1)

       This assures accurate translation of colors for all  displays.   In

  the  cases  of  creating  GIF images from hardware without color palette

  capability, a fixed palette should be created  based  on  the  available

  display  colors for that hardware.  If no Global Color Map is indicated,

  a default color map is generated internally  which  maps  each  possible

  incoming  color  index to the same hardware color index modulo <n> where

  <n> is the number of available hardware colors.

Graphics Interchange Format (GIF)                                    Page 6

Specification

IMAGE DESCRIPTOR

       The Image Descriptor defines the actual placement  and  extents  of

  the  following  image within the space defined in the Screen Descriptor.

  Also defined are flags to indicate the presence of a local color  lookup

  map, and to define the pixel display sequence.  Each Image Descriptor is

  introduced by an image separator  character.   The  role  of  the  Image

  Separator  is simply to provide a synchronization character to introduce

  an Image Descriptor.  This is desirable if a GIF file happens to contain

  more  than  one  image.   This  character  is defined as 0x2C hex or ','

  (comma).  When this character is encountered between images,  the  Image

  Descriptor will follow immediately.

       Any characters encountered between the end of a previous image  and

  the image separator character are to be ignored.  This allows future GIF

  enhancements to be present in newer image formats and yet ignored safely

  by older software decoders.

             bits

        7 6 5 4 3 2 1 0  Byte #

       +---------------+

       |0 0 1 0 1 1 0 0|  1    ',' - Image separator character

       +---------------+

       |               |  2    Start of image in pixels from the

       +-  Image Left -+       left side of the screen (LSB first)

       |               |  3

       +---------------+

       |               |  4

       +-  Image Top  -+       Start of image in pixels from the

       |               |  5    top of the screen (LSB first)

       +---------------+

       |               |  6

       +- Image Width -+       Width of the image in pixels (LSB first)

       |               |  7

       +---------------+

       |               |  8

       +- Image Height-+       Height of the image in pixels (LSB first)

       |               |  9

       +-+-+-+-+-+-----+       M=0 - Use global color map, ignore 'pixel'

       |M|I|0|0|0|pixel| 10    M=1 - Local color map follows, use 'pixel'

       +-+-+-+-+-+-----+       I=0 - Image formatted in Sequential order

                               I=1 - Image formatted in Interlaced order

                               pixel+1 - # bits per pixel for this image

       The specifications for the image position and size must be confined

  to  the  dimensions defined by the Screen Descriptor.  On the other hand

  it is not necessary that the image fill the entire screen defined.

LOCAL COLOR MAP

Graphics Interchange Format (GIF)                                    Page 7

Specification

       A Local Color Map is optional and defined here for future use.   If

  the  'M' bit of byte 10 of the Image Descriptor is set, then a color map

  follows the Image Descriptor that applies only to the  following  image.

  At the end of the image, the color map will revert to that defined after

  the Screen Descriptor.  Note that the 'pixel' field of byte  10  of  the

  Image  Descriptor  is used only if a Local Color Map is indicated.  This

  defines the parameters not only for the image pixel size, but determines

  the  number  of color map entries that follow.  The bits per pixel value

  will also revert to the value specified in the  Screen  Descriptor  when

  processing of the image is complete.

RASTER DATA

       The format of the actual image is defined as the  series  of  pixel

  color  index  values that make up the image.  The pixels are stored left

  to right sequentially for an image row.  By default each  image  row  is

  written  sequentially, top to bottom.  In the case that the Interlace or

  'I' bit is set in byte 10 of the Image Descriptor then the row order  of

  the  image  display  follows  a  four-pass process in which the image is

  filled in by widely spaced rows.  The first pass writes every  8th  row,

  starting  with  the top row of the image window.  The second pass writes

  every 8th row starting at the fifth row from the top.   The  third  pass

  writes every 4th row starting at the third row from the top.  The fourth

  pass completes the image, writing  every  other  row,  starting  at  the

  second row from the top.  A graphic description of this process follows:

  Image

  Row  Pass 1  Pass 2  Pass 3  Pass 4          Result

  ---------------------------------------------------

    0  **1a**                                  **1a**

    1                          **4a**          **4a**

    2                  **3a**                  **3a**

    3                          **4b**          **4b**

    4          **2a**                          **2a**

    5                          **4c**          **4c**

    6                  **3b**                  **3b**

    7                          **4d**          **4d**

    8  **1b**                                  **1b**

    9                          **4e**          **4e**

   10                  **3c**                  **3c**

   11                          **4f**          **4f**

   12          **2b**                          **2b**

  . . .

       The image pixel values are processed as a series of  color  indices

  which  map  into the existing color map.  The resulting color value from

  the map is what is actually displayed.  This series  of  pixel  indices,

  the  number  of  which  is equal to image-width*image-height pixels, are

  passed to the GIF image data stream one value per pixel, compressed  and

  packaged  according  to  a  version  of the LZW compression algorithm as

  defined in Appendix C.

Graphics Interchange Format (GIF)                                    Page 8

Specification

GIF TERMINATOR

       In order to provide a synchronization for the termination of a  GIF

  image  file,  a  GIF  decoder  will process the end of GIF mode when the

  character 0x3B hex or ';' is found after an image  has  been  processed.

  By  convention  the  decoding software will pause and wait for an action

  indicating that the user is ready to continue.  This may be  a  carriage

  return  entered  at  the  keyboard  or  a  mouse click.  For interactive

  applications this user action must  be  passed  on  to  the  host  as  a

  carriage  return  character  so  that the host application can continue.

  The decoding software will then typically leave graphics mode and resume

  any previous process.

GIF EXTENSION BLOCKS

       To provide for orderly extension of the GIF definition, a mechanism

  for  defining  the  packaging  of extensions within a GIF data stream is

  necessary.  Specific GIF extensions are to be defined and documented  by

  CompuServe in order to provide a controlled enhancement path.

       GIF Extension Blocks are packaged in a manner similar to that  used

  by the raster data though not compressed.  The basic structure is:

        7 6 5 4 3 2 1 0  Byte #

       +---------------+

       |0 0 1 0 0 0 0 1|  1       '!' - GIF Extension Block Introducer

       +---------------+

       | function code |  2       Extension function code (0 to 255)

       +---------------+    ---+

       |  byte count   |       |

       +---------------+       |

       :               :       +-- Repeated as many times as necessary

       |func data bytes|       |

       :               :       |

       +---------------+    ---+

       . . .       . . .

       +---------------+

       |0 0 0 0 0 0 0 0|       zero byte count (terminates block)

       +---------------+

       A GIF Extension Block may immediately preceed any Image  Descriptor

  or occur before the GIF Terminator.

       All GIF decoders must be able to recognize  the  existence  of  GIF

  Extension  Blocks  and  read past them if unable to process the function

  code.  This ensures that older decoders will be able to process extended

  GIF   image   files   in  the  future,  though  without  the  additional

  functionality.

Graphics Interchange Format (GIF)                                    Page 9

Appendix A - Glossary

                                GLOSSARY

Pixel - The smallest picture element of a  graphics  image.   This  usually

  corresponds  to  a single dot on a graphics screen.  Image resolution is

  typically given in units of  pixels.   For  example  a  fairly  standard

  graphics  screen  format  is  one 320 pixels across and 200 pixels high.

  Each pixel can  appear  as  one  of  several  colors  depending  on  the

  capabilities of the graphics hardware.

Raster - A horizontal row of pixels representing one line of an  image.   A

  typical method of working with images since most hardware is oriented to

  work most efficiently in this manner.

LSB - Least Significant Byte.  Refers to a convention for two byte  numeric

  values in which the less significant byte of the value preceeds the more

  significant byte.  This convention is typical on many microcomputers.

Color Map - The list of definitions of each color  used  in  a  GIF  image.

  These  desired  colors are converted to available colors through a table

  which is derived by assigning an incoming color index (from  the  image)

  to  an  output  color  index  (of  the  hardware).   While the color map

  definitons are specified in a GIF image, the output  pixel  colors  will

  vary  based  on  the  hardware used and its ability to match the defined

  color.

Interlace - The method of displaying a GIF image in which  multiple  passes

  are  made,  outputting  raster  lines  spaced  apart to provide a way of

  visualizing the general content of an entire image  before  all  of  the

  data has been processed.

B Protocol - A CompuServe-developed error-correcting file transfer protocol

  available  in  the  public  domain  and implemented in CompuServe VIDTEX

  products.  This error checking mechanism will be used  in  transfers  of

  GIF images for interactive applications.

LZW - A sophisticated data compression algorithm  based  on  work  done  by

  Lempel-Ziv  &  Welch  which  has  the feature of very efficient one-pass

  encoding and decoding.  This allows the image  to  be  decompressed  and

  displayed  at  the  same  time.   The  original  article from which this

  technique was adapted is:

         Terry  A.   Welch,  "A  Technique  for  High   Performance   Data

         Compression", IEEE Computer, vol 17 no 6 (June 1984)

       This basic algorithm is also used in the  public  domain  ARC  file

  compression  utilities.   The  CompuServe  adaptation  of LZW for GIF is

  described in Appendix C.

Graphics Interchange Format (GIF)                                   Page 10

Appendix B - Interactive Sequences

          GIF Sequence Exchanges for an Interactive Environment

       The following sequences are defined for use  in  mediating  control

  between a GIF sender and GIF receiver over an interactive communications

  line.  These  sequences  do  not  apply  to  applications  that  involve

  downloading  of  static  GIF  files and are not considered part of a GIF

  file.

GIF CAPABILITIES ENQUIRY

       The GCE sequence is issued from a host and requests an  interactive

  GIF  decoder  to  return  a  response  message that defines the graphics

  parameters for the decoder.  This involves returning  information  about

  available screen sizes, number of bits/color supported and the amount of

  color detail supported.  The escape sequence for the GCE is defined as:

       ESC [ > 0 g     (g is lower case, spaces inserted for clarity)

                        (0x1B 0x5B 0x3E 0x30 0x67)

GIF CAPABILITIES RESPONSE

       The GIF Capabilities Response message is returned by an interactive

  GIF  decoder  and  defines  the  decoder's  display capabilities for all

  graphics modes that are supported by the software.  Note that  this  can

  also include graphics printers as well as a monitor screen.  The general

  format of this message is:

    #version;protocol{;dev, width, height, color-bits, color-res}... <CR>

  '#'          - GCR identifier character (Number Sign)

  version      - GIF format version number;  initially '87a'

  protocol='0' - No end-to-end protocol supported by decoder

                 Transfer as direct 8-bit data stream.

  protocol='1' - Can use an error correction protocol to transfer GIF data

              interactively from the host directly to the display.

  dev = '0'    - Screen parameter set follows

  dev = '1'    - Printer parameter set follows

  width        - Maximum supported display width in pixels

  height       - Maximum supported display height in pixels

  color-bits   - Number of  bits  per  pixel  supported.   The  number  of

              supported colors is therefore 2**color-bits.

  color-res    - Number of bits  per  color  component  supported  in  the

              hardware  color  palette.   If  color-res  is  '0'  then  no

              hardware palette table is available.

       Note that all values in the  GCR  are  returned  as  ASCII  decimal

  numbers and the message is terminated by a Carriage Return character.

Graphics Interchange Format (GIF)                                   Page 11

Appendix B - Interactive Sequences

       The  following   GCR   message   describes   three   standard   EGA

  configurations  with  no  printer;  the GIF data stream can be processed

  within an error correcting protocol:

       #87a;1 ;0,320,200,4,0 ;0,640,200,2,2 ;0,640,350,4,2<CR>

ENTER GIF GRAPHICS MODE

       Two sequences are currently defined to invoke  an  interactive  GIF

  decoder into action.  The only difference between them is that different

  output media are selected.  These sequences are:

    ESC [ > 1 g   Display GIF image on screen

                  (0x1B 0x5B 0x3E 0x31 0x67)

    ESC [ > 2 g   Display image directly to an attached graphics  printer.

                  The  image  may optionally be displayed on the screen as

                  well.

                  (0x1B 0x5B 0x3E 0x32 0x67)

       Note that the 'g' character terminating each sequence is  in  lower

  case.

INTERACTIVE ENVIRONMENT

       The assumed environment for the transmission of GIF image data from

  an  interactive  application  is  a  full 8-bit data stream from host to

  micro.  All 256 character codes must be transferrable.  The establishing

  of  an 8-bit data path for communications will normally be taken care of

  by the host application programs.  It is however  up  to  the  receiving

  communications programs supporting GIF to be able to receive and pass on

  all 256 8-bit codes to the GIF decoder software.

Graphics Interchange Format (GIF)                                   Page 12

Appendix C - Image Packaging & Compression

       The Raster Data stream that represents the actual output image  can

  be represented as:

        7 6 5 4 3 2 1 0

       +---------------+

       |   code size   |

       +---------------+     ---+

       |blok byte count|        |

       +---------------+        |

       :               :        +-- Repeated as many times as necessary

       |  data bytes   |        |

       :               :        |

       +---------------+     ---+

       . . .       . . .

       +---------------+

       |0 0 0 0 0 0 0 0|       zero byte count (terminates data stream)

       +---------------+

       The conversion of the image from a series  of  pixel  values  to  a

  transmitted or stored character stream involves several steps.  In brief

  these steps are:

  1.  Establish the Code Size -  Define  the  number  of  bits  needed  to

      represent the actual data.

  2.  Compress the Data - Compress the series of image pixels to a  series

      of compression codes.

  3.  Build a Series of Bytes - Take the  set  of  compression  codes  and

      convert to a string of 8-bit bytes.

  4.  Package the Bytes - Package sets of bytes into blocks  preceeded  by

      character counts and output.

ESTABLISH CODE SIZE

       The first byte of the GIF Raster Data stream is a value  indicating

  the minimum number of bits required to represent the set of actual pixel

  values.  Normally this will be the same as the  number  of  color  bits.

  Because  of  some  algorithmic constraints however, black & white images

  which have one color bit must be indicated as having a code size  of  2.

  This  code size value also implies that the compression codes must start

  out one bit longer.

COMPRESSION

       The LZW algorithm converts a series of data values into a series of

  codes  which may be raw values or a code designating a series of values.

  Using text characters as an analogy,  the  output  code  consists  of  a

  character or a code representing a string of characters.

Graphics Interchange Format (GIF)                                   Page 13

Appendix C - Image Packaging & Compression

       The LZW algorithm used in  GIF  matches  algorithmically  with  the

  standard LZW algorithm with the following differences:

  1.  A   special   Clear   code   is    defined    which    resets    all

      compression/decompression parameters and tables to a start-up state.

      The value of this code is 2**<code size>.  For example if  the  code

      size  indicated  was 4 (image was 4 bits/pixel) the Clear code value

      would be 16 (10000 binary).  The Clear code can appear at any  point

      in the image data stream and therefore requires the LZW algorithm to

      process succeeding codes as if  a  new  data  stream  was  starting.

      Encoders  should output a Clear code as the first code of each image

      data stream.

  2.  An End of Information code is defined that explicitly indicates  the

      end  of  the image data stream.  LZW processing terminates when this

      code is encountered.  It must be the last code output by the encoder

      for an image.  The value of this code is <Clear code>+1.

  3.  The first available compression code value is <Clear code>+2.

  4.  The output codes are of variable length, starting  at  <code size>+1

      bits  per code, up to 12 bits per code.  This defines a maximum code

      value of 4095 (hex FFF).  Whenever the LZW code value  would  exceed

      the  current  code length, the code length is increased by one.  The

      packing/unpacking of these codes must then be altered to reflect the

      new code length.

BUILD 8-BIT BYTES

       Because the LZW compression  used  for  GIF  creates  a  series  of

  variable  length  codes, of between 3 and 12 bits each, these codes must

  be reformed into a series of 8-bit bytes that  will  be  the  characters

  actually stored or transmitted.  This provides additional compression of

  the image.  The codes are formed into a stream of bits as if  they  were

  packed  right to left and then picked off 8 bits at a time to be output.

  Assuming a character array of 8 bits per character and using 5 bit codes

  to be packed, an example layout would be similar to:

        byte n       byte 5   byte 4   byte 3   byte 2   byte 1

       +-.....-----+--------+--------+--------+--------+--------+

       | and so on |hhhhhggg|ggfffffe|eeeedddd|dcccccbb|bbbaaaaa|

       +-.....-----+--------+--------+--------+--------+--------+

       Note that the physical  packing  arrangement  will  change  as  the

  number  of  bits per compression code change but the concept remains the

  same.

PACKAGE THE BYTES

       Once the bytes have been created, they are grouped into blocks  for

  output by preceeding each block of 0 to 255 bytes with a character count

  byte.  A block with a zero byte count terminates the Raster Data  stream

  for  a  given  image.  These blocks are what are actually output for the

Graphics Interchange Format (GIF)                                   Page 14

Appendix C - Image Packaging & Compression

  GIF image.  This block format has the side effect of allowing a decoding

  program  the  ability to read past the actual image data if necessary by

  reading block counts and then skipping over the data.

Graphics Interchange Format (GIF)                                   Page 15

Appendix D - Multiple Image Processing

       Since a  GIF  data  stream  can  contain  multiple  images,  it  is

  necessary  to  describe  processing and display of such a file.  Because

  the image descriptor allows  for  placement  of  the  image  within  the

  logical  screen,  it is possible to define a sequence of images that may

  each be a partial screen, but in total  fill  the  entire  screen.   The

  guidelines for handling the multiple image situation are:

  1.  There is no pause between images.  Each is processed immediately  as

      seen by the decoder.

  2.  Each image explicitly overwrites any image  already  on  the  screen

      inside  of  its window.  The only screen clears are at the beginning

      and end of the  GIF  image  process.   See  discussion  on  the  GIF

      terminator.