Provided by: pngnq_1.0-2ubuntu1_amd64 bug


       pngnq - quantize png images


       pngnq  [-vfhV][-s sample_factor ][-Q dither ][-g gamma ][-e extension ][-d dir ][-n colors
       ][ inputfiles ]


       pngnq quantizes a 32-bit RGBA PNG image to an 8 bit RGBA palette PNG  using  the  neuquant
       algorithm.  The  output  file  name  is  the input file name extended with "-nq8.png" or a
       specified extension.


       -v     Verbose mode. Prints status messages.

       -f     Force overwriting of files.

       -s sample factor
              Sample factor. The neuquant algorithm samples pixels stepping by this  value.   The
              default  value  of  3  gives  good  results. Higher values sample less of the image
              pixels and thus are faster but less accurate. A factor of  1  samples  every  image

       -n colors
              Specifies  the  number  of  colors  to  quantize  to.  Defaults to 256 which is the
              maximum.  The minimum here is 2.

       -Q dither
              Choose a dithering method: n = no dither (default), f = Floyd Steinberg dithering.

       -g gamma
              Set the image gamma correction. If not  present,  uses  the  png  file's  gamma  or
              defaults to 1.0.

       -e extension
              Specifies  the  new  filename  extension. Defaults to "-nq8.png".  Pngnq drops .png
              from the original filenames. If you set the argument of the -e option to  .png  and
              choose the -f option the input file will be overwritten.

       -d dir Tells  pngnq  to put output files in a directory other than the one the input files
              are in.

       input files
              The png files to be processed. Defaults to standard input  if  not  specified.   If
              standard input is being processed the output is sent to standard output.

       -h     Print program help.

       -V     Print version number and library versions.


       Does  not  deal  correctly with greyscale alpha images with low bit depths, but these wont
       benefit from quantizing.


       Stuart Coyle <>