From: albertnash@ro.ru   
      
   A reader might wonder, why care about the arithmetic encoding? Recently, I   
   tried to compress a rather big scan (private content, an A4 page at 1200 dpi   
   with black foreground text and a nontrivial green-white background) into a   
   JPEG file of a possibly    
   small size. I ran:   
   cjpeg -quality 0 -arithmetic -dct float -outfile output.arithmetic.jpg -report   
   -strict -verbose input.pnm   
   cjpeg -quality 0 -optimize -dct float -outfile output.optimized.jpg -report   
   -strict -verbose input.pnm   
   cjpeg -quality 0 -dct float -outfile output.unoptimized.jpg -report -strict   
   -verbose input.pnm   
   (Quality 1 would do as well. The real file names have been replaced by   
   “input” and “output” for privacy here.)   
   The sizes are this:   
   output.arithmetic.jpg: 239224 bytes   
   output.optimized.jpg: 807963 bytes   
   output.unoptimized.jpg: 1928235 bytes   
   The foregrounds of the three files are readable and have visually comparable   
   quality. The savings in size between output.arithmetic.jpg and o   
   tput.optimized.jpg are a factor exceeding 3.3. (And NOT a tiny one-digit   
   percentage, as reported in “A Review    
   on JPEG2000 Image Compression” by Maini and Mehra and in the Wikipedia   
   article on JPEG.) These savings do make a difference if you use more images of   
   this kind and your storage or bandwidth is restricted.   
      
   --==bound.0.ca1e715c64d489b57073e516db55d2d3.mail.rambler.ru=Con   
   ent-Transfer-Encoding: quoted-printable   
   Content-Type: text/html; charset=utf-8   
      
   
A reader might wonder, why care about the arithmetic encoding? Recently,   
   I tried to compress a rather big scan (private content, an A4 page at 1200 dpi   
   with black foreground text and a nontrivial green-white background) into a   
   JPEG file of a    
   possibly small size. I ran:

cjpeg -quality 0 -arithmetic -dct float -outfile   
   output.arithmetic.jpg -report -strict -verbose input.pnm
<   
   div>cjpeg -quality 0 -optimize -dct float -outfile output.optimized.jpg   
   -report -strict -verbose input.pnm

cjpeg -quality 0 -dct float -outfile output.unoptimized.jpg   
   -report -strict -verbose input.pnm


The    
   sizes are this:

outp   
   t.arithmetic.jpg: 239224 bytes

   
   /div>
output.optimized.jpg: 807963 bytes

output.unoptimized.jpg: 1928235    
   bytes

The foregrounds of the three   
   files are readable and have visually comparable quality. The savings in size   
   between output.arithmetic.jpg and output.optimized.jpg are a factor exceeding   
   3.3. (And NOT a tiny    
   one-digit percentage, as reported in “A Review on JPEG2000 Image   
   Compression” by Maini and Mehra and in the Wikipedia article on JPEG.) These   
   savings do make a difference if you use more images of this kind and your   
   storage or bandwidth is restricted.   
   
   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)