home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1993 #3 / NN_1993_3.iso / spool / comp / compress / 4791 < prev    next >
Encoding:
Internet Message Format  |  1993-01-25  |  1.0 KB
  1. Path: sparky!uunet!math.fu-berlin.de!uni-paderborn.de!urmel.informatik.rwth-aachen.de!messua!dak
  2. From: dak@messua.informatik.rwth-aachen.de (David Kastrup)
  3. Newsgroups: comp.compression
  4. Subject: Re: JPEG compression
  5. Date: 25 Jan 93 14:06:27 GMT
  6. Organization: Rechnerbetrieb Informatik - RWTH Aachen
  7. Lines: 20
  8. Distribution: world
  9. Message-ID: <dak.727970787@messua>
  10. References: <peterb.727926960@wambenger>
  11. NNTP-Posting-Host: messua.informatik.rwth-aachen.de
  12. Keywords: JPEG, compression
  13.  
  14. peterb@cs.uwa.oz.au (Peter Bousfield) writes:
  15.  
  16. >Hello fellow compressors,
  17.  
  18. >It is my understanding that JPEG compression
  19. >works in basically two parts.
  20. >Step 1: DCT (Discrete Cosine Transform)
  21. >and
  22. >Step 2: Huffman Coding (for those of us
  23. >            who don't wish to
  24. >            pay for a license.)
  25.  
  26. >I was wanting to know which of these
  27. >two steps contributes more to the compression.
  28. >That is, how good is the compression without
  29. >Huffman coding (eg is it 50% worse or only 10%
  30. >worse.)
  31.  
  32. Of course you forgot the most important Step 1.5:
  33. Throwing away information. THAT really contributes.
  34.