Yamaha RX-V3900 User Manual

Page of 169
xi
The emphasis in designing this software has been on achieving portability and 
flexibility, while also making it fast enough to be useful. In particular, the software is not 
intended to be read as a tutorial on JPEG. (See the REFERENCES section for 
introductory material.) Rather, it is intended to be reliable, portable, industrial-strength 
code. We do not claim to have achieved that goal in every aspect of the software, but we 
strive for it.
We welcome the use of this software as a component of commercial products. No 
royalty is required, but we do ask for an acknowledgement in product documentation, as 
described under LEGAL ISSUES.
LEGAL ISSUES
In plain English:
1.
We don't promise that this software works.  (But if you find any bugs, please let us 
know!)
2.
You can use this software for whatever you want. You don't have to pay us.
3.
You may not pretend that you wrote this software. If you use it in a program, you 
must acknowledge somewhere in your documentation that you've used the IJG 
code.
In legalese:
The authors make NO WARRANTY or representation, either express or implied, with 
respect to this software, its quality, accuracy, merchantability, or fitness for a particular 
purpose. This software is provided "AS IS", and you, its user, assume the entire risk as 
to its quality and accuracy.
This software is copyright (C) 1991-1998, Thomas G. Lane.
All Rights Reserved except as specified below.
Permission is hereby granted to use, copy, modify, and distribute this software (or 
portions thereof) for any purpose, without fee, subject to these conditions:
(1) If any part of the source code for this software is distributed, then this README file 
must be included, with this copyright and no-warranty notice unaltered; and any 
additions, deletions, or changes to the original files must be clearly indicated in 
accompanying documentation.
(2) If only executable code is distributed, then the accompanying documentation must 
state that "this software is based in part on the work of the Independent JPEG 
Group".
(3) Permission for use of this software is granted only if the user accepts full 
responsibility for any undesirable consequences; the authors accept NO LIABILITY 
for damages of any kind.
These conditions apply to any software derived from or based on the IJG code, not just 
to the unmodified library. If you use our work, you ought to acknowledge us.
Permission is NOT granted for the use of any IJG author's name or company name in 
advertising or publicity relating to this software or products derived from it. This 
software may be referred to only as "the Independent JPEG Group's software".
We specifically permit and encourage the use of this software as the basis of commercial 
products, provided that all warranty or liability claims are assumed by the product 
vendor.
ansi2knr.c is included in this distribution by permission of L. Peter Deutsch, sole 
proprietor of its copyright holder, Aladdin Enterprises of Menlo Park, CA.
ansi2knr.c is NOT covered by the above copyright and conditions, but instead by the 
usual distribution terms of the Free Software Foundation; principally, that you must 
include source code if you redistribute it. (See the file ansi2knr.c for full details.)  
However, since ansi2knr.c is not needed as part of any program generated from the IJG 
code, this does not limit you more than the foregoing paragraphs do.
The Unix configuration script "configure" was produced with GNU Autoconf. It is 
copyright by the Free Software Foundation but is freely distributable. The same holds 
for its supporting scripts (config.guess, config.sub, ltconfig, ltmain.sh). Another support 
script, install-sh, is copyright by M.I.T. but is also freely distributable.
It appears that the arithmetic coding option of the JPEG spec is covered by patents 
owned by IBM, AT&T, and Mitsubishi. Hence arithmetic coding cannot legally be used 
without obtaining one or more licenses. For this reason, support for arithmetic coding 
has been removed from the free JPEG software. (Since arithmetic coding provides only 
a marginal gain over the unpatented Huffman mode, it is unlikely that very many 
implementations will support it.) So far as we are aware, there are no patent restrictions 
on the remaining code.
The IJG distribution formerly included code to read and write GIF files. To avoid 
entanglement with the Unisys LZW patent, GIF reading support has been removed 
altogether, and the GIF writer has been simplified to produce "uncompressed GIFs".  
This technique does not use the LZW algorithm; the resulting GIF files are larger than 
usual, but are readable by all standard GIF decoders.
We are required to state that
"The Graphics Interchange Format(c) is the Copyright property of CompuServe 
Incorporated. GIF(sm) is a Service Mark property of CompuServe Incorporated."
REFERENCES
We highly recommend reading one or more of these references before trying to 
understand the innards of the JPEG software.
The best short technical introduction to the JPEG compression algorithm is 
Wallace, Gregory K. "The JPEG Still Picture Compression Standard", 
Communications of the ACM, April 1991 (vol. 34 no. 4), pp. 30-44.
(Adjacent articles in that issue discuss MPEG motion picture compression, applications 
of JPEG, and related topics.) If you don't have the CACM issue handy, a PostScript file 
containing a revised version of Wallace's article is available at ftp://ftp.uu.net/graphics/
jpeg/wallace.ps.gz. The file (actually a preprint for an article that appeared in IEEE 
Trans. Consumer Electronics) omits the sample images that appeared in CACM, but it 
includes corrections and some added material. Note: the Wallace article is copyright 
ACM and IEEE, and it may not be used for commercial purposes.
A somewhat less technical, more leisurely introduction to JPEG can be found in "The 
Data Compression Book" by Mark Nelson and Jean-loup Gailly, published by M&T 
Books (New York), 2nd ed. 1996, ISBN 1-55851-434-1. This book provides good 
explanations and example C code for a multitude of compression methods including 
JPEG.  It is an excellent source if you are comfortable reading C code but don't know 
much about data compression in general. The book's JPEG sample code is far from 
industrial-strength, but when you are ready to look at a full implementation, you've got 
one here...
The best full description of JPEG is the textbook "JPEG Still Image Data Compression 
Standard" by William B. Pennebaker and Joan L. Mitchell, published by Van Nostrand 
Reinhold, 1993, ISBN 0-442-01272-1. Price US$59.95, 638 pp. 
The book includes the complete text of the ISO JPEG standards (DIS 10918-1 and draft 
DIS 10918-2).  This is by far the most complete exposition of JPEG in existence, and we 
highly recommend it.
The JPEG standard itself is not available electronically; you must order a paper copy 
through ISO or ITU. (Unless you feel a need to own a certified official copy, we 
recommend buying the Pennebaker and Mitchell book instead; it's much cheaper and 
includes a great deal of useful explanatory material.)
In the USA, copies of the standard may be ordered from ANSI Sales at (212)642-4900, 
or from Global Engineering Documents at (800) 854-7179. (ANSI doesn't take credit 
card orders, but Global does.) It's not cheap: as of 1992, ANSI was charging $95 for Part 
1 and $47 for Part 2, plus 7% shipping/handling. The standard is divided into two parts, 
Part 1 being the actual specification, while Part 2 covers compliance testing methods.  
Part 1 is titled "Digital Compression and Coding of Continuous-tone Still Images, Part 
1: Requirements and guidelines" and has document numbers ISO/IEC IS 10918-1, ITU-
T T.81.  Part 2 is titled "Digital Compression and Coding of Continuous-tone Still 
Images, Part 2: Compliance testing" and has document numbers ISO/IEC IS 10918-2, 
ITU-T T.83.
Some extensions to the original JPEG standard are defined in JPEG Part 3, a newer ISO 
standard numbered ISO/IEC IS 10918-3 and ITU-T T.84. IJG currently does not support 
any Part 3 extensions.
The JPEG standard does not specify all details of an interchangeable file format. For the 
omitted details we follow the "JFIF" conventions, revision 1.02. A copy of the JFIF spec 
is available from:
Literature Department
C-Cube Microsystems, Inc.
1778 McCarthy Blvd.
Milpitas, CA 95035
phone (408) 944-6300,  fax (408) 944-6314
A PostScript version of this document is available by FTP at 
ftp://ftp.uu.net/graphics/jpeg/jfif.ps.gz. There is also a plain text version at 
ftp://ftp.uu.net/graphics/jpeg/jfif.txt.gz, but it is missing the figures.
The TIFF 6.0 file format specification can be obtained by FTP from ftp://ftp.sgi.com/
graphics/tiff/TIFF6.ps.gz. The JPEG incorporation scheme found in the TIFF 6.0 spec 
of 3-June-92 has a number of serious problems. IJG does not recommend use of the 
TIFF 6.0 design (TIFF Compression tag 6). Instead, we recommend the JPEG design 
proposed by TIFF Technical Note #2 (Compression tag 7). Copies of this Note can be 
obtained from ftp.sgi.com or from ftp://ftp.uu.net/graphics/jpeg/. It is expected that the 
next revision of the TIFF spec will replace the 6.0 JPEG design with the Note's design. 
Although IJG's own code does not support TIFF/JPEG, the free libtiff library uses our 
library to implement TIFF/JPEG per the Note. libtiff is available from 
ftp://ftp.sgi.com/graphics/tiff/.
ARCHIVE LOCATIONS
The "official" archive site for this software is ftp.uu.net (Internet address 192.48.96.9).  
The most recent released version can always be found there in directory graphics/jpeg.  
This particular version will be archived as 
ftp://ftp.uu.net/graphics/jpeg/jpegsrc.v6b.tar.gz. If you don't have direct Internet access, 
UUNET's archives are also available via UUCP; contact help@uunet.uu.net for 
information on retrieving files that way.
Numerous Internet sites maintain copies of the UUNET files. However, only ftp.uu.net 
is guaranteed to have the latest official version.
You can also obtain this software in DOS-compatible "zip" archive format from the 
SimTel archives (ftp://ftp.simtel.net/pub/simtelnet/msdos/graphics/), or on CompuServe 
in the Graphics Support forum (GO CIS:GRAPHSUP), library 12 "JPEG Tools". Again, 
these versions may sometimes lag behind the ftp.uu.net release.
The JPEG FAQ (Frequently Asked Questions) article is a useful source of general 
information about JPEG.  It is updated constantly and therefore is not included in this 
distribution.  The FAQ is posted every two weeks to Usenet newsgroups 
comp.graphics.misc, news.answers, and other groups. It is available on the World Wide 
Web at http://www.faqs.org/faqs/jpeg-faq/ and other news.answers archive sites, 
including the official news.answers archive at rtfm.mit.edu: ftp://rtfm.mit.edu/pub/
usenet/news.answers/jpeg-faq/.
If you don't have Web or FTP access, send e-mail to mail-server@rtfm.mit.edu with 
body