
This document only mentions what has changed since exomizer 2.0. For
more general exomizer info please read the exo20info.txt file which also
might contain updated information about previously existing features.

Exomizer 3.1.2
--------------

New features since 3.1.1
------------------------
o  The license has been changed from a home grown license based on the zlib license to
   a pure zlib license which makes us compatible with GPL. Please observe that the
   distribution also contains contributed decruncher source code in rawdecrs/6809/,
   rawdecrs/8080/, rawdecrs/8086/ and rawdecrs/z80/ using different licenses which have
   not changed.
o  The exodecr/exodecrunch.s 6502 decruncher now also exists in a version for kick-
   assembler. See exodecrs/kick for the files. As the existing versions for dasm
   and acme it implement the same feature set as the ca65 version.

For bugfixes and minor improvements see the file changelog.txt for more details.

Exomizer 3.1.1
--------------

New features since 3.1
----------------------
o New -y option to the sfx command to add text to the end of the generated
  SYS-line (or CALL).
o new target, -t 65, for the Commodore c65 and the MEGA65 computer.


Exomizer 3.1
------------

New features since 3.0
----------------------
o  Improved compression on average due to offset reuse. The bit stream format
   for crunched files has been extended with a new bit to control this:

   bit 5  Decides if we may reuse the latest sequence offset after a single
          literal byte or a single literal sequence: 1=enable, 0=disable

   This featue is enabled by default so the raw, mem and level modes now
   default to -P39.

o  The flags -e and -E have been extended to support split encoding. This
   means that the encoding used by a crunched file is provided by a file of
   its own and no longer included in the crunched file itself.

   This can save space if several crunched files contain similar data and can
   share the encoding. To facilitate this The -E flag now enables the mem,
   level and raw sub command to generate a header that is optimized for all
   the given input files. See the -E documentation in exo20info.txt for more
   details. Support for this has been added to exodecr/exodecrunch.s but is
   disabled by default.

o  The exodecr/exodecrunch.s 6502 decruncher now supports forward decrunching.
   There are also versions for dasm and acme too. They implement the same
   feature set as the ca65 version.

o  Changes to contributed decrunchers:
   Added ARM 32bit thumb2 source code by ALeX Kazik
   Added Intel 8080 source code by Ivan Gorodetsky
   Updated Zilog Z80 source code by Antonio Villena

o  The -P flag has been extended to make it possible to only affect certain
   bits by prefixing a value with a - or a + character. This will modify the
   current value using AND or OR respectively instead of replacing it. E.g.
   -P+16 will set bit 4 but leave the other bits as they were. Likewise, -P-16
   clears bit 4 but leave the others. The syntax also allows for chaining
   multiple operations. E.g. -P-32+16 is allowed. An initial absolute value is
   also accepted like this -P43-32+16 .

o  Miscellaneous bug fixes, See changelog.txt for details.

New features since 2.0
---------------------

o  New bit stream format for crunched files that allow for faster decrunching.

o  The bit stream format is controlled by a new -P<bit-field> flag. The bits
   control different aspects of the bit stream format. -P0 disables all new
   aspects and will generate exomizer 2.0 compatible files. The raw, mem and
   level modes default to -P7. (-P39 and new bit 5 since 3.1)

   bit 0  Controls bit stream bit orientation, 1=big endian, 0=little endian
   bit 1  Controls how more than 7 bits are shifted 1=split into a shift of
          less than 8 bits + a byte, 0=all bits are shifted
   bit 2  Implicit first literal byte: 1=enable, 0=disable
   bit 3  Align bit stream towards start without shift flag bit included:
          1=enable, 0=disable
   bit 4  Decides if we are to have two lengths (1 and 2) or three lengths
          (1, 2 and 3) using offset tables of their own: 0=two, 1=three

   Please note that many decrunchers has not yet been ported to the new bit
   stream format and requires a -P0 on the command-line. However,
   exodecrs/exodecrunch.s, rawdecrs/exodecr.c and rawdecrs/exodecrunch.c will
   work out of the box. Only the built in decruncher src/exodec.c handles all
   possible bit combinations. Please check it out for implementation details.

o  There is a new -T<bit-field> flag that controls traits of the bit stream
   that don't affect the bit stream format. Typically traits are disallowed to
   make the decruncher faster and/or shorter. The raw mode defaults to -T0.
   The mem and level modes default to -T4.

   bit 0  Disallow literal sequences, 0 allow, 1 disallow
   bit 1  Disallow sequences with length 1, 0 allow, 1 disallow
   bit 2  Disallow sequences with lengths (256*m + n) where m > 0 and n is in
          the interval [0 - p] and p is 2 or 3 depending on bit 4 of the
          -P<bit-field>, 0 allow, 1 disallow

o  The sfx decruncher and exodecrs/exodecrunch.s take advantage of the new bit
   stream format for decrunching speed improvements. They also become faster
   if the sequence length is limited to 256 by using -M256 when crunching.
   (exodecrs/exodecrunch.s has to be told by a define, see the comments in the
   file for more info. It also needs the bit 2 trait (-T4) which is used by
   default by the level and mem sub commands but not by raw)

o  The sfx decruncher now recognizes a new symbol, i_perf, that controls the
   decruncher size/speed performance trade-off. The valid values are -1 to 2
   where -1 is slowest/shortest and 2 is fastest/largest. Default is 0. They
   all are faster than exomizer 2.11 by some margin. None of the decrunchers
   will destroy zero-page.

o  The sfx decruncher supports the -P+16 flag to enable a dedicated offset
   table for sequences of length 3. Using this flag might increase compression
   a bit for some files but will also make the decrunch table larger, 204
   bytes instead of the default 156 bytes.

o  Very slightly improved compression.


Contributing
------------

   You are, as before, very welcome to contribute decruncher source code for
   unsupported targets and/or improving existing decrunchers.

   However, The unclear licensing of some of the contributions so far has not
   been ideal. So please decide on what license to use before contributing. If
   you are unsure about the licensing then I recommend using the same license
   that I use for this project, the zlib license. See exodecrs/exodecrunch.s
   for how it can be included in the source file as a comment block.

Sources of inspiration
----------------------

   The new bit stream format features/traits and the 6502 decruncher speed
   improvements have been inspired from several sources that I would like to
   mention (in no particular order):

   * The exoopt tool developed by Metalbrain and Antonio Villena
     (z80 decruncher contributors, the new -P and -T flags should be able to
      replicate all bit protocol variants that exoopt generates).
   * The Doynamite 6502 decruncher by Doynax and the subsizer cruncher by TLR,
     for raising the bar.
   * Feedback from users

Benchmarks for the improved 6502 decrunchers
--------------------------------------------

The measured time is the number of cycles the decruncher needs to run.
Interrupt processing and other kinds of overhead that exists in a typical
computer environment is not included. (M=*1000000 and k=*1000)

* The benchmark for the sfx decruncher use the file zorrounpacked.prg file
(https://web.archive.org/web/20050102175732/http://www.swolff.dk/cruncher/zorrounpacked.prg) also used in the README.txt of subsizer for comparisons:

Sorted on size:
| File name                      | Size  | Reduced | Cycles  | C/B    | B/kC  |
|--------------------------------|-------|---------|---------|--------|-------|
| zorro_exo31_-1-P+16.prg        | 30231 |  44.12% | 6073989 | 112.27 |  8.91 |
| zorro_exo31_0-P+16.prg         | 30246 |  44.10% | 5933882 | 109.68 |  9.12 |
| zorro_exo31_1-P+16.prg         | 30254 |  44.08% | 5670418 | 104.81 |  9.54 |
| zorro_exo31_0-M256-P+16.prg    | 30267 |  44.06% | 5659124 | 104.60 |  9.56 |
| zorro_exo31_1-M256-P+16.prg    | 30279 |  44.03% | 5450866 | 100.75 |  9.93 |
| zorro_exo31_2-M256-P+16.prg    | 30305 |  43.99% | 5271433 |  97.43 | 10.26 |
| zorro_exo31_2-M256.prg         | 30365 |  43.88% | 5268500 |  97.38 | 10.27 |
| zorro_exo31_2-M256-P-32+16.prg | 30697 |  43.26% | 5123329 |  94.70 | 10.56 |
| zorro_exo31_2-M256-P-32.prg    | 30758 |  43.15% | 5120580 |  94.65 | 10.57 |

Sorted on cycles:
| File name                      | Size  | Reduced | Cycles  | C/B    | B/kC  |
|--------------------------------|-------|---------|---------|--------|-------|
| zorro_exo31_2-M256-P-32.prg    | 30758 |  43.15% | 5120580 |  94.65 | 10.57 |
| zorro_exo31_2-M256-P-32+16.prg | 30697 |  43.26% | 5123329 |  94.70 | 10.56 |
| zorro_exo31_2-M256.prg         | 30365 |  43.88% | 5268500 |  97.38 | 10.27 |
| zorro_exo31_2-M256-P+16.prg    | 30305 |  43.99% | 5271433 |  97.43 | 10.26 |
| zorro_exo31_1-M256-P+16.prg    | 30279 |  44.03% | 5450866 | 100.75 |  9.93 |
| zorro_exo31_0-M256-P+16.prg    | 30267 |  44.06% | 5659124 | 104.60 |  9.56 |
| zorro_exo31_1-P+16.prg         | 30254 |  44.08% | 5670418 | 104.81 |  9.54 |
| zorro_exo31_0-P+16.prg         | 30246 |  44.10% | 5933882 | 109.68 |  9.12 |
| zorro_exo31_-1-P+16.prg        | 30231 |  44.12% | 6073989 | 112.27 |  8.91 |

The information in the two tables above has been generatedby the exomizer
desfx -S command.

* The benchmark for the memory decruncher exodecrunch.s uses the "Pearls for
Pigs" files (http://codebase64.org/doku.php?id=base:compression_benchmarks):

(get_bits not in-lined, unlimited sequence lengths)
| File name | Size  | Reduced | Cycles   | C/B    | B/kC  |
|-----------|-------|---------|----------|--------|-------|
| pfp_1.cru |  2934 |  73.35% |   744298 |  67.61 | 14.79 |
| pfp_2.cru |  2157 |  56.63% |   468603 |  94.23 | 10.61 |
| pfp_3.cru |  1727 |  56.27% |   349649 |  88.54 | 11.29 |
| pfp_4.cru |  3427 |  51.15% |   662205 |  94.38 | 10.59 |
| pfp_5.cru | 19186 |  44.80% |  3517541 | 101.20 |  9.88 |
| pfp_6.cru |  8076 |  74.45% |  1974737 |  62.48 | 16.00 |
| pfp_7.cru |  8644 |  57.61% |  1771969 |  86.90 | 11.51 |
| pfp_8.cru |  2926 |  48.78% |   519864 |  91.00 | 10.99 |
| pfp_9.cru |  5166 |  42.34% |   976430 | 108.98 |  9.18 |
| Total     | 54243 |  57.75% | 10985296 |  85.57 | 11.69 |

(get_bits in-lined, sequence lengths limited to 256 or less, no offset reuse)
| File name          | Size  | Reduced | Cycles   | C/B   | B/kC  |
|--------------------|-------|---------|----------|-------|-------|
| pfp_1.cruiM256P-32 |  2948 |  73.22% |   672500 | 61.09 | 16.37 |
| pfp_2.cruiM256P-32 |  2199 |  55.78% |   419715 | 84.40 | 11.85 |
| pfp_3.cruiM256P-32 |  1784 |  54.82% |   318707 | 80.71 | 12.39 |
| pfp_4.cruiM256P-32 |  3429 |  51.13% |   583501 | 83.17 | 12.02 |
| pfp_5.cruiM256P-32 | 19586 |  43.65% |  3239878 | 93.21 | 10.73 |
| pfp_6.cruiM256P-32 |  8392 |  73.45% |  1838881 | 58.18 | 17.19 |
| pfp_7.cruiM256P-32 |  8763 |  57.03% |  1607365 | 78.82 | 12.69 |
| pfp_8.cruiM256P-32 |  3073 |  46.21% |   489281 | 85.64 | 11.68 |
| pfp_9.cruiM256P-32 |  5314 |  40.69% |   889859 | 99.31 | 10.07 |
| Total              | 55488 |  56.78% | 10059687 | 78.36 | 12.76 |

You can make these tables yourself by naming the pfp files pfp[1-9].raw,
dropping them in the exodecrs/perf folder and in that folder execute the
following command line: $ make pfp.stat pfp.statiM256P-32

(You need to build exomizer in the src folder first and you also need to have
cc65, gcc and gnu make on the path) It works in Linux and in a Windows
mingw/msys2 environment in Windows.
