Tile compression: Revision history

2 May 2016

• curprev 16:0816:08, 2 May 2016‎ Bisqwit talk contribs‎ m 18,253 bytes +163‎ →‎Tokumaru: rearrange text a little bit
• curprev 12:3312:33, 2 May 2016‎ Bisqwit talk contribs‎ m 18,090 bytes +18‎ →‎Tokumaru: variable name
• curprev 12:3212:32, 2 May 2016‎ Bisqwit talk contribs‎ m 18,072 bytes +96‎ →‎Tokumaru: reword
• curprev 12:2612:26, 2 May 2016‎ Bisqwit talk contribs‎ 17,976 bytes +133‎ →‎Tokumaru: tile size
• curprev 12:2412:24, 2 May 2016‎ Bisqwit talk contribs‎ 17,843 bytes +2,772‎ →‎Tokumaru: Spell out the format. Depending on external links for textual content is bad wiki practice.
• curprev 11:5611:56, 2 May 2016‎ Bisqwit talk contribs‎ 15,071 bytes −263‎ →‎RLEINC: Remove some text that is not really useful for the reader

1 May 2016

• curprev 01:2101:21, 1 May 2016‎ Bisqwit talk contribs‎ m 15,334 bytes −13‎ →‎Codemasters: rewording
• curprev 01:1301:13, 1 May 2016‎ Bisqwit talk contribs‎ m 15,347 bytes +27‎ Add anchor for Tokumaru compression
• curprev 00:4600:46, 1 May 2016‎ Bisqwit talk contribs‎ 15,320 bytes +36‎ After analyzing the Codemasters algorithm in detail, I don't see any evidence of Run-Length encoding. The closest semblance is a bit that indicates whether to change data for new row or to repeat the previous one. I'm moving it under a separate header.

30 April 2016

• curprev 23:3223:32, 30 April 2016‎ Bisqwit talk contribs‎ m 15,284 bytes +1‎ →‎Tokumaru: typo fix
• curprev 23:3223:32, 30 April 2016‎ Bisqwit talk contribs‎ 15,283 bytes +149‎ Add Tokumaru explanation article
• curprev 21:5721:57, 30 April 2016‎ Bisqwit talk contribs‎ 15,134 bytes +56‎ →‎Bit-based RLE: Add Tokumaru compression benchmark
• curprev 21:4621:46, 30 April 2016‎ Bisqwit talk contribs‎ m 15,078 bytes −1‎ →‎RLEINC: typo
• curprev 21:4421:44, 30 April 2016‎ Bisqwit talk contribs‎ 15,079 bytes +229‎ →‎RLEINC: explain purpose
• curprev 21:3621:36, 30 April 2016‎ Bisqwit talk contribs‎ 14,850 bytes +37‎ →‎Bit-based RLE: clarify
• curprev 21:2421:24, 30 April 2016‎ Bisqwit talk contribs‎ m 14,813 bytes +5‎ →‎Bit-based RLE: missing word
• curprev 21:2321:23, 30 April 2016‎ Bisqwit talk contribs‎ m 14,808 bytes −1‎ →‎Bit-based RLE: typo
• curprev 21:2321:23, 30 April 2016‎ Bisqwit talk contribs‎ 14,809 bytes −510‎ →‎Bit-based RLE: In hindsight, less data is better in this case.
• curprev 21:0921:09, 30 April 2016‎ Bisqwit talk contribs‎ 15,319 bytes +76‎ →‎Chrono Trigger LZ: Rename the instruction byte into mask byte for consistency in the LZSS chapter
• curprev 20:5420:54, 30 April 2016‎ Bisqwit talk contribs‎ 15,243 bytes +175‎ →‎Bit-based RLE: Reason
• curprev 20:4920:49, 30 April 2016‎ Bisqwit talk contribs‎ m 15,068 bytes +28‎ →‎Bit-based RLE: comments
• curprev 20:4820:48, 30 April 2016‎ Bisqwit talk contribs‎ 15,040 bytes +88‎ →‎Bit-based RLE: correct data (there was a bug in the benchmark)
• curprev 18:3818:38, 30 April 2016‎ Bisqwit talk contribs‎ 14,952 bytes +1,386‎ →‎Run-length encoding: Added an example bit-based RLE (profiled and optimized exhaustively using a brute force algorithm)

28 April 2016

• curprev 12:5112:51, 28 April 2016‎ Bisqwit talk contribs‎ m 13,566 bytes +11‎ →‎RLEINC: wording

27 April 2016

• curprev 15:5615:56, 27 April 2016‎ Bisqwit talk contribs‎ 13,555 bytes +45‎ →‎Chrono Trigger LZ: elaborations
• curprev 15:4315:43, 27 April 2016‎ Bisqwit talk contribs‎ m 13,510 bytes −1‎ →‎Chrono Trigger LZ: clarifications
• curprev 15:4115:41, 27 April 2016‎ Bisqwit talk contribs‎ 13,511 bytes +128‎ →‎Chrono Trigger LZ: clarifications
• curprev 15:4015:40, 27 April 2016‎ Bisqwit talk contribs‎ m 13,383 bytes +1‎ →‎Chrono Trigger LZ: typo
• curprev 15:3715:37, 27 April 2016‎ Bisqwit talk contribs‎ 13,382 bytes +1,683‎ →‎LZSS: Add Chrono Trigger method
• curprev 15:1115:11, 27 April 2016‎ Bisqwit talk contribs‎ 11,699 bytes +1,678‎ →‎LZSS: Add Pokémon LZ method
• curprev 14:5214:52, 27 April 2016‎ Bisqwit talk contribs‎ 10,021 bytes +29‎ →‎RLEINC: +notes
• curprev 14:4014:40, 27 April 2016‎ Bisqwit talk contribs‎ 9,992 bytes +92‎ →‎RLEINC: elaborations
• curprev 13:5113:51, 27 April 2016‎ Bisqwit talk contribs‎ 9,900 bytes +1,005‎ Add RLEINC

25 April 2016

• curprev 14:1814:18, 25 April 2016‎ Tepples talk contribs‎ 8,895 bytes +87‎ →‎PB53: Where to get PB53

17 June 2015

• curprev 17:2117:21, 17 June 2015‎ Tepples talk contribs‎ 8,808 bytes +12‎ →‎PB53: clarify which "run lengths" are unary coded, as LZSS uses a 4-bit binary code for each back-reference's length

27 May 2015

• curprev 01:3001:30, 27 May 2015‎ Tepples talk contribs‎ 8,796 bytes −66‎ →‎LZ77UnComp: Clarify that I'm deferring to GBATEK for details'
• curprev 01:1701:17, 27 May 2015‎ Tepples talk contribs‎ 8,862 bytes +119‎ →‎LZ77UnComp: mention the header, as well as the fact that the DS BIOS supports it too
• curprev 01:0501:05, 27 May 2015‎ Tepples talk contribs‎ 8,743 bytes −127‎ →‎LZ77UnComp: don't imply existence of other restrictions
• curprev 00:5300:53, 27 May 2015‎ Tepples talk contribs‎ 8,870 bytes +86‎ →‎LZ77UnComp: If RFC 2119 is jarring, then perhaps the standard for the C language might be a better source of terms

26 May 2015

• curprev 17:2217:22, 26 May 2015‎ Rainwarrior talk contribs‎ 8,784 bytes −14‎ →‎LZSS: toning down the "MUST NOT" again, RFC2119 box is just noise inhibiting reading comprehension

20 April 2015

• curprev 06:0506:05, 20 April 2015‎ Tepples talk contribs‎ m 8,798 bytes +23‎ →‎Oracle common byte: more concrete description of a worst case block
• curprev 06:0406:04, 20 April 2015‎ Tepples talk contribs‎ 8,775 bytes +49‎ →‎PB53: number of planes in a tile is theoretically open
• curprev 05:4805:48, 20 April 2015‎ Tepples talk contribs‎ m 8,726 bytes +30‎ →‎LZSS: Found two MUST NOTs; added {{RFC 2119}}

30 June 2014

• curprev 00:5400:54, 30 June 2014‎ Tepples talk contribs‎ m 8,696 bytes +78‎ →‎LZSS: fix incomplete refactors while I was writing the Oracle LZ part

28 June 2014

• curprev 16:5316:53, 28 June 2014‎ Tepples talk contribs‎ 8,618 bytes +3,117‎ A couple LZ codecs, plus common byte

26 June 2014

• curprev 18:1518:15, 26 June 2014‎ Tepples talk contribs‎ 5,501 bytes +144‎ →‎Codemasters: improved version
• curprev 17:0417:04, 26 June 2014‎ Tepples talk contribs‎ 5,357 bytes +206‎ Ad Hoc Compression Methods: RLE
• curprev 00:0300:03, 26 June 2014‎ Tepples talk contribs‎ 5,151 bytes +881‎ PackBits arrived earlier (Mac OS 1 in 1Q 1984, FDS in 1Q 1986); add GBA for the sake of completeness

25 June 2014

• curprev 21:4421:44, 25 June 2014‎ Tepples talk contribs‎ m 4,270 bytes +89‎ oops
• curprev 21:4221:42, 25 June 2014‎ Tepples talk contribs‎ 4,181 bytes +4,181‎ Created page with "'''Tile compression''' refers to techniques that allow fitting more graphics data into a smaller space. Programs using CHR ROM cannot use compressed ti..."