(Link to AcmlmWiki) Offline: thank ||bass
Register | Login
Views: 13,040,846
 Main | Memberlist | Active users | Calendar | Chat | Online users
Ranks | FAQ | ACS | Stats | Color Chart | Search | Photo album 		05-05-24 04:30 AM
0 users currently in ROM Hacking.
Acmlm's Board - I3 Archive - ROM Hacking - Compression broken, problems caused New poll | |
Add to favorites | Next newer thread | Next older thread
User Post
Gideon Zhi

Keese








Since: 12-05-05
From: ...behind you! Boo!

Last post: 6287 days
Last view: 6285 days
Posted on 04-06-06 03:04 AM Link | Quote
Okay, so I've hit a bit of a problem. I'm working on this game whose text originally was compressed. I broke the compression, preferring to store my data decompressed in-rom.

The problem is that each of the compressed strings stored the decompressed size with the string, PASCAL style. Now, the game's still expecting those bytes to be at the beginning of my strings, and it really doesn't like it when those values aren't there - or when the values that are there are much bigger than it's used to handling. Here's some example code...

From the original... 
$81/F86E C2 20       REP #$20                A:F99A X:9821 Y:0000 P:eNvMxdizc

$81/F870 E6 23       INC $23    [$00:0023]   A:F99A X:9821 Y:0000 P:eNvmxdizc

$81/F872 E6 23       INC $23    [$00:0023]   A:F99A X:9821 Y:0000 P:eNvmxdizc

$81/F874 BD 00 00    LDA $0000,x[$B0:9821]   A:F99A X:9821 Y:0000 P:eNvmxdizc

$81/F877 85 0E       STA $0E    [$00:000E]   A:0021 X:9821 Y:0000 P:envmxdizc

$81/F879 E8          INX                     A:0021 X:9821 Y:0000 P:envmxdizc

$81/F87A E8          INX                     A:0021 X:9822 Y:0000 P:eNvmxdizc

$81/F87B 80 1E       BRA $1E    [$F89B]      A:0021 X:9823 Y:0000 P:eNvmxdizc

$81/F89B E2 20       SEP #$20                A:0021 X:9823 Y:0000 P:eNvmxdizc

$81/F89D 64 0D       STZ $0D    [$00:000D]   A:0021 X:9823 Y:0000 P:eNvMxdizc

$81/F89F A5 25       LDA $25    [$00:0025]   A:0021 X:9823 Y:0000 P:eNvMxdizc

$81/F8A1 85 AE       STA $AE    [$00:00AE]   A:00A1 X:9823 Y:0000 P:eNvMxdizc

$81/F8A3 A0 00 00    LDY #$0000              A:00A1 X:9823 Y:0000 P:eNvMxdizc

$81/F8A6 A9 FF       LDA #$FF                A:00A1 X:9823 Y:0000 P:envMxdiZc

$81/F8A8 85 07       STA $07    [$00:0007]   A:00FF X:9823 Y:0000 P:eNvMxdizc

$81/F8AA BD 00 00    LDA $0000,x[$B0:9823]   A:00FF X:9823 Y:0000 P:eNvMxdizc

$81/F8AD 85 06       STA $06    [$00:0006]   A:003F X:9823 Y:0000 P:envMxdizc

$81/F8AF E8          INX                     A:003F X:9823 Y:0000 P:envMxdizc

$81/F8B0 06 06       ASL $06    [$00:0006]   A:003F X:9824 Y:0000 P:eNvMxdizc

$81/F8B2 90 11       BCC $11    [$F8C5]      A:003F X:9824 Y:0000 P:envMxdizc

$81/F8C5 BD 00 00    LDA $0000,x[$B0:9824]   A:003F X:9824 Y:0000 P:envMxdizc

$81/F8C8 20 96 F9    JSR $F996  [$81:F996]   A:000D X:9824 Y:0000 P:envMxdizc


From the new build, with the direct-from-ROM load in place... 
$81/F86E C2 20       REP #$20                A:F99A X:B0A5 Y:0000 P:eNvMxdizc

$81/F870 E6 23       INC $23    [$00:0023]   A:F99A X:B0A5 Y:0000 P:eNvmxdizc

$81/F872 E6 23       INC $23    [$00:0023]   A:F99A X:B0A5 Y:0000 P:eNvmxdizc

$81/F874 BD 00 00    LDA $0000,x[$E8:B0A5]   A:F99A X:B0A5 Y:0000 P:eNvmxdizc

$81/F877 85 0E       STA $0E    [$00:000E]   A:47F4 X:B0A5 Y:0000 P:envmxdizc

$81/F879 E8          INX                     A:47F4 X:B0A5 Y:0000 P:envmxdizc

$81/F87A E8          INX                     A:47F4 X:B0A6 Y:0000 P:eNvmxdizc

$81/F87B 80 1E       BRA $1E    [$F89B]      A:47F4 X:B0A7 Y:0000 P:eNvmxdizc

$81/F89B E2 20       SEP #$20                A:47F4 X:B0A7 Y:0000 P:eNvmxdizc

$81/F89D 64 0D       STZ $0D    [$00:000D]   A:47F4 X:B0A7 Y:0000 P:eNvMxdizc

$81/F89F A5 25       LDA $25    [$00:0025]   A:47F4 X:B0A7 Y:0000 P:eNvMxdizc

$81/F8A1 85 AE       STA $AE    [$00:00AE]   A:47A1 X:B0A7 Y:0000 P:eNvMxdizc

$81/F8A3 A0 00 00    LDY #$0000              A:47A1 X:B0A7 Y:0000 P:eNvMxdizc

$81/F8A6 A9 FF       LDA #$FF                A:47A1 X:B0A7 Y:0000 P:envMxdiZc

$81/F8A8 85 07       STA $07    [$00:0007]   A:47FF X:B0A7 Y:0000 P:eNvMxdizc

$81/F8AA BD 00 00    LDA $0000,x[$E8:B0A7]   A:47FF X:B0A7 Y:0000 P:eNvMxdizc

$81/F8AD 85 06       STA $06    [$00:0006]   A:4716 X:B0A7 Y:0000 P:envMxdizc

$81/F8AF E8          INX                     A:4716 X:B0A7 Y:0000 P:envMxdizc

$81/F8B0 06 06       ASL $06    [$00:0006]   A:4716 X:B0A8 Y:0000 P:eNvMxdizc

$81/F8B2 90 11       BCC $11    [$F8C5]      A:4716 X:B0A8 Y:0000 P:envMxdizc

$81/F8C5 BD 00 00    LDA $0000,x[$E8:B0A8]   A:4716 X:B0A8 Y:0000 P:envMxdizc

$81/F8C8 20 96 F9    JSR $F996  [$81:F996]   A:47D4 X:B0A8 Y:0000 P:eNvMxdizc


I'm sure you can see what the problem is. Those values control MVNs, and 0006 specifically causes a lot of loops. I've run a full-execution trace without squelching in geiger snes9x, and that BCC never fails to resolve with the new script in place. To compound on things, this code is used for other compression-related things as well, so breaking it to load an absolute doesn't work. (It did elsewhere, though.)

Any suggestions?

(edited by Gideon Zhi on 04-06-06 02:04 AM)
Add to favorites | Next newer thread | Next older thread
Acmlm's Board - I3 Archive - ROM Hacking - Compression broken, problems caused |


ABII

Acmlmboard 1.92.999, 9/17/2006
©2000-2006 Acmlm, Emuz, Blades, Xkeeper

Page rendered in 0.032 seconds; used 338.14 kB (max 419.46 kB)