| Age | Commit message (Collapse) | Author |
|---|
|
|
|
- in nightmare sequence in dungeon level -4
- can not always be noticed
|
|
(this has to be done a bit different for PC-98, since - unlike the other currently supported platforms - PC-98 has ingame music)
|
|
|
|
|
|
The original adds a delay after track 28.
|
|
Don't use y-coords from LOL-PC98 (which are all 4/8 aligned due to its use of the text mode instead of graphics mode)
|
|
(eliminate some duplicate code for dirty rect handling)
|
|
|
|
Seems that I accidently removed some of digitall's changes while rebasing (I moved that particular code block to a new function).
|
|
The PC98 version has as slightly different layout for the character stats page.
|
|
Object now destructs cleanly even before init() has been called. This has no practical relevance, but should be fixed nonetheless.
|
|
|
|
(The PC-98 version has its own unique ending sequence)
|
|
|
|
|
|
(unlike LOL/PC98 we have the same key codes as in the DOS version)
|
|
|
|
|
|
|
|
(no block drawing yet)
|
|
sjis font
The assumption till now was that PC-9801 games in 16 color mode will always use text mode for text display. However, EOB1 uses text mode for intro and ending sequence and graphics mode for ingame.
Add new font variant for this and also do some cleanup, since the text displayer code really needs it.
(The problem is that at least 3 different Japanese publishers did the Japanese ports for the various games and every one of them did his own hacks for his specific target.)
|
|
|
|
|
|
The PC98 version has a modified version of decodeFrame4() and also requires planar decoding (similiar to Amiga).
|
|
- hook up static resources, sound code, etc
- nothing really working yet, though (except the music)
|
|
|
|
|
|
|
|
(The Amiga version has as special warning message if the spell has already been cast on a character. This warning was not displayed).
|
|
Seems that some font glyphs that were used for the print_map command aren't present any more. I replaced them with more conventional glyphs.
|
|
These are flagged by GCC if -Wswitch-default is enabled.
|
|
|
|
(EOB 1 has a higher chance to receive a random item from certain monsters than EOB2 - this fixes that)
|
|
|
|
The game would actually allow placing the 4 horns or the large red rings from the ringmaster riddle into the inventory ring slots. Simply, because these items are flagged that way in the data file (itemtype.dat). I wasn't aware of this until I happened to come across some youtube video. The video was from DosBox, but this doesn't make a difference.
This patch prevents placing these items into the ring slots.
|
|
|
|
This fixes an original bug that recently got my attention.
While I never managed to duplicate items like the person in the video I did experience lockups.
Maybe the exploit works only with the original executable...
https://www.youtube.com/watch?v=fkJ2KilULco&feature=youtu.be
https://www.youtube.com/watch?v=4oRESbQLDig&feature=youtu.be
|
|
|
|
(mostly paranoia fixes)
|
|
|
|
The memset call to clear the Common::Event is unecessary as all fields
are cleared either directly by the class constructor or for fields which
are struct/class objects, by their constructors.
|
|
|
|
Get rid of alternative anim frame triggers in cases where animations stop too early (e. g. the tree in the first scene not vanishing completely).
|
|
The game would error out when trying to return to the launcher via GMM right at the beginning (due to trying to load files based on incompletely initialized data).
|
|
|
|
|
|
|
|
|
|
|
