First of all, this specification is rejected.
Instead, we will provide the same functionality as comment.
For details, please see the following.
The latest version has the ability to read the sym for the no$gba debugger placed in “FEBuilderGBA/Config/etc/YOURROMNAME/”.
Please place the symbol generated by buildfile.
The file name can be anything as long as it meets the requirements of directory and extension sym.
“etc” is the directory where ROM-specific settings are recorded.
As for the name, the name up to the first period is used as well as the no$gba debugger.
In the case of FE7U, a UTF-8 patch is not required for LAT1 strings.
LAT1 is a character code that refers to languages derived from Latin.
It refers to languages, including the alphabet, that are mainly used in Europe.
In the latest version, we have removed the restriction on symbol file generation for the no$gba debugger.
The old no$gba debugger would crash and not start if 30,000-40,000 symbols were loaded, so it avoided generating some data.
The current no$gba debuugger seems to have fixed this bug, so I decided to remove the restriction.
If you are using the old no$gba debugger, please update to the latest version.
I’m very sorry, but I don’t know what you mean with “add font”. Actually, I know nothing about hacking - I only wanted to translate some hack games. (I have translated other games before, with another tool, but I’m not well versed in programming and so on) I’m open to learning though. Also, I have found this font: Scraiza's Crazy ASM - #2 by Scraiza and I wondered if, somehow, I could add it to FEBuilder. Thank you and I apologize for my ignorance.
Limitations:
The comment cannot be attached to RAM areas below 0x08000000.
Therefore, all symbols for RAM will be ignored.
Also, please do not put any tabs in the file.
This is because FEBuilderGBA files are in TSV format and are separated by tabs.
no$gba is separated by spaces.
This is to move to the process of reading as no$gba symbol as a failback.
Note that if there is even one tab, it will be processed incorrectly.
Benefits:
User comments are always read in their entirety at startup.
Therefore, they are displayed with the highest priority compared to symbols.
There is no need to wait for FELint to finish.
Compared to symbols, comments are displayed in a special color that stands out.
You no longer have to work hard to find the relevant label or address.
You can now repoint unallocated data in Unit Placer.
Until now, it has been improved that the number of cases could not be increased unless the data written as NEW was assigned to LOAD1 and LOAD2.
First of all, I don’t think I’m talking to you about the program.
Since this is a thread related to FEBuilder GBA, we are also announcing new features.
Adding fonts is must to display text characters that are not in ROM.
This is because it cannot be drawn without a font.
It’s more like a basic computer knowledge than a hackrom.
If you want to know more, please search on google.
In the new version, the reverse lookup routine was buggy and has been fixed.
The reverse lookup data of where TextID and SongID are used could not be completely created by this glitch.
I released the modified version earlier, so I recommend updating again.
I believe the FE7 versions of GOTO_IFCL (branch if character is alive) and GOTO_IFNUF (branch if unit is not fielded) are incorrect due to missing values.
GOTO_IFCL is missing the additional event flag that is checked and triggered if the character is dead. It should be a 32-bit Word value. This is how the game knows to play a post-chapter death scene only once. The event ID can optionally be 0.
(should be 55000000XXXX0000YYYY0000ZZZZ0000)
GOTO_IFNUF has the unit specification in the wrong area. The character ID should be moved two bytes to the right. In other words, the unit specification should lead the next 32-bit area.
(should be 5700XXXXYYYY0000ZZZZ0000)
For GOTO_IFCL, the overall event that occurs in any chapter when someone dies in Lyn mode has many inconsistent oddities immediately after the GOTO_IFCL code. The event address is CC0928. The same oddities can be seen in the Eliwood/Hector chapter endings at the event address CC0C14.
For GOTO_IFNUF, the event in chapter 0x25 (Battle Before Dawn) where Harken and Karel greet the player tactician shows non-event Word values after the GOTO_IFNUF. It also lists the character ID as 0. The event address is CBA1E0.
I noticed this because l use Event Assembler for my eventing. When l inject my events, the two codes behave as expected, but they appear oddly in FEBuilder.
The Terrain BG Table and the Terrain Floor Table in FE7 appears to list the background sets incorrectly. Numbers 0x6, 0xA, 0xB, and 0xC cannot be selected, but they are valid and used in the vanilla game. And starting at 0x7, the terrain editors do not match the chosen background set. This ultimately means 0xA, 0xB, 0xC, 0xD, and 0xE cannot be viewed or edited. However, the chapter editor is always correct. The maps that use these unlisted background sets are 0x15, 0x1C, 0x1E, 0x1A, and 0x1B.
It seems like the data is still in order, so the terrain editors just need to be extended by four entries and have the combobox renumbered/renamed.
Also, there is another, separate issue with the Terrain Floor editor. The floors start at 1 when selecting a floor, but when editing the graphics of floors, they start at 0. This causes a desync of information when editing terrain floors. With 1 selected, you will see the plains, but when you go to the graphics editor the caves floor is shown instead. The enumerator for the “Battle Platforms” graphics editor may need to be offset by +1.