Commit Graph

768 Commits

Author SHA1 Message Date
nmlgc 765eae82e8 [Maintenance] [th05] Minimize #includes for the shot type translation units
Since a few annoying alignment bytes suggested more translation units
than previously expected, using many small headers has proved to be
better than one big shared TH04/TH05 header file. Or should we *really*
pepper the code with lots of `#pragma codestring`? 😛

And in case we have multiple translation units which all #include the
same set of headers, we'll just go with situational shared headers,
using a common prefix.

Part of P0062, funded by Touhou Patch Center.
2019-12-22 15:32:44 +01:00
nmlgc c5b07b746e [Maintenance] `#pragma once` has no effect?
Should have remembered this from all the times I've looked at the list
of #pragma directives supported by Turbo C++… Still, I would have at
least expected a warning, but not even the `ill` warning covers this.

Part of P0062, funded by Touhou Patch Center.
2019-12-22 15:31:24 +01:00
wintiger0222 1d6fbb8108 [Reverse-engineering] [th05] Masked PI display
As used for the title screen fade-in effect. Another function that
apparently was deliberately written to run not that fast, by blitting
each row individually to the 400th VRAM row just so that it can then
turn on the EGC, perform the *actual* masked blit to the VRAM
destination, and then turn the EGC off before moving to the next now.
The same effect could have entirely been accomplished by copying
graph_pack_put_8() and applying the mask there; it's not like ZUN
didn't know how to modify master.lib…

(See also 44ad3eb4) --Nmlgc
2019-12-17 23:27:02 +01:00
nmlgc a5e714b165 [Maintenance] [th05] Clarify pi_slot_put() a bit
Can't blame WindowsTiger for writing hard-to-review code when he just
adapted what I wrote in 2014…
2019-12-17 23:27:01 +01:00
nmlgc 75a779e82a [Maintenance] Clean up PI function declarations and comments 2019-12-17 23:27:01 +01:00
nmlgc deb1b73016 [Maintenance] Use a single DEFCONV definition file to cover all games
This makes a lot more sense in C land, where we can then #include this
tiny file in front of all DEFCONV declarations.
2019-12-17 23:27:01 +01:00
wintiger0222 13e4476cc8 [Naming] [th04/th05] MPTN functions 2019-12-17 23:27:01 +01:00
wintiger0222 67ab2f4a41 [Naming] [th01] Game init and exit functions 2019-12-17 23:27:01 +01:00
wintiger0222 990a1e2c94 [Naming] [th03/th04/th05] Main and option menu functions 2019-12-17 23:27:00 +01:00
wintiger0222 04f0491885 [Naming] [th03/th04/th05] .cfg file functions 2019-12-17 23:27:00 +01:00
wintiger0222 e219527964 [Naming] [th03/th04/th05] High score menu functions 2019-12-17 23:27:00 +01:00
wintiger0222 b95a625d87 [Naming] [th04/th05] Character select screens 2019-12-17 23:27:00 +01:00
wintiger0222 02d1c04858 [Reverse-engineering] [th05] Character selection and unlock variables 2019-12-17 23:27:00 +01:00
nmlgc 50c36f5551 [Reverse-engineering] Player character constants 2019-12-17 23:26:59 +01:00
wintiger0222 6e741bb1e1 [Reverse-engineering] [th04] cdg_put_plane_roll
Used for the falling blue stars in TH04 Stage 5 --Nmlgc
2019-12-17 23:26:59 +01:00
nmlgc 417ba80c00 [Decompilation] Add a separate segment type
At least wherever Turbo C++ and master.lib want us to use a
non-pointer, since both use uint16_t for segment values throughout
their APIs instead of the more sensible void __seg*. Maybe, integer
arithmetic on segment values was widely considered more important than
dereferencing?
2019-12-17 23:26:59 +01:00
nmlgc 59bbe313ad [Decompilation] Add separate types for 1bpp planar pixel lines 2019-12-17 23:26:59 +01:00
nmlgc 9cb6cc527a [Decompilation] Finally declare <stdint.h> types
*Finally*. We already used `(unsigned) int` in quite a few places where
we actually want a 16-bit value, which was bound to annoy future port
developers.
2019-12-17 23:26:58 +01:00
wintiger0222 cb45bf5a80 [Reverse-engineering] [th04/th05] EGC-powered page 1→0 region copies 2019-12-17 23:26:58 +01:00
wintiger0222 0dcfc445b8 [Reverse-engineering] [th03] input_wait_for_ok_or_measure()
Moving this into hardware/, since we already have an
input_wait_for_change() function there. --Nmlgc
2019-12-17 23:26:58 +01:00
wintiger0222 c140281ba7 [Naming] [th01] graph_putsa_fx 2019-12-17 23:26:58 +01:00
wintiger0222 590436dfa8 [Reduction] [th03] graph_putsa_fx
Fine, let's drag that ASM slice along, instead of immediately reusing
the already decompiled TH02 version… The website really needs a
decompilation metric soon, anyway. --Nmlgc
2019-12-17 23:26:58 +01:00
wintiger0222 9173833a8d [Reverse-engineering] [th01] grcg_setcolor() C reimplementations 2019-12-17 23:26:57 +01:00
wintiger0222 88a9143309 [Reverse-engineering] [th01] JIS X 0208 東方★靈異伝 code points 2019-12-17 23:26:57 +01:00
wintiger0222 c0705c395c [Naming] [th03/th04/th05] Music Room functions changed from TH02 2019-12-17 23:26:57 +01:00
wintiger0222 d7906ef78e [Reverse-engineering] [th04/th05] ZUN Soft logo 2019-12-17 23:26:57 +01:00
nmlgc c6ab779897 [Maintenance] [th04/th05] Correctly declare the ZUN Soft explosion functions 2019-12-17 23:26:57 +01:00
wintiger0222 38e8fd6a8d [Reverse-engineering] [th04/th05] Music Room / ending background image buffers
"bgimage" really was the best name I could come up with here, given
what it's used for. Everything else sounded way too ambiguous. Even
something like "mempage", in the meaning of "a third memory-backed
graphics page"… --Nmlgc
2019-12-17 23:26:56 +01:00
wintiger0222 9684cf5312 [Reduction] [th03] Music Room comment back buffer
I think cmt_back_* type functions need more universal name
Because at th04 and th05, it used in zunsoft opening animation
--WindowsTiger

And in MAINE.EXE. So I agree for TH04 and TH05, but TH03 still only
allocates, snaps, and recovers only the VRAM area occupied by the music
comments. --Nmlgc
2019-12-17 23:26:56 +01:00
wintiger0222 1317273aa0 [Reduction] [th03/th04/th05] Music Room code identical to TH02 2019-12-17 23:26:56 +01:00
wintiger0222 680af1afd9 [Position independence] [th03/th04/th05] Music Room file names and titles 2019-12-17 23:26:56 +01:00
wintiger0222 2a722db215 [Reverse-engineering] [th05] Music Room piano functions
No need to declare most of these in C land if 1) it isn't even clear
yet whether we can even decompile these functions with their 2)
complex __usercall conventions, which 3) aren't even called from
outside this slice. --Nmlgc
2019-12-17 23:26:56 +01:00
wintiger0222 53f316792f [Reverse-engineering] [th05] Music Room piano sprite data 2019-12-17 23:26:56 +01:00
wintiger0222 b25ec3d138 [Reverse-engineering] [th05] PMD work area and piano note data
Good job identifying what is bascially the QQ structure from PMDWin!
--Nmlgc
2019-12-17 23:26:55 +01:00
wintiger0222 2b617dfa75 [Reverse-engineering] [th05] Music Room piano coordinates 2019-12-17 23:26:55 +01:00
wintiger0222 b74b7f9bce [Reverse-engineering] [th05] Music Room piano label font 2019-12-17 23:26:55 +01:00
wintiger0222 d0c4de0c7f [Naming] [th04/th05] GameCore 2019-12-17 23:26:55 +01:00
wintiger0222 895884fc7a [Naming] [th02/th04/th05] HUD display functions 2019-12-17 23:26:55 +01:00
wintiger0222 25903719a3 [Naming] [th02] Current and high score 2019-12-17 23:26:55 +01:00
wintiger0222 1ffa121821 [Naming] [th02] Gaiji loading
Renaming these to gaiji_load() and gaiji_free() for consistency with
other functions that load files. --Nmlgc
2019-12-17 23:26:54 +01:00
wintiger0222 0b084b83c3 [Reverse-engineering] [th04/th05] graph_putsa_fx 2019-12-17 23:26:54 +01:00
wintiger0222 75f05c8b90 [Reverse-engineering] [th04/th05] graph_putsa_fx tables 2019-12-17 23:26:54 +01:00
wintiger0222 6e09a2c15f [Naming] [th04/th05] graph_putsa_fx 2019-12-17 22:44:30 +01:00
wintiger0222 811ba84e3f [Reverse-engineering] [th03/th04/th05] snd_delay_until_measure 2019-12-17 22:44:27 +01:00
wintiger0222 e8c9876bef [Reverse-engineering] [th05] Current BGM measure
Split off from the delay function in this game, since the ending syncs
a bunch of things independent from any frame delays. -Nmlgc
2019-12-17 22:44:23 +01:00
nmlgc a23dab3154 [Reverse-engineering] MMD ticks-per-quarter-note constant
And then, ZUN hardcodes all measure-related functions to 4/4…
2019-12-17 22:40:05 +01:00
nmlgc 5f4f5d87dc [Decompilation] [th03] Shot update and render functions
Meh, can't overload arithmetic operators that take a Subpixel without
generating a needless load and store, even with -Z. But heck, slightly
uglifying subpixel/subpixel arithmetic is exactly the right trade-off.

Completes P0061, funded by Touhou Patch Center.
2019-12-05 21:41:31 +01:00
nmlgc 8b01c60f44 [Maintenance] Move PC-98 hardware constants to a separate include file
SPRITE16 would also like to refer to those later.

Part of P0061, funded by Touhou Patch Center.
2019-12-05 21:38:39 +01:00
nmlgc 5c1a864432 [Reverse-engineering] [th03] Shot pair structure
Yup, one structure instance describes one *pair* of shot sprites in
TH03.

Part of P0061, funded by Touhou Patch Center.
2019-12-05 21:38:32 +01:00
nmlgc 296f9f61c2 [Reverse-engineering] [th03] Playfield shaking
Which only applies to the set of sprites considered to be in the
foreground, so no EGC acceleration.

Part of P0061, funded by Touhou Patch Center.
2019-12-05 21:36:02 +01:00