Soldier of Fortune (Dreamcast) #117
Replies: 4 comments 12 replies
-
|
NVM I located 117 |
Beta Was this translation helpful? Give feedback.
-
|
Looks like there's some large recurring structures that interrupt file data This means #21 alone won't be enough, the |
Beta Was this translation helpful? Give feedback.
-
|
The recurring structure is |
Beta Was this translation helpful? Give feedback.
-
|
Might've found tools for browsing the GD-ROM filesystem: |
Beta Was this translation helpful? Give feedback.
-
|
Couldn't figure out fuseiso3 once I got it to compile |
Beta Was this translation helpful? Give feedback.
-
|
I did also manage to locate some map name strings in the data (also some C headers?) Removed a recurring struct thing from the middle, a little confused by the repetition |
Beta Was this translation helpful? Give feedback.
-
|
Just found 105 Footnotes |
Beta Was this translation helpful? Give feedback.
-
|
The map name strings data looks like |
Beta Was this translation helpful? Give feedback.
-
|
The I could probably determine |
Beta Was this translation helpful? Give feedback.
-
How does a portal file contain the map source? Do you mean the brushes of a map? |
Beta Was this translation helpful? Give feedback.
-
|
Finding a bunch of It's unlikely any uncompiled |
Beta Was this translation helpful? Give feedback.
-
|
The C code is maybe like Quake 3 LCC bot code? |
Beta Was this translation helpful? Give feedback.
-
|
Just some // if we wind up #define'ing more than 500 anims, update ANIMLISTSIZE in ds.exe/ds.cpp
#define SOLID_NOT 0
#define SOLID_SOLID 2 // erm, why wasn't this made SOLID_BBOX? grrrr
#define SOLID_BSP 3This could be for scripting actually // template file for common directory
#include "../common/header.ds"
output "p:/base/ds/common"
parameter entity kill_thing
remove entity kill_thing // removes whatever was in parm1 |
Beta Was this translation helpful? Give feedback.
-
|
Extracted all import struct
def substring_indices(file, sub: bytes, buf_max: int = 2048):
# hopefully buffer alignment won't split the substring
cur = 0
while True:
file.seek(cur) # resume scanning
buf = file.read(2048)
cur = file.tell()
if sub in buf:
yield cur - 2048 + buf.index(sub)
elif len(buf) == 0:
break # reached EOF
with open("Track 06.iso", "rb") as gd_file:
for i, offset in enumerate(substring_indices(gd_file, b"IBSP\x2E\x00\x00\x00")):
gd_file.seek(offset + 8) # after ("IBSP", 46)
# assuming we only have the same 19 headers as raven.soldier_of_fortune
lump_headers = struct.iter_unpack("2I", gd_file.read(8 * 19))
bsp_size = sum(sorted(lump_headers)[-1])
gd_file.seek(offset)
print(f"{i:03d} {offset:08X} writing to 'maps/entry_{i + 1:03d}.bsp' ...")
with open(f"maps/entry_{i + 1:03d}.bsp", "wb") as bsp_file:
bsp_file.write(gd_file.read(bsp_size))An initial glance at the extracted maps with
So
|
Beta Was this translation helpful? Give feedback.
-
Scratch that, I forgot I will run the |
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
Notes
Support for this format was confirmed in c08a75b
Keeping the thread up for future reference
Original Post
Trawling through a
.gdidump, there are enough instances of"classname" "worldspawn"to account for 121.bspsHowever, I can't seem to find a way to split the
.imgholding this dataMight have to reverse engineer the executable to figure out the filesystem
I also haven't been able to locate
b"IBSP"headers, so it's possible the format is uniqueBeta Was this translation helpful? Give feedback.
All reactions