sepp256
941dfe086e
Edited wiki page through web user interface.
2008-08-13 20:26:00 +00:00
emukidid
6ee88c6612
Moved TLB dumping to separate functions. Made TLB cache LUT's static once again
2008-08-12 12:53:07 +00:00
emukidid
066d58a1c3
Fixed savestate filenames, fixed DVD directories glitch (removed trailing '/'), fixed minimum DVD entries to 1024, removed savestates.c (PC) file
2008-08-12 08:13:17 +00:00
emukidid
08238a2be2
TLB LUT clearing on Init since they are no longer statically allocated as NULL initialised arrays.
2008-08-09 00:42:13 +00:00
emukidid
490feea15d
fixed multiple declaration of tlb luts
2008-08-08 23:31:17 +00:00
emukidid
bef9c0c5fb
TLB Saving/Loading in savestates, tested it on a handful of ROMS and it seems to work
2008-08-08 13:21:51 +00:00
tehpola
860fb958fd
Dynarec now successfully executes the standard N64 apploader and works on a (extremely simple, controlled) demo that we created. SLT/SLTI/SLTU/SLTIU instructions were not correct in every case and now should work in all cases. Only dirty registers are flushed and when flushed they are properly sign extended. r0 and decodeNInterpret are now mapped as constant so fewer instructions are generated with respect to those operations. Still no commercial games will run, and no already existing demos that I tried ran so there's still plenty of work to do.
2008-08-05 02:49:01 +00:00
tehpola
179d82fbe3
Dynarec: too much to name. The dynarec matches the interpreter for memory accesses through the 0xa4000000 block and proceeds into 0x80000000 where the CRC check for the DMA transfer of the game code off the cart fails.
2008-07-28 13:14:28 +00:00
emukidid
82a6085f3b
Mempaks should format every time a new game is loaded (to avoid Corrupt Mempak warnings)
2008-07-06 06:08:51 +00:00
emukidid
0f6311962c
Savestate issue with expansion pak fixed (for later on)
2008-07-05 13:38:20 +00:00
sepp256
cf7a8eb211
- Fixed texture interleaving (SSB textures).
...
- Cleaned up texture code.
- Added PolygonOffset switching for near clipped triangles.
- Cleaned up combiner code.
- Updated glN64 makefiles.
2008-07-04 05:43:16 +00:00
tehpola
131f3554a4
Making progress on both dynarecs
2008-07-03 20:27:32 +00:00
sepp256
f785afcd76
- Reworked VI to use the PreRetrace callback in GX_gfx plugin, which improves the framerate a little.
2008-07-03 02:27:26 +00:00
sepp256
db8964e9fc
- Implemented Fog using GX hardware.
...
Note: Fog requires the patched gx.c included in this update.
- Implemented Polygon Offset using matrix transforms.
Note: Let me know if you still see any flashing co-planar polygons.
- Cleaned up stuff in VI.
- Other various cleanup stuff.
2008-07-03 02:09:40 +00:00
tehpola
8df5c5053f
Scheme: Making some good progress, but getting an exception prematurely.
2008-07-01 23:03:48 +00:00
tehpola
6d6d498ea5
Grr...
2008-06-28 05:46:20 +00:00
tehpola
87c68897ea
Scheme dynarec in a compilable state (however scheme doesn't think 'dynarec' is defined).
2008-06-28 01:28:22 +00:00
tehpola
192f4aa235
More Scheme work (interfacing mostly which is a lot less fun)
2008-06-27 04:38:31 +00:00
tehpola
bdf09faacd
Scheme dynarec (still incomplete), still needs C interface to run and interact with
2008-06-24 13:12:33 +00:00
tehpola
ed30f38901
Added NO_BT flag so bluetooth can be disabled to allow stepi to work in gdb.
2008-06-17 12:12:59 +00:00
tehpola
cb50ab26cb
Latest dynarec work. Recompiles the code, but disappears after it branches into the recompiled code.
2008-06-16 02:19:36 +00:00
emukidid
7321c40a61
the 0x40/0x80 loops make a difference for us since we define 4mb sometimes
2008-06-15 12:43:03 +00:00
emukidid
8c4a299693
4mb fix now works on more games
2008-06-15 12:38:39 +00:00
tehpola
6b8f335471
Latest changes to the dynarec. I'm not exactly sure what's going on, but something is being corrupted in the recompilation stage it seems.
2008-06-15 06:17:27 +00:00
tehpola
e01e4a1ed4
Getting close to finishing all the work for overhauling out the 1-1 register mapping. Same thing as before, compiles, but I'm fairly certain it won't work (yet).
2008-06-14 22:15:15 +00:00
tehpola
cc15c6fda7
Committing all my latest sources to my branch.
2008-06-12 23:39:13 +00:00
tehpola
32b0f261ba
Moving closer to eliminating 1-1 register mapping. Compiles but does not run (yet).
2008-06-12 23:35:25 +00:00
emukidid
c528abf3dd
TLB cache is now working faster
2008-06-12 15:14:24 +00:00
tehpola
99c3a519d5
Fix to TLB instructions (no more redundant TLB sets).
2008-06-12 02:26:12 +00:00
tehpola
9abf46654d
I've significantly improved the TLB Cache hash function (it was just terrible!), but strangely, that was not the issue: TLB Cache dumping (which is added) has shown Goldeneye's TLB to be empty (although Mario 64 had it slightly populated).
2008-06-11 04:31:40 +00:00
emukidid
d537eaf78d
TLB Should now be saved in save states for MEM2 and should still regardless, States made before this commit won't work anymore
2008-06-10 05:16:40 +00:00
sepp256
7774ed01a1
Added/Fixed viewport and scissoring.
2008-06-09 15:48:12 +00:00
tehpola
00f50dd18e
Got my branch compiling with some new (unfinished) dynarec code.
2008-06-08 22:54:58 +00:00
tehpola
80c36993c6
Created tehpola branch
2008-06-08 21:55:45 +00:00
emukidid
007578b1b7
tlb mem2 pointer fixup
2008-06-08 15:44:10 +00:00
emukidid
4757ba1012
TLB MEM2 clearing for every new game loaded
2008-06-08 14:42:04 +00:00
emukidid
eeb185008a
Put old reset checking method back
2008-06-08 07:59:38 +00:00
sepp256
da064d5724
Fixed a bunch of things in glN64_GX, such as:
...
- 3D transform handled in hardware when possible (mainly Mario64)
- Fixed clipping and texture perspective correction by hacking the Projection matrix to reproduce the proper z and w coords (GX can't handle homogeneous coordinates :P). Note that this isn't guaranteed to work all the time so I'll have to do more testing.
- Added code to handle primeDepth source.
- Added Ztexture code for primeDepth.
- Fixed AlphaCompare flag.
- Other stuff...
2008-06-08 07:45:23 +00:00
emukidid
076bfa0751
moved reset button stop=1 to interrupt. Untested
2008-06-06 14:02:46 +00:00
emukidid
fc3d1c61ee
-DUSE_TLB_CACHE added to all makefiles, you can remove it to use the MEM2 uncached version but obviously only on Wii. We later should change it to be malloc'd from MEM2 but for now it's fine.
2008-06-06 13:42:48 +00:00
emukidid
87c70cd398
Edited wiki page through web user interface.
2008-06-05 10:09:55 +00:00
emukidid
4c568b6407
Debug.c updated to proper function names
2008-06-05 05:51:35 +00:00
sepp256
daf2d67193
Updating sepp256 branch to trunk and adding retrace callback fix.
2008-06-05 03:54:28 +00:00
emukidid
6e91a46755
save/load state checks to see if rom is loaded
2008-06-05 03:50:43 +00:00
sepp256
2520ca43e5
Moved xfb switching to a retrace callback. Now we don't need to wait for vsync in order to avoid framebuffer tearing.
2008-06-05 03:28:01 +00:00
emukidid
dd61c7bf73
Save/Load states (needs TLB fixup), -lz added to all makefiles
2008-06-05 00:36:05 +00:00
sepp256
dbc43b4d3c
2008-06-05 00:21:36 +00:00
sepp256
5a9fef5520
Untested modifications to VI to remove not limit FPS by WaitVsynch(). Please test to see if it works and make sure you don't crash when returning to the menu or switching ROMs.
2008-06-04 21:07:55 +00:00
sepp256
d76dbd292d
Updating sepp256 branch to trunk/
2008-06-04 21:05:31 +00:00
emukidid
900571b1c1
Mupen64 HLE RSP Plugin 0.2d replaced rsp-hle-ppc, F-zero sounds 100% now, haven't tested the rest!
2008-06-04 07:03:40 +00:00
