Mednafen 0.9.28-WIP [message #2854] |
Sun, 13 January 2013 10:06 |
|
This is primarily a bug-fix release, including bug-fixes for some nasty subtle regressions introduced in recent releases.
In addition to bug-fixes, a new "WYSIWYG" screen snapshot function has been added, activated by SHIFT+F9, along with a new GPU-intensive pixel shader, "sabr".
ChangeLog
-- 0.9.28-WIP: --
January 13, 2013:
PSX: Corrected lcm_width and lcm_height emulation module structure values(affects QuickTime movie recording).
Fixed a bug in the netplay sound buffer underrun mitigation code that was causing it to only write 1/2 the data it should
be writing when the emulated system has stereo sound output.
Make netplay text line command processing only able to make the netplay text area visible or refresh its visibility, and not
able to hide it; fixes a problem of the netplay text area disappearing before the command response text
(like the disconnect message after /quit) can be seen.
Un-documented Musepack support for ripped CD images as the feature doesn't get much testing, and the support will eventually
be removed(someday ;)).
Removed some ancient unused members of a debugger-related address space data structure.
Fixed a major bug in the CD image audio decoding code that broke repeated playback of
CD audio tracks that are encoded in files handled by libsndfile(e.g. FLAC, WAV), although it may not have affected all
formats, the problem was noted to occur with FLAC at least.
(The bug is a regression apparently introduced in 0.9.25-WIP)
January 12, 2013:
PCE_FAST: Fixed some problems with the 8bpp palette generation routine.
Don't compile with -ffast-math anymore, as it was noted to mess up the NES sound resampler code when
__attribute__((cold)) was used on the filter class constructor, to the extent that it was clearly audible.
January 10, 2013:
Don't needlessly clear the software video buffers to black when changing video modes; fixes the problem of the screen going
black when in frame advance mode and toggling fullscreen.
January 8, 2013:
Fixed a crash(due to a buffer overflow) in Mednafen's test CD-DA player when sound is disabled.
Fixed recording of QuickTime movies when sound is disabled(it was producing unplayably broken QuickTime files before).
PSX: Added a pseudorandom component to command processing and seek timing, to prevent a lockup in "Legend of Dragoon".
January 6, 2013:
NES, PCE_FAST: Made limited use of MDFN_COLD.
Added "MDFN_COLD" define, expanding to "__attribute__((cold))" on sufficiently-recent versions of gcc.
NES, PCE_FAST: Implemented 8-bit rendering internally(for ports, not used by Mednafen proper), using static(in the sense
that the palettes won't change while emulation is in progress) palettes. The limitations of palettized 8bpp don't
really affect the NES too much in practice, except for at least one homebrew demo. The PC Engine, on the other hand, is
a different story...
January 5, 2013:
PC-FX, VB, PSX: Fixed debugger trace log PC address recording.
VB: Corrected (lack of) disassembly of "di" and "ei" instructions.
PC-FX: Fixed broken VDC-A/VDC-B VRAM read/write aux debugger breakpoint support.
PC-FX: Fixed debugger's graphics viewer weirdness when viewing transparent KING BG pixels(by forcing them to black).
PC-FX: Fixed totally broken VDC VRAM debugger memory editor functionality
January 4, 2013:
Only rebuild and re-run the 68K core generator program when necessary, hopefully fixing one big problem when doing
a pure cross-compiled build.
January 3, 2013:
Added "SABR"( http://board.byuu.org/viewtopic.php?f=3&t=3416 ) pixel shader, as an option for the *.pixshader settings.
December 24, 2012:
Added "WYSIWYG"-style(except for the "overlay" video.driver) scaled, filtered, maimed, diced, and so forth screen snapshots,
taken with the SHIFT+F9 key
Marked "Insert/Eject disk/disc" key function assigned to key F8 as being "dangerous" and subject to the delay specified by
the "ckdelay" setting, since many games respond very poorly to the disk/disc being ejected in the middle of a disc/disk
operation.
December 23, 2012:
Simplified(less multithreaded message moving madness) and moved some driver-side netplay code into the Mednafen core.
mednafen-0.9.28-wip.tar.bz2
SHA1: 6948b0f06d72f8056690d229581e94408d40c918
mednafen-0.9.28-wip-win32.zip
SHA1: 6abe0b26dae77fb3824fad15a46531f70854e339
[Updated on: Fri, 14 November 2014 21:58]
|
|
Re: Mednafen 0.9.28-WIP [message #2859 is a reply to message #2854 ] |
Mon, 14 January 2013 11:36 |
|
I think there's one regression still in 0.9.28
From 0.6.4 changelog:
Fixed a bug that was causing flickering OSD remnants to remain onscreen in fullscreen mode if the OSD was drawn outside of the game screen area.
That's happening again, I've tried versions down to 0.9.21
Happens only with these settings:
video.driver opengl
video.glvsync 1
My video card (if that's important) is a Radeon 6770 with, Catalyst drivers version 12.10
|
|
Re: Mednafen 0.9.28-WIP [message #2860 is a reply to message #2854 ] |
Mon, 14 January 2013 12:30 |
|
Do you have triple buffering forcibly enabled in your driver control panel/software?
That's a known current cause of the problem you're describing; it happens on Linux with the Intel open-source driver reportedly too(though it didn't used to happen, seems that the triple-buffering was a recent addition to the driver).
|
|
Re: Mednafen 0.9.28-WIP [message #2861 is a reply to message #2860 ] |
Mon, 14 January 2013 13:11 |
|
Yes, disabling triple buffering in the control panel fixes that error. Thank you.
|
|
Re: Mednafen 0.9.28-WIP [message #2883 is a reply to message #2854 ] |
Sat, 19 January 2013 07:01 |
|
I compiled the source as usual but getting 'Unable to initialize SDL' error can anyone help?
|
|
Re: Mednafen 0.9.28-WIP [message #2884 is a reply to message #2854 ] |
Sat, 19 January 2013 08:36 |
|
can`t build.
gcc 4.5
Making all in snes
make[2]: Entering directory `/tmp/mednafen/src/snes'
depbase=`echo interface.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
g++ -DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H -DNOMINMAX -I../../include -I../../intl -I../../src/snes/src/lib -fsigned-char -Wall -Winline -Wshadow -Wempty-body -Wignored-qualifiers -fno-fast-math -fno-unsafe-math-optimizations -fomit-frame-pointer -finline-limit=6000 --param large-function-growth=800 --param inline-unit-growth=175 --param max-inline-insns-single=10000 -fno-strict-overflow -Wno-unused -Wno-inline -Wno-shadow -Wno-sign-compare -Wno-ignored-qualifiers -Wno-uninitialized -Wno-parentheses -Wno-switch -g -O2 -MT interface.o -MD -MP -MF $depbase.Tpo -c -o interface.o interface.cpp &&\
mv -f $depbase.Tpo $depbase.Po
In file included from ../../src/snes/src/lib/nall/string.hpp:12:0,
from ../../src/snes/src/lib/../base.hpp:32,
from interface.cpp:27:
../../src/snes/src/lib/nall/string/strl.hpp: In function 'size_t strlcpy(char*, const char*, size_t)':
../../src/snes/src/lib/nall/string/strl.hpp:7:58: error: declaration of 'size_t strlcpy(char*, const char*, size_t)' throws different exceptions
../../src/snes/src/lib/nall/string/base.hpp:37:15: error: from previous declaration 'size_t strlcpy(char*, const char*, size_t) throw ()'
../../src/snes/src/lib/nall/string/strl.hpp: In function 'size_t strlcat(char*, const char*, size_t)':
../../src/snes/src/lib/nall/string/strl.hpp:25:58: error: declaration of 'size_t strlcat(char*, const char*, size_t)' throws different exceptions
../../src/snes/src/lib/nall/string/base.hpp:38:15: error: from previous declaration 'size_t strlcat(char*, const char*, size_t) throw ()'
make[2]: *** [interface.o] Error 1
make[2]: Leaving directory `/tmp/mednafen/src/snes'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/mednafen/src'
make: *** [all-recursive] Error 1
|
|
Re: Mednafen 0.9.28-WIP [message #2885 is a reply to message #2854 ] |
Sat, 19 January 2013 10:37 |
|
What OS and what version of gcc exactly(gcc --version)?
[Updated on: Sat, 19 January 2013 10:37]
|
|
Re: Mednafen 0.9.28-WIP [message #2886 is a reply to message #2885 ] |
Sat, 19 January 2013 11:00 |
|
I m under mountain lion and using lib via macports. Gcc version 4.2 apple by default. I m wondering if it is caused by the libsdl from macports. It compiles just fine but when I run it sadly cannot be initialized. When I try compiling again I got architecture error... Now I m using the compiled binary by libdiy using old libs seems working.
|
|
Re: Mednafen 0.9.28-WIP [message #2887 is a reply to message #2885 ] |
Sat, 19 January 2013 11:32 |
|
$ gcc --version
x86_64-alt-linux-gcc (GCC) 4.5.3 20120403 (ALT Linux 4.5.3-alt1.M60P.1)
ALT Linux t6
|
|
Re: Mednafen 0.9.28-WIP [message #2888 is a reply to message #2854 ] |
Sat, 19 January 2013 12:07 |
|
try old gcc
$ gcc --version
x86_64-alt-linux-gcc-4.1 (GCC) 4.1.2 20070626 (ALT Linux, build 4.1.2-alt10)
Making all in snes
make[2]: Entering directory `/tmp/mednafen/src/snes'
depbase=`echo interface.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
g++ -DLOCALEDIR=\"/usr/local/share/locale\" -DHAVE_CONFIG_H -DNOMINMAX -I../../include -I../../intl -I../../src/snes/src/lib -fsigned-char -Wall -Winline -Wshadow -fno-fast-math -fno-unsafe-math-optimizations -fomit-frame-pointer -finline-limit=6000 --param large-function-growth=800 --param inline-unit-growth=175 --param max-inline-insns-single=10000 -fwrapv -Wno-unused -Wno-inline -Wno-shadow -Wno-sign-compare -Wno-uninitialized -Wno-parentheses -Wno-switch -g -O2 -MT interface.o -MD -MP -MF $depbase.Tpo -c -o interface.o interface.cpp &&\
mv -f $depbase.Tpo $depbase.Po
../../src/snes/src/lib/nall/any.hpp:35: warning: 'struct nall::any::placeholder' has virtual functions but non-virtual destructor
../../src/snes/src/lib/nall/string/strl.hpp: In function 'size_t strlcpy(char*, const char*, size_t)':
../../src/snes/src/lib/nall/string/strl.hpp:7: error: declaration of 'size_t strlcpy(char*, const char*, size_t)' throws different exceptions
../../src/snes/src/lib/nall/string/base.hpp:37: error: from previous declaration 'size_t strlcpy(char*, const char*, size_t) throw ()'
../../src/snes/src/lib/nall/string/strl.hpp: In function 'size_t strlcat(char*, const char*, size_t)':
../../src/snes/src/lib/nall/string/strl.hpp:25: error: declaration of 'size_t strlcat(char*, const char*, size_t)' throws different exceptions
../../src/snes/src/lib/nall/string/base.hpp:38: error: from previous declaration 'size_t strlcat(char*, const char*, size_t) throw ()'
../../src/snes/src/lib/../interface.hpp: At global scope:
../../src/snes/src/lib/../interface.hpp:8: warning: 'struct SNES::ChipDebugger' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:1: warning: 'struct SNES::Memory' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:7: warning: 'struct SNES::MMIO' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:12: warning: 'struct SNES::UnmappedMemory' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:18: warning: 'struct SNES::UnmappedMMIO' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:23: warning: 'struct SNES::StaticRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:40: warning: 'struct SNES::MappedRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:60: warning: 'struct SNES::MMIOAccess' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/memory.hpp:68: warning: 'struct SNES::Bus' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../memory/smemory/smemory.hpp:1: warning: 'class SNES::sBus' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../cpu/core/core.hpp:1: warning: 'class SNES::CPUcore' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../smp/core/core.hpp:1: warning: 'class SNES::SMPcore' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../dsp/dsp.hpp:1: warning: 'class SNES::DSP' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../dsp/sdsp/sdsp.hpp:1: warning: 'class SNES::sDSP' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../system/interface/interface.hpp:1: warning: 'class SNES::Interface' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/supergameboy/supergameboy.hpp:1: warning: 'class SNES::SuperGameBoy' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/bus/bus.hpp:1: warning: 'struct SNES::SuperFXBus' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/bus/bus.hpp:5: warning: 'struct SNES::SuperFXGSUROM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/bus/bus.hpp:11: warning: 'struct SNES::SuperFXGSURAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/bus/bus.hpp:17: warning: 'struct SNES::SuperFXCPUROM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/bus/bus.hpp:23: warning: 'struct SNES::SuperFXCPURAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/superfx/superfx.hpp:3: warning: 'class SNES::SuperFX' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:1: warning: 'struct SNES::VBRBus' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:5: warning: 'struct SNES::SA1Bus' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:9: warning: 'struct SNES::VectorSelectionPage' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:16: warning: 'struct SNES::CPUIRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:22: warning: 'struct SNES::SA1IRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:28: warning: 'struct SNES::SA1BWRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:34: warning: 'struct SNES::CC1BWRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/bus/bus.hpp:41: warning: 'struct SNES::BitmapRAM' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sa1/sa1.hpp:3: warning: 'class SNES::SA1' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/bsx/bsx.hpp:1: warning: 'class SNES::BSXBase' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/bsx/bsx.hpp:25: warning: 'class SNES::BSXCart' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/bsx/bsx.hpp:46: warning: 'class SNES::BSXFlash' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/srtc/srtc.hpp:1: warning: 'class SNES::SRTC' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/sdd1/sdd1.hpp:3: warning: 'class SNES::SDD1' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/spc7110/spc7110.hpp:20: warning: 'class SNES::SPC7110' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/cx4/cx4.hpp:1: warning: 'class SNES::Cx4' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/dsp1/dsp1.hpp:3: warning: 'class SNES::DSP1' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/dsp2/dsp2.hpp:1: warning: 'class SNES::DSP2' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/dsp3/dsp3.hpp:1: warning: 'class SNES::DSP3' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/dsp4/dsp4.hpp:1: warning: 'class SNES::DSP4' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/obc1/obc1.hpp:1: warning: 'class SNES::OBC1' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/st010/st010.hpp:1: warning: 'class SNES::ST010' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/st018/st018.hpp:1: warning: 'class SNES::ST018' has virtual functions but non-virtual destructor
../../src/snes/src/lib/../chip/21fx/21fx.hpp:1: warning: 'class SNES::S21fx' has virtual functions but non-virtual destructor
../PSFLoader.h:31: warning: 'class PSFLoader' has virtual functions but non-virtual destructor
interface.cpp:39: warning: 'class MeowFace' has virtual functions but non-virtual destructor
make[2]: *** [interface.o] Error 1
make[2]: Leaving directory `/tmp/mednafen/src/snes'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/mednafen/src'
make: *** [all-recursive] Error 1
|
|
Re: Mednafen 0.9.28-WIP [message #2889 is a reply to message #2854 ] |
Sat, 19 January 2013 14:51 |
|
Try replacing the corresponding files in src/snes/src/lib/nall/string with these.
Attachment: base.hpp
(Size: 4.61KB, Downloaded 312 time(s))
Attachment: split.hpp
(Size: 1.31KB, Downloaded 259 time(s))
Attachment: strl.hpp
(Size: 1.12KB, Downloaded 260 time(s))
Attachment: utility.hpp
(Size: 2.25KB, Downloaded 288 time(s))
|
|
Re: Mednafen 0.9.28-WIP [message #2890 is a reply to message #2889 ] |
Sun, 20 January 2013 02:05 |
|
it working!
thanks!
|
|
Re: Mednafen 0.9.28-WIP [message #2898 is a reply to message #2854 ] |
Wed, 23 January 2013 13:17 |
|
Hopefully this hasn't been asked a thousand times:
Does mednafen use any kind of version control system? SVN/CVS/GIT
|
|
Re: Mednafen 0.9.28-WIP [message #2899 is a reply to message #2854 ] |
Wed, 23 January 2013 14:03 |
|
Compile fails at gb core withIn file included from gb.cpp:33:0:
memory.h:25:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperMBC1’
In file included from gb.cpp:28:0:
./memory.h:25:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperMBC1’
In file included from gb.cpp:33:0:
memory.h:32:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperMBC2’
In file included from gb.cpp:28:0:
./memory.h:32:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperMBC2’
In file included from gb.cpp:33:0:
memory.h:37:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperMBC3’
In file included from gb.cpp:28:0:
./memory.h:37:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperMBC3’
In file included from gb.cpp:33:0:
memory.h:57:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperMBC5’
In file included from gb.cpp:28:0:
./memory.h:57:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperMBC5’
In file included from gb.cpp:33:0:
memory.h:66:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperMBC7’
In file included from gb.cpp:28:0:
./memory.h:66:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperMBC7’
In file included from gb.cpp:33:0:
memory.h:83:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperHuC1’
In file included from gb.cpp:28:0:
./memory.h:83:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperHuC1’
In file included from gb.cpp:33:0:
memory.h:92:8: error: redefinition of ‘struct MDFN_IEN_GB::mapperHuC3’
In file included from gb.cpp:28:0:
./memory.h:92:8: error: previous definition of ‘struct MDFN_IEN_GB::mapperHuC3’
make[2]: *** [gb.o] Error 1 GNU Compiler 4.7.2
|
|
Re: Mednafen 0.9.28-WIP [message #2900 is a reply to message #2854 ] |
Wed, 23 January 2013 15:15 |
|
Provide more details about your environment.
That shouldn't happen unless your toolchain is broken or you passed weird *FLAGS to the configure script.
[Updated on: Wed, 23 January 2013 15:16]
|
|
Re: Mednafen 0.9.28-WIP [message #2905 is a reply to message #2900 ] |
Fri, 25 January 2013 12:31 |
|
Administrator wrote on Thu, 24 January 2013 06:15 | you passed weird *FLAGS to the configure script.
| Ah, this. Thanks for the tip.
I was working on a tuned and profiled build. It's taking an obnoxiously long time to figure out which flag(s) break compile. I know it's mostly unnecessary, but I wanted to give it a try for... fun?
So far, these are safe (as in don't break compile, their effect on the emulator seems also to be safe):
to make the profile ./configure CFLAGS="-pipe -Ofast -msahf -m64 -mabm -march=corei7-avx -msse4.2 -mavx -fprofile-generate" CXXFLAGS="-pipe -Ofast -msahf -m64 -mabm -march=corei7-avx -msse4.2 -mavx -fprofile-generate" LDFLAGS="-lgcov -s" (tuned for core i7 2700k)
to apply the profile ./configure CFLAGS="-pipe -Ofast -msahf -m64 -mabm -march=corei7-avx -msse4.2 -mavx -fprofile-generate" CXXFLAGS="-pipe -Ofast -msahf -m64 -mabm -march=corei7-avx -msse4.2 -mavx -fprofile-use -fprofile correction -Wno-coverage-mismatch" LDFLAGS="-lgcov -s" (-fprofile-correction for multicore kludge and -Wno-coverage-mismatch because it asked for it)
one, two or all of these are unsafe (haven't tested each one individually... yet; batch testing failed to isolate the problem): -floop-flatten -floop-interchange -floop-parallelize-all -floop-strip-mine -flto -fmath-errno -fmerge-all-constants -fmodulo-sched -fmodulo-sched-allow-regmoves -fno-enforce-eh-specs -fno-threadsafe-statics -fnothrow-opt -fomit-frame-pointer -fopenmp -freschedule-modulo-scheduled-loops -fsched-pressure -fsched-spec-load -fsched-spec-load-dangerous -fsched-stalled-insns=0 -fsched2-use-superblocks -fschedule-insns -ftree-loop-distribution -ftree-loop-linear -funroll-loops -funroll-all-loops -funsafe-loop-optimizations -fvariable-expansion-in-unroller -maccumulate-outgoing-args (flags I've used before with higan and pcsxr)
|
|
Re: Mednafen 0.9.28-WIP [message #2924 is a reply to message #2854 ] |
Sat, 02 February 2013 12:04 |
|
Nice work, runs flawless in mint 14. Can you support Neo Geo AES roms, because there are only one emulator (gngeo) and I can't use it. Greetings.
|
|
SNES won't start under Ubuntu Linaro 12.11 [message #2972 is a reply to message #2854 ] |
Sun, 10 March 2013 06:36 |
|
compiling mednafen went without any trouble it seems most of the emulators run fine except SNES.
It crashes with:
*** longjmp causes uninitialized stack frame ***: mednafen terminated
It's an ARM based platform with Debaian based - Ubuntu Linaro 12.11.
Please tell me what files/logs you need and i provide you with it.
I would love to have a working SNES version within mednafen.
|
|
Re: Mednafen 0.9.28-WIP [message #3003 is a reply to message #2854 ] |
Tue, 19 March 2013 09:16 |
|
I tried the new SABR shader and there are some artifacts. Horizontal and vertical lines. Thanks to the new screenshot feature, I don't have to strain myself to explain it.
A screenshot is attached.
I thought it was related to non-integer-scaling somehow, but the lines still appear with aspect_int and aspect_mult2.
Running Mednafen with a Sapphire HD 7750 with latest drivers on Win7 64 bit.
EDIT: I noticed another bug, that also appears with the previous version.
Using the shader "autoipsharper", it works fine in fullscreen, but in windowed mode, there will be a distortion along a diagonal line from top left corner to bottom right corner. With no shader selected, there are no distortions in windowed or fullscreen mode.
[Updated on: Tue, 19 March 2013 09:30]
|
|
Re: Mednafen 0.9.28-WIP [message #3005 is a reply to message #2854 ] |
Tue, 19 March 2013 15:12 |
|
I think there's something precision-related wrong with GLSL or OpenGL on ATI/AMD graphics cards. I've noticed that problem for a while, but wasn't ever able to figure out how to fix it.
The weird thing is that it happens with my Sapphire HD 5670 under Windows XP, but not Windows 7 64-bit. Though I think the driver versions installed are different...
[Updated on: Tue, 19 March 2013 15:17]
|
|
Re: Mednafen 0.9.28-WIP [message #3006 is a reply to message #3003 ] |
Tue, 19 March 2013 15:55 |
|
Dryer Lint wrote on Tue, 19 March 2013 18:16 | I tried the new SABR shader and there are some artifacts. Horizontal and vertical lines.
|
What's interesting, these lines remind me of some artifacts of Haali Video Renderer. It's also an ATI bug
|
|
Re: Mednafen 0.9.28-WIP [message #3034 is a reply to message #2854 ] |
Sat, 20 April 2013 16:58 |
|
Just FYI for anyone trying to get netplay to work, if you pass two arguments(host AND port) to the /server command in the netplay console, Mednafen will crash. Passing one or no arguments shouldn't cause a crash. This issue will be fixed in the next release.
|
|
Re: Mednafen 0.9.28-WIP [message #3085 is a reply to message #2854 ] |
Mon, 03 June 2013 02:12 |
|
I hate asking this, but is there a pending release coming soon (tm)? We're quickly coming up to 6 months since the last WIP and it seems unusually long. I'm just hoping development hasn't halted.
Thanks much!
|
|