Hey guys.. it's been a long time.. but I am finally ready to release a new test version of DAPHNE. The goal is to identify the major issues that people are having and eliminate them before the official release.

To start, I am releasing a very minimal incremental build (that will require you to first download v0.99.6, or v0.99.6c) that includes a batch file to run through a series of self-tests. I suspect that some ATI users may fail some of these self tests, and I would like to hear about it. My goal is to have all hardware setups passing these self-tests before the official release.

The URL is http://www.daphne-emu.com/download/daphne-0.99.7pre1-win32.zip.

You can post your daphne_log.txt files and any bug reports to this thread (everyone should have access to post to this thread). Or you can report problems to us live in the chatroom (http://www.daphne-emu.com/chatroom.php). If you report your results to the chatroom (or bugs) the chance of getting your issues solved is quite a bit higher, because we will be able to make inquiries to you live instead of having to communicate back and forth via posts.

The new Daphne Loader will not be included (yet) in this release because I want to focus mainly on the self-test. But if you know how to use the command-line, you can unlock all of the new stuff. I will make a complete changelog once the official release is ready.

I'm going to keep track of test results for different vid cards as they come in:

PASSED ALL TESTS:
-----------------------
NVIDIA GeForce 6200 TurboCache(TM)
GeForce 3
GeForce 4 MX
GeForce 6800 Ultra
ATI 8500
ATI Radeon VE
ATI Radeon 9800 Pro
ATI X300
Intel(R) 82852/82855 GM/GME Graphics Controller
ALL-IN-WONDER 9800 SERIES

FAILED AT LEAST 1 TEST:
------------------------------
Radeon 9200 (failed windowed w/ HW acceleration)
SAPPHIRE RADEON 9200SE ATLANTIS (failed fullscreen w/ acceleration)
Radeon 9600 (failed fullscreen w/ HW acceleration)
MOBILITY RADEON 9000 (acceleration doesn't work at all)
S3 Graphics Inc. SavageIX (failed fullscreen w/ accel)

Hi.
There is the report of my selftest.bat:

Here's my report from the self test.bat.

--DAPHNE version 0.99.7pre1
--Command line is: daphne releasetest noldp
--CPU : GenuineIntel 3400 MHz || Mem : 2048 megs
--OS : Windows XP/.NET || Video : NVIDIA GeForce 6800 Ultra
Set 640x480 at 32 bpp with flags: 0
Initializing sound system ...
No joysticks detected
Booting ROM ...
Beginning RGB2YUV exerciser (this may take a long time) ...
Beginning BLEND accuracy test...
YUV overlay is done in software (ie unaccelerated).
Beginning VLDP BLEND sanity test...
Beginning VLDP Render test...
Beginning YUV hardware acceleration test (this test should be last since it may mess up the video modes beyond use)
--- RELEASETEST SUMMARY
-----------------------
LineParse #1 passed.
LineParse #2 passed.
LineParse #3 passed.
LineParse #4 passed.
LineParse #5 passed.
LineParse #6 passed.
FrameFile Parse #1 passed.
FrameFile Parse #2 passed.
FrameFile Parse #3 passed.
FrameFile Parse #4 passed.
FrameFile Parse #5 passed.
FrameFile Bad Parse #1 passed.
FrameFile Bad Parse #2 passed.
FrameFile Bad Parse #3 passed.
FrameFile Bad Parse #4 passed.
RGB2YUV Complete Exerciser passed.
BLEND accuracy test passed.
VLDP BLEND Sanity Test passed.
VLDP Overlay w/ Vertical Offset Render passed.
Env Var Test #1 passed.
Env Var Test #2 passed.
Env Var Test #3 passed.
YUV Hwaccel (windowed = 0) and (accel = 0) passed.
YUV Hwaccel (windowed = 1) and (accel = 0) passed.
YUV Hwaccel (windowed = 0) and (accel = 1) passed.
YUV Hwaccel (windowed = 1) and (accel = 1) passed.
# of passed tests: 26
# of failed tests: 0
Shutting down sound system...
Shutting down video display...

Hello Matt,

Not sure if you want the logs from people that have passed the test but I'll post mine anyway just in case.

--DAPHNE version 0.99.7pre1
--Command line is: daphne releasetest noldp
--CPU : GenuineIntel 3300 MHz || Mem : 1024 megs
--OS : Windows XP/.NET || Video : ALL-IN-WONDER 9800 SERIES
Set 640x480 at 32 bpp with flags: 0
Initializing sound system ...
No joysticks detected
Booting ROM ...
Beginning RGB2YUV exerciser (this may take a long time) ...
Beginning BLEND accuracy test...
YUV overlay is done in software (ie unaccelerated).
Beginning VLDP BLEND sanity test...
Beginning VLDP Render test...
Beginning YUV hardware acceleration test (this test should be last since it may mess up the video modes beyond use)
--- RELEASETEST SUMMARY
-----------------------
LineParse #1 passed.
LineParse #2 passed.
LineParse #3 passed.
LineParse #4 passed.
LineParse #5 passed.
LineParse #6 passed.
FrameFile Parse #1 passed.
FrameFile Parse #2 passed.
FrameFile Parse #3 passed.
FrameFile Parse #4 passed.
FrameFile Parse #5 passed.
FrameFile Bad Parse #1 passed.
FrameFile Bad Parse #2 passed.
FrameFile Bad Parse #3 passed.
FrameFile Bad Parse #4 passed.
RGB2YUV Complete Exerciser passed.
BLEND accuracy test passed.
VLDP BLEND Sanity Test passed.
VLDP Overlay w/ Vertical Offset Render passed.
Env Var Test #1 passed.
Env Var Test #2 passed.
Env Var Test #3 passed.
YUV Hwaccel (windowed = 0) and (accel = 0) passed.
YUV Hwaccel (windowed = 1) and (accel = 0) passed.
YUV Hwaccel (windowed = 0) and (accel = 1) passed.
YUV Hwaccel (windowed = 1) and (accel = 1) passed.
# of passed tests: 26
# of failed tests: 0
Shutting down sound system...
Shutting down video display...


Allister Fiend

Hi Matt,

First off, The Batch Test Passed All 26 Tests on my Lap Top Computer.

--DAPHNE version 0.99.7pre1
--Command line is: daphne releasetest noldp
--CPU : GenuineIntel 1500 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : Intel(R) 82852/82855 GM/GME Graphics Controller
Set 640x480 at 32 bpp with flags: 0
Initializing sound system ...
No joysticks detected
Remapping input ...
Booting ROM ...
Beginning RGB2YUV exerciser (this may take a long time) ...
Beginning BLEND accuracy test...
YUV overlay is done in software (ie unaccelerated).
Beginning VLDP BLEND sanity test...
Beginning VLDP Render test...
Beginning YUV hardware acceleration test (this test should be last since it may mess up the video modes beyond use)
--- RELEASETEST SUMMARY
-----------------------
LineParse #1 passed.
LineParse #2 passed.
LineParse #3 passed.
LineParse #4 passed.
LineParse #5 passed.
LineParse #6 passed.
FrameFile Parse #1 passed.
FrameFile Parse #2 passed.
FrameFile Parse #3 passed.
FrameFile Parse #4 passed.
FrameFile Parse #5 passed.
FrameFile Bad Parse #1 passed.
FrameFile Bad Parse #2 passed.
FrameFile Bad Parse #3 passed.
FrameFile Bad Parse #4 passed.
RGB2YUV Complete Exerciser passed.
BLEND accuracy test passed.
VLDP BLEND Sanity Test passed.
VLDP Overlay w/ Vertical Offset Render passed.
Env Var Test #1 passed.
Env Var Test #2 passed.
Env Var Test #3 passed.
YUV Hwaccel (windowed = 0) and (accel = 0) passed.
YUV Hwaccel (windowed = 1) and (accel = 0) passed.
YUV Hwaccel (windowed = 0) and (accel = 1) passed.
YUV Hwaccel (windowed = 1) and (accel = 1) passed.
# of passed tests: 26
# of failed tests: 0
Shutting down sound system...
Shutting down video display...

More Good News. The 11 Second delay on Coin Up that was in previous version of DAPHNE while playing any version of Dragon's LAir is gone in this New Daphne 0.99.7-Pre version.

Now the Bad News.


Galaxy Ranger and Star Blazers doesn't work.

--DAPHNE version 0.99.7pre1
--Command line is: C:\Documents and Settings\Tony M. Victorino\Desktop\Games\daphne-0.99.7pre1-win32\daphne.exe galaxy vldp -framefile D:\DAPHNE\framefile\DAPHNE Prototype\mpeg2\galaxy.txt -fullscreen -blank_skips -blank_searches -bank 0 00000000 -bank 1 00000110
--CPU : GenuineIntel 1500 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : Intel(R) 82852/82855 GM/GME Graphics Controller
Set 640x480 at 32 bpp with flags: 80000001
Initializing sound system ...
No joysticks detected
Remapping input ...
Loading compressed ROM image gr5633.bin...16384 bytes read.
Loading compressed ROM image gr5634.bin...16384 bytes read.
Loading compressed ROM image gr5592.bin...16384 bytes read.
Loading compressed ROM image gr5592.bin...16384 bytes read.
Loading compressed ROM image gr5593.bin...16384 bytes read.
Loading compressed ROM image gr5594.bin...16384 bytes read.
Loading compressed ROM image gr5611.bin...16384 bytes read.
Loading compressed ROM image gr5595.bin...16384 bytes read.
Loading compressed ROM image gr5612.bin...16384 bytes read.
Loading compressed ROM image 5280...2048 bytes read.
Loading compressed ROM image 5281...2048 bytes read.
Loading compressed ROM image 5279...512 bytes read.
Loading compressed ROM image pr-5278.bin...256 bytes read.
Loading compressed ROM image pr-5277.bin...256 bytes read.
Loading compressed ROM image pr-5276.bin...32 bytes read.
Loading compressed ROM image pr-5275.bin...32 bytes read.
total used colors 0xcd - compressing palette
WARNING : Astron Belt transparent color changed but it isn't 'wired' to anything
Framefile parse succeeded.
Could not open file : D:/DAPHNE/framefile/DAPHNE Prototype/mpeg2/./galaxy/galaxy.m2v!
Could not initialize laserdisc player!
Shutting down sound system...
Shutting down video display...


DLE 1.1 while using DVD2DAPH DVD Video Files. DAPHNE does a Parse
check everytime it starts up. Only checks the Prototype files.
Otherwise, Game seems to work fine.

--DAPHNE version 0.99.7pre1
--Command line is: C:\Documents and Settings\Tony M. Victorino\Desktop\Games\daphne-0.99.7pre1-win32\daphne.exe dle11 vldp -framefile D:\DAPHNE\framefile\DAPHNE Prototype\mpeg2\DL20\lair_dl20.txt -fullscreen -blank_skips -blank_searches -useoverlaysb -bank 0 11011001 -bank 1 00100011
--CPU : GenuineIntel 1500 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : Intel(R) 82852/82855 GM/GME Graphics Controller
Set 640x480 at 32 bpp with flags: 80000001
Initializing sound system ...
No joysticks detected
Remapping input ...
Loading compressed ROM image dle11u1l.bin...8192 bytes read.
Loading compressed ROM image dle11u2l.bin...8192 bytes read.
Loading compressed ROM image dle11u3l.bin...8192 bytes read.
Loading compressed ROM image dle11u4l.bin...8192 bytes read.
Framefile parse succeeded.
YUV overlay is hardware accelerated.
YUV overlay is hardware accelerated.
LDP-VLDP: Could not parse video because file D:/DAPHNE/framefile/DAPHNE Prototype/mpeg2/DL20/./dlproto16.vob.m2v¡ could not be opened.
YUV overlay is hardware accelerated.
Booting ROM ...
dle11 : Re-allocated overlay surface (360 x 240)...
Play
LDP : disc is already playing, play command ignored
Play
Display disable received
Search to 323 received - [25] Attract Mode, Sequence 0 Attract Mode
YUV overlay is hardware accelerated.
NOTE: converting FPS from 23.976000 to 29.970000. This may be less accurate.
dle11 : Re-allocated overlay surface (320 x 240)...
Play
Display disable received
Search to 306 received - [25] Attract Mode, Sequence 0 Attract Mode
YUV overlay is hardware accelerated.
Freeing memory used to store audio stream...
dle11 : Re-allocated overlay surface (360 x 240)...
Search to 306 received - [25] Attract Mode, Sequence 0 Attract Mode
Shutting down sound system...
Shutting down video display...


DLE 2.0 while using DVD2DAPH DVD Video Files. DAPHNE does a Parse
check everytime it starts up. Only checks the Prototype files.
No matter if you play original or prototype game, or change the Coin Settings or use Daphne/Factory checked, the Game Plays as 4 Coins for 5 Lives at all times.
Otherwise, Game seems to work fine.

--DAPHNE version 0.99.7pre1
--Command line is: C:\Documents and Settings\Tony M. Victorino\Desktop\Games\daphne-0.99.7pre1-win32\daphne.exe dle20 vldp -framefile D:\DAPHNE\framefile\DAPHNE Prototype\mpeg2\DL20\lair_dl20.txt -fullscreen -blank_skips -blank_searches -useoverlaysb -bank 0 11011000 -bank 1 01110011
--CPU : GenuineIntel 1500 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : Intel(R) 82852/82855 GM/GME Graphics Controller
Set 640x480 at 32 bpp with flags: 80000001
Initializing sound system ...
No joysticks detected
Remapping input ...
Loading compressed ROM image DLE20_U1.bin...8192 bytes read.
Loading compressed ROM image DLE20_U2.bin...8192 bytes read.
Loading compressed ROM image DLE20_U3.bin...8192 bytes read.
Loading compressed ROM image DLE20_U4.bin...8192 bytes read.
Framefile parse succeeded.
YUV overlay is hardware accelerated.
YUV overlay is hardware accelerated.
LDP-VLDP: Could not parse video because file D:/DAPHNE/framefile/DAPHNE Prototype/mpeg2/DL20/./dlproto16.vob.m2v¡ could not be opened.
YUV overlay is hardware accelerated.
Booting ROM ...
dle20 : Re-allocated overlay surface (360 x 240)...
Play
LDP : disc is already playing, play command ignored
Play
Display disable received
Search to 323 received - [25] Attract Mode, Sequence 0 Attract Mode
YUV overlay is hardware accelerated.
NOTE: converting FPS from 23.976000 to 29.970000. This may be less accurate.
dle20 : Re-allocated overlay surface (320 x 240)...
Play
Display disable received
Search to 306 received - [25] Attract Mode, Sequence 0 Attract Mode
YUV overlay is hardware accelerated.
Freeing memory used to store audio stream...
dle20 : Re-allocated overlay surface (360 x 240)...
Shutting down sound system...
Shutting down video display...


Space Ace and SAE 1.0 while using DVD2DAPH DVD Video Files. Parse Error. Game won't work.

--DAPHNE version 0.99.7pre1
--Command line is: C:\Documents and Settings\Tony M. Victorino\Desktop\Games\daphne-0.99.7pre1-win32\daphne.exe ace vldp -framefile D:\DAPHNE\framefile\DAPHNE Prototype\mpeg2\SADVDROM\ace_dvdrom.txt -fullscreen -blank_skips -blank_searches -useoverlaysb -bank 0 00000000 -bank 1 00100001
--CPU : GenuineIntel 1500 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : Intel(R) 82852/82855 GM/GME Graphics Controller
Set 640x480 at 32 bpp with flags: 80000001
Initializing sound system ...
No joysticks detected
Remapping input ...
Loading ./roms/ace/sa_a3_u1.bin ... 8192 bytes read into memory
Loading ./roms/ace/sa_a3_u2.bin ... 8192 bytes read into memory
Loading ./roms/ace/sa_a3_u3.bin ... 8192 bytes read into memory
Loading ./roms/ace/sa_a3_u4.bin ... 8192 bytes read into memory
Loading ./roms/ace/sa_a3_u5.bin ... 8192 bytes read into memory
Framefile Parse Error
Expected a number followed by a string, but on line 350, found this: ð

Hi Matt,

Bega's Battle Continue problem.

I see that Bega's Battle no longer crashes at the end of a game.

Great Job!

But the Continue goes by super fast. You can continue, but you just need to be really fast.


Yours Truly,

Tony M. Victorino

I won't waste time posting stuff you already know.

I have a radeon 9600 and like the others, it fails the fullscreen test with hardware acceleration.

Good work all around though, I just don't see the point in posting a whole page of data that somebody else has already posted.

I found some more oddities.......


There seems to be errors in both the video parser and the actual parsing of the framefile.

Here's what happens with cliffy, notice the garbage tacked on to the path.

--DAPHNE version 0.99.7pre1
--Command line is: E:\daphne\daphne.exe cliff vldp -framefile g:\roms\daphne\cliff.txt -fullscreen -nohwaccel
--CPU : AuthenticAMD 1900 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : RADEON 9600 SERIES
Set 640x480 at 32 bpp with flags: 80000000
Initializing sound system ...
Joystick #0 was successfully opened
Remapping input ...
Loading compressed ROM image cliff_u1.bin...8192 bytes read.
Loading compressed ROM image cliff_u2.bin...8192 bytes read.
Loading compressed ROM image cliff_u3.bin...8192 bytes read.
Loading compressed ROM image cliff_u4.bin...8192 bytes read.
Loading compressed ROM image cliff_u5.bin...8192 bytes read.
Framefile parse succeeded.
Could not open file : G:/mpeg/cliff/ch.m2v(q1Â'q1
Could not initialize laserdisc player!
Shutting down sound system...
Shutting down video display...


Also, although these games aren't officially working, Dragon's Lair 2, Us vs Them and Mach3 all error out on the sound initilization. This is odd as thayer's quest and other games are just fine. Also they ran fan on my pc for the cvs build prior to this.

Here's Us vs Them, notice that it's trying to call the mach3-null.ogg????

--DAPHNE version 0.99.7pre1
--Command line is: E:\daphne\daphne.exe uvt vldp -framefile g:\roms\daphne\uvt.txt -fullscreen -blank_skips -blank_searches -x 1024 -y 768 -nohwaccel -nocrc
--CPU : AuthenticAMD 1900 MHz || Mem : 512 megs
--OS : Windows XP/.NET || Video : RADEON 9600 SERIES
Set 1024x768 at 32 bpp with flags: 80000000
Initializing sound system ...
Can't load sound sample sound/mach3-01.ogg
ERROR: one or more required sound sample files could not be loaded!
Sound initialization failed!
Shutting down video display...

I figured out a few of the issues, one of which is an error on my end (that you'll all make) and the other is some odd glitch that's been newely introduced afaik.

Mach3 has a new sample set (of oggs oddly enough) that goes in your daphne/sound folder. It's floating around in the usual places. Without these samples, several games will fail.

Now for the odd one. The famefile parser is getting a bit picky....

Use to, you could put in a path like:

g:\mpeg\lair

or

g:/mpeg/lair

or even

g:\mpeg\lair\

now daphne only accepts:

g:/mpeg/lair

Notice that the slashes are reverse the usual way and there isn't a trailing slash at the end.

I had to go in and alter many of my framefiles, but most work now.

I hope this helps.

HAC

Heh, ok take three.....


Cliffy is still acting up

It isn't parsing the framefile properly, but cliffalt is. I even pointed to the alternate version's framefile and cliff still errored out. So something is wrong there.


To summarize: Cliff isn't working, it gives the error I posted above.

cliffalt, on the otherhand, is

I'll keep looking into this but it appears at first galnce that the issue might be related to whatever function daphne has written to read the framefiles.

Hi Matt,

The selftest failed on my Dell D600 laptop, results below. It also states wrong CPU, should be 1700 MHz.

--DAPHNE version 0.99.7pre1
--Command line is: daphne releasetest noldp
--CPU : GenuineIntel 600 MHz || Mem : 1024 megs
--OS : Windows XP/.NET || Video : MOBILITY RADEON 9000
Set 640x480 at 32 bpp with flags: 0
Initializing sound system ...
No joysticks detected
Booting ROM ...
Beginning RGB2YUV exerciser (this may take a long time) ...
Beginning BLEND accuracy test...
YUV overlay is done in software (ie unaccelerated).
Beginning VLDP BLEND sanity test...
Beginning VLDP Render test...
Beginning YUV hardware acceleration test (this test should be last since it may mess up the video modes beyond use)
--- RELEASETEST SUMMARY
-----------------------
LineParse #1 passed.
LineParse #2 passed.
LineParse #3 passed.
LineParse #4 passed.
LineParse #5 passed.
LineParse #6 passed.
FrameFile Parse #1 passed.
FrameFile Parse #2 passed.
FrameFile Parse #3 passed.
FrameFile Parse #4 passed.
FrameFile Parse #5 passed.
FrameFile Bad Parse #1 passed.
FrameFile Bad Parse #2 passed.
FrameFile Bad Parse #3 passed.
FrameFile Bad Parse #4 passed.
RGB2YUV Complete Exerciser passed.
BLEND accuracy test passed.
VLDP BLEND Sanity Test passed.
VLDP Overlay w/ Vertical Offset Render passed.
Env Var Test #1 passed.
Env Var Test #2 passed.
Env Var Test #3 passed.
YUV Hwaccel (windowed = 0) and (accel = 0) passed.
YUV Hwaccel (windowed = 1) and (accel = 0) passed.
# of passed tests: 24
YUV Hwaccel (windowed = 0) and (accel = 1) FAILED!
YUV Hwaccel (windowed = 1) and (accel = 1) FAILED!
# of failed tests: 2
Shutting down sound system...
Shutting down video display...

ok, the problem w/ the framefile parser was it didn't terminate the end of the file properly, so it was continuing to parse garbage. I've fixed this. Here is a quick 'patch' for you testers. I'll bundle it up into a real archive later.

http://www.daphne-emu.com/download/daphne.exe-0.99.7pre2.zip

Tony M. Victorino reported that this solved most of his framefile woes, so hopefully it will solve yours too.

This probably won't fix any of the failed tests, but if you did fail any tests and want to try selftest.bat again on this new exe, AND if you notice any changes, feel free to report them here.

This is not what is supposed to happen. It is supposed to accept \ characters and convert them to / characters (since win32 accepts those too and linux only accepts / characters). I'll try to reproduce this on my latest build (pre2) just to make sure the framefile non-termination bug isn't intefering...

Yeah I figured as such. Upon further inspection, I found out that the framefile garbage problem you mentioned was the real culprit. I found that if you put a space at the end of the framefile it doesn't screw up, otherwise it does.

But it looks like you've already fixed it so nevermind.