GroovyMAME 0.171 - SwitchRes v0.015m

[Calamity]

I was afraid to hear that. Something in the way we're forcing video switching has broken input. SDL sucks so much. Only idea left... build with xinput enabled?

[Doozer]

Stock vanilla mame have xinput disabled and can handle sdl2. I am now digging inside the src/emu changes.

Edit: nothing useful discovered here...

Edit2: damn! really annoying issue. Just to report here that I have tested previous mame versions up to mame64_0162.015g and they all behave identically. Something have changed inside SDL from 2.0.3 to 2.0.4 and groovy patch s impacted.

@ALL. could a Windows user with sdl 2.0.4 installed confirm that this issue is not OS dependant? Just to narrow down the search area

[Calamity]

Ok I've uploaded new binaries, the "_fix2" ended. These should work fine, finally. I've deleted the previous, broken ones. I'd appreciate the users that had been testing the previous versions could test this one, combined with the CRT Tools beta 4 should solve previous issues.

Now working on a tutorial to configure VMMaker for HD 4xxx, step by step.

The Linux binaries have still broken input, read Doozer's instructions regarding SDL 2.0.3.

[haynor666]

Is it beta 4 and fix2 versions still for people that need Taito Type X games running at 100% speed?

What exactly was changed in fix2 ?

[Calamity]

Is it beta 4 and fix2 versions still for people that need Taito Type X games running at 100% speed?

What exactly was changed in fix2 ?

"fix2" is simply a working version of the "fix" version I posted yesterday. Just follow my instructions and enable positive sync if you want 100% refresh speed on interlaced games. Unfortunately this setting is not supported by some monitors.

[keropi]

[...]

Now working on a tutorial to configure VMMaker for HD 4xxx, step by step.

[...]

Excellent news , can't wait for it! I do have questions but don't want to bother you fine folks that much, hopefully the tutorial will be of great help.
Haven't tested it yet but my arcade monitor is a "POLO 25" according to a sticker on the control board, I assume it's a 15/25khz one and hopefully it will work great with GM 

[Calamity]

Uploaded "fix3", solves crash related to new cards (legacy cards not affected).

[Jollywest]

Is GM v0.151 Linux available anywhere please?

All links I've found are either dead ends or lead to the latest version.

Thanks.

[Calamity]

Is GM v0.151 Linux available anywhere please?

All links I've found are either dead ends or lead to the latest version.

Thanks.

Nobody built that exact version for Linux, the closest you have is 0.152, here:

https://code.google.com/p/groovyarcade/downloads/list

However the diff is there, in case you want to roll your own build.

[Jollywest]

Is GM v0.151 Linux available anywhere please?

All links I've found are either dead ends or lead to the latest version.

Thanks.

Nobody built that exact version for Linux, the closest you have is 0.152, here:

https://code.google.com/p/groovyarcade/downloads/list

However the diff is there, in case you want to roll your own build.

Cheers for that, I'll have a look into building it.

[Doozer]

Just to clarify, it's not that I get a black screen, its that I get nothing at all, mame creates no window or viewport or whatever it creates.

@RobertJ,

I have uploaded versions named *_SDL2.0.4+_fix to the share drive. Could you kindly check if it solves your issue?

[Calamity]

I've uploaded 0169_groovymame_015l_fix4.diff, including new Doozer's patch for SDL 2.0.4 input.

(Windows users don't need to update)

[RobeeJ]

I have uploaded versions named *_SDL2.0.4+_fix to the share drive. Could you kindly check if it solves your issue?

The groovymame inside the tar.bz2 doesn't appear to be an executable. Whenever I try and launch it bash says command not found.

I tried the diff Calamity just posted and rebuilt from scratch, and that has the same issue as before. :/ Tried various sdl2 versions, no difference.

I appreciate the time both of you are putting into this, it's a thankless task, but thank you all the same.

[Jonny G]

Just wanted to say thank you for all the hard work you've put into this new version, Calamity.

Got my cab back up and running tonight using XP-x64 9.3 Emu Drivers with CRT Tools 2.0 Beta 4. A bit of trial and error required getting the right settings in VMM, but just stuck with dynamic mode for now. Deffo ran into the blurry problem at one point and a few cases where I had to check if thottle was enabled or not because some games were running at extreme speeds.

Anyway, happy now. Thanks again.

[Doozer]

The groovymame inside the tar.bz2 doesn't appear to be an executable. Whenever I try and launch it bash says command not found.

I tried the diff Calamity just posted and rebuilt from scratch, and that has the same issue as before. :/ Tried various sdl2 versions, no difference.

I appreciate the time both of you are putting into this, it's a thankless task, but thank you all the same.

In any case you should see groovymame picture. Have you already tried to launch mame64 without any patch?

[RobeeJ]

In any case you should see groovymame picture. Have you already tried to launch mame64 without any patch?

Just literally built a clean mame169 with just the hiscore diff patch, and that runs straight away, no 100% cpu, however it does run in completely the wrong aspect ratio for reasons I don't understand as its the same config as groovymame is using.

So the issue is somehow related to the groovymame patch. :/

[Calamity]

Just literally built a clean mame169 with just the hiscore diff patch, and that runs straight away, no 100% cpu, however it does run in completely the wrong aspect ratio for reasons I don't understand as its the same config as groovymame is using.

So the issue is somehow related to the groovymame patch. :/

Try forcing -lcd_range 60-60
(default lcd range is the only change I did in 169)

Also, consider posting a full log.

[Doozer]

So the issue is somehow related to the groovymame patch. :/

I am wondering if your issue is not linked to some configuration parameters. For example, the shader model is having issue when both tilt angle components are set to zero.

Code: [Select]

const vec2 angle = vec2(0.0,0.0);

# == to be changed to =>

const vec2 angle = vec2(0.0,0.01);

But as Calamity said, please provide your logs and ini to have a better diagnostic.

[Calamity]

Uploaded "fix5":

 - (Linux) Fixes input for SDL 2.0.4 (Doozer)
 - (Windows) Fixes Powerstrip support

[haynor666]

Calamity, is fix5 version ready for Windows XP ? If yes, I have to use latest tools or I can stay on old ones ?

[Calamity]

Calamity, is fix5 version ready for Windows XP ? If yes, I have to use latest tools or I can stay on old ones ?

It should be ready. Always use the latest tools.

[RobeeJ]

Try forcing -lcd_range 60-60
(default lcd range is the only change I did in 169)

Same issue with that either in mame.ini or on the command line.

Also, consider posting a full log.

I did.

http://forum.arcadecontrols.com/index.php/topic,135823.msg1552569.html#msg1552569

[RobeeJ]

For comparison, here is the working log from 168, if I run 169 with the lcd_range setting the SwitchRes monitor range line matches the one below, but still hangs before the rng(0) line:

Code: [Select]

SwitchRes: v0.015k, Monitor: lcd, Orientation: horizontal, Modeline generation: disabled
SwitchRes: Using default vfreq range for LCD 60.000000-60.000000
SwitchRes: Found output connector 'DVI-I-1'
SwitchRes: Creating automatic specs for LCD based on VESA GTF
SwitchRes: Monitor range 74520.00-74520.00,60.00-60.00,0.696,1.044,1.740,0.013,0.040,0.510,0,1,1200,1200,0,0
SwitchRes: -resolution was set at command line or in .ini file as 1600x1200@60

SwitchRes: Entering switchres_modeline_setup
SwitchRes: v0.015k:[pacman] Calculating best video mode for 224x288@60.606060 orientation: rotated

SwitchRes: [1600]x[1200]_(60=0.0000Hz)
   rng(0): 1600 x1200_0.000p 0.000 [integ] scale(4, 4, 1) diff(0.44, 3.95, -0.6061) ratio(7.143, 4.167)

SwitchRes: [pacman] (1) vertical (224x288@60.61)->(1600x1200@0.00)
   rng(0): 1600 x1200_0.000p 0.000 [integ] scale(4, 4, 1) diff(0.44, 3.95, -0.6061) ratio(7.143, 4.167)
SwitchRes: Setting option -rotate
SwitchRes: Setting option -noror
SwitchRes: Setting option -noautoror
SwitchRes: Setting option -norol
SwitchRes: Setting option -noautorol
SwitchRes: Setting option -noblack_frame_insertion
SwitchRes: Setting option -multithreading
SwitchRes: Setting option -syncrefresh
SwitchRes: Setting option -waitvsync
SwitchRes: Setting option -keepaspect
SwitchRes: Setting option -nounevenstretch
SwitchRes: Setting option -nofilter
SwitchRes: Setting option -prescale 3
Available videodrivers: x11 wayland dummy
Current Videodriver: x11
        Display #0
                Renderdrivers:
                            opengl (0x0)
                         opengles2 (0x0)
                          opengles (0x0)
                          software (0x0)
Available audio drivers:
        pulseaudio         
        alsa               
        dsp                 
        disk               
        dummy               
Build version:      0.168 (Dec 20 2015)
Build architecure: 
Build defines 1:    SDLMAME_UNIX=1 SDLMAME_X11=1 SDLMAME_LINUX=1
Build defines 1:    LSB_FIRST=1 PTR64=1 SYNC_IMPLEMENTATION=tc
SDL/OpenGL defines: SDL_COMPILEDVERSION=2003 USE_OPENGL=1
Compiler defines A: __GNUC__=5 __GNUC_MINOR__=3 __GNUC_PATCHLEVEL__=0 __VERSION__="5.3.0"
Compiler defines B: __amd64__=1 __x86_64__=1 __unix__=1
Compiler defines C: __USE_FORTIFY_LEVEL=0
Enter init_monitors
Adding monitor screen0 (1600 x 1200)
Leave init_monitors
Enter sdlwindow_init
Using SDL multi-window OpenGL driver (SDL 2.0+)
/dev/dri/card0 successfully opened

Hints:
        SDL_FRAMEBUFFER_ACCELERATION             (null)
        SDL_RENDER_DRIVER                        (null)
        SDL_RENDER_OPENGL_SHADERS                (null)
        SDL_RENDER_SCALE_QUALITY                 (null)
        SDL_RENDER_VSYNC                         (null)
        SDL_VIDEO_X11_XVIDMODE                   (null)
        SDL_VIDEO_X11_XINERAMA                   (null)
        SDL_VIDEO_X11_XRANDR                     (null)
        SDL_GRAB_KEYBOARD                        (null)
        SDL_VIDEO_MINIMIZE_ON_FOCUS_LOSS         (null)
        SDL_IOS_IDLE_TIMER_DISABLED              (null)
        SDL_IOS_ORIENTATIONS                     (null)
        SDL_XINPUT_ENABLED                       (null)
        SDL_GAMECONTROLLERCONFIG                 (null)
        SDL_JOYSTICK_ALLOW_BACKGROUND_EVENTS     (null)
        SDL_ALLOW_TOPMOST                        (null)
        SDL_TIMER_RESOLUTION                     (null)
        SDL_RENDER_DIRECT3D_THREADSAFE           (null)
        SDL_VIDEO_ALLOW_SCREENSAVER              (null)
        SDL_ACCELEROMETER_AS_JOYSTICK            (null)
        SDL_MAC_CTRL_CLICK_EMULATE_RIGHT_CLICK   (null)
        SDL_VIDEO_WIN_D3DCOMPILER                (null)
        SDL_VIDEO_WINDOW_SHARE_PIXEL_FORMAT      (null)
        SDL_VIDEO_MAC_FULLSCREEN_SPACES          (null)
        SDL_MOUSE_RELATIVE_MODE_WARP             (null)
        SDL_HINT_RENDER_DIRECT3D11_DEBUG         (null)
        SDL_VIDEO_HIGHDPI_DISABLED               (null)
        SDL_HINT_WINRT_PRIVACY_POLICY_URL        (null)
        SDL_HINT_WINRT_PRIVACY_POLICY_LABEL      (null)
        SDL_HINT_WINRT_HANDLE_BACK_BUTTON        (null)
Leave sdlwindow_init
Enter sdl_info::create
OpenGL: nouveau
OpenGL: Gallium 0.4 on NVC8
OpenGL: 3.0 Mesa 11.1.0
OpenGL: texture rectangle supported
OpenGL: non-power-of-2 textures supported (new method)
OpenGL: vertex buffer supported
OpenGL: pixel buffers supported
OpenGL: framebuffer object supported
OpenGL: GLSL supported, but disabled
OpenGL: max texture size 16384 x 16384
Leave sdl_info_ogl::create
Keyboard: Start initialization
Input: Adding Kbd #0: System keyboard
Keyboard: Registered System keyboard
Keyboard: End initialization
Mouse: Start initialization
Input: Adding Mouse #0: System mouse
Mouse: Registered System mouse
Mouse: End initialization
Joystick: Start initialization
Input: Adding Joy #0: Ultimarc IPAC 2 Ultimarc IPAC 2
Joystick: Ultimarc IPAC 2 Ultimarc IPAC 2
Joystick:   ...  4 axes, 32 buttons 1 hats 0 balls
Joystick:   ...  Physical id 0 mapped to logical id 1
Joystick: End initialization
output: unable to open output notifier file /tmp/sdlmame_out
Audio: Start initialization
Audio: Driver is pulseaudio
Audio: frequency: 48000, channels: 2, samples: 256
sdl_create_buffers: creating stream buffer of 18432 bytes
Audio: End initialization
Region ':maincpu' created
Region ':gfx1' created
Region ':proms' created
Region ':namco' created
Starting Pac-Man (Midway) ':'
Optional shared pointer 'patched_opcodes' not found
Optional shared pointer 'rocktrv2_prot' not found
Optional shared pointer 's2650_tileram' not found
Optional shared pointer 's2650_spriteram' not found
  (missing dependencies; rescheduling)
Starting Z80 ':maincpu'
Starting gfxdecode ':gfxdecode'
Starting palette ':palette'
Starting Video Screen ':screen'
Starting Speaker ':mono'
  (missing dependencies; rescheduling)
Starting Namco ':namco'
Starting Pac-Man (Midway) ':'
Optional shared pointer 'patched_opcodes' not found
Optional shared pointer 'rocktrv2_prot' not found
Optional shared pointer 's2650_tileram' not found
Optional shared pointer 's2650_spriteram' not found
  (missing dependencies; rescheduling)
Starting Speaker ':mono'
Starting Pac-Man (Midway) ':'
Optional shared pointer 'patched_opcodes' not found
Optional shared pointer 'rocktrv2_prot' not found
Optional shared pointer 's2650_tileram' not found
Optional shared pointer 's2650_spriteram' not found
OpenGL: VBO supported
OpenGL: PBO supported
OpenGL: FBO supported
OpenGL: using vid filter: 0
GL texture: copy 1, shader 0, dynamic 1, 864x672 864x672 [PALETTE16, Equal: 0, Palette: 1,
            scale 3x3, border 0, pitch 384,864/16384], bytes/pix 4
GL texture: copy 1, shader 0, dynamic 1, 864x672 864x672 [PALETTE16, Equal: 0, Palette: 1,
            scale 3x3, border 0, pitch 384,864/16384], bytes/pix 4
GL texture: copy 1, shader 0, dynamic 1, 864x672 864x672 [PALETTE16, Equal: 0, Palette: 1,
            scale 3x3, border 0, pitch 384,864/16384], bytes/pix 4
Average speed: 81.70% (10 seconds)
sdl_kill: closing audio
Sound buffer: overflows=3 underflows=6
Enter sdlwindow_exit
Leave sdlwindow_exit
Joystick: Start deinitialization
Joystick: End deinitialization
SwitchRes: Restoring desktop resolution: 1600x1200
SwitchRes: Running 'xrandr --output DVI-I-1 --mode 1600x1200'

[Calamity]

I'll have a look RobertJ.

(It's much better to attach the log file to your post)

[Calamity]

Try forcing -lcd_range 60-60
(default lcd range is the only change I did in 169)

Same issue with that either in mame.ini or on the command line.

Also, consider posting a full log.

I did.

http://forum.arcadecontrols.com/index.php/topic,135823.msg1552569.html#msg1552569

That's not a full log, just a few lines.

[RobeeJ]

That's not a full log, just a few lines. Did you try the lcd_range thing I suggested?

No, really, that's the full log! That's as far as it gets with -v, after that the process hits 100% cpu and never gets any further.

I did try the lcd_range, it just changed the text in the Monitor Range line to match the previous working version, it still hangs in the same place.

[Calamity]

No, really, that's the full log! That's as far as it gets with -v, after that the process hits 100% cpu and never gets any further.

I did try the lcd_range, it just changed the text in the Monitor Range line to match the previous working version, it still hangs in the same place.

Ok thanks! I didn't imagine it'd be that bad  I have a slight idea of the cause of it.

[Calamity]

Uploaded "fix6":

 - (Linux) Attempt to fix bug reported by RobertJ (untested, please check if possible).
 - (Windows, new AMD cards) Fix crash reported by intealls, (still won't turn a progressive mode to interlaced in the same session, probably not possible at all).
- (Windows) Properly tell apart system modes (surprised this hasn't cause issues yet, or has it?).

[intealls]

Edit: Only seems to work with ddraw and d3d9ex, not d3d?? D3D still selects the progressive mode.

[Calamity]

I accidentally edited your post trying to answer.

I meant if what only seems to work is your workaround or my new build.

[intealls]

The new build shows the same result for D3D and DDraw, such that it (I presume) selects the native/progressive mode. D3D9Ex still shows 'Error during present call'.

With the old build with the workaround, D3D9Ex and DDraw selects 720x480@60i. D3D selects 720x480@60p.
With the new build and the workaround D3D and DDraw selects 720x480@60p. D3D9Ex selects 720x480@60i.

Sorry for the confusion, attached logs for the new build + workaround. Fastforward is used so the speed is not reported.

[Calamity]

So new build without wa + d3d9ex crashes? I cloned your modelines here this morning and it worked fine for all apis. Although mode was not updated properly to interlace as I said.

[intealls]

So new build without wa + d3d9ex crashes? I cloned your modelines here this morning and it worked fine for all apis. Although mode was not updated properly to interlace as I said.

Yes, that is correct. It doesn't crash per se but locks up with an empty black window covering most of the screen, when ESC is pressed there are a lot of Error during present call in the log.

[Calamity]

That didn't happen here, will test again on monday.
edit: It might be related with -mt

[intealls]

That didn't happen here, will test again on monday.
edit: It might be related with -mt

Ok, let me know if you need anything.

MT does not seem to affect the result, attaching two logs of the latest build with/without mt without the workaround.

Edit: Also, there seems to be a forum bug related to the attachments. I *just* posted this and the download counter for these files must be corrupt (24/27). Now 38/38. I saw the issue once before in the original fix2 crash post. Who should we contact about this?

[RobeeJ]

Uploaded "fix6":

 - (Linux) Attempt to fix bug reported by RobertJ (untested, please check if possible).

Wooooooop!!!! \o/ Thanks Calamity, fix6 now works for me!!!

What was the cause?

[Calamity]

Ok, let me know if you need anything.

MT does not seem to affect the result, attaching two logs of the latest build with/without mt without the workaround.

Edit: Also, there seems to be a forum bug related to the attachments. I *just* posted this and the download counter for these files must be corrupt (24/27). Now 38/38. I saw the issue once before in the original fix2 crash post. Who should we contact about this?

Thank for the logs. I've searched for error 08760877, it's S_PRESENT_MODE_CHANGED: https://msdn.microsoft.com/es-es/library/windows/desktop/bb172554(v=vs.85).aspx

It looks like we should reset the device if that happens, but I remind you reporting another issue about resetting the device in d3d9ex.

No idea about the download counter thing 

[Calamity]

Wooooooop!!!! \o/ Thanks Calamity, fix6 now works for me!!!

What was the cause?

The specific combination you're using: -nomodeline_generation, -monitor lcd, was forcing a call to the modeline generator with vfreq = 0, causing the call to hang. Old version didn't expose the bug because it forced refresh to be editable regardless the -modeline_generation option.

[RobeeJ]

Ah, that makes sense, though surprising it eats CPU rather than bombs out.

Thanks for your time and patience fixing this release, it seems to have been a difficult one.

[Calamity]

Edit: Only seems to work with ddraw and d3d9ex, not d3d?? D3D still selects the progressive mode.

Because I accidentally edited your post I missed the part where you describe the workaround. By looking at your logs I can't see any relevant difference with/without workaround, apart from the crash obviously.