740 replies to this topic

[kiwi]

I have already read in the past discussions about whether to delete or make the wall at the bottom of the playfield invisible.
This wall could be annoying in non-standard tables, for example in cocktail tables.

 

Thanks

Something about this, if it's not a big deal.

 

[fuzzel]

But you can do that by setting the wall's top/side to invisible!? Or do I miss something?

[fuzzel]

rev3282 is up:

 

- add size column to collection manager and fix sorting of collections

- tweak add/remove to collection option

- check for potential null pointer 

- update nudging test table with mods by Sir Cheddar

- core.vbs package patch by nFozzy:

o Prevent 'object not a collection' errors if vpmNudge.TiltObj isn't set

o Support for double leaf flipper switches

o For now, keybinds for these staged flippers are defined in VPMKeys.vbs. By default they are set to LeftFlipperKey and RightFlipperKey, disabling them.

o Adapting older tables requires vpmFlips: Create upper flipper subs and point SolCallback(sULFlipper) and SolCallback(sURFlipper) to them.

o This may break compatibility with some older WPC tables that use the 'cSingleLFlip' method (More info in WPC.vbs)

o Integrated FastFlips, (new object vpmFlips): Low latency flipper response for games with pre-solid state flippers

o Ensure 'vpmInit me' is called in the table _init section

o UseSolenoids = 2 enables and auto sets the game-on solenoid (based on GameOnSolenoid in the system .vbs script)

o Important info on supported WPC games is documented in WPC.vbs

[jimmihenry]

Got the first copy! Halleluja! #1

[kiwi]

 

I have already read in the past discussions about whether to delete or make the wall at the bottom of the playfield invisible.
This wall could be annoying in non-standard tables, for example in cocktail tables.

 

Thanks

Something about this, if it's not a big deal.

 

 

Effectively, I cut the image a bit too much and it is not very clear what it actually is.
I'm talking about the gray wall that is part of the playfield.

 

[toxie]

 

rev3282 is up:

 

- add size column to collection manager and fix sorting of collections

- tweak add/remove to collection option

- check for potential null pointer 

- update nudging test table with mods by Sir Cheddar

- core.vbs package patch by nFozzy:

o Prevent 'object not a collection' errors if vpmNudge.TiltObj isn't set

o Support for double leaf flipper switches

o For now, keybinds for these staged flippers are defined in VPMKeys.vbs. By default they are set to LeftFlipperKey and RightFlipperKey, disabling them.

o Adapting older tables requires vpmFlips: Create upper flipper subs and point SolCallback(sULFlipper) and SolCallback(sURFlipper) to them.

o This may break compatibility with some older WPC tables that use the 'cSingleLFlip' method (More info in WPC.vbs)

o Integrated FastFlips, (new object vpmFlips): Low latency flipper response for games with pre-solid state flippers

o Ensure 'vpmInit me' is called in the table _init section

o UseSolenoids = 2 enables and auto sets the game-on solenoid (based on GameOnSolenoid in the system .vbs script)

o Important info on supported WPC games is documented in WPC.vbs

 

 

Please guys, test all kinds of machine generations/manufacturers, just to be sure that nothing broke by this..

[toxie]

 

 

I have already read in the past discussions about whether to delete or make the wall at the bottom of the playfield invisible.
This wall could be annoying in non-standard tables, for example in cocktail tables.

 

Thanks

Something about this, if it's not a big deal.

 

 

Effectively, I cut the image a bit too much and it is not very clear what it actually is.
I'm talking about the gray wall that is part of the playfield.

 

 

 

is removed with the next build.. it's just a leftover from ancient times actually..

[nFozzy]

Quick setup for the new flipper script:

 

Go into the table script and change UseSolenoids = 2, add the line 'VpmInit me' to the table _init sub (if it's not there already). That's it.

 

This bypasses pinmame for a faster flipper response. If it's working, you should see a very low (<1ms) 'Left Flipper keypress to rotate' response in the F11 performance info.

The game compatibility list:

- Pre-solid-state flipper games (except Zaccaria and LTD) should work perfectly. This includes Bally/Williams WPCs up to Terminator 2 / Party Zone

- Data East / early Segas will work perfectly, unless they have rom-controlled flipper effects.

- Fliptronics and WPC-S games (Addams Family through Jack Bot / WHO Dunnit) will work with caveats (no rom controlled flipper effects, beware stuck balls. More info in WPC.vbs)

- Sega Whitestar, (Apollo 13 / Goldeneye) WPC95, (Congo/AFM) and Capcom onward will not work.

 

There's also a debug test command which may be useful if it's not working properly. Open the debug window (Accessible from the escape menu, press the ">" button to bring up the text field) and type in 'vpmflips.DebugTest'

[fuzzel]

rev3286 is up:

 

- add paranoia re-import sound patch by DJRobX

- remove rendering of lower playfield border/quad

[DJRobX]

Question for DJRobX: Does the new physics loop routine potentially speed up latency even if...

1. You’ve added nFozzy’s fastflips routine to a table?
2. Bypassed flipper vpinmame polling (leaving flippers “always on” even when game is over)?

Thanks!

 

Sorry for the slow reply, it's been a busy week with the upcoming holiday...

 

1 & 2 = Yes.   

 

The amount of reduction will be less dramatic but still present.   The purpose of the change is space out the input reads and subsequent physics responses before a frame is rendered.    The issue is present regardless of whether not VPM is in use, but the cost of a missed flipper read is a lot bigger when VPM needs time to turn it back around. 

 

In the case of non-VPM, this reduction will not show up on the F11 flipper test.  Why?  Because that test begins measuring from the time VP read the flipper input.    The problem that this attempts to address is that input is mostly only being read once per frame (it may read it ~16 times, but is mostly bunching them up in the same time period which isn't terribly useful).    There may have been up to 16.6ms (in the case of 60fps rendering) of additional delay than what that test shows, and this attempts to reduce that.

 

It will also help nudge sensors, because those input reads will also be read at more time periods over the course of a frame.   This is especially true with PinScape controllers that currently sum up movements for 10ms increments.    Using 450-500 loop delay should keep the VP reads under that amount, resulting in more accurate nudging and less interference from shakers and sounds.

Edited by DJRobX, 22 December 2017 - 11:18 PM.

[Ben Logan]

Great, highly detailed response, DJRobX. Thanks so much. Now if we can just get a simple way to speed up usb polling on the Windows OS front!

Your development in conjunction with nfozzy’s represents a big leap forward for flipper snappiness. Totally appreciated. You guys are champs. And thanks to Toxie and Fuzzel for baking all these changes into the latest 10.4 update.

[senseless]

Quick setup for the new flipper script:

 

Go into the table script and change UseSolenoids = 2, add the line 'VpmInit me' to the table _init sub (if it's not there already). That's it.

 

This bypasses pinmame for a faster flipper response. If it's working, you should see a very low (<1ms) 'Left Flipper keypress to rotate' response in the F11 performance info.

The game compatibility list:

- Pre-solid-state flipper games (except Zaccaria and LTD) should work perfectly. This includes Bally/Williams WPCs up to Terminator 2 / Party Zone

- Data East / early Segas will work perfectly, unless they have rom-controlled flipper effects.

- Fliptronics and WPC-S games (Addams Family through Jack Bot / WHO Dunnit) will work with caveats (no rom controlled flipper effects, beware stuck balls. More info in WPC.vbs)

- Sega Whitestar, (Apollo 13 / Goldeneye) WPC95, (Congo/AFM) and Capcom onward will not work.

 

There's also a debug test command which may be useful if it's not working properly. Open the debug window (Accessible from the escape menu, press the ">" button to bring up the text field) and type in 'vpmflips.DebugTest'

 

Tried the new flipper setup but failed . With UseSolenoids = 2 the flipper stop working. Tried the debugtest:

 

--DebugTest--
Game:(t2_l8) Flipper Sol=31(Auto) Flippers on?: False
Tiltsol problem, flippers should be on right now!
no flipper key inputs detected (press some buttons, or check system vbs!)
--errors detected!--

 

Running the latest v3286

Edited by senseless, 22 December 2017 - 11:37 PM.

[nFozzy]

If the table already has it's own solution then don't bother

 

That's something else to mention, the script works off of UseSolenoids, the core solenoid handling script. So if the table has it's own solenoid scripting (UseSolenoids = 0) then it won't work, and turning UseSolenoids on will cause a script conflict and break solenoid handling altogether. Most tables use UseSolenoids though.

[lizard]

Probably a bit late in the day for 10.4 but as mentioned before about adding control points to walls. Being able to add a nominated number of control points to a wall would be great. So you could click on add control point then enter an amount of points to add would be good for when doing rounded corners where you add 10 or so points to achieve the 1 rounded corner, would save a heap of time when building walls.

 

Cheers

[chepas]

Probably a bit late in the day for 10.4 but as mentioned before about adding control points to walls. Being able to add a nominated number of control points to a wall would be great. So you could click on add control point then enter an amount of points to add would be good for when doing rounded corners where you add 10 or so points to achieve the 1 rounded corner, would save a heap of time when building walls.

 

Cheers

 

Use the ramps, don't look as good but save time and most are hidden anyway.

 

----------------------------------------------------------------------------------------------------------------

 

I'm playing on my jukebox (touchscreen), how to remove that message telling me i can tap the screen?

[chepas]

Haven't opened this for a long while now and was kind of blaming the PROC crashing as it fully loaded, but you can clearly see here something strange going on. No crash dumps or anything on this.

 

This is from a build 10.4 in September sometime but the table is actually doing it with all versions. The only fix I can think of is to export into a new table for now but will wait and see if anyone shed any light on it.

 

https://youtu.be/njPfncqK2iY

 

 

------------------------

Ok found out what it is. I do remember making Physics sets for this game and had some beautys. Posted in another thread that something fishy was going on with these ages ago, like ghost in the machine.

 

Just setting the override physics to disable and I'm able to launch it.

Edited by chepas, 23 December 2017 - 02:44 AM.

[DJRobX]

Ok, that should do.

 

I deleted unused images and sound samles to reduce size.

 

@arngrim can you take a look at it and make the changes to zzz public?

 

Finally got a chance to look at this.

The update is great - I like the sound test.   Interestingly the high pitched knocker sometimes triggers unwanted movement from my nudge.  

The only problem is "zzz" is not defined at all.   It's an empty DOF config.  Could we use the "DOF test table" instead which has some clean definitions for the various DOF features?    It seems like an appropriate choice for this purpose. 

Edited by DJRobX, 24 December 2017 - 12:54 AM.

[Thalamus]

I thought zzz was there because of a bug and that bug would be that the last table, here zzz isn't exported. So, if we start to use it we would need to fix the bug or give it a new definition.

Edited by Thalamus, 24 December 2017 - 01:07 AM.

[Sir Cheddar]

 

The update is great - I like the sound test.   Interestingly the high pitched knocker sometimes triggers unwanted movement from my nudge.  

The only problem is "zzz" is not defined at all.   It's an empty DOF config.  Could we use the "DOF test table" instead which has some clean definitions for the various DOF features?    It seems like an appropriate choice for this purpose. 

Finally got a chance to look at this.

 

True that - zzz won't work and the DOF test table would be an obvious choice.

However having an dedicated entry like "Nudge Test and Calibration VP10.4" would be even better, DOF test table has a duration on shaker and gear and, well, I'd like to put the hotkeys on blink.

 

Angrim, it's your call - do you want to create a new configuration or should I use the DOF test table configuration?

 

Since the current version needs replacing I've extended the sound test today and set a few samples to bg - many cabs have a sub for the backbox sound system which might influence the acceleration sensors too.

 

Also, documentation may be for wusses but I put some basic instructions into Table Info anyway.

Edited by Sir Cheddar, 24 December 2017 - 02:10 AM.

[DJRobX]

 

Since the current version needs replacing I've extended the sound test today and set a few samples to bg - many cabs have a sub for the backbox sound system which might influence the acceleration sensors too.

 

Good idea.  I'd suggest for now use DOF test table code.   I think the end of the 10.4 cycle is imminent, so to get a working table into that release is prudent.   Maybe could set up blinking control lights as part of the DOF test table profile too (Exxx controlled by script), as to not pollute the DOF table list with too much test stuff.