3889 replies to this topic

[unclewilly]

is the core.vbs file being updated with the new builds with the recent changes and c;ass additions?

[KieferSkunk]

is the core.vbs file being updated with the new builds with the recent changes and c;ass additions?

The dictionary change is live, but I haven't finished anything else yet.

[Talantyyr]

Scientists from the TU Wien published a paper regarding a new rendering method called Seperable Subsurface Scattering. This seems to be very performant. Fuzzel, toxie, you might want to take a look:

 

http://cg.tuwien.ac....ision-blizzard/

 

Source code, documentation and everything is published there and it's opensource, maybe VP could benefit from this method?

[toxie]

yeah, i already know that..

but this is more targeted at the subtle SSS effects like found in skin or marble, not rather simple stuff like backlit plastics..

the principle i implemented in VP so far is much simpler (e.g. much faster), but very very roughly based on the same idea (blurring out light contribution).

 

this kind of advanced shader magic is currently not so easy anyhow in VP, as we still have this concept of static rendering mixed with dynamic rendering (and no depth buffer reads due to that :/).

Edited by toxie, 10 June 2015 - 09:37 AM.

[Talantyyr]

Thx for the explanation. When i read this, i instantly thought about VP... (addicted... ) 

[fuzzel]

Request: I'm searching for a good top down image of a kicker cup. I would like to update the texture for the build-in kicker cup so I need a good image of such a kicker. If someone could provide such image that would be great

[randr]

Most of mine have cliffys but my woz don't have a cliffy protector I could provide pic of that one

[vanlion]

Most likely you have some form of anti-aliasing enabled/forced on in the driver control panel.
Set it to default there, and it should work. Otherwise try to update your gfx driver to the latest version.
 
@CK: Did you already rule out that the VBS changes are the actual problem (i haven't looked at the differences there in detail, so don't know if that could cause such behavior)?

It worked :-)

[KieferSkunk]

 

Most likely you have some form of anti-aliasing enabled/forced on in the driver control panel.
Set it to default there, and it should work. Otherwise try to update your gfx driver to the latest version.
 
@CK: Did you already rule out that the VBS changes are the actual problem (i haven't looked at the differences there in detail, so don't know if that could cause such behavior)?

It worked :-)

 

 

vanlion: Since I'm concerned about the VBS stuff and asked folks to try an older version to see if that had any performance impacts, could you clarify what "it worked" means here?  Setting your gfx anti-aliasing to "default", updating your driver, or running with an older core.vbs?

[vanlion]

Hi KieferSkunk,

I have a Nvidia card and what worked for me was to set the 3d quality from high to "Let the graphic card deside" whats the best in the control panel.

[KieferSkunk]

Hi KieferSkunk,

I have a Nvidia card and what worked for me was to set the 3d quality from high to "Let the graphic card deside" whats the best in the control panel.

 

Cool, thanks for clarifying.   Based on ClarkKent's message from earlier, it sounds like the VBS files aren't contributing to the perf issues at the moment.

[ClarkKent]

It's still weird as I played Kingpin today with constant 120 fps. No changes at all. No other settings. I hope that there will be further optimizations to have some room for more performance...

[Ben Logan]

Just want to give a quick word of encouragement to you VP10 developers and beta-testers.  I just discovered Visual Pinball.  Our whole family has been "gathering around the table" nightly, thanks to you folks.  It is wonderful to see development continuing on this amazing platform.  You guys are making Randy Davis proud. 

 

 

[fuzzel]

Thanks Ben

[Slydog43]

I was wondering what Randy would think of VP10.  When its "done" someone should contact him and find out his thoughts. Great job Devs

[toxie]

at some point he was quickly around for some weeks, but that was still in VP9.2 times.. and he even wanted to commit more stuff back then (which turned out to be just some tiny snippets though )..

 

but maybe he still checks the forum now and then..

[KieferSkunk]

Good news, everyone!  Core.vbs 3.46 is now ready.  Find the new version as an attachment to this message.  I will also send a copy to toxie, who will make sure it gets into the next VP10 update.

 

 core346.zip   69.92KB   55 downloads

 

I recommend that you back up your current core.vbs and vbsdoc.html files before applying the new files, so that if you find something breaking you, you can quickly restore your current working versions.  If you find bugs, please let me know and I'll try to fix them ASAP.

 

 

What's new in this version?

 

New class: cvpmTrough, which takes over for cvpmBallStack in "trough mode".  Provides accurate simulation of trough switches, stacking balls in the exit, and can handle any number of balls.  Also more flexible to setup - you can initialize trough switches with an array of numbers, or one at a time.  (You can also set ".isDebug = True" on a trough and watch what it's doing internally!  Requires a TextBox named DebugBox on your backglass. )

 

New class: cvpmSaucer, which takes over for cvpmBallStack in "saucer mode".  Provides all the same support for saucer functions as cvpmBallStack, but should be easier to work with.

 

cvpmBallStack is now marked as DEPRECATED.  It's easy to migrate from this class to the new classes - only your initialization calls should need to change.  Tables will continue to work with BallStack, including with all of its known issues.

 

Updated vbsdoc.html with the new classes and complete instructions on how to use them.

 

Two new global helper functions: "vpMin(a,b)" and "vpMax(a,b)" - returns the lesser or greater of two numbers, respectively.  Can also be used with strings.

 

 

What was the purpose of this change?

 

There were some known issues with cvpmBallStack that I wanted to fix.

• Your script would crash if you tried to add more balls to a ballstack than it could handle.
• Balls in a trough didn't really "move" - they simply changed position.  In some cases, this would cause one or more trough switches to be permanently "stuck down", resulting in the ROM throwing test errors after a while.  (For example, this happens in Bride of Pinbot because the machine requires 3 balls in the trough but only ever uses 2 of them.)
• The trough couldn't accurately simulate "stacking" balls at the exit.  In some machines, a subway kicker kicks only one ball out even if there are multiple balls that had been added from an entrance, while it will kick out two or more balls if they came in from the exit.  BallStack cannot handle this scenario - it does allow you to set a "KickBalls" value, but this unconditionally kicks out up to that number of balls no matter how they were added.

When I tried to address these issues in the existing BallStack class, I found this work to be very difficult for several reasons:

• The "trough" and "saucer" concepts were so intermingled in this class that it was difficult to separate them enough to make this change elegantly.
• The new features I wanted to add had a high risk of breaking compatibility with existing tables.
• Conceptually, troughs and saucers are very different things with only a little bit in common.

After asking here and getting some consensus from the community, it made more sense to split cvpmBallStack into two separate classes that each simulate their own things.  Not only does this give you purpose-built classes that make it easier to understand what's going on, it also makes it possible to take advantage of these new features without breaking your table.

 

 

Let me know what you think, and if you find any bugs in this code.

Edited by KieferSkunk, 13 June 2015 - 06:17 AM.

[nicco84]

Another brick in vpx wall!!
I know it's not elegant like answer but could we have an idea of the release of vpx alpha?

[Talantyyr]

It's already beta. Go to the first post and download it
It's already beta. Go to the first post and download it

[gtxjoe]

Thanks for the core vbs update. Will try switching to the new classes on some table wips and see how it goes