3889 replies to this topic

[ClarkKent]

The new physics of the kickers look mch more realistic now! Thank you! What a great update!

[Swisslizard]

Hi guys

 

Unfortunately I havent found time to test VP 10 yet, but there is a small issue in earlier VP versions. VP does (or at least did) not release the Pinmame object properly when exiting back to the editor.

 

It seems that VP is still holding some handle for the Pinmame object, which stops the object from being discarded which in turns also stops other dependant objects like B2S.Server and DOF from being disposed as well. This is always a bit annoying when developing and testing DOF, since VP always needs to be closed (at least the table) before I DOF can be updated to the newest test version. In particular while hunting bugs, frequent updates are common.

 

So, if something can be done about this easily, I'd be grateful for a fix.

 

All the best

 

Tom

[hmueck]

We had similar problem with db2s tables for a long time. The solution was to add a table_exit subroutine to the table script.
 

sub table1_exit
  controller.stop
end sub

[lodger]

regarding b2s implementation, is there any reason that backglasses aren't managed within the vp format itself (with a checkbox option on whether to render the backglass, etc.)? This seems like it would lend itself to much more robust control over the backglass, etc.

 

in the name of customizability you could hypothetically choose to import your own (being that the vp files act as sort of a container for all the graphics, sound, code, etc.)

[hmueck]

regarding b2s implementation, is there any reason that backglasses aren't managed within the vp format itself ?

 

Sure. Nobody programmed it yet.

 

if herweh would make his sourcecode public, maybe there would be somebody to implement his editor in VP directly. But that's probably crazy much work.

Edited by hmueck, 02 June 2015 - 02:02 PM.

[lodger]

 

regarding b2s implementation, is there any reason that backglasses aren't managed within the vp format itself ?

 

Sure. Nobody programmed it yet.

 

if herweh would make his sourcecode public, maybe there would be somebody to implement his editor in VP directly. But that's probably crazy much work.

 

touché That would probably help ease the learning curve a bit

[ICPjuggla]

regarding b2s implementation, is there any reason that backglasses aren't managed within the vp format itself ?

 
Sure. Nobody programmed it yet.
 
if herweh would make his sourcecode public, maybe there would be somebody to implement his editor in VP directly. But that's probably crazy much work.

I could be wrong but I thought Herweh did release the source code!

[hmueck]

I could be wrong but I thought Herweh did release the source code!

He did? I thought i read somewhere, that he left and never released it?! Hmm...

[arngrim]

no he didn't release as me and swisslizard were asking him for over a year now and no response, we want to extend DOF functionalities and we are now stuck
or if the code is available let us know where

[toxie]

If somebody feels like doing that, go ahead.. Its not -that- much work actually.. Most of it would be to extend VP to multi-monitor rendering and then make all the BG elements respect that..

(i guess we would not need new elements there, the existing ones should be good enough)

 

As a bonus a b2s importer would be nice of course (don't know how hard that would be though)..

[freneticamnesic]

these new kickers cause some serious lag, setting to legacy fixes a little bit but i notice stutter when the ball is near the kicker still, i have one table that is actually unplayable because of it

[toxie]

do you have a repro table that easily triggers that (so that i could profile it without having to 'play' a table)?

[freneticamnesic]

unfortunately not, need multiple balls on the table......i'll pm you a link incase you feel like trying it. i got report from another person that it lagged as well. otherwise let me know if there's another way i can figure out what's going on for you

i notice it on drain and ball release mostly, if you change those kickers from legacy you'll see it slow down immediately once it drains

[toxie]

ah, okay, so it could be good enough to just put a bunch of balls near the drain..

[toxie]

hmmm.. i guess a link to your test table cannot hurt.. on the default table i don't see it..

[Swisslizard]

Pinmame resp Com object dispose problem:
If I'm not mistaken Controller.Stop does not solve the problem. Stop does stop all processes of the Controller, but the controller does not get disposed. Maybe we should try Set controller=nothing after controller.stop. Anyhow, that fact that the com object is still locked after the table quits to the editor shows that there are still references to the com object. Maybe VP does not properly dispose the VBS host, which in turn still holds a reference to Punmame resp. B2S.Server.

B2S.Server:
I'm not sure how easy and clever it would be to integrate the Backglass editor into VP. After all B2S.Server, the B2S editor and also DOF are all .net technology (B2S is VB.net, DOF is C#). Cluttering the VP project which is real C++, with a lot of .net code is maybe not the best idea.

Secondly I guess VP doesnt know anything about Pinmame and the related objects, since Pinmame is handled from VBS and not directly from VP. B2S.Server is just a proxy for Pinmame, which allows to intercept all data which is sent to and received from Pinmame. If the B2S.Server would be integrated in VP it would be necessary to ensure that B2S does really receive all Pinmame data, which would require that VP has some knowledge about Pinmame resp. the B2S.Server.

As far as I know the source code of B2S has not been released. The last thing I heard from Herweh when I asked about a extra feature for B2S.Server, was that he is not that interested in continuing dev work on B2S (that was more than half a year ago). I dont know if he would be willing to release his source code (at least the code for the B2S.Server would be very valuable).

Edited by Swisslizard, 03 June 2015 - 08:55 AM.

[KieferSkunk]

Hi guys.  Toxie invited me to help out with various improvements to the VBScripts that are provided with the VP10 distribution, so look for some improvements in the core VPM classes.

 

Here's a preview of things I intend to work on:

 

cvpmBallStack: Unless someone fixed it recently already, trough-style ballstacks do not pulse all switches that have balls occupying them.  This means that on tables that require 3 balls to be installed but only ever use two of them (eg. Bride of Pinbot), you eventually get a "Stuck trough switch" error since under normal conditions there will always be a ball in the right-most trough position.  I'm planning to fix this so that when balls "move" inside a ballstack, the associated switches briefly pulse open before being closed again.

 

cvpmTurnTable: This class doesn't currently allow for multiple turn speeds - more specifically, it can't handle slowing down (eg. going from fast to slow but not just stopping).  I have a replacement implementation in the latest VP9 version of World Cup Soccer '94 that handles this case, so I plan to implement that functionality into the core version as well.

 

cvpmLightManager: I've just finished an early version of a light-manager script for VP9 that simplifies things like handling fading lamps, multiple lamps per controller index, lamps of various types (Lights, walls, primitives, etc.).  Since it looks like VP10 has much of the fading stuff built into it, I imagine a LightManager will be considerably simpler for regular Lights, though it could still benefit light effects implemented with other objects.  Look for this soon.

 

Internal: There's a note on the "cvpmDictionary" class that it was created because Microsoft somehow managed to make its own Scripting.Dictionary object "unsafe for scripting".  I'm pretty sure it's been a LONG time since that was the case - all modern Windows systems should have a script host that includes a safe Dictionary object.  And MS's Dictionary is *much* more efficient and flexible than cvpmDictionary (no offense to whoever wrote it).  As such, I will be removing this class and replacing all uses of it with Scripting.Dictionary, which should significantly improve overall performance.  (It will also allow use of dictionaries where the key is a primitive - eg. a String or Integer.)

 

I'm also planning to take a stab at improving the VBS documentation - mostly to make it more readable and to add some navigation to it.

 

Lemme know if there are other bugs or feature requests you know of for the core objects.  I'll happily take a look at them as time allows.  (Note: I'm starting a new job soon, so I can't guarantee quick turnarounds on this stuff. )

[toxie]

Please just make sure that its somehow backwards compatible with already existing VP9.X.X and its VP9 tables, as we want to have the vbs files as a one-works-for-all solution, so that people don't get confused where to put what, etc (currently it works to have VP8, VP9, VP10 all in the same directory, including the tables and vbs)..

 

Thanks for the help already!

[fuzzel]

I think there are some confusions about the new kicker behavior so here is a rule of thumb how to use the new kicker:

- use the new kicker physics if you habe a visible kicker cup or hole. With visible I mean it's visible for the player.

- if the kicker is more a special trigger because you have to delete/create balls like the release or drain kicker, use the legacy mode

 

The new kicker physics have an impact on the ball calculation (velocity and position). If multiple balls touch the kicker at the same time it might produce slow downs. I know that's not perfect but until I find a solution for this situation you have to deal with that

[The Loafer]

Secondly I guess VP doesnt know anything about Pinmame and the related objects, since Pinmame is handled from VBS and not directly from VP. B2S.Server is just a proxy for Pinmame, which allows to intercept all data which is sent to and received from Pinmame. If the B2S.Server would be integrated in VP it would be necessary to ensure that B2S does really receive all Pinmame data, which would require that VP has some knowledge about Pinmame resp. the B2S.Server.

Not 100% sure but I think the intent is to bring VPM emulation into VPX at some point... I think?