1002 replies to this topic

[Pinball999]

 

The new physics really change the experience, BUT.. I'm having some serious stutter with it, and also with the latest non official DX9 builds.

It's not the case though with the official 9.2.1 or the DX9test 13.

 

I've read somewhere that someone else had the same problem, but wanted to check if more people were in this situation.

 

 

All the latest DX9 versions (beyond Test13) use their own registry key, so you'll have to set up all the options again. Make sure that everything is identical.

 

 

Yep, I have realized that the options had to be set again since no sound was coming from my second sound system.

But I've solved the problem, it looks like the version does not like when the table Vsync is set to 60Hz.

[Sunskift]

Please Help !

I have tested every dx9 version up to .980 without any problems and I had .980 as my standard VP.

I couldn't resist trying out also the phys mod (version 2)  and played this without issues with the updated Whitewater.

Tonight, I also tried Abracadabra with updated settings and it worked well, but since then I can not go back to .980 without VP crashing (VP9.9.0 has stopped working...... error message).

VP starts up OK, but as soon as I try to load a table, I get the error message.

I have tried everything I can without success. I have cleared the registry key for Visual Pinball, started over with the settings, been able to start a table, but then it crashes again. (same error message).

 

Sorry for bugging this magnificent thread with this issue, but I don't know where to go!

Can anybody suggest what to do to get back to normal ?

Edited by Sunskift, 16 April 2014 - 10:28 PM.

[mukuste]

I hate COM...

 

Have you tried running rev980 as admin/not as admin (whichever you currently don't do)?

[Sunskift]

No difference, still same crash... Tried both

[mukuste]

I just posted a new release in the DX9 thread, try that please.

[EalaDubhSidhe]

I don't recall if I've asked this before, but a big help to me with the updated physics engine would be if the ball could roll along the outside wall of a ramp as smoothly as it does the inside wall. The reason for this is that I make guide walls out of ramps, as well as rails - it guarantees a uniform width along the length of the wall and requires only half as many points, plus you're not limited to flat-topped guides. But having to make them non-collidable, and overlay them with an invisible physical wall is a pain in the arse and offers no physical benefit. There's a lot you could do if guide wall physics were as malleable as the look of the ramps could suggest.

[EalaDubhSidhe]

And on a similar note (although this is more related to the editor, so should probably go in the VP10 thread), could we have the ability to curve flat ramps vertically as well as horizontally? Ramps on a normal machine taper off when they reach the top, and VP can't replicate this very well. It would also give us the means to create proper realistic-looking VUKs where the offshoot doesn't stick out at a 90 degree angle.

[Sunskift]

I just posted a new release in the DX9 thread, try that please.

THANK YOU ! I was totally lost!
Back in business! :-)

[mukuste]

I don't recall if I've asked this before, but a big help to me with the updated physics engine would be if the ball could roll along the outside wall of a ramp as smoothly as it does the inside wall. The reason for this is that I make guide walls out of ramps, as well as rails - it guarantees a uniform width along the length of the wall and requires only half as many points, plus you're not limited to flat-topped guides. But having to make them non-collidable, and overlay them with an invisible physical wall is a pain in the arse and offers no physical benefit. There's a lot you could do if guide wall physics were as malleable as the look of the ramps could suggest.

 

Aah, that's good feedback, I didn't know that doesn't work. So this has always been like this or it's a problem with the new physics?

 

 

And on a similar note (although this is more related to the editor, so should probably go in the VP10 thread), could we have the ability to curve flat ramps vertically as well as horizontally? Ramps on a normal machine taper off when they reach the top, and VP can't replicate this very well. It would also give us the means to create proper realistic-looking VUKs where the offshoot doesn't stick out at a 90 degree angle.

 

Yes, I've thought about that myself. It shouldn't be too hard to do, the main problem is really how to do it on the GUI side since the VP editor doesn't have much to offer in the way of verticality... any ideas how the interface for that could look like?

 

 

 

I just posted a new release in the DX9 thread, try that please.

THANK YOU ! I was totally lost!
Back in business! :-)

 

At your service

[EalaDubhSidhe]

 

I don't recall if I've asked this before, but a big help to me with the updated physics engine would be if the ball could roll along the outside wall of a ramp as smoothly as it does the inside wall. The reason for this is that I make guide walls out of ramps, as well as rails - it guarantees a uniform width along the length of the wall and requires only half as many points, plus you're not limited to flat-topped guides. But having to make them non-collidable, and overlay them with an invisible physical wall is a pain in the arse and offers no physical benefit. There's a lot you could do if guide wall physics were as malleable as the look of the ramps could suggest.

 

Aah, that's good feedback, I didn't know that doesn't work. So this has always been like this or it's a problem with the new physics?

 

 

And on a similar note (although this is more related to the editor, so should probably go in the VP10 thread), could we have the ability to curve flat ramps vertically as well as horizontally? Ramps on a normal machine taper off when they reach the top, and VP can't replicate this very well. It would also give us the means to create proper realistic-looking VUKs where the offshoot doesn't stick out at a 90 degree angle.

 

Yes, I've thought about that myself. It shouldn't be too hard to do, the main problem is really how to do it on the GUI side since the VP editor doesn't have much to offer in the way of verticality... any ideas how the interface for that could look like?

 

 

 

It's always been like that - the ball doesn't recognize the curve of the outside wall as being a curve, so it rattles around in a very sticky way. I think it's linked to how if two walled ramps were set close to each other side by side, the collision area of the outer walls would clash in such a way that they could block off the inside of each other. I'm not sure if that was ever fixed.

 

You'd probably have to create a VUK subset to the ramp object, with its own set of properties: flat or wire type, ramp width, minimum height, maximum height, curve radius. Assuming that the curve starts off vertical and finishes horizontal (and really, there are very few times when you wouldn't do this), this is as much as you probably need to lay the basic geometry down. You could also create two of these and place them together to make a semi-circular arc for depositing the ball on a habitrail or upper playfield, the sort of device Stern employ on Spider-Man and Lord Of The Rings for example.

Edited by EalaDubhSidhe, 17 April 2014 - 07:07 PM.

[slashbot]

Anyone knows if its permitted to upload modified Physics tables, i send a PM to Fuzzel but waiting for an answer?

[ringorian]

Maybe you can use the patch system ?

[fuzzel]

Anyone knows if its permitted to upload modified Physics tables, i send a PM to Fuzzel but waiting for an answer?

I vote for a clean "No way"! The reason is that the only version which supports that is just a special demo version Mukuste did. I understand that all here are excited about the new physics and behavior but "releasing" physics mods here officially will rise more attention and we land there where FP is today. Another reason is that you have to make this special beta version somewhat official and I think I speak for all devs here, we don't want to support beta versions. But I'm not the admin here that is only my personal opinion.

[ringorian]

Anyone knows if its permitted to upload modified Physics tables, i send a PM to Fuzzel but waiting for an answer?

I vote for a clean "No way"! The reason is that the only version which supports that is just a special demo version Mukuste did. I understand that all here are excited about the new physics and behavior but "releasing" physics mods here officially will rise more attention and we land there where FP is today. Another reason is that you have to make this special beta version somewhat official and I think I speak for all devs here, we don't want to support beta versions. But I'm not the admin here that is only my personal opinion.

Fuzzel, isnt it about playing virtual pinball? and if you can get a better play with that physic mod edition of the nicely developed dx version of vp by all you guys, why not.
You can clearly state in the release for what phys mod version it is, whats the problem ? Its not a hack, its not a new stern table.

There are so much tables available with these "old" physics, why not make them much better if possible in an easier way than using bmpr, which are the only tables i play since dx9 make em possible to play more fluid?

Pushing vp to its limits like you guys are doing it brings it to this possibilty. Some people like to have 500 tables in their collection and dont care how they play , others have 10 of em cause they are the only which nearly play like a pinball. If we can have more of em before vp10, why not.

[DJRobX]

I really wish VPF had a table patching system.  This would make sharing physics tweaks for test purposes a total non-issue, the files would be tiny.    That would also make them more "disposable" when future releases that break them come out.

 

Such a patching system exists, unfortunately there's no way it will be used here due to the way it was named.

 

Ringorian - Agreed, I think the problem is that full VPT files are very big, and we probably don't want the download sections getting clogged with new versions of physics mod tables every time Mukuste releases a new version.   

[The Loafer]

Modified tables need author permission, not dev permission. I think some are missing the boat here, but if authors allow the mod, then anything can be released. But devs permission should not and I think will never trump author permission.

Now SHOULD physics modded tables be done, that is a good debatable issue. My take is no, it's too early, but is getting there. I think because many members don't understand the editor very well, its a good idea to have a few additional modded tables as "tests" so more members can provide feedback to devs to assist in tweaking the new physics to a released candidate, but IMHO I would label these "test physics" so it is clear these are only tests and not an official release and yes, they REQUIRE author permission to release. No exceptions.

I really wish VPF had a table patching system.  This would make sharing physics tweaks for test purposes a total non-issue, the files would be tiny.    That would also make them more "disposable" when future releases that break them come out.
 
Such a patching system exists, unfortunately there's no way it will be used here due to the way it was named.
 
Ringorian - Agreed, I think the problem is that full VPT files are very big, and we probably don't want the download sections getting clogged with new versions of physics mod tables every time Mukuste releases a new version.

If a test mod is authorized, there's nothing that prevents a downloading service from being used. I think because of the test nature of these physics mods, VPF should be free of these downloads until a dev says "ok, nothing will change substantially that would negate this table from loading in a newer VP.exe".

[fuzzel]

I'm with The Loafer here. It's not my intention to censor something here. I only see a problem with support questions we will face if a bunch of tables are released for an early test version. Not every user follows the development threads but will have issues here and there with the beta builds. And of course the author's permission was given.

[Sunskift]

I really think The Loafer is right. It would be nice to have a few more test tables with the physics mod. As a beginner, I'm at the same time worried for the forum getting swamped with "hot wired" alpha phys tables. It is easy to get lost. Good to hear from the Loafer that authors are totally in controll of their tables. That make me feel better. We should let the proos concentrate on getting VP9 finished and then started on VP10. And after that ,the authors could start updating tables with totally mindblowing content! Thanks everyone for putting so much effort and energy in to this !

[mukuste]

I also find myself agreeing with The Loafer on most points here.

• The permission to release a table mod should always come from the original author. I can see an exception being made if the original author isn't active in the VP community anymore.
• The physics mod isn't stable enough yet to produce official tables for it. It's still under development, and future releases can and will break previously released modded tables. This is only a test version.
• The official release channels (VPF download section or external hosts) should under no circumstances be used to release tables modded for the new physics at this point. It would make it even harder for newcomers to figure out what's what (see the FP situation).
• I like the idea to provide more test tables so that a somewhat wider public can try the new physics and potentially report bugs. Whether this happens through a few "sanctioned" test tables (I could add them to the first post if that happens) or a more free exchange of tables doesn't matter much to me. What I find important is that these downloads must be confined to this thread and always clearly labeled "as-is" with explicitly no guarantee of future support.

[mukuste]

I don't recall if I've asked this before, but a big help to me with the updated physics engine would be if the ball could roll along the outside wall of a ramp as smoothly as it does the inside wall. The reason for this is that I make guide walls out of ramps, as well as rails - it guarantees a uniform width along the length of the wall and requires only half as many points, plus you're not limited to flat-topped guides. But having to make them non-collidable, and overlay them with an invisible physical wall is a pain in the arse and offers no physical benefit. There's a lot you could do if guide wall physics were as malleable as the look of the ramps could suggest.

 

Can you provide a test table for this, it seems to work fine here:

Attached Files

•  rampoutertest.vpt   128KB   13 downloads