2087 replies to this topic

[BigBoss]

 

For now the best step would be to disable this UAC setting in the VS project since it's generally not needed for most people.

 

 

 

Sadly it seems people do still need to run as administrator - there are tables that write high scores to files, that fail if VP isn't run as administrator.   

 

See this post for an example.   I saw a few people complaining about it on and off, but it took me a while to correlate this to the fact that your build didn't trigger UAC the same way as the official builds.  

They could simply adjust the folder permissions where the files need to be saved.

[mukuste]

I checked in a change which should hopefully fix the weird behavior/crashes when running recent VP versions as administrator. Also removed the UAC flag since it's usually not needed now.

 

 

Bump.

How's progress going on this.
Looking forward to the mesh ball.

Thanks for all you devs hard work.

Where's toxie been? Haven't heard much from him on the forum lately

 

Last I heard he had to take a one month break for real-life reasons. That was... wait for it... one month ago! So he should pop in any minute now

[fuzzel]

Bump.

How's progress going on this.
Looking forward to the mesh ball.
 

I hope that I find the time to complete it this weekend

[mukuste]

 

Bump.

How's progress going on this.
Looking forward to the mesh ball.
 

I hope that I find the time to complete it this weekend

 

 

What do you think about keeping the 3D ball for the initial VP10 release? Since you said it's a breaking change, it seems like it would be a better fit there. I think we should keep the compatibility release focused on compatibility -- and this way we have more exciting features to put into the shiny new version! See also the discussion surrounding this post: http://www.vpforums....e=4#entry261414

[fuzzel]

I thought about that too, but it's not a big deal to add the 3D ball as an optional feature.You have to set it for each table because you have to use different textures. The 3D ball fixes some odd behaviour with decals and perhaps fixes the kicker cup issue. For VP10 I would like to add real reflection mapping on the ball but this is only possible using shaders so a new exciting feature still exists

[gigalula]

Don't you think that always trying to get older table compatible on newer build could be more simple by adding some kind of way to detect the version required inside .VPT and simply launch the proper code for example if a table require vp8 or vp9.15 or 9.2.1 looking at this specific zone inside .VPT if auto detection fail with no version stamped then ask to select proper version to be store for next time auto detection and running some kind of DLL version of older existing VP version. I'm not a coder but I think this is the best way to ensure compatibility with old table by playing them it their original VP version.  

 

Just a suggestion to keep all process more easy and stop request to adjust table to new VP level so that way if maker of the table or any other with permission want to upgrade a table then they upgrade tag version so any user could be able to use it without headache.  

 

If i'm wrong with the suggestion sorry in advance. It's just that I seriously think that currently keeping track of all old VPinball.x.x.x.exe version to keep the compatibility perfect and manually select any required version is getting complicate for nothing and could be managed for the pleasure of playing

 

Hope this will not offend anyone because it's only a suggestion for the best interest of all.

 

Please be nice my English is not perfect    

Edited by gigalula, 11 April 2014 - 10:06 AM.

[mukuste]

I thought about that too, but it's not a big deal to add the 3D ball as an optional feature.You have to set it for each table because you have to use different textures. The 3D ball fixes some odd behaviour with decals and perhaps fixes the kicker cup issue. For VP10 I would like to add real reflection mapping on the ball but this is only possible using shaders so a new exciting feature still exists

 

But let's face it, introducing a new feature always introduces new bugs. So then we have to support this version further, have additional bugfix releases, and so on. This whole procedure is already taking forever and I just want to finish the VP9 branch at this point. At this pace, if we keep adding new stuff, it will be months until we can finally start on VP10.

 

As you said, introducing a mesh ball is only the first step, the real visual benefit will only come when using shaders and reflection mapping. On its own, I guess the mesh ball won't even look significantly better than the current sprite-based one; maybe even worse in some respects because it can't be antialiased. So I don't know how much benefit there will even come from adding this now.

 

I think the best way would be to have a true 3D ball in VP10, not as an optional per-table setting, but as a core part of the engine such that it can be integrated in the new lighting system that will be developed. There are already too many per-table settings, I don't believe adding yet another is the right approach.

[unclewilly]

I'm happy whatever you guys decide.

I have to admit though, I am eager to see vp10 development to begin.

I also have an update and a new table in progress waiting on the dx9 port.

Much appreciation goes to all of you devs for bringing new life to this community

[melon]

Although I'd like to use all these new features today, I think the best idea is to incorporate them in VP10 version. Yes we'll have to wait, but it will be worth. Otherwise we're going to have dozens of versions of each table (table to be played with DX version revxxx, table to be played with Physics versionxxx...) and it's going to be a terrible headache for you programmers to have allways to work in compatibility, a lot of settings here and there... you'd waste more time with this than develloping VP10 itself.

That's my opinion.

[boiydiego]

you also hit the nial with that quote MELON it may not never be a mess of running , with that version , enable that options etc etc just keep it simple for us all ..

[fuzzel]

 

I thought about that too, but it's not a big deal to add the 3D ball as an optional feature.You have to set it for each table because you have to use different textures. The 3D ball fixes some odd behaviour with decals and perhaps fixes the kicker cup issue. For VP10 I would like to add real reflection mapping on the ball but this is only possible using shaders so a new exciting feature still exists

 

But let's face it, introducing a new feature always introduces new bugs. So then we have to support this version further, have additional bugfix releases, and so on. This whole procedure is already taking forever and I just want to finish the VP9 branch at this point. At this pace, if we keep adding new stuff, it will be months until we can finally start on VP10.

 

As you said, introducing a mesh ball is only the first step, the real visual benefit will only come when using shaders and reflection mapping. On its own, I guess the mesh ball won't even look significantly better than the current sprite-based one; maybe even worse in some respects because it can't be antialiased. So I don't know how much benefit there will even come from adding this now.

 

I think the best way would be to have a true 3D ball in VP10, not as an optional per-table setting, but as a core part of the engine such that it can be integrated in the new lighting system that will be developed. There are already too many per-table settings, I don't believe adding yet another is the right approach.

 

Well good point. I'm going to make some tests with my 3D ball in VP9 especially how that one behaves with the kicker cup which cuts the ball in half at the moment, an issue you can't fix with a 2D sprite under DX9 I guess.

[mukuste]

There are other ways to fix it... the main problem right now is that the sprite is angled in a weird way, I think ideally it should always face the viewer.

 

The other way that could work would be to use a stencil buffer to keep the playfield from overwriting the kicker hole instead of abusing the Z buffer for that.

 

So far I haven't come up with any other way, though there might be more options.

[ClarkKent]

What I do not fully understand with VP10 is if it will be full 3D or not? If it will be full 3D how will the tables solves where 3d objects are missing/not available (like Roadshow, White Water, etc.)? Or can tables still be 2.5D as they are now?

[pmarak]

When running BBB under DX9 the tube dancer doesn't appear for me. Anyone else having this issue ? Works fine on any other executable (9.21, 9.15)

[fuzzel]

actually all the walls, gates, ramps, and so on are 3d objects. They are just pre rendered from a fixed camera view that's all. If VP10 renders them as dynamic objects you get a complete 3d table. You don't need real 3d objects made in Blender for the normal elements but that should be possible too.

[mukuste]

I think CK was referring to "fake" 3D table elements like the heads on Funhouse, Roadshow, Whitewater, etc which are now done using textured walls or alpha ramps.

 

My opinion: those will still be supported, however the preferred way in VP10 should be to do as much as possible using true 3D primitives. Only that way can we have a freely movable camera and improve the quality of the lighting. Unclewilly's MB is an excellent model of how I imagine future VP tables to be built: AFAIK, almost everything there is a 3D mesh.

[unclewilly]

Yep pretty much everything

[DJRobX]

I think CK was referring to "fake" 3D table elements like the heads on Funhouse, Roadshow, Whitewater, etc which are now done using textured walls or alpha ramps.

 

My opinion: those will still be supported, however the preferred way in VP10 should be to do as much as possible using true 3D primitives. Only that way can we have a freely movable camera and improve the quality of the lighting. Unclewilly's MB is an excellent model of how I imagine future VP tables to be built: AFAIK, almost everything there is a 3D mesh.

 

Is supporting it going to be an issue as you try to improve the 3d environment?   This is what has me scratching my head a bit about the physics mod being released as a "10.0.0".

 

Now that people have had a drink of that magic sauce, we want to retrofit it onto our library of old tables (and it's something those of us without artistry skills can do ).   If we don't have an interim "physics only" VP9 release (which is, effectively the test you shared this week), I'm concerned you'll be stuck trying to keep old hacky 2D stuff around, as people will expect forward VP10.x releases to run "VP10.0" tables.    I'd rather see you guys be able to make big changes in VP10 without worry of the way things were done in VP9.  

Edited by DJRobX, 11 April 2014 - 10:43 PM.

[mukuste]

I don't think it's a problem for now. Walls and alpha ramps will still exist, and we can't really stop people from abusing them They ARE true 3D objects, people just use them as makeshift billboards, and that's ok.

 

It only becomes a problem once we have a more sophisticated lighting engine which really relies on normal vectors. Then those billboard graphics might stick out like sore thumbs because they won't get lit properly. But that's further down the road, I don't think we'll have anything of the sort in VP 10.0.

 

It may be possible that, say, the 10.1 releases won't play 10.0 tables perfectly. Hence the version check which I suggested. The only way to ensure this compatibility would be to postpone 10.0 for a very long time (basically until we have the whole feature set for all of VP10 mapped out, which may take years). But people want new physics in an official version soon, and new physics require (IMO) a major version number increase. It's all a bit of a catch-22, but it's the best scheme I can come up with.

 

Basically, we'll rely on the table authors (or modders) keeping tables up to date to the latest 10.X release.

[ClarkKent]

I think CK was referring to "fake" 3D table elements like the heads on Funhouse, Roadshow, Whitewater, etc which are now done using textured walls or alpha ramps.
 
My opinion: those will still be supported, however the preferred way in VP10 should be to do as much as possible using true 3D primitives. Only that way can we have a freely movable camera and improve the quality of the lighting. Unclewilly's MB is an excellent model of how I imagine future VP tables to be built: AFAIK, almost everything there is a 3D mesh.

Yes, they are 3d mesh - but only because a member here has the table and the equipment to scan those objects. So if you do not have the specific table and equipment it will be difficult to create appropriate 3D objects...