1353 replies to this topic

[toxie]

maybe we could settle on some super-easy intermediate solution?

if a specific option is set, then the file open dialog is automatically opened on VP start and also acts like open and play?

and the default would be that this option is enabled??

 

opinions??

[BorgDog]

maybe we could settle on some super-easy intermediate solution?

if a specific option is set, then the file open dialog is automatically opened on VP start and also acts like open and play?

and the default would be that this option is enabled??

 

opinions??

 

sounds like just what I was looking for, I wasn't asking for anything fancy.  default either way, doesn't matter to me.

[dark]

I'm all for improving work flow, making things easier and having features and options that allow authors creativity to be unhindered but maybe just maybe we should stop looking for unnecessary things to add to VPX and start looking towards DX11 so we can as Toxie said really 'pimp vp".  I know it's always easier to continue to polish a mostly finished work rather than start something from scratch but I can't help but feel like we need to get a start on getting caught up with the times.

 

I know this may sound critical, but how old is FP now?  FP had full 3D viewing and some very limited shaders, and it's really old already now.  I would argue vpx finally looks better than FP and has much better game play but it took a very long time to get there and I worry that it will take even longer to actually surpass this level and get to DX11.  Everyone wants to see their favourite tables built with passion in an environment that's realistic like Time Shock is a great example.  Unity3D was looking promising in this regard but has sadly fell flat, so from where I see it, you guys (fuzzel/toxie) are the only hope of an open source VP platform using Unreal4 game engine or similar.  VPX in it's current state could last us a few more years at least and developing a new vp platform could more than likely also take a few years to develop. 

 

VPX tables already blow away a lot of their commercial counter parts, if we were working with PBL (physical based lighting) and active shaders and lighting from UR4 game engine or unity we'd be leaving the commercial tables in the dust.  I've seen time and time again VP authors just put more passion into their tables than commercial companies do.....and VP authors do it for free.

[toxie]

I do VP programming because i like to do 3D graphics, so just using an existing engine wouldn't be any fun for me anymore.

 

But i agree that over time we need to switch over to something newer than DX9. For efficiency reasons and to make use of newer features there.

Note that so far we were not really limited by DX9 (except for some minor technical details) though.

Even with most of the plans on my todo list, DX9 would be still good enough, simply because it already exposes most of the things that we need for this.

 

Time Shock, btw, could be done with DX9 without problems, btw, too (i don't know what they're using currently).

 

Although, one thing that i would be most interested in currently, HDR and WCG support is definetly not possible with DX9.  :/

Also the multi-monitor support will be a bit tricky with that.

[gStAv]

I'm with you Darkie! 

 

I was so excited when the kickstarter of "Pinball Labs" launched for a year ago. It was build with UR4 engine and ha proof of concept connectivity to VPinMAME allready. 

Too bad it never reached its goal... 

 

I would for sure chip in a steady amount if the community would get a build like that started. VP but in true 3D and all lightning + shaders and textures that comes with it plus the 

gain of it using the GPU and CPU to its full potential. *dreamin* 

 

Anyways! The creators is the one thing we could not live without, so I'm happy. All though I too see them growing out of VP in a period of time with this quality releases! 

 

Cheers!

[dark]

I do VP programming because i like to do 3D graphics, so just using an existing engine wouldn't be any fun for me anymore.

 

But i agree that over time we need to switch over to something newer than DX9. For efficiency reasons and to make use of newer features there.

Note that so far we were not really limited by DX9 (except for some minor technical details) though.

Even with most of the plans on my todo list, DX9 would be still good enough, simply because it already exposes most of the things that we need for this.

 

Time Shock, btw, could be done with DX9 without problems, btw, too (i don't know what they're using currently).

 

Although, one thing that i would be most interested in currently, HDR and WCG support is definetly not possible with DX9.  :/

Also the multi-monitor support will be a bit tricky with that.

Thanks for acknowledging my concerns so long as DX11 is in your plans some where I can rest easier.    I realise we can still get a lot out of DX9 but we can get that much more out of DX11, with the gpu and cpu boost we'd be able to push tables further visually with less performance issues.  I can understand you wanting to build something yourself, I only mention UR4 and unity as prime examples of commonly used modern game engines.

[TerryRed]

I'm all for improving work flow, making things easier and having features and options that allow authors creativity to be unhindered but maybe just maybe we should stop looking for unnecessary things to add to VPX and start looking towards DX11 so we can as Toxie said really 'pimp vp".  I know it's always easier to continue to polish a mostly finished work rather than start something from scratch but I can't help but feel like we need to get a start on getting caught up with the times.

 

I know this may sound critical, but how old is FP now?  FP had full 3D viewing and some very limited shaders, and it's really old already now.  I would argue vpx finally looks better than FP and has much better game play but it took a very long time to get there and I worry that it will take even longer to actually surpass this level and get to DX11.  Everyone wants to see their favourite tables built with passion in an environment that's realistic like Time Shock is a great example.  Unity3D was looking promising in this regard but has sadly fell flat, so from where I see it, you guys (fuzzel/toxie) are the only hope of an open source VP platform using Unreal4 game engine or similar.  VPX in it's current state could last us a few more years at least and developing a new vp platform could more than likely also take a few years to develop. 

 

VPX tables already blow away a lot of their commercial counter parts, if we were working with PBL (physical based lighting) and active shaders and lighting from UR4 game engine or unity we'd be leaving the commercial tables in the dust.  I've seen time and time again VP authors just put more passion into their tables than commercial companies do.....and VP authors do it for free.

 

This is coming from a guy who has been a huge PC Gamer for many years, and loves to tweak graphics for all my games. I've also dabbled in 3d animation in the past with good old Lightwave 3D.

 

I truly love what I've seen with VPX. The new tables are looking wonderful and the lighting, glow and (bloom?) effects are a vast improvement. Ambient Occlusion can also look great.

 

However, in terms of graphics, I would argue that VPX only beats out FP in the glow / radiant effects that can be used on objects. Please correct me if I'm wrong here guys:

 

-VPX still seems to be limited with frame rate of certain table objects. Or other words...the ball moves smooth, but toys look like they move at 15fps? (I always wondered why this is the case...help me out here)

-no true per-pixel type of lighting and shaders?  The new lighting is amazing, but does it effect everything on the playfield? Aren't alot of the lighting effects "faked"?  (amazing what table authors pull off)

-the entire 3d environment still seems to not be rendered in real-time?

 

Future Pinball (which uses OpenGL instead of DX9) hasn't been updated for many years, but ravarcade has done some amazing things with BAM (Better Arcade Mode). When a Future Pinball table has been created and setup with BAM in mind...the graphics, lighting, post processing can look stunning. With BAM and Future Pinball you get:

 

-fully real-time rendered 3d environment 

-Oculus VR support

-REAL normal / bump map support, which can look amazing when utilized

-real per-pixel lighting and rendering

-ambient occlusion

-post processing effects, bloom, etc..

-mini playfields (Indy table - Path of Adventure)

-realtime 3d head tracking for cabinet use (looks soooo cool)

-all lighting and post processing can be adjusted in real-time.

 

Future Pinball also now has full cabinet support through DOF with DOFLinx. Many tables now support all the main cabinet toys and lighting. They have some really cool animated RGB flasher effects and animations (courtesy of yours truly) combined with shaker, gear, fan,beacons, strobes, fire - launch - start - coin buttons, rgb undercab, etc.... Combine all of this together with the Jersey Jack type of videos and lcd display graphics, and it makes for a really stunning looking cabinet experience.

 

I know...but the physics sucks!  Well the original physics do suck...but a newer table that properly supports 2.7 or Zed physics does play MUCH better. Matching the correct physics on a FP table makes a huge difference. Is it perfect...no...but I personally have a harder time playing VP than I do FP. I know....I'm weird.

 

I'm not hi-jacking this thread or VP bashing (I hate that). Anyone who has seen my cabinet videos knows I love VPX and especially the addressable leds I have setup. I just know alot of people dropped FP a long time ago and may have never seen a modern setup with BAM properly utilized..let alone the new cabinet hardware support it has gotten.

 

If I want arcade reproductions of pinball...then VPX is the choice for me. If I want some amazing original tables with all the best graphics, much better sound, videos and bling...then FP is my choice.

 

 

Also...has anyone noticed that Pinball FX 2 for Oculus Rift seems to be using a much nicer lighting engine? (maybe DX11, like PBA?) I hope the normal PC version will get that one day.

 

 

Again...great work on VPX!

[sliderpoint]

I know...but the physics sucks! Well the original physics do suck...but a newer table that properly supports 2.7 or Zed physics does play MUCH better. Matching the correct physics on a FP table makes a huge difference. Is it perfect...no...but I personally have a harder time playing VP than I do FP. I know....I'm weird.

I've tried to go back many times to FP with all the latest physics improvements.  Yes, they are " much better".  But I think they have polished that turd as much as possible and they are still pretty terrible.
 
On a VP note.  I have seen with the last build that I am randomly having the entire editor close out when I hit "esc" when playing a table, not all the time though.  No errors, no crash files, just boop.... gone.
 
-Mike

edit: quote broken

Edited by sliderpoint, 29 November 2016 - 08:15 PM.

[Ben Logan]

 
Time Shock, btw, could be done with DX9 without problems, btw, too (i don't know what they're using currently

I know we discussed this a bit before on a dedicated Time Shock thread when it was released on Steam a while back. How does Time Shock achieve such high resolution of playfield elements? Pre-rendered objects? Static camera angles?

I certainly enjoy playing with different perspective angles in camera / light mode. But, table authors seem to be nailing perspective so consistently lately that I rarely feel the need to adjust perspective.

Could we have an option to pre-render a single static viewpoint in high resolution as with Timeshock? I asked Dark a similar question after seeing his amazing Gilligan's 3D renders (which I realize take hours for a single frame), and I drew the impression that there's more to the puzzle than just pre-rendering to get Time Shock level of detail. But Toxie's quote has me excited about the idea again, nonetheless.

Edited by Ben Logan, 29 November 2016 - 09:11 PM.

[vampirolatino2]

Its almost 2017... in my opinion, the UT4 Engine is the best option to move on. But is just my opinion. Even if is a "already made engine" its not just plug and play, developers need to build and tweak the engine to get what they want. I'm in favor to UT4 Engine because is very good in physics, lightning and graphics department. 3 options that VP need to be strong at for a fluent virtual recreation of a physical thing. Remember guys that VP is trying to emulate a physical thing that exist in real life. It's not only a game. Its a simulation.

 

If only we can just move on...

[dark]

Exactly why we have to look forward,  FP is stuck and can't be made much better than it already is, which is what SliderMike put so eloquently.  Terry outlines a lot of the 'visual' drawbacks of vp, yes it's not real time rendering, but the advantage to that is we can have 'static' rendered primitives that are very high poly with no effect on performance (such as wire ramps).  VP really lacks lighting that really just automatically effects it's surrounding objects, ie illuminates surrounding objects and casts shadows from those objects, FP has it beat there sure but with new vp light it's possible to shape lights and play with illumination values to make objects to 'appear as if' they are being hit by light.  Dynamic camera views are great (as a result of real time rendering) but they really aren't necessary to enjoy a game of pinball, but of course this is limiting for future VR applications.

[Ben Logan]

TicTox has a few pinball tables made with Unity: Tommy and Elvis. They look great. The physics are good (remarkable, considering he's developing on his own).

That said, I'd hate to lose our physics with a move to Unity. VP's physics are the best in pinball simulation, in my opinion. Time Shock is a close second (missing backspin).

[dark]

Its almost 2017... in my opinion, the UT4 Engine is the best option to move on. But is just my opinion. Even if is a "already made engine" its not just plug and play, developers need to build and tweak the engine to get what they want. I'm in favor to UT4 Engine because is very good in physics, lightning and graphics department. 3 options that VP need to be strong at for a fluent virtual recreation of a physical thing. Remember guys that VP is trying to emulate a physical thing that exist in real life. It's not only a game. Its a simulation.

 

If only we can just move on...

Yeah this is true, UR4 would still take a lot of work to get working nicely as a pinball simulator.  Unity 3D probably would of taken off if people besides the developers can make tables for it.  If we had something similar to the VP editor with UR4 engine we'd be cranking out top notch games.  My only worry with this course of action is that it may become so good that the commercial virtual pinball developers may feel threatened by it and be inclined to pursue legal action.

 

I'm not saying we need to move on immediately or anything, but we need to plan ahead, like I said we can still get a few good years of life out of VPX easy, but I just don't want to come to the end of that lifespan and find ourselves still stuck using it because we were still content to stick with vpx and some of us...FP.

[ICPjuggla]

 

I know...but the physics sucks! Well the original physics do suck...but a newer table that properly supports 2.7 or Zed physics does play MUCH better. Matching the correct physics on a FP table makes a huge difference. Is it perfect...no...but I personally have a harder time playing VP than I do FP. I know....I'm weird.

I've tried to go back many times to FP with all the latest physics improvements.  Yes, they are " much better".  But I think they have polished that turd as much as possible and they are still pretty terrible.
 
On a VP note.  I have seen with the last build that I am randomly having the entire editor close out when I hit "esc" when playing a table, not all the time though.  No errors, no crash files, just boop.... gone.
 
-Mike

edit: quote broken

 

 

I've been getting the same problem when hitting "esc". It seems to happen at random...

[MrKnister]

Sorry guys, dont want to Interrupt the discussion... I have a question or Feature request. Maybe its already covered then Just ignore my ignorace , but would it be possible to Set the true fullscreen Function as a per Table Option within the Table options ?

Maybe some Kind of "viewset 1" or "viewset 2" which you predefine in the General options before ?

Im still Not 100% happy on all tables with true fs, and for me i sometimes like the Overall Play experience better with the regular windowed mode :-) .. yeah freaky i know ..

Edited by MrKnister, 29 November 2016 - 10:00 PM.

[toxie]

I was so excited when the kickstarter of "Pinball Labs" launched for a year ago. It was build with UR4 engine and ha proof of concept connectivity to VPinMAME allready. 

Too bad it never reached its goal...

 

I would for sure chip in a steady amount if the community would get a build like that started. VP but in true 3D and all lightning + shaders and textures that comes with it plus the 

gain of it using the GPU and CPU to its full potential. *dreamin* 

 

Anyways! The creators is the one thing we could not live without, so I'm happy. All though I too see them growing out of VP in a period of time with this quality releases!

 

True 3D is not that far out, it just requires more GPU power then (see the already existing camera adjustment mode (start with F6)).

Full PBR shaders+lighting+textures is also already a reality, but 'just' support for multiple 'real' lights is still missing. So that the lighting from flashers + normal lights does not need to be faked.

 

And getting more benefit from CPU + GPU also does not come for free, as the majority of work done in the drivers with DX9 now gets shifted into the hands of the engine with newer versions.  :/

Edited by toxie, 29 November 2016 - 10:38 PM.

[toxie]

 

-VPX still seems to be limited with frame rate of certain table objects. Or other words...the ball moves smooth, but toys look like they move at 15fps? (I always wondered why this is the case...help me out here)

-no true per-pixel type of lighting and shaders?  The new lighting is amazing, but does it effect everything on the playfield? Aren't alot of the lighting effects "faked"?  (amazing what table authors pull off)

-the entire 3d environment still seems to not be rendered in real-time?

 

Future Pinball (which uses OpenGL instead of DX9) hasn't been updated for many years, but ravarcade has done some amazing things with BAM (Better Arcade Mode). When a Future Pinball table has been created and setup with BAM in mind...the graphics, lighting, post processing can look stunning. With BAM and Future Pinball you get:

 

-fully real-time rendered 3d environment 

-Oculus VR support

-REAL normal / bump map support, which can look amazing when utilized

-real per-pixel lighting and rendering

-ambient occlusion

-post processing effects, bloom, etc..

-mini playfields (Indy table - Path of Adventure)

-realtime 3d head tracking for cabinet use (looks soooo cool)

-all lighting and post processing can be adjusted in real-time.

 

Stuttery table objects: thats because the animation usually is dependent on a timer. Using the new timer interval of -1 that i included can fix this, as it is called exactly once per frame, and that could steer the animation perfectly then, with the help of getting the exact time via GameTime or SystemTime.

Lighting+Shaders: This is all there, as said 'just' multiple light support is missing which would help the playfield bulb/flasher lighting to become automatic. One of the major todos on my list. Nothing to complicated, too, it just requires some time to fully implement, maybe over the xmas holidays i'll start with some experiments, who knows.

 

The rest of the FP list is also almost there already except for the VR and head tracking, as we still pre-process as much as possible. But this is mainly done because of performance gains on older hardware, but could be rather easily disabled to allow for full 3D (just take the camera adjustment mode (start with F6) as an example where you basically already get this now, and performance is far from bad IMHO).

 

 

 
On a VP note.  I have seen with the last build that I am randomly having the entire editor close out when I hit "esc" when playing a table, not all the time though.  No errors, no crash files, just boop.... gone.
 

 

i screwed that up with the latest latency reduction changes. will look into it. sorry!

 

 

 
Time Shock, btw, could be done with DX9 without problems, btw, too (i don't know what they're using currently

I know we discussed this a bit before on a dedicated Time Shock thread when it was released on Steam a while back. How does Time Shock achieve such high resolution of playfield elements? Pre-rendered objects? Static camera angles?

I certainly enjoy playing with different perspective angles in camera / light mode. But, table authors seem to be nailing perspective so consistently lately that I rarely feel the need to adjust perspective.

Could we have an option to pre-render a single static viewpoint in high resolution as with Timeshock? I asked Dark a similar question after seeing his amazing Gilligan's 3D renders (which I realize take hours for a single frame), and I drew the impression that there's more to the puzzle than just pre-rendering to get Time Shock level of detail. But Toxie's quote has me excited about the idea again, nonetheless.

 

 

The basic concept is actually super simple what they do, so in a (simplified) nutshell: Prerender a single, full resolution, image for each light being turned on (separately). Then 'compress' all these tons of single images somehow so that for example only regions with a lot of changes (per light) need to be stored explicitly. Then all of that is composited during gameplay (lighting is additive, so one can have each light and each brightness level perfectly included), depending on which lights are on (and how bright these are).

In addition, there are special hacks for the animated objects and the ball itself.

 

So nothing that could be easily faked or done in VP right now, but requires really a highly specialized engine for exactly this usecase and a lot of education for the guys building the table PLUS a lot of preprocessing time PLUS the table size becomes gigantic.

Edited by toxie, 29 November 2016 - 10:42 PM.

[Ben Logan]

Great answer. Thanks, Toxie. It's amazing how great VPX tables are looking in our current engine. Kudos to you, fuzzel, and our authors.

[toxie]

When we did VPX we really thought a lot about what makes sense to do with new features. And thats why we decided to still go for example with the fake lighting and keep as much preprocessing as possible.

Otherwise the hardware requirements would have been even more harsh than it already was the case with going from VP9.9.X vs VPX (e.g. a lot of users would have been out with their existing hardware). Plus some authors would have been pissed because of the missing artistic freedom with the lighting.

 

But nowadays i think that we could really go to the next level and have some more dynamic elements, e.g. full lighting and full camera movement. At least its not impossible to do, given the right hardware.

[nFozzy]

I avoid always-on timers whenever I can, but still one of the most common complaints I get is stutter related to script timers. And it kills me, because people mess with the timer intervals and that wrecks the timing for everything.

 

Highest on my wishlist is code fading for Flasher objects. I know Lampstates have to be handled by a timer, I'm not under any delusions that JP's light routine is slower than uselights=1 or whatever, but the fact that the timer has to handle fading as well just seems outdated to me.

 

Also now that we're in the age of modulated solenoids, it'd be great if there was a light state that could fade both up and down dynamically based on an input value. I've made a script that does this, but the timer has to be really fast or a lot of detail gets lost.