Physics

[etamme]

First things first - I love PBA, and I love that I can play real tables on my Nexus 10.

I have one "major" complaint with PBA, physics. I joined a pinball league at my local pinball arcade and practiced Theater of Magic on PBA b/c I had never played the table before - all great, learned the table and set the second highest score on the machine when I played it at the arcade. That said the physics in PBA are nothing close to resembling reality.

Three physics issues that are most prevalent to me are the following

Float: The ball seems to behave as if it were a really light weight plastic ball it seems to lack significant momentum/mass.

Spring/bounce: This one is insane. You can't even catch the ball in PBA. The flippers must be made of flubber (that hyper springy material that comedic mad scientest created).

Weight: this may be the single element that affects the other two - I dont know. The ball just moves too freely, it just seems to not have the mass to be controlled.

Ok - so I know that one of the joys of pinball is that every table is different. I want to be clear that I am not trying to say that PBA should model the exact TOM table that I have locally - but the physics are generally so so so far off that I think it is a more generic issue. Also - I want to be clear that I am not just complaining here, I am trying to provide constructive feed back.

Does any one have thoughts or comments on what I have said?

-Eric

 

[Fungi]

I agree with everything you say. Here's the kicker. ToM is one of the first tables FarSight created for TPA, therefore it's physics engine is among the earliest released. Compare that to the latest, being Attack from Mars, and you'll notice a remarkable improvement. Now, if FarSight would retrofit the original tables with the AFM physics, that would be incredible.

 

[Nik Barbour]

I'd agree too, I used to play more real pinball 20yrs ago, then didn't really play much till TPA released 1yr ago. I did a fair bit of virtual practice and got to the point where I thought I was ok-ish. Then feeling confident I went to a local expo with about 70 real pins.
It was then I realised how unrealistic TPA was - this was in October last year.
Initially I was fairly disappointed, and a little angry, and switched to playing Visual Pinball which is a lot more realistic, but of varying quality (not that I'm complaining as I couldn't create what these people have - thank you VP creators for all your work and kindness sharing).
I do however prefer the polish TPA has, and I think more recently with STTNG & AFM the difficulty tuning and physics have been getting far closer to the real tables.
I'm enjoying TPA again, I nearly walked away in November, but it is improving. I hope the earlier tables will eventually be caught up with the latest.

 

[Sean DonCarlos]

Having access to the physical versions of 16 of the 22 released tables, I'd say the physics are "98% accurate". There are some general inaccuracies, like the excessive bounciness you noticed; some limitations, especially with fine flipper control; and some table-specific trouble spots. But most of the time, I can play a game of the real table and a game of the TPA version and nothing comes up that makes me go "why the @#$%! did it do that?".

That being said, the later tables are much improved physics-wise compared to the original 4 tables. As FarSight releases more tables, they've learned how to better handle the physics (and the lighting!) and so each new table pack is a little better than what came before.

They do occasionally go back and revisit the older tables as they have time - Black Hole got a major improvement, and Medieval Madness had its ramp acceleration issues squelched and its bounciness toned down recently.

 

[Day]

I noticed only today that whilst the physics and the feel of AFM feels great, maybe CFTBL has suffered. with this latter table and with release PA ver 1.8.0 I could use a dead flipper strategy to reliably decelerate a ball emerging at very high speed from a loop, and know it would reliably pass across to other flipper. it's not behaving like this any more. A ball emerging from a loop at speed hits the flipper, all momentum is killed completely - no bounce at all - and the ball shoots down the inactive flipper to a centre drain. I'm also finding ball passes using a flip from cradle to the slingshot above to get the ball onto other flipper more tricky now I'm on 1.10.0

I personally like flippers/rubbers with some bounce

Sent from my Nexus 7 using Tapatalk HD

 

[etamme]

Having access to the physical versions of 16 of the 22 released tables, I'd say the physics are "98% accurate".
...
...
They do occasionally go back and revisit the older tables as they have time - Black Hole got a major improvement, and Medieval Madness had its ramp acceleration issues squelched and its bounciness toned down recently.

Yeah I have access to a fair number - not 16 .. probably 5-10. Either the rubber is ancient on the TOM table I played, or the flippers are excessivly bouncey in the TPA version. I have access to black hole as well - ill start playing that on TOM. I think i have mars attacks as well - ill have to check. Any way I am glad to hear they updated the physics, but I am confused why they dont run the same physics engine on all tables?

 

[Bonzo]

It's the same engine. But optimizing it is done on a per table basis. Hence all the experience gained by FS accumulates in the best physics on the newest releases.

 

[ER777]

While I agree that on some tables the ball can feel just a little bit on the light and floaty side, I also agree that overall its pretty darn close to the real thing. I can't complain.

 

[Jeff Strong]

They're definitely being improved all the time. Thumbs up to Bobby and Mike.

 

[Kevlar]

I think it's pretty close. The only thing that stands out to me is the lack of ball spin. I've seen some freakish ball behaviour on my table at times due to spin, and when I bought it TPA was my only prior reference so I was quite shocked to see the ball deviating so wildly. I've seen the ball take a little bounce off a flipper then spin all the way up and over a slingshot and escape down the outlane

 

[Pinballfan69]

I find that bride and funhouse seemed pretty close. The tables that annoy me with the 'floatyness' is MM, MB, TOTAN, CV. Admittedly enough I thought Zen was pretty good Physics wise but the balls had too much weight and not enough bounce. To me TPA need more weight to the ball and not so bouncy. I know it would probably be tough to really get it right but the recent tables feel like such an improvement. Like many I hope that they go back and tweak their earlier tables. T'would be ashamed not to play those tables again with the new version of the physics of the recent tables.

 

[Nik Barbour]

I find that bride and funhouse seemed pretty close.

I found Funhouse to be one of the worse ones. I could play it on Xbox to the point where 1 game would go over the 1 hour mark, and I'd let the balls drain deliberately.

The real table not only didn't feel or play the same, but.... Back on the difficulty issue.... After a few warm up games, the best I could manage, really working, was about 8mins.

 

[Pinballfan69]

I found Funhouse to be one of the worse ones. I could play it on Xbox to the point where 1 game would go over the 1 hour mark, and I'd let the balls drain deliberately.

The real table not only didn't feel or play the same, but.... Back on the difficulty issue.... After a few warm up games, the best I could manage, really working, was about 8mins.

Well that may be true about the length of time on Funhouse in TPA but I think the ball acts with the correct amount of weight as opposed to other floaty tables. My opinion though.

 

[chemmerling]

I've reported the bug in a different forum, but I have found that the physics are definitely changed by the viewing angle of the tables. You can see it on Android with ST:TNG, the ball launch from Medieval Madness.

The physics are definitely improving. I've seen it already in a couple of updated tables. I do wish, however, that they were consistent among camera angles.

 

[Sumez]

I've reported the bug in a different forum, but I have found that the physics are definitely changed by the viewing angle of the tables. You can see it on Android with ST:TNG, the ball launch from Medieval Madness.

I find it seriously difficult to believe that a bug like that would exist. Why would they even bother programming their physics code to take the viewing angle into consideration? The table physics and the camera are two entirely separate entities, and I've never met a 3D game developer who doesn't follow that basic philosophy.

 

[Willi Cicci]

I find it seriously difficult to believe that a bug like that would exist. Why would they even bother programming their physics code to take the viewing angle into consideration? The table physics and the camera are two entirely separate entities, and I've never met a 3D game developer who doesn't follow that basic philosophy.

I have to agree with Sumez. I don't know anything about programming, but I've played a lot of Star Trek and I've never noticed anything even remotely close to what you're describing, chemmerling. I've noticed sometimes it seems like different camera angles give the game a little different 'feel,' and it might take a few minutes to learn where to hit the ball on the flippers to make the shots you're already accustomed to hitting on another camera angle, but I definitely wouldn't say that it feels like the physics have actually been changed.

Perhaps we're not understanding exactly what you mean. Can you provide any more clarification? (I've read your bug report also and I'm still not sure what you mean.)

 

[epthegeek]

Being able to control the ball on the flippers is a big sticking point for me. I don't know if it's true for every table, but you can't back-angle back-hand by letting a ball roll part way down a flipper, and quickly dropping the flipper and hitting it again. It always goes off as if you'd hit it normally. This bugs me just beacuse it's so unrealistic. It's mostly impossible to trap/stop balls too.

Oh, and the ramp shots on TOM that have no business making it that seem to just drive themselves around ... WAT?

 

[netizen]

Backhand shots are possible on Android.

TOM is scripted and uses and older Physics engine so it's behaviour tends to be more on the sketchy, WTF, did I just see side of things in my experience.

As we get newer and more improved tables I personally find TOM to be one of the Tables that I loved to be more painful to go back and play again with any regularity.

 

[Sean DonCarlos]

TOM is scripted and uses and older Physics engine so it's behaviour tends to be more on the sketchy, WTF, did I just see side of things in my experience.

Theatre of Magic is fully emulated. The only scripted tables on TPA are Black Hole, Black Knight, Big Shot and Genie.

 

[Nik Barbour]

Theatre of Magic is fully emulated. The only scripted tables on TPA are Black Hole, Black Knight, Big Shot and Genie.

& Gorgar