27 replies to this topic

[Csilk]

it doesn't take to much messing around to see that the ball bounces as it rolls down an inlane and across the flipper when elasticity is set to a value that makes deadflipper passes possible and accurate, change the elasticity to lower and then the ball stops bouncing as it crosses the threshold of the inlane to the flipper and has correct behavior but then deadflips don't look right.

 

I would love to write a script that handles when a ball comes from the inlane versus down the middle but i'm unsure if it would work, what if the ball is rolling down the inlane when a deadflipper bounce happens?

 

what is everyones thoughts on this? wouldn't it be nice if the behavior was correct in both instances?

[Csilk]

was able to write a simple script that changes the elasticity when a trigger is rolled through... I made it as small as possible so hopefully the chances of it being triggered when a ball is falling in multi ball is lower.

 

but a hack workaround like this highlights why a standard elasticity setting is meh, by changing the flipper elasticity for the one ball at the time of triggering it also will change it for the other balls.

https://www.twitch.t...deos/1628306469

 

shows the correction versus the non-corrected behavior (left is corrected, right lane is not)

Edited by Csilk, 19 October 2022 - 06:35 AM.

[Rajo Joey]

Have you tested it with nFozzy's flipper physics?

 

[Edit]

Forget the question. Just saw the other thread.

Edited by Rajo Joey, 19 October 2022 - 09:01 AM.

[wiesshund]

 

what is everyones thoughts on this? wouldn't it be nice if the behavior was correct in both instances?

 

Have not really had a problem with it.

I generally get more bounce from a ball plowing down the center for the simple fact
it deflects the flipper itself, which springs back adding bounce
which the ball coming from the inlane does not do.

mind you i usually wind up tossing out the physics on a table when i get it and replace them

[Csilk]

yeah but to get the proper inlane rolling behavior i must use 0.4 elasticity, no ball comming down the middle is going to bounce enough with 0.4 elasticity and get the accurate bounce.

 

its as I said if you want inlane rolls to be correct you must use a script hack, otherwise you wont get the proper inlane roll on many tables across the flipper, it only takes looking at table footage and watching how the ball goes across the flipper region on said table to see something isn't right, the ball shouldn't be bouncing up as it crosses the threshold of the inlane and the flipper. this is i'm assuming as having the correct slope and ball speed that the real table would have.

 

the script hack works fairly well but it would be nice if the flipper elasticity wasn't being applied to the immediate base of the flipper

Edited by Csilk, 19 October 2022 - 11:48 PM.

[wiesshund]

Odd
I have not run into anything like that, and .4 elasticity is 
well it might as well be lead.

 

not sure what physics you are using though

[Csilk]

what do you mean physics set? flipper settings?

 

physics set has nothing to do with what I'm talking about (look at the video) thats using 0.85 elasticity 0.2 falloff. pretty much nFozzy recommended range. VP just doesn't do the inlane behavior correctly with appropriate flipper elasticity, it will plague all tables or at least ones with a decent run off from the inlane. It doesn't matter to much because my script bandaid works to correct it but it would just be better if visual pinball handled it correctly.

Edited by Csilk, 20 October 2022 - 01:51 AM.

[wiesshund]

I did watch the video

 

but i dont have that issue, at all.

 

look at ltd alien warrior for example, you wont see that happening at all
and i dont use nfozzy's stuff, or any switches to change values
and elasticity is 0.85 with a falloff of 1.2

[Csilk]

I'll take a look at it, its possible its a problem with the inlane shaping / angle of some tables too.

[Csilk]

yeah I notice that it does bounce slightly but not significantly but maybe it has to do with the ball not gaining enough velocity before hitting the area. if you wouldn't mind looking at my table mod and seeing if there is a problem with the section where it leads to the flippers I would be greatly appreciative.

[wiesshund]

It should not float over the flipper like that regardless

in your video, it is just like floating over it

 

that aint right

 

but it isnt really a failing of VPX causing it, more of a lack of implementation i think?

 

feel free to pull apart alien warrior, or haunted hotel
both are recent, and both use the same materials (different ones for physics / visual)

both are based of JP's method, and use minimal script hoodoo.

Actually, the only simple script hoodoo is to manage ball max speed cause you can get a ball moving unrealistically fast in vpx
if you just let it go, and mad playfield friction isnt the right answer.

[Csilk]

I'm basing the behavior off of real tournament play, it does slide across the flipper with minimal bounce (video is a bit lower than what I settled on for the behavior I wanted) on a inlane return.

Edited by Csilk, 20 October 2022 - 03:54 AM.

[wiesshund]

yeah I notice that it does bounce slightly but not significantly but maybe it has to do with the ball not gaining enough velocity before hitting the area. if you wouldn't mind looking at my table mod and seeing if there is a problem with the section where it leads to the flippers I would be greatly appreciative.

 

I could take a look at it but

1st thing i would have to do is strip it of any script driven physics
because those can vary widely from one machine to the next and vary with FPS also

 

If i tinker with script operated physics
and i do it on my desktop, and get what seems good, at 376FPS

and then give it to someone to run on a system that get 60 to 120 fps
it isnt going to work right for them, they wont get the same thing that i had and they cant just nudge something and fix it.

If i do the physics within VPX, then they could have a slow machine cranking out 32fps
and while they wont get great visuals, they will get the same exact physics as the machine that was running 376fps
up to the point where their machine struggles so much that vpx is basically doing a lot of pausing, waiting for cpu time and then doing fast forward catchup
but the only fix for that is get another pc of course.

 

But i can look at it if you like, as long as you realize that you would be getting back a different table



 

[Csilk]

if the solution is to not use any script driven physics then I'll pass, it could be something to do with flipper polarity but i can check version 1.0 before nFozzy / VPW and stuff was added and i'm pretty sure it has the same hop. default vpx flippers are poop, nFozzy flippers are far superior and while I do agree that solutions should try to avoid using complex scripts when possible for me its more about making something as authentic as possible to my reference footage with whatever means necessary.

 

I just checked and the bounce is still there just not as pronounced because the table slope is like 4 on version 1.... the newest version of my table is night and day difference of look and feel, i don't think i could handle the slo mo of version 1 anymore.

Edited by Csilk, 20 October 2022 - 04:15 AM.

[bigus1]

I had 4 Firepower machines across decades and twice I had 2 at once. I was able to have a game on one and then another game on the one next to it and I can tell you they played differently. Pinball machines are not a precision instrument. "Furry" flipper tips make a big difference, for example, and millimetre differences here and there combine to alter the gameplay slightly. Maintenance matters too. So I wouldn't get too hung-up trying to precisely emulate a single example of a table from a video, because it may not match the next guys' table anyway. Things get bent, twisted, worn, warped, filthy, rusty, wobbly legs - you get the picture. As a player, you need to adapt to the machine at hand. So, provided the physics are relatively life-like in general, you should be able to make a fist of it and have some fun.  It's all relative anyway - if you get a whopping score on a particular version, it just means you and anyone else have to get a more whopping score to beat it. So long as it's fun, it doesn't really matter.

[Thalamus]

Csilk, I told you NOT to use high values for table difficulty - believe I saw 50 in bsd. Reason for saying so is this : https://github.com/v...lipper.cpp#L279

I don't read c++ well yet, so, I'm not sure exactly what this does, but, I do know that I don't like to see ANYTHING named table difficultly related to the flippers. It just doesn't sound right to me.

 

Maybe someone that do read c++ easy as a book, could explain exactly what this does.

Edited by Thalamus, 20 October 2022 - 04:28 AM.

[Csilk]

https://youtu.be/cR5mvi_edM4?t=1879

 

watch how the ball goes across the flipper.

Csilk, I told you NOT to use high values for table difficulty - believe I saw 50 in bsd. Reason for saying so is this : https://github.com/v...lipper.cpp#L279

I don't read c++ well yet, so, I'm not sure exactly what this does, but, I do know that I don't like to see ANYTHING named table difficultly related to the flippers. It just doesn't sound right to me.

 

Maybe someone that do read c++ easy as a book, could explain exactly what this does.

 

I'm glad you mentioned it as I thought 50 was just the default setting which would select the table slope 50 percent between the minimum table slope and max slope. If 50 is not the default I will set to 20 (which appears to be the default) and work off of that then.

[wiesshund]

if the solution is to not use any script driven physics then I'll pass

 

 

i don't think i could handle the slo mo of version 1 anymore.

 

Have you ever played Heavy Metal Rowament?

In the default "Heavy Metal" way i set up the table, the table plays ungoddly fast, and violent
when i say ungoddly, i mean way unrealistically, the ball would be beating a real pin to a pile of parts.

(the script does have an option to play in actual planet earth physics
No magic hoodoo, it just changes values at load up, and swaps in some other parts that are normally under the table)

It still has none of the ball float \ hopping that you showed in the video.

 

And for what it is worth, i dont think nfozzzy physics would do that either if set up right
but i dont use them because it is too much niggling to work out, and then they are machine dependent.

 

But i am not sure why you say 

 

 

if the solution is to not use any script driven physics then I'll pass

Tell me a single table JP has made, that plays like shit?
And JP uses no script physics, he works only with what VPX has inside it, that will play the same on an old pc or a new fast one.
 

[Csilk]

considering I'm modding a JPsalas table and in my opinion making it play better... I'm not going to lie and say JPsalas doesn't make great tables that play well using low friction environment or that he hasn't probably advanced past an old dracula vp9x table but I'm sure even JPsalas would chime in and explain that there are aspects that each method of table creation (low friction, high friction, script driven, vpx only flippers) all have their own drawbacks, pros and cons.

 

I also don't really think its a matter of setting nFozzy up incorrectly either, and you clearly haven't looked under the hood of the table if you believe its using a lot of script based physics. It has stuff that does sling correction (which can be disabled) and checks ball speed for the same reasons you mentioned, it dampens rubbers but dampening would not cause a rebound as the ball flows over the top if anything it would remove rebounding.

 

that leaves flipper tricks and flipper polarity which are governed by scripts but I can show you the behavior using stock flippers. if you can show me a way to have the ball flow over like the video i linked of PAPApinball while maintaining the correct inlane slope and using elasticity above .8 then by all means I'm all ears (eyes)

[wiesshund]

Oh how JP does tables now it entirely different from his vp9 tables
it is entire different from lets say Jan 2021 tables even.

It is why he is redoing and updating all of his old tables now

 

What table is it in your video whel the ball is being understandably annoying and launching right over the flipper
(sorry vid is small, i cant read it)

 

Have a link to that table?