Register    Login    Forum    FAQ    PinSimDB.org

Board index » Future Pinball » BAM corner




Post new topic Reply to topic  [ 29 posts ]  Go to page 1, 2, 3  Next
Author Message
 Post subject: the quest for the most accurate physics
 Post Posted: Wed Nov 20, 2013 8:18 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
Sorry for my poor english


IN the real life:
===================
ball: Standard carbon steel 1-1/16 inch (27mm)diameter, 80 gram
recommended pinball table slope is 6.5 to 7.0 degrees.
below 6.5 the game is easy and boring. up to 7.0 the game is more chalenging. More than 7 is a money rip off.





IN FUTURE pinball
====================

future pinbal use Newton Game Dynamics library.
An Open-source zlib-licensed integrated solution for real time simulation of physics environments.



Newton engine don't use SI units. Is up to the user to do the some conversion.
The International System of Units (abbreviated SI from French: Le Système international d'unités) is the modern form of the metric system and is the world's most widely used system of measurement, used in both everyday commerce and science.


Unit name Unit symbol Quantity name
metre m length
kilogram kg mass
second s time

gravity : 9.810 m/s2


My goal is to try to do the most accurate physics with using the most real world values.
In the following i assume that we use a table slope of 6.5 degree.


THE BALL
=========
First let's discover if future pinbal use SI units.
So let's try <ball mass="0.80" gravity="9.810" damping="0.80"></ball>
what's going on? it's like playing pinball in the international space station. The ball is floating !!!

Maybe the author of future pinbal doesn't use kg and m units?
let's try to change the mass value to gram and length to millimeter. That's could make sense as future pinball editor use millimeters for placing objects.


let try <ball mass="80.0" gravity="9810.0" damping="0.80"></ball>
Hmmmmmm much better no ?


You may find that the flipper may move when hitted by the ball. To correct that you must rise the flipper mass value.
Why? Because immovables components (that must not move when collided by a ball like plunger, flipper, diverter and emkicker) must have an enormous mass value compared to the ball mass.
So we could try to use this values:

<flipper mass="99999.0" omega="40.0" moeMethod="0"
leftXoff="0" leftYoff="1500" leftZoff="0"
rightXoff="0" rightYoff="1500" rightZoff="0"></flipper>
<autoplunger mass="99999.0" force="60000.0"></autoplunger>
<diverter mass="99999.0" omega="33.0"></diverter>
<plunger mass="99999.0" force="60000.0"></plunger>
<spindisk mass="99999.0" angularDamp="0.33" linearDamp="0.25"></spindisk>
<emkicker mass="99999.0" omega="100.0"></emkicker>


THE SPINNER
===========

Spinner mass has no effect on the ball. Yes it's weird!
It's just used in the spin effect only. To try to reset the spinner to its original vertical position.
We could think it's a good idea to put a low value like 10 grams.
But a low mass value will raise the risk to have a spinner in a stuck position.
I think because the spinner mass model is badly designed
That's why in some table like back to the future or star trek, the spinner was replaced by a toy and the spinning is simulated with 2 triggers (one before and one after the fake spinner).
It was done to be sure to reset the spinner to a vertical position.


So we couldn't use real world values for the spinners, that a fact!
After some trial an error the following seems to be ok:
<spinner mass="99999.0" gravity="9810.0" angularDamp="0.5" angularAccel="5.25"
spinDampLoose="0.30" spinBackLoose="0.65"
spinDampNorm="0.40" spinBackNorm="0.65"
spinDampTight="0.70" spinBackTight="0.65"></spinner>




Your ball may sometime be stopped by a spinner.
It's normal if the spinner has a specific angle with the ball at the imapct point.

......###..................impact point with a spinner angle that could stop the ball
............###..........|
spinner.-->.O.........|
..axe...........###....\/..+++++
......................###.+..........+
...........................+....ball.....+
............................+...........+
..............................++++++

One last thing, the suggested spinner height is 28 millimeters. Less is bug, more and the spinner may not spin!
Check the top height of the used surface for spinners!

THE GATE
========


Some weird things happens when sometime the ball couldn't pass the gate even in the good direction.
Unlike spinners, the gate mass affect the ball. So we put a low value compared to ball mass

After some trials and error i found this values:
<gate mass="0.50" gravity="3000.0" damping="1.0"></gate>




THE MATERIALS
==============

THE BIG PART!!!!!
there is two thing to deal with:
REBOUND and FRICTION

all material except rubber have 1.0 in softnessCoef (very hard).
a softnessCoef lower than 0.3 is for jelly...


REBOUND aka coefficient of restitution
=======================================
ressources on internet
http://hypertextbook.com/facts/2006/restitution.shtml
http://en.allexperts.com/q/Rubber-2470/bounce.htm
In real life, a steel ball could perfectly rebound when hitting a very hard surface.

as you see nobody is agreed on values. we have to take them as starting values and do some fine tunning....


FRICTION
=========
ressources on internet
http://www.tribology-abc.com/abc/cof.htm
http://www.engineeringtoolbox.com/frict ... d_778.html

Again as you see nobody is agreed on values. so again we have to take them as starting values and do some fine tunning....

in future pinball, all friction value are for a ball (a steel ball) on:
wood
plastic
rubber (default rubber material use the rubberIntMat value)
metal (steel)

A ball can slide on a playfield (very well if waxed). But also ROLLS!!!
An a spinning ball can "curve" it's movement.

So we have to find out the good friction value for having a rolling ball without changing a steel ball to a very grippy rubber ball!!!
Or changing the playfield to an ice field....


after many tries here some good values

<!-- Default Materal Settings to act as defaults incase we forget to set an object during development.
There shouldn't be any objects set to this material so these are just a fail safe values -->
<defaultMat softnessCoef="1.0" elasticCoef="0.1" staticFriction="0.92" kineticFriction="0.92"></defaultMat>

<!-- Material Settings for surfaces marked as playfields -->
<playfieldMat softnessCoef="1.0" elasticCoef="0.25" staticFriction="0.02" kineticFriction="0.02"></playfieldMat>

<!-- Primary Material Types -->
<metalMat softnessCoef="1.0" elasticCoef="0.20" staticFriction="0.02" kineticFriction="0.01"></metalMat>
<plasticMat softnessCoef="1.0" elasticCoef="0.20" staticFriction="0.02" kineticFriction="0.01"></plasticMat>
<woodMat softnessCoef="1.0" elasticCoef="0.20" staticFriction="0.02" kineticFriction="0.01"></woodMat>
<rubberHardMat softnessCoef="0.90" elasticCoef="0.42" staticFriction="0.13" kineticFriction="0.10"></rubberHardMat>
<rubberIntMat softnessCoef="0.90" elasticCoef="0.47" staticFriction="0.13" kineticFriction="0.10"></rubberIntMat>
<rubberSoftMat softnessCoef="0.90" elasticCoef="0.55" staticFriction="0.13" kineticFriction="0.10"></rubberSoftMat>

<!-- Object Specific Material Settings -->
<gateMat softnessCoef="1.0" elasticCoef="0.24" staticFriction="0.00001" kineticFriction="0.00001"></gateMat>
<kickerMat softnessCoef="1.0" elasticCoef="0.24" staticFriction="1.0" kineticFriction="0.80"></kickerMat>
<rampMat softnessCoef="1.0" elasticCoef="0.05" staticFriction="0.02" kineticFriction="0.01"></rampMat>
<plungerMat softnessCoef="1.0" elasticCoef="0.55" staticFriction="0.15" kineticFriction="0.10"></plungerMat>
<spindiskMat softnessCoef="1.0" elasticCoef="0.50" staticFriction="1.50" kineticFriction="2.00"></spindiskMat>

that's all for now
I'll give a link to zip which contains some basic tables to test rebound, friction ans spinners with a full physics XML.
For know i need some rest ....


Last edited by smoke on Thu Nov 21, 2013 2:52 am, edited 5 times in total.

Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Wed Nov 20, 2013 8:22 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
one more thing i have already do some test on table like LOTR,BTTF,IRONMAN and the result is very good if setting the table slope between 6,5->7.0. and maybe changing the flipper strength.

LOTR need some tweacking on the right wirered ramp because of a collision bug with a ball flying protection on the ramp that lead to the sword.


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Wed Nov 20, 2013 8:41 pm 
Offline

Joined: Fri Nov 08, 2013 12:13 am
Posts: 58
The physics system is all interconnected. You change one value and it directly affects most other values, sometimes radically so. Alter primary material elasticity coefficients and all of a sudden your gravity, ball weight, flipper strength, etc, and all wrong. Even a change of 0.1 can sometimes drastically alter the balance you're trying to find. You have to take into account literally every single thing the ball touches during play, including all surfaces.

Also, this balance you're looking for can still be achieved with almost any table slope by altering gravity, ball weight, etc, within the xml. For example, in 2.7 I've taken to low (4~) table slope with high gravity and ball weight. This gives several advantages, the most obvious of which is an immediate decrease in unwanted vertical motion: the ball stops leaving the playing field during extreme power moments or glitches. Of course that's a simplification; there's actually several other additional factors that influence these values and the end result, and these unorthodox settings have to be counteracted with other variables to achieve balance.

I'm going to write a detailed post about FP's Newton limitations and the necessary compromises later, but I just don't have time right now. Play Taxi 2.7 or Centaur 2.7 to see what I'm talking about. If you use the editor to look at what I'm doing (and contrast it directly with the xml) you might start to figure it out, but I doubt it. You have to fundamentally understand how Newton's FP settings work together before you can improve the physics of FP or even work out what's going wrong. Otherwise you're just stabbing in the dark.


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Sun Nov 24, 2013 2:47 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
I've done some more adjustement to the values after playing for real a spiderman pinball.
Essentially to the rubber elasticCoef.
And i'm very happy of the result.

I'will soon post the XML physics file with the updated values.
i have really good result by only changing the table slope to a more real value (6,1 degrees), lowering the flipper power. And adjusting the plunger power.
And i feel that the flipper aiming/targeting is more accurate without changing the flipper model.

So far i've verified the following tables with my physics :

lord of the ring (some bug in the script and some collision glitch with the right ramp that explain why sometime the ball stucks). i've done some fix to my version but i have to contact the author.

jhonny mnemonic
jurassic park
star war empire dition
ironman
spiderman

and more to come

i will post soon the slope+flipper power setting for each pinball tables


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Thu Nov 28, 2013 4:56 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
just to say i'm still testing my tables.
the ball move with a natural ways (no weird hyper fast lightning ball effect)
I have good result so far for:

AddamsFamily
atack from mars
aliens
appollo13
back to the future (my modded slamtilt version)
big bang bar
black night 2k
f14 tomcat dmd edition
jurassic park
indianna jones
ironman
jonny mnemonic
lord of the ring (heavy modifications done, the table has many flaws)
master of univers
spiderman (my reference table because i can play it for real every day after lunch)
star wars imperial edition
tron legacy (with playfield change when in game battle)


As reminder the philosophy of my physics is to be as close as possible to real world values (gravity, table slope, ball mass). And these values are the same for every table.
Except for 4 tables all share the same value for flipper power, plunger power, slingshot power.

Because in the real life, a ball on spiderman react the same as a ball on demolition man, ironman, black knight 2k ....



i'm keeping validating my physics. You will get the file and instructions soon.


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Sat Nov 30, 2013 7:50 am 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
Here a link to my physic for future pinball :

http://www.filedropper.com/testphys



You must have BAM (ravarcade you are a genius for this tool !!!) installed.
In this archive you will have the default.xml file that contains the modified physics.
Just copy the default.xml file into the BAM folder.
make sure that your tables haven't a special xml file with them or you will not load my physics.


The first time you will have some difficuties to perform ramps.
It's normal because you have taken the bad habit with the over future pinball physics that you could perform ramps easily with a no velocity ball.
Now, like in real life, you will have to have some velocity to perform the ramps.


you will notice some test table in my archive:

perfect_plunguer_guide
----------------------
show how to make a perfect plunguer guide for the ball (they are painted red and yellow you can't miss them).
These guides resolve a nasty thing when some time the ball have low speed or lightning speed with the same power applied to the plunguer.
The reasons is that the ball is not perfectly aligned with the plunguer, so the plunguer doesn't hit the ball on the center but a little off.
If you played snooker game you see what I mean.


phys3
----------------------
only show how the spinner react this my physics with the 3 possible setting (a counter show how many rotation are performed for each spinner).
You will see that the stuck spinner position is a thing of the past.
You may think that sometime the spinners seems to be stuck but wait 1-2 seconds an you will see that it slowy return to a vertical position.
That the best I can do because other spinBack values on the physics file could result in an infinite balancing motion of the spinner (like there is no friction applied on the rotating axe)


flipper & plunger & bumper strength
======================================
0=minimum=cursor to all left
8=maximum=cursor to all right

general values for all tables (special setting on some table explained at the end)
====================================================================================

rubber for slingshot & flipper is: soft
table slope : 6.1 degrees
flipper power is: 4
slingshot power is: 6
manual plunger power : 1 with ball guide or 7 without
auto plunger power : 7

recommended (for best experience but you can pass it) bumper power:
===================================================================

bumpers in triangular pattern
--------------------
......5.....5........5..5........5..5
.........8.............8..........8


bumpers in diamond pattern
---------------------------
.........5......
......5.....5...
.........8......



special TABLES settings
=================

B2k
=====================
left flipper =4
right flipper =5
upper right flipper =5
autoplunguer =3


f14 tomcat
=====================
the vertical kicker TopRightKicker strength need to be set to 8


masters of universe
=====================
upper left flipper =4
left flipper =4
right flipper =5

medieval madness
=====================
left flipper =5 or 6 ( the right ramp has some design flaws, i've modifed my table version and i could use a strength of 4)
right flipper =4

the vertical kicker named "catapkicker" strength need to be lowered to 6 or 7.
Don't try to set to 8 because the ball will have to much velocity and hit with too much force the invisible wired ramp and will go back to the kicker.


reel it in
=====================
you need to move away of the table the invisible surface named "slower1" at the end of the fish ramp. If you choose to delete it, don't forget to remove all reference to "slower1" in the script.
left flipper =4 I recommend to set the swing to -55
right flipper =4 I recommend to set the swing to 55

simpson pinball party
=====================
left flipper =4
right flipper =4
the 2 upper left flippers =4
upper right flipper =4

the vertical kicker named "Kicker4" strength need to be lowered to 5.
Don't try to set to 7 or 8 because the ball will have to much velocity and hit with too much force the scoop above and will go back to the kicker.


spiderman
=====================
left flipper =4
right flipper =4
upper right flipper =6 I recommend to set the swing to 48

star trek
=====================
left flipper =4
right flipper =4
upper right flipper =5 I recommend to set the swing to 40
set the left kickback autoplunger strength to 3 or 4 or the ball will go lightning speed



more special settings to come if necessary


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Mon Dec 02, 2013 6:09 am 
Offline

Joined: Sat Jan 07, 2012 1:14 pm
Posts: 180
DAMMM!!!!!

this is great news, tnx for the explanation of some things, cause i didn't knew how to change them and now i do :) :) :)

also tnx for the file i have tested the file with some tables and soem of them plays much better..

Hope that there is some one that can make a program of file that alle tables are working as they should be in real life ;)

Keep up the good work smoke, love it..

CHEERS!!


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Mon Dec 02, 2013 2:33 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
thanks that you find this new physic more enjoyable.

I found that some tables require a table slope of 5.8 instead of 6.1.
This way you can keep the other parameters to the recommended one and it will be a little more easier.

The following table are good with a slope of 5.8 (and set the other parameters to the recommended ones):

reel it in
3 angels
TutenKham
dead hunter (even if I personnally let it to 6.1 for more challenge)


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Mon Dec 02, 2013 2:35 pm 
Offline

Joined: Fri Jun 28, 2013 4:46 pm
Posts: 348
If you have some recommendations on my physics or if you are happy, please reply on this topic.
Don't be shy :p


Top 
 Profile  
 
 Post subject: Re: the quest for the most accurate physics
 Post Posted: Mon Dec 09, 2013 3:10 am 
Offline

Joined: Sat Jan 07, 2012 1:14 pm
Posts: 180
well i have played it on a few tables, but on some tables i think the ball is to bouncy on the bumpers, it shoot very fast over the table so fast i cant even see it ;)

so the i have changed some things in de xml but i have screwd it up and it doesn't work properly anymore ;)

also the flippers are not strong enough on some tables cause i cant hit the ramp, i have changed the strenght and then they work..

for now im playing with the standard and de zedonius and hoping that there will be some xml that plays for all tables as it should be..

cheers.


Top 
 Profile  
 
Display posts from previous:  Sort by  
 
Post new topic Reply to topic  [ 29 posts ]  Go to page 1, 2, 3  Next

Board index » Future Pinball » BAM corner


Who is online

Users browsing this forum: No registered users and 6 guests

 
 

 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to: