This topic contains 82 replies, has 13 voices, and was last updated by  Clark Kent 5 days, 6 hours ago.

Currently, there are 0 users and 1 guest visiting this topic.
Viewing 20 posts - 1 through 20 (of 83 total)
  • Author
    Posts
  • #143437
     rothbauerw 
    Participant
    ContributorMembervip

    I’m moving the physics discussion to it’s own thread.  Here we’ll discuss changes that can be made to the nFozzy physics that are used on the new Tales of the Arabian Nights table.  The goal will be to improve the physics set that can be used in an update and on other tables.

    There are three items I’d like to explore:

    • flipper tricks
    • playfield friction
    • ball mass

    I’d like to try to tackle these one at a time.  We’ll focus on flipper tricks first.

    @clarkkent

    Thanks for the video link to the flipper tricks.  I’d like to request a video of you performing some of these flipper tricks on a real machine.  Here’s why I ask.  I’d like to gauge the difficulty level between a real pin and your vp settings.  Your vp settings feel much too easy to me as I can easily perform all these tricks using a keyboard.  I’ve never been able to perform tap passes or stop a ball with a tap the way I can on your updated table.  I think you’re on the right track, but it shouldn’t be that easy to perform these tricks.  Ideally a video on the same era pinball would be best.

    A couple of other observations.  A coil ramp up of 3 takes 30 ms to reach full flipper strength.  On my cab, the flippers reach EOS in about 17-20 ms with no coil ramp up.  So a setting of 3 seems much too high.  In addition, without adjusting strength, it slow the EOS timing to about 27-30 ms.  You’ve adjusted for this by increasing the flipper strength.  But that results in unrealistic ball speeds on many shots.

    I think if we use a lower coil ramp up setting and lower flipper strength, we can enable flipper tricks without sacrificing game play.  Tricks will probably be more difficult, but I think they should be.

    I will try to post a couple of test tables for folks to play with in the next day or so.  One with Clark Kent’s flippers and one with nFozzy flippers.

    Current Project: Perpetual updates of VPX physics.

    1 user thanked author for this post.
    #143438
     Clark Kent 
    Participant
    MemberContributor

    That’s a good idea. My physics adjustments are not cut in stone, it’s in a constant evolution. You know, my settings didn’t come out of the blue, it was just progress – and it‘s still going on.

    I was never that happy that the flipper rubbers are not bouncy enough in my adjustment and backhands are a little bit too weak – but I haven’t found a way to combine both with sensitive ball control.

    Maybe a slightly lower ball mass than 1.7 (but I do not want to loose the steel ball feeling) with slightly lower strength and coil ramp up and relative to that the return strength. But there are so many different variables that it‘s almost impossible to tell which combination is the best, it’s more like luck combined with some experience. The biggest shame is that the physics mastermind mukuste left before finishing the physics…

    I‘ll do another video on my real machine but it will take a while, I have no time until next week.

    #143439
     Thalamus 
    Moderator
    ContributorMemberModerator

    I agree with all your observations Roth. When I was a toddler, I had a toy pinball where the speed of the keypress decided how fast the flipper would move. Too slow coil ramp up and it starts to feel like that. Like you say, CK’ got some valid points and I agree with him that currently, flipper skills is really suffering from what I else would call great physics. I remember especially two releases where the table became too easy to play. Bad Cats and Red and Ted’ Roadshow. So, I’ve kept both version. Have I been able to improve on it ? No, this is something I hope you two are able to fine tune further. I think you are going to find that is very hard. nFozzy must have spent a huge amount of time on the subject to come up with these settings, and I would be very surprised if he didn’t attempt to do something about the control of the flipper while the ball is cradled or returning from their upper position. Personally. I believe that if we are able to do a drop catch like the pros do, then settings needs to be damn close.

    #143442
     Clark Kent 
    Participant
    MemberContributor

    What I want to mention – and I think this is very important: I set the flipper strength that high because on a real pinball machine you can do backhands onto ramps. This is not possible with regular settings (at least I was never able to do so). On my Dialed In I can backhand on the left ramp (and it is a steep ramp) and on the right side I can backhand into the phone scoop (even if it’s a slightly negative angle, and even if the ball is rolling down the inlane – not even if I stop it before). Same on my other machine Roadshow. Left side backhand onto the ramp.

    #143448
     rothbauerw 
    Participant
    ContributorMembervip

    That’s one of the primary advantages of the nfozzy flippers. Back hands are more realistic without super high flipper strength.

    @thalamus

    I believe nFozzy was more concerned with shot profile than flipper tricks. I didn’t spend time on flipper tricks and knew it was an opportunity. I believe we can make the necessary adjustments for post passes, tap passes, and cradle separation. Not sure yet on drop catched or stalls. Those might be more difficult.

    Current Project: Perpetual updates of VPX physics.

    1 user thanked author for this post.
    #143456
     Clark Kent 
    Participant
    MemberContributor

    Hmmmm. That was one thing I didn’t find realistic. The backhands are THAT strong that I could do fast orbits with nFozzy’s physics. But maybe that was just luck.

    Nevertheless I forgot to mention another thing: I improved the flipper, slingshot and return lane angle/size/position in my adjusted version. As a pinball machine is a mathematical construction I think it is very important that the sides are mirrored, exactly on the same position and angle as the other side.

    screen-1

    • This reply was modified 1 month, 2 weeks ago by  Clark Kent.
    • This reply was modified 1 month, 2 weeks ago by  Clark Kent.
    #143459
     Drybonz 
    Participant
    Member

    Great to see these types of discussions going on… people are always saying that VP has the advantage over other pinball games with physics… so it’s nice to see this physics fine tuning.

    I always thought it would be cool if VPX could have a slider with some physics “presets” (EM, 70’s, 80’s Bally, Modern Stern, etc and whatever) so that the player could quickly switch between a different physics settings depending on what they were playing.

    #143462
     rothbauerw 
    Participant
    ContributorMembervip

    Hmmmm. That was one thing I didn’t find realistic. The backhands are THAT strong that I could do fast orbits with nFozzy’s physics. But maybe that was just luck.

    Nevertheless I forgot to mention another thing: I improved the flipper, slingshot and return lane angle/size/position in my adjusted version. As a pinball machine is a mathematical construction I think it is very important that the sides are mirrored, exactly on the same position and angle as the other side.

    screen-1

    • This reply was modified 1 month, 2 weeks ago by  Clark Kent.
    • This reply was modified 1 month, 2 weeks ago by  Clark Kent.

    The good news is the backhands are adjustable with the nFozzy physics.  What’s a fast orbit?

    I find video references are best to tune things.  All the shots I’ve observed are consistent with video evidence I have.  If you have video examples to the contrary, please share and we can make adjustments.

    Current Project: Perpetual updates of VPX physics.

    #143469
     The Loafer 
    Participant
    Member

    I appreciate this discussion but I also appreciate there is an understanding that the ability to do every type of shot does not equate automatically to success.  I like CK’s physics a lot but a few of those tables end up being too easy.  That doesn’t mean it’s all on the physics, maybe some table geometry are too open or a missed shot doesn’t cost the player enough like it would on a real table.

    Ie physics are important but it’s not all about the physics (not negating it’s importance here!)

    #143532
     rothbauerw 
    Participant
    ContributorMembervip

    Here’s the first test file.  These are clarkkent’s flippers and physics.  Basically, I copied the flippers, slings, and return lanes from TOTAN over to the example table.

    @clarkkent

    Please check to make sure this behaves as you’d expect.  I think I moved everything correctly and it feels right to me.

    I’m currently working on some adjustments to the nFozzy flipper physics.  Once I feel like it’s simulated flipper tricks well, I’ll post a similar table with those physics for comparison and feedback.

     

    Current Project: Perpetual updates of VPX physics.

    Attachments:
    1 user thanked author for this post.
    #143573
     Clark Kent 
    Participant
    MemberContributor

    Seems to be ok. Bumpers are too weak.

    #143583
     Thalamus 
    Moderator
    ContributorMemberModerator

    So. I did try these flippers for maybe 20 minutes. Surprised to see that I could actually from time to time accomplish drop catches. Not perfectly, and I’m no pro at it either. However. I still have the feeling that there is something wrong. But, I’m having hard time figuring out WHAT the problem is. It is just feels a bit to easy some way.  Is it that flippers absorb the ball too much ? Friction ? Sorry, I can’t see it, I just feel it. Looking forward to see if you guys are able to find something in between this and the nFozzy physic.

    #143592
     rothbauerw 
    Participant
    ContributorMembervip

    Seems to be ok. Bumpers are too weak.

    I didn’t modify any of the physics above the slings/flippers.  Playfield, ballmass, slings, flippers, return lane, and rubber posts in the sling/return lane area.

    Current Project: Perpetual updates of VPX physics.

    #143659
     chihua07 
    Participant
    Member

    Can we export and import the physics into totan 1.1 to see how it plays on the table itself, or is that not a good indication of the physics? Just wanted to find out if that is something that would be desired to evaluate the physics?

     

    #144100
     chihua07 
    Participant
    Member

    Any Nfozzy physics adjustments?  Just curious.

    #144107
     rothbauerw 
    Participant
    ContributorMembervip

    They are coming.  Randr has been helping me with measurements to dial-in the right behavior.

    Current Project: Perpetual updates of VPX physics.

    #144154
     Clark Kent 
    Participant
    MemberContributor

    The right approach would be playing on a real machine and feel the correct behavior. Just watching videos won’t do the job.

    #144156
     rothbauerw 
    Participant
    ContributorMembervip

    It’s a combination of both. The videos get us close to the correct behavior. Trajectories, velocities, etc. I have randr to test and provide feedback on feel. It would be much easier if I had regular access to real machine for sure.

    Current Project: Perpetual updates of VPX physics.

    #144291
     Clark Kent 
    Participant
    MemberContributor

    It is very important to feel the behavior at first hand. Just take ANY b/w DMD machine and play a while.

    #144387
     rothbauerw 
    Participant
    ContributorMembervip

    Here’s my first pass at adding flipper tricks to the nFozzy flippers.  Post passes, drop catches, and multi-ball separation all work.  Post passes and drop catches both require practice to perform consistently.

    Thanks to @randr for helping me dig into the flipper behavior.  I’m fairly happy with where they are at currently, but I would love feedback from the community.  I’m more than happy to tweak things if folks think it’s necessary.

    Here are a few findings in my research:

    • It typically takes between 25-30 ms for the flipper to reach a full flip angle once it begins to move.  Coil-ramp up impacts this timing.  I used 3, same as Clark.  It takes a tad longer than 30 ms with this setting.
    • It typically takes about 60 ms for the flipper to fall back to the start position from a full flip once it starts to move.  Return strength is the primary setting impacting this timing.  It was adjusted to provide a reasonable return time which is currently slightly higher than 60 ms.
    • End of stroke torque needs to be much lower than what most have been using.  Clark uses 0.22 on his physics.  I used 0.3.  With 0.3, you get a springy flipper when a ball hits it while in the up position.  To prevent this, I adjust end of stroke torque when the flipper is in the full up position.
    • We measured the end of stroke switch and found it kicks in somewhere between 4.5 and 6 degrees.  I used 6 to give a bit more control of the ball.
    • On the newer Bally/Williams, post passes need to be performed from the bottom side (see image below) of the sling post instead of the inside.  The flippers are too strong to control a pass from the inside of the sling post.  In addition, the flipper needs to fall 2/3’s to 3/4’s of the way down for a good pass.  This is consistent with the behavior of the current physics.
    • Other than some minor tweaks to the trajectory profile, the remainder of the nfozzy flipper settings remain unchanged.

    postpass-optimize

    Current Project: Perpetual updates of VPX physics.

    Attachments:
Viewing 20 posts - 1 through 20 (of 83 total)

You must be logged in to reply to this topic.

©2019 VPinBall.com

WP2Social Auto Publish Powered By : XYZScripts.com
Vpinball.com

Log in with your credentials

or    

Forgot your details?

Create Account