Where randomness is good in digital pinball.

[Zaphod77]

There are certain things that are effectively random in real pinball, and can be simulated well in TPA with randomness.

1) small variations in scoop kickouts. they do noo kick out with precisely the same strength every time. and there can be slight trajectory adjustments as well.
2) ramp feeds. they rattle differently sometimes depending on spin and design. so the ball will not always reach the flippers at the same speed.
3) rebounds off standups. they deform a bit, and unless hit dead on, rebounds off them are somewhat unpredictable. the angle should be randomized a little.
4) jet bumper rebounds. randomization there is a good thing. they are very unpredictable.
5) eject hole kickouts. these are highly random in real life in most cases.

things that should definitely NOT be random.

1) flippers
2) bounce off of rubbers
3) bounce off of metal
4) bounce off of drop targets. these are highly predictable in real life.
5) autoplungers, when plunger timing determines the skillshot (like dracula) with games like monster bash, randomness in autoplungers is okay, and desired)

 

[vpalmer]

here no randomness in the real world - only physic laws.

 

[superballs]

here no randomness in the real world - only physic laws.

</Thread> right there

 

[Heretic]

pffffttt

 

[Squid]

I'm all for harder tables!

Bring on the randomness!

 

[clembo2021]

here no randomness in the real world - only physic laws.

Yes, however the physics behind the phenomena that make pinball appear random are likely too intensive to be simulated by most of the TPA hardware. Instead some amount of forced randomization on the kickouts and such could give us the same effect with less hardware processing needed.

I mean kickouts are just a joke in this game right now. Hold up right flipper on Whitewater for safe kickout. Let the ball dead bounce every time on Twilight Zone from Slot Machine kickout. There's a reason why professional player are scared of kickouts, and its something TPA doesn't capture even a little bit of.

 

[LurkerCreep]

Add to the list of things that should not be random:

- Balls returning from the orbits should never kick out from perfectly smooth orbit walls.
-- This "randomness" popped up in Attack from Mars (right orbit) and Cactus Canyon (both left and right orbits). It was fixed in AFM, but remains in CC. When this was pointed out, FS said CC is working as designed. Whatever...

 

[Zaphod77]

cactus canyon games do sometimes do that on location. the ball is spinning the wrong way and sometimes takes a hop off the rubber at the end of the orbit because of it. its pretty rare in real life though. (it only happens to me in a 4 player game and not the other players.)

 

[brakel]

I don't think we need randomness as much as we need a higher resolution of physical attributes. The plunger shouldn't be random. The harder you pull back the harder the ball should be shot out of the plunger lane. I think there's only maybe a few hundred increments of strength that TPA can register on a plunger pull. There should be a million or more for a truer real life equivalent.

When you start adding "random" events you start getting weird things happening. Real pinball has billions of outcomes when you start adding up all of the variables. But none of them are random.

The randomness that FarSight has tried to add to the game (plunger strength, AFM right orbit) have come off feeling like glitches and not true to life unpredictable outcomes.

I think it would be better for FarSight to work on increasing the resolution of the physics and the 3D models of the table parts. The higher the physics resolution, in theory, will give you more real world outcomes.

 

[Zaphod77]

non autoplungers should not be random at all. the randomness that's ok is in auto-plungers where it plunges up to some rollovers.

Ones where randomness is okay in the autoplunge.

Monster Bash (and it is random somewhat)
Indiana Jones
Attack from Mars
Medieval Madness
Star Trek, the next generation.
and any similar game that lets you auto plunge up to some rollovers
and any game with a combo plunger, as long as it's never too weak.

Games where it's not ok

1) dracula. but in this case, some should be introduced off of the drop target, contrary to what i said before, to affect what lane it drops into.
2) t2 (the randomness here should be off the standups)
3) LW3 (same thing here)

My post was an outline of where it's okay to cheat with randomness, and where it's not okay.

 

[Espy]

I don't think we need randomness as much as we need a higher resolution of physical attributes. The plunger shouldn't be random. The harder you pull back the harder the ball should be shot out of the plunger lane. I think there's only maybe a few hundred increments of strength that TPA can register on a plunger pull. There should be a million or more for a truer real life equivalent.

When you start adding "random" events you start getting weird things happening. Real pinball has billions of outcomes when you start adding up all of the variables. But none of them are random.

The randomness that FarSight has tried to add to the game (plunger strength, AFM right orbit) have come off feeling like glitches and not true to life unpredictable outcomes.

I think it would be better for FarSight to work on increasing the resolution of the physics and the 3D models of the table parts. The higher the physics resolution, in theory, will give you more real world outcomes.

Plungers can be pretty random, especially when they're not brand-spanking new. The angle you pull the plunger out at can lead to a lot of variation.

 

[JPelter]

I don't think we need randomness as much as we need a higher resolution of physical attributes. The plunger shouldn't be random. The harder you pull back the harder the ball should be shot out of the plunger lane. I think there's only maybe a few hundred increments of strength that TPA can register on a plunger pull. There should be a million or more for a truer real life equivalent.

When you start adding "random" events you start getting weird things happening. Real pinball has billions of outcomes when you start adding up all of the variables. But none of them are random.

The randomness that FarSight has tried to add to the game (plunger strength, AFM right orbit) have come off feeling like glitches and not true to life unpredictable outcomes.

I think it would be better for FarSight to work on increasing the resolution of the physics and the 3D models of the table parts. The higher the physics resolution, in theory, will give you more real world outcomes.

Plunger strength is about the only thing I agree with here, but I think it's already predictable enough. We already know we're not going to get more fidelity in physics modeling because mobiles can't handle the cpu load. Randomness done right can simulate it just as well. For example those orbit variations in quite a few tables are about the best parts about TPA's modeling as far as I'm concerned, although I feel like those are probably not even intentionally randomized. Kickouts being more unpredictable and ramps/outlanes on most tables behaving less like vacuums would be lovely too.

 

[brakel]

Plunger strength is about the only thing I agree with here, but I think it's already predictable enough. We already know we're not going to get more fidelity in physics modeling because mobiles can't handle the cpu load. Randomness done right can simulate it just as well. For example those orbit variations in quite a few tables are about the best parts about TPA's modeling as far as I'm concerned, although I feel like those are probably not even intentionally randomized. Kickouts being more unpredictable and ramps/outlanes on most tables behaving less like vacuums would be lovely too.

But FarSight hasn't been able to do random subtly enough. That's why I bring up the plunger. I think the plunger on the PS3 version is borked because of attempted randomness. Yes, plungers can be strange in real life. I've played plenty of real machines to know. Real plungers don't plunge with full power when pulled back 10 percent and they don't plunge 10 percent when pulled all the way back. These things happen on the PS3 version in the name of randomness. I'd rather not see more random behavior like that in the future.

 

[JPelter]

Real plungers don't plunge with full power when pulled back 10 percent and they don't plunge 10 percent when pulled all the way back. These things happen on the PS3 version in the name of randomness. I'd rather not see more random behavior like that in the future.

I haven't seen the former happen, but I agree that the latter needs to be fixed. It's already been reported and probably being looked into. The problem is there's often some problems with the contact physics between the plunger and the ball when a new ball enters the plunger lane, instead of being intentionally randomized.

 

[Zaphod77]

There's also a hilarious feature. plunge the ball sorta weakly. after it bounces off the plunger plunge again before it comes to rest. The plunger will still provide impulse to the ball!

 

[Mike Reitmeyer]

But FarSight hasn't been able to do random subtly enough. That's why I bring up the plunger. I think the plunger on the PS3 version is borked because of attempted randomness. Yes, plungers can be strange in real life. I've played plenty of real machines to know. Real plungers don't plunge with full power when pulled back 10 percent and they don't plunge 10 percent when pulled all the way back. These things happen on the PS3 version in the name of randomness. I'd rather not see more random behavior like that in the future.

It's actually not randomness, it's an issue with hardware. On computer hardware we read the input 60 times per second. In real life the input is read in an infinite amount.

Lets say you pull all the way back, release, and the timing is such that the controller input reads the controller at 100% and then 90% and at 40% and then at 0% in the subsequent frames. How does it know where you had it pulled to when you release? You can assume 100%.

But then lets say you pull all the way back, then bring forward before releasing. Then the input is 100, 97, 94, 90, 60, 40, 0. Did you release at 100? 97? 94? or 90?

We had to smooth the values to give the feel that its doing what you want in most instances. Unfortunately it doesn't always work correctly.

 

[brakel]

It's actually not randomness, it's an issue with hardware. On computer hardware we read the input 60 times per second. In real life the input is read in an infinite amount.

Lets say you pull all the way back, release, and the timing is such that the controller input reads the controller at 100% and then 90% and at 40% and then at 0% in the subsequent frames. How does it know where you had it pulled to when you release? You can assume 100%.

But then lets say you pull all the way back, then bring forward before releasing. Then the input is 100, 97, 94, 90, 60, 40, 0. Did you release at 100? 97? 94? or 90?

We had to smooth the values to give the feel that its doing what you want in most instances. Unfortunately it doesn't always work correctly.

Thanks Mike! That makes so much more sense now. It seemed to me though that when it first came out a real gentle pull on TotAN's plunger would always get you the first hole but now half the time it goes to the third. This is on the PS3 version. If I pull just a bit back, not fully back and then let up some, but just pull back a bit, how does that hit the third hole?

 

[Zaphod77]

It's actually not randomness, it's an issue with hardware. On computer hardware we read the input 60 times per second. In real life the input is read in an infinite amount.\

And this, in a nutshell is the problem with digital pinball, and it's very hard to solve.

It is why certain shots under certain conditions become totally impossible, even though they wold work in real life. one frame too soon, next frame too late.

It goes the other way too, where shots that should have muhc tighter timing instead are makable 100% of the time because it rounds to the nearest frame.

When i made tables for visual pinball i made an effort to make sure that any shot was shootable under circumstances people were likely to try to shoot it. WHen i played a VP table by someone who did't do this, i'd just get frustrated.

But as i said in certain parts of the game, a small amount of randomness can really go a long way towards making the game feel more real, if it's done with the right things..

 

[invitro]

here no randomness in the real world - only physic laws.

The laws of quantum physics inherently include randomness.

 

[invitro]

There are certain things that are effectively random in real pinball, and can be simulated well in TPA with randomness.

Great stuff, and I admire you folks who are keeping on the physics tip.