Hacker News new | ask | show | jobs
by toast0 1299 days ago
> it is interesting just how much irl pinball physics differ from their virtual counterparts, there really is nothing quite like it.

I think there's probably a couple components.

#1 is virtual pinball physics is usually too simple and it plays too deterministically. Real pinball plays differently when the machine clean vs dirty, and it gets (minutely) dirtier as you play, parts wear, etc. Sometimes the ball jumps or otherwise moves in unexpected ways.

#2 is flipper timing variability. In virtual pinball, the controller is usually sampled once a frame, but pinball machines pre-fliptronics had the switches connected to the flippers through a relay, post fliptronics, I'm not sure if there's a sampling delay, but if so, I think the sampling rate is higher than 60Hz. That really increases the possibilities, even if a couple ms here or there doesn't make a big difference.

#3 tilting on virtual pinball is very precise, but I haven't found it nearly as precise in the real world.

2 comments

If you've seen the Slow-Mo Guys pinball episode you will quickly realize that video game physics still aren't really up to the task; they're very good at getting an effect, but still approximate.

Pinball is full of violent, three-dimensional forces: the pop bumpers launch the ball by crushing it from above, the slingshot bumpers deform the rubber rapidly enough to create visible resonance. Nudging and shaking the game creates a vibration across the whole playfield. And so on.

Most video pinballs outside of the dedicated simulators don't even try to get close, and opt for an implementation that treats the flippers as simple point-and-shoot devices that can always return the ball in any direction; the reality is that the available shots are all dependent on how the ball was tumbling, and the precise details of the flipper coil mechanism(games since the solid state era switch from a high-power coil to a lower-power one to prevent overheating when the button is held; this means that the flipper in the low power state bounces back when hit). High spin and low spin balls are categorically different and many games are designed so that you have to set up difficult shots by first creating a high spin scenario. Add in all the vibrations and deformations of a real game - the exact behavior of rolling down a wireform ramp, for example, is easy to dictate and hard to simulate - and you get to the parts that a simulator can't touch.

> games since the solid state era switch from a high-power coil to a lower-power one to prevent overheating when the button is held

Is this really limited to solid state machines? Pre-fliptronics, this happens under the board, with end of stroke switches, no electronics needed, there's no reason it couldn't have happened on an EM, and lots of good reasons to do it...

My personal take is that even if you could simulate the physics perfectly, there's just something about the physical movement of a ball across a table, that you track with your physical eyes, that just can't be replicated.