1K ball with regular physics

[FireDragon76]

Are there any negative effects from using a 1K ball with default physics tables?

 

[AnonTet]

It is actually recommended to use the 1k. Feel free to report if you find something weird.

Keep in mind that not every single physics version was tested but even with default FP physics (no XML or whatever) the table should play fine with the extra resolution of the ball.

 

[FireDragon76]

It is actually recommended to use the 1k. Feel free to report if you find something weird.

Keep in mind that not every single physics version was tested but even with default FP physics (no XML or whatever) the table should play fine with the extra resolution of the ball.

I had a ball sticking on the ramp in Super Android. I haven't tried the default ball. I did try it with and without FizX physics code. My guess is that the ball would stick whether or not I used the 1K ball.

Otherwise, the 1K ball has been perfectly fine.

I have experimented with running FizX Lite tables with FPS at 296. I haven't noticed anything odd happening. It might matter with FizX scatter and bounce code, which FizX Lite doesn't have.

 

[AnonTet]

the higher the fps setting the best physics will behave since you have more simulation steps running. 512 I think was the last and highest recommended.
For fun fact only FP's internal physics engine max resolution is 600 (even if you set it it higher it won't go past that)

 

[FireDragon76]

the higher the fps setting the best physics will behave since you have more simulation steps running. 512 I think was the last and highest recommended.
For fun fact only FP's internal physics engine max resolution is 600 (even if you set it it higher it won't go past that)

It doesn't surprise me that I don't see a difference. 296 FPS means there's over 3 frames of physical interactions being composed every ten miliseconds. That's much faster than the time it takes to setup and draw a high quality image.

 

[AnonTet]

Just in case, let me mention that you should not mistake physics speed with refresh rate of a monitor.

Sure, at 60Hz you might not "see" a difference but at higher refresh rates you'll notice it.

In any case, the highest cycle rate for physics, the more accurate they'll be so you want them running at the fastest possible.

There is no reason for not running 512 unless you have a really weak CPU in which case I'd recommend to upgrade

Also, the physics rate impacts some objects on the table. You should run 512Hz with FizXz, period.

 

[FireDragon76]

Just in case, let me mention that you should not mistake physics speed with refresh rate of a monitor.

Sure, at 60Hz you might not "see" a difference but at higher refresh rates you'll notice it.

In any case, the highest cycle rate for physics, the more accurate they'll be so you want them running at the fastest possible.

The accuracy of the physics isn't necessarily directly related to the amount of frames per second. Those are two different things. Given this is a simulation, the goal is to produce a convincing approximation of the experience of playing pinball. That doesn't necessarily require a particularly fast framerate.

A few weeks ago there was a discussion of a physics paper about how after a couple of bounces, the position of a ball on a table is essentially indeterminate. Which means after a certain point, the accuracy of any pinball simulation will also be indeterminate.

There is no reason for not running 512 unless you have a really weak CPU in which case I'd recommend to upgrade

It has less to do with the CPU, and more to do with the overall system. I have a Ryzen 3600 and a Geforce 1650 Super, and I usually play at 1440p. It's not a slow system, but it's not capable of really feeding a 144 Hz monitor. Graphics cards that can do that aren't cheap.

Furthermore, there are externalities involved with needlessly pushing extra FPS, like higher energy consumption. Sure, in the case of using 512 vs. 196, the amount of processor power used is relatively small, but it isn't insignificant, with the higher FPS using about 40-50 percent more power from the CPU.

Also, the physics rate impacts some objects on the table. You should run 512Hz with FizXz, period.

Which objects? Does it affect the scatter and bounce physics for rubber and drop targets?

 

[AnonTet]

You repeated what I said regarding physics and monitor frame rate. I was just adding to the discussion because FPS really is a bad choice for an acronym in XML :)

Physics math run on the CPU only.

The CPU usage can be barely none or more intensive. Academic stuff: same amount of stuff to to will generste more load as you have to run it more times.

Point is not that. Point is, although you can use lower rate for physics, you'll also have less accuracy, obviously.

Plus, like I said, newton engine used by FP, alters objects weight starting with the ball.

You can probably find post from me here pasting Rafal answer to one of my questions about this via email a long time ago. I won't discuss this further because I don't have a full list either. Just trust on this because this was tested by me and JLou (and many others over the years I'm sure) but also because it was Rafal writing it :)

And since FizX was create with some basis in mind, like I said, you should use 512.

It makes a difference, especially at higher frames rates and as in all simulations you are already at a disadvantage if you have to run under optimal settings. The feel is quite obvious for me.