# Physics engine for proof of concept

## Recommended Posts

5 minutes ago, DandelionTheory said:

So solar panels to motor to masses right?

5 minutes ago, DandelionTheory said:

I've seen the same kind of motion with one mass being forced upon about the pin in one direction, and another script with equal but opposite force on the center pin. I'm really trying to make it not work...

And where did the forces come from? AddRelativeForce?

Edited by Ghideon

• Replies 116
• Created

#### Posted Images

29 minutes ago, Ghideon said:

And where did the forces come from? AddRelativeForce?

I never said it was perpetual motion. I said it made the center of gravity shift.

##### Share on other sites

10 hours ago, DandelionTheory said:

Yay, now look to the script and ask yourself how I represented an equal but opposite force.

Maybe I should mention the lines connecting the masses are rigid and rotate around the common mass.

I hope this also answers swansont's question about what a pin is.

A "pin" in this scenario is the common connected mass to all other masses by a rigid body with the pin as the pivot point.

So it has a mass. Why is it a pivot point? Objects tend to rotate about their center of mass, and your pin isn’t it. If you force an animation to do otherwise, there’s a reason why the physics it represents wouldn’t be working.

##### Share on other sites

2 minutes ago, DandelionTheory said:

I never said it was perpetual motion. I said it made the center of gravity shift.

Did an external force move the center of mass? Did the little machine eject mass to propel itself?

##### Share on other sites

53 minutes ago, Dagl1 said:

Let your force be air or photons, 'shot' from a point a little away from it? Forces have to come from somewhere, so if you will apply force, then something has to apply that force. That thing (let's say a laser), will undergo an equal but opposite force when 'shooting' the photons at your construct.

I don’t understand why a reaction force would be represented here. An object’s acceleration is the result of the forces acting on it, not the force it exerts. The laser, in your example, is actually irrelevant. How the force got there is also irrelevant - the object in question doesn’t care. If it feels 1N of force, the acceleration doesn’t rely on  how that force is exerted

47 minutes ago, DandelionTheory said:

I wrote a calculation for the center of mass. I've seen the same kind of motion with one mass being forced upon about the pin in one direction, and another script with equal but opposite force on the center pin. I'm really trying to make it not work...

Where is that calculation? How did you determine the CoM?

##### Share on other sites

5 minutes ago, swansont said:

I don’t understand why a reaction force would be represented here. An object’s acceleration is the result of the forces acting on it, not the force it exerts. The laser, in your example, is actually irrelevant. How the force got there is also irrelevant - the object in question doesn’t care. If it feels 1N of force, the acceleration doesn’t rely on  how that force is exerted

I thought, but maybe completely misunderstood, that the issue at hand is that there is a force exerted from nowhere. OP asked how he could model this in a realistic manner, so (I thought) he would need something that exerts said force. So I thought for his (mental picture and) experiment, he could introduce something akin to a laser which pushed his object, thus making it more realistic. Did I completely misunderstand the issue at hand? My apologies...

##### Share on other sites

10 minutes ago, swansont said:

Where is that calculation? How did you determine the CoM?

Apologies, it's something I've been working on currently due to the need for specificity.

showed me how to do it.

Edited by DandelionTheory
##### Share on other sites

5 hours ago, swansont said:

How the force got there is also irrelevant - the object in question doesn’t care. If it feels 1N of force, the acceleration doesn’t rely on  how that force is exerted

True. Let's name your object "O".

5 hours ago, Dagl1 said:

the issue at hand is that there is a force exerted from nowhere.

True.

In my opinion both are right. An object O pushed by a force in Unity3D seems to behave as Swansont states, it will gain momentum and behave pretty realistic within the limits of the physical model. There is zero total momentum when simulation starts and a non-zero total momentum at some time later; total momentum in the "universe" in the Unity3D model is not, and not designed to be, conserved. OP seems to argue that this lack of momentum conservation tells us something about physics outside the 3D software. Dagl1's suggestion is to model a complete little "universe" where there is zero momentum at time=0 and zero total momentum at any later time due to correctly modelled forces and counter forces.

Sorry if this confuses the discussion further. But the question seems to directly address possible misunderstandings of OP regarding the modelling in the physics engine.

Edited by Ghideon
grammar, spelling
##### Share on other sites

5 hours ago, Dagl1 said:

I thought, but maybe completely misunderstood, that the issue at hand is that there is a force exerted from nowhere. OP asked how he could model this in a realistic manner, so (I thought) he would need something that exerts said force. So I thought for his (mental picture and) experiment, he could introduce something akin to a laser which pushed his object, thus making it more realistic. Did I completely misunderstand the issue at hand? My apologies...

I will note that physics textbooks are full of problems that fits into the category.

The issue as I understand it is that an object starts to move when the net force on it is zero. But it also apparently rotates around a point that isn’t the CoM, so there is an implied force that is not being accounted for.

##### Share on other sites

4 hours ago, swansont said:

I will note that physics textbooks are full of problems that fits into the category.

The issue as I understand it is that an object starts to move when the net force on it is zero. But it also apparently rotates around a point that isn’t the CoM, so there is an implied force that is not being accounted for.

Thank you.

Can you be more specific?

I do want to point out I wasn't trying to show perpetual motion or some nonsense. It just shifts it's gravity.

##### Share on other sites

3 hours ago, DandelionTheory said:

I do want to point out I wasn't trying to show perpetual motion or some nonsense.

I don't think anyone intended suggested that perpetual motion was your purpose. It is just that it is possible to build a working* perpetual motion machine in Unity3d. And it is possible to build a reaction-less drive or some other nonsense by mistake since the software is intended to allow it.

3 hours ago, DandelionTheory said:

It just shifts it's gravity.

What does that mean? Should it be "shifts it's center of gravity"?

(Downvote for critical questions in speculations section, that's a first  )

*) Displaying perpetual motion inside software. Not

Edited by Ghideon
##### Share on other sites

26 minutes ago, Ghideon said:

What does that mean? Should it be "shifts it's center of gravity"?

Now you're getting it.

26 minutes ago, Ghideon said:

And it is possible to build a reaction-less drive or some other nonsense by mistake

Okay, did you assume I made that mistake?

Or did you not read or look at anything I claim?

Because you keep bringing up how I'm missing something, yet your argument is with the engine not the method. Elaborate...

Opposite forces can be represented with unity if it is set up correctly, right? Are they set up correctly? Or did I miss a reaction force? Tell me. Don't argue language when you don't know the sentence structure.

Also, downvotes are the only thing on this forum open for opinion. Or did you forget that

##### Share on other sites

32 minutes ago, DandelionTheory said:

Okay, did you assume I made that mistake?

If the resulting model behaves in a way that is agains the laws of physics that could be a mistake or it could be a bug in the software.

32 minutes ago, DandelionTheory said:

Or did you not read or look at anything I claim?

The claim that a Unity3D model can shift it's center of gravity? Yes I did read that.

32 minutes ago, DandelionTheory said:

Because you keep bringing up how I'm missing something, yet your argument is with the engine not the method. Elaborate...

To elaborate we might need to discuss the setup on a more abstract level so that properties of the software does not get in the way. Ok?

32 minutes ago, DandelionTheory said:

Opposite forces can be represented with unity if it is set up correctly, right?

From what I understand of the engine, yes it can. But it does not have to. And correct setup does not guarantee a model that is true to a real situation.

32 minutes ago, DandelionTheory said:

Or did I miss a reaction force?

Not necessarily. It is hard to tell how the script behaves when the engine runs it. Example: Are the two AddRelativeForce calls below intended to be equal and opposite?

##### Share on other sites

1 hour ago, Ghideon said:

Example: Are the two AddRelativeForce calls below intended to be equal and opposite?

Yes, I'll ask you to check which one is the parent object. As the mass rotates relative to the pins position, the proper way to describe "down" to a 180°rotated object is with a negative vector.

The idea is to change the Mass's direction and apply force to the center pin, the variables were entered into the computer to be stimulated by a physics engine.

1 hour ago, Ghideon said:

To elaborate we might need to discuss the setup on a more abstract level so that properties of the software does not get in the way. Ok?

So I have to explain the phenomenon in exact words before it is allowed to be posted on a physics forum in speculation? That's absurd, what if I don't know how it works? How would I bring forth the phenomenon to your attention? PS, do some work too.

Edited by DandelionTheory
##### Share on other sites

1 hour ago, DandelionTheory said:

Yes, I'll ask you to check which one is the parent object. As the mass rotates relative to the pins position, the proper way to describe "down" to a 180°rotated object is with a negative vector.

The idea is to change the Mass's direction and apply force to the center pin, the variables were entered into the computer to be stimulated by a physics engine.

Without investigating the internals of the engine, can we be sure that the engine applies the exact same force in exact opposite direction? (Floating point calculations, rounding of numbers, limited precision etc that adds up as the simulation runs)

1 hour ago, DandelionTheory said:

So I have to explain the phenomenon in exact words before it is allowed to be posted on a physics forum in speculation?

No. Math would also be ok.

1 hour ago, DandelionTheory said:

That's absurd, what if I don't know how it works?

You could post questions in the mainstream sections, for instance computer help?

1 hour ago, DandelionTheory said:

How would I bring forth the phenomenon to your attention? PS, do some work too.

I was thinking of bringing the discussion to a more abstract level, without the internal details about the 3D engine as a collaborative effort. Within the forum rules of course.
Question; How would the phenomenon be described outside the physics engine, in reality? That allows us to check an implementation against the predictions from mainstream physics.

##### Share on other sites

9 hours ago, DandelionTheory said:

Thank you.

Can you be more specific?

I do want to point out I wasn't trying to show perpetual motion or some nonsense. It just shifts it's gravity.

I said nothing about perpetual motion

Your “pin” is not the center of mass. The system should not freely rotate about that point. The fact that it does is unphysical. There should be no expectation that the laws of physics will be followed by this simulation.

You say it shifts its gravity, but didn’t you also say there is no gravity? You don’t show gravity as a force in your diagrams.

##### Share on other sites

woopsie

Edited by DandelionTheory
##### Share on other sites

8 hours ago, DandelionTheory said:

woopsie

This is less than illuminating.

##### Share on other sites

20 hours ago, DandelionTheory said:

woopsie

Why are you not just pasting the code?

##### Share on other sites

2 hours ago, pzkpfw said:

Why are you not just pasting the code?

The code is meaningless unless you post the mathematical code that regulates it. The last code link didn't provide that detail.

Edited by Mordred
##### Share on other sites

2 hours ago, Mordred said:

The code is meaningless unless you post the mathematical code that regulates it. The last code link didn't provide that detail.

I don't think it'll ever have meaning.

I just thought it was weird they were posting code via (unedited) screenshots.

##### Share on other sites

Posting the code is only helpful if you are fluent in that code.

##### Share on other sites

Posting code as a screenshot so that the code have be retyped manually into an editor by all members interested in the code's behaviours is not the best way to communicate. Even with those fluent in the code.

(I guess OP identified issues with the post and removed it, this is fine, but forgot to remove the picture.)

##### Share on other sites

if you have 3 masses A, B & C

A & C are rigidly attached to each other

B Pivots around C via a rigid bearing.

If a force on any of these masses is to be represented correctly, opposite forces need to be shown for every force applied.

so if mass A acts on Mass B(represented by F1 and F2), the work done on mass B will be done oppositely to Mass C (F3)correct? see picture below.

##### Share on other sites

On 4/1/2020 at 7:12 PM, DandelionTheory said:

If I plug values into a physics engine, would that be enough of a proof of concept?

Typical game engine uses flat geometry i.e. gravity force vector is toward the ground in straight line in Y axis when in the reality it is toward center of mass, ~ core of the Earth. Game playfields are flat. Earth is spherical.

Typical game engine has no air resistance, so if you would make e.g. pendulum it could work forever. Potential energy of bob is transformed to kinetic energy and then back to potential energy with the same amount. Without air resistance only imprecision of floating point numbers could lead to stopping it.

Pay special attention to imprecision of the floating point math.

Edited by Sensei
##### Share on other sites

This topic is now closed to further replies.

×