# Gravity Simulation

Status
Not open for further replies.
L

#### LacheLimbo

##### Guest
<p>&nbsp;</p><p>I am seeking an efficient way to simulate gravity in a computer system without relying on object-to-object or particle-to-particle math.</p><p>Instead of&nbsp;using objects'&nbsp;center of gravity, I am favoring a hi/lo space pressure system to create&nbsp;the simulation of gravity. This should allow for objects that have inconsistent densities to interact at far and close or&nbsp;range.</p><p>There would be any number of objects with mass in the "field". Each object of mass creates an area of low space pressure. Areas that do not contain mass are areas of high space pressure.&nbsp;Influenced bodies would be forced to move from an area of high pressure to an area of low pressure. </p><p>Performance and accuracy&nbsp;of the simulation should be dependant on the resolution of the field rather than the number of influenced bodies. The field would have to be recalculated at each interval.</p><p>Is there a simulation that works this way already? What math would be most efficient to calculate the vectors of the influenced bodies over the pressure field?</p><p>&nbsp;</p>

D

#### DrRocket

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>&nbsp;I am seeking an efficient way to simulate gravity in a computer system without relying on object-to-object or particle-to-particle math.Instead of&nbsp;using objects'&nbsp;center of gravity, I am favoring a hi/lo space pressure system to create&nbsp;the simulation of gravity. This should allow for objects that have inconsistent densities to interact at far and close or&nbsp;range.There would be any number of objects with mass in the "field". Each object of mass creates an area of low space pressure. Areas that do not contain mass are areas of high space pressure.&nbsp;Influenced bodies would be forced to move from an area of high pressure to an area of low pressure. Performance and accuracy&nbsp;of the simulation should be dependant on the resolution of the field rather than the number of influenced bodies. The field would have to be recalculated at each interval.Is there a simulation that works this way already? What math would be most efficient to calculate the vectors of the influenced bodies over the pressure field?&nbsp; <br />Posted by LacheLimbo</DIV></p><p>Your question is a bit vague and perhaps would be better understood if you would explain a bit more about what you are trying to do and why.&nbsp; The motivation for this approach&nbsp;would also be of interest.&nbsp; <br /></p> <div class="Discussion_UserSignature"> </div>

M

#### Mee_n_Mac

##### Guest
<p>It's not clear to me how the proposed approach would be any more efficient computationally.&nbsp; Aren't you stuck calculating the "pressure slope" for each body whose motion you wish to predict ?&nbsp; Wouldn't that calculation involve the same variables&nbsp;that a Newtonian gravity field would ?&nbsp; I can begin to see the difficulty in handling an irregular&nbsp;body with differing density (even if it's rigid) but&nbsp;I don't see how a pressure based model helps. &nbsp;I guess I'm not quite understanding what you're proposing ....</p><p>&nbsp;</p> <div class="Discussion_UserSignature"> <p>-----------------------------------------------------</p><p><font color="#ff0000">Ask not what your Forum Software can do do on you,</font></p><p><font color="#ff0000">Ask it to, please for the love of all that's Holy, <strong>STOP</strong> !</font></p> </div>

D

#### DrRocket

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>It's not clear to me how the proposed approach would be any more efficient computationally.&nbsp; Aren't you stuck calculating the "pressure slope" for each body whose motion you wish to predict ?&nbsp; Wouldn't that calculation involve the same variables&nbsp;that a Newtonian gravity field would ?&nbsp; I can begin to see the difficulty in handling an irregular&nbsp;body with differing density (even if it's rigid) but&nbsp;I don't see how a pressure based model helps. &nbsp;I guess I'm not quite understanding what you're proposing ....&nbsp; <br />Posted by mee_n_mac</DIV></p><p>He needs to explain this a bit better.&nbsp; I have this idea in the back of my mind that what he is contemplating is simply using a potential function, which might be thought of as analagous to a pressure, but is nothing new.&nbsp; I am also a little suspicious that this might have someting to do with a homework assignment.<br /></p> <div class="Discussion_UserSignature"> </div>

L

#### LacheLimbo

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>He needs to explain this a bit better.&nbsp; I have this idea in the back of my mind that what he is contemplating is simply using a potential function, which might be thought of as analagous to a pressure, but is nothing new.&nbsp; I am also a little suspicious that this might have someting to do with a homework assignment. <br />Posted by DrRocket</DIV><br /><br />I have been a&nbsp;software professional for the past 15 years, not&nbsp;a student, who happens to like noodling around with computer simulations. Every once in a while this subject has crossed my mind and I thought I would pursue it this time.</p><p>Hopefully I can explain it better...</p><p>I am looking at empty space as a field. With no mass present the field is smooth, or there are no space pressure differences. Now introduce to the field an object of some mass. Due to the presence of mass, the area in the field closest to the object has a lower space &nbsp;pressure. Or less space for the same distance.&nbsp;</p><p>Now taking a step back from the introduction of mass from elsewhere. I could imagine using this simulation to condense mass out of the smooth field by creating a point of low space pressure and&nbsp;assigning some threshold to high pressure that defines the spawning of a new point of mass. The low space pressure attenuation for a single point of mass would be a constant. More points of mass could be created in the same manner or by the overlapping ripples of high pressure triggered by the mass generation. The individual points should clump together, forming larger objects. Over time, there&nbsp;could be many&nbsp;objects, and&nbsp;the field's empty space would be of much lower pressure than it started but the average pressure should be the same as when it started.</p><p>The&nbsp;pressure slope sounds like the right fit.</p><p>Has this approach to gravity simulation been done? If so where can I find it?</p><p>Thanks</p>

D

#### DrRocket

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>I have been a&nbsp;software professional for the past 15 years, not&nbsp;a student, who happens to like noodling around with computer simulations. Every once in a while this subject has crossed my mind and I thought I would pursue it this time.Hopefully I can explain it better...I am looking at empty space as a field. With no mass present the field is smooth, or there are no space pressure differences. Now introduce to the field an object of some mass. Due to the presence of mass, the area in the field closest to the object has a lower space &nbsp;pressure. Or less space for the same distance.&nbsp;Now taking a step back from the introduction of mass from elsewhere. I could imagine using this simulation to condense mass out of the smooth field by creating a point of low space pressure and&nbsp;assigning some threshold to high pressure that defines the spawning of a new point of mass. The low space pressure attenuation for a single point of mass would be a constant. More points of mass could be created in the same manner or by the overlapping ripples of high pressure triggered by the mass generation. The individual points should clump together, forming larger objects. Over time, there&nbsp;could be many&nbsp;objects, and&nbsp;the field's empty space would be of much lower pressure than it started but the average pressure should be the same as when it started.The&nbsp;pressure slope sounds like the right fit.Has this approach to gravity simulation been done? If so where can I find it?Thanks <br />Posted by LacheLimbo</DIV></p><p>You still are not very clear.&nbsp; And mostly I think you are way off the mark. &nbsp;But here is what I think you are trying to say</p><p>You can model space as ordinary Euclidean 3-space.&nbsp; A distribution of mass in 3-space creates a scalar potential field in 3-space.&nbsp; It also creates a vector field that is called the gravitational field.&nbsp; The gravitational field is the gradient of the scalar potential field.</p><p>This is well known and the potential field associated with a point particle is mu*m/r where mu is the universal gravitational constant, m is the mass of the object and r is the distance to the object.&nbsp; But this really has nothing to do with pressure or with creation of more mass.</p><p>Associated with this notion there are no waves, and no ripples.&nbsp; It is classical Newtonian gravitational theory.&nbsp; Gravity waves are part of general relativity, and that is much more complicated mathematically.<br /><br /></p> <div class="Discussion_UserSignature"> </div>

L

#### LacheLimbo

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>You still are not very clear.&nbsp; And mostly I think you are way off the mark. &nbsp;But here is what I think you are trying to sayYou can model space as ordinary Euclidean 3-space.&nbsp; A distribution of mass in 3-space creates a scalar potential field in 3-space.&nbsp; It also creates a vector field that is called the gravitational field.&nbsp; The gravitational field is the gradient of the scalar potential field.This is well known and the potential field associated with a point particle is mu*m/r where mu is the universal gravitational constant, m is the mass of the object and r is the distance to the object.&nbsp; But this really has nothing to do with pressure or with creation of more mass.Associated with this notion there are no waves, and no ripples.&nbsp; It is classical Newtonian gravitational theory.&nbsp; Gravity waves are part of general relativity, and that is much more complicated mathematically. <br />Posted by DrRocket</DIV></p><p>&nbsp;Thanks for your response.</p><p>&nbsp;Sorry I wasn't clear enough. Really wasn't trying to hit a mark, rather I was&nbsp;asking a question.&nbsp;Your reply has helped me simplify the question... </p><p>- I know&nbsp;the mass and position of a single object, <br />- I do NOT know the mass or position of others.<br />-&nbsp;I also know the gradient map for the gravitational field.</p><p>With the above information how do I determine the vector of the single object?</p><p>&nbsp;I am specifically&nbsp;looking for non-Newtonian methods of simulating the affect of gravity in a computer system. Pressure, space,&nbsp;condensation, ripples&nbsp;are just rules in my simulation and I am not trying to match Newtonian or GR. So let's pretend for a moment that pressure is important in this particular approach. The original question I was trying to&nbsp;ask, however clear,&nbsp;is how do I determine the vector of a particle in a field of varying highs and low pressures?</p><p>&nbsp;Why is pressure import in this approach? Maybe it is more like fluid dynamics as opposed to gravity. The difference being that instead of mass displacing fluid outward it contracts or condenses&nbsp;space inward. Creating a low pressure region of space. Mass would move from high to low pressure,&nbsp;having an affect like Newtonian gravity except you don't need to compare objects to each other to calculate a simulation. You just need to know the pressure slope (?) of the immediate region around each object.</p><p>Would like to see how it works when a small object approaches a large object. This would be much different than Newtonian gravity calculations,&nbsp;which would rely on&nbsp;both objects' center of mass, or particle-to-particle. Perhaps this could also simulate how objects, spin&nbsp;and/or&nbsp;break apart due to differences in the field.</p><p>&nbsp;</p>

D

#### DrRocket

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>&nbsp;Thanks for your response.&nbsp;Sorry I wasn't clear enough. Really wasn't trying to hit a mark, rather I was&nbsp;asking a question.&nbsp;Your reply has helped me simplify the question... - I know&nbsp;the mass and position of a single object, - I do NOT know the mass or position of others.-&nbsp;I also know the gradient map for the gravitational field.With the above information how do I determine the vector of the single object?&nbsp;I am specifically&nbsp;looking for non-Newtonian methods of simulating the affect of gravity in a computer system. Pressure, space,&nbsp;condensation, ripples&nbsp;are just rules in my simulation and I am not trying to match Newtonian or GR. So let's pretend for a moment that pressure is important in this particular approach. The original question I was trying to&nbsp;ask, however clear,&nbsp;is how do I determine the vector of a particle in a field of varying highs and low pressures?&nbsp;Why is pressure import in this approach? Maybe it is more like fluid dynamics as opposed to gravity. The difference being that instead of mass displacing fluid outward it contracts or condenses&nbsp;space inward. Creating a low pressure region of space. Mass would move from high to low pressure,&nbsp;having an affect like Newtonian gravity except you don't need to compare objects to each other to calculate a simulation. You just need to know the pressure slope (?) of the immediate region around each object.Would like to see how it works when a small object approaches a large object. This would be much different than Newtonian gravity calculations,&nbsp;which would rely on&nbsp;both objects' center of mass, or particle-to-particle. Perhaps this could also simulate how objects, spin&nbsp;and/or&nbsp;break apart due to differences in the field.&nbsp; <br />Posted by LacheLimbo</DIV></p><p>If you know the gravitatinal potential function, you take the gradiane to get the vector field for the gravitational field.&nbsp; To get the force on a mass you mutiply the gravitational vector at the point at which the object resides by the mass of the object.&nbsp; But this is nothing more than a Newtonian gravity calculation.&nbsp; It what you are doing is n"much different than Newtonian gravity calculation" then it has nothing to do with gravity, and whatever you are simulating it is not gravity.<br /></p> <div class="Discussion_UserSignature"> </div>

L

#### LacheLimbo

##### Guest
<p><BR/>Replying to:<BR/><DIV CLASS='Discussion_PostQuote'>If you know the gravitatinal potential function, you take the gradiane to get the vector field for the gravitational field.&nbsp; To get the force on a mass you mutiply the gravitational vector at the point at which the object resides by the mass of the object.&nbsp; But this is nothing more than a Newtonian gravity calculation.&nbsp; It what you are doing is n"much different than Newtonian gravity calculation" then it has nothing to do with gravity, and whatever you are simulating it is not gravity. <br />Posted by DrRocket</DIV></p><p>Thanks for your response!</p>

Status
Not open for further replies.