Universe Sandbox
Universe Sandbox Legacy => Universe Sandbox 2008 | Discussion => Topic started by: SuperNova on September 01, 2009, 12:03:57 PM
-
i to have a idea, like timed screenshot
Timed Impact
Object A will impact Object B in ??:?? (sec,hours,years)
Object A's path will cross Object B's path and will form a collision
isn't that a good idea?
-
But what if another object interferes?
-
But what if another object interferes?
It can be Dynamic
-
It can be Dynamic
Um, what does tht mean in context???
???
I like the idea.
-
It can be Dynamic
Um, what does tht mean in context???
???
I like the idea.
What I mean by Dynamic is the 'Time' can change depending on what is effecting it.
So let's say Impact is in 10 seconds it gets to close to planet and slows down the time will increase.
or if the planet changes it's trajectory to miss the impact the time can display "No Impact Detected"
-
There are many problems with that. There will have to be a setting that detects the soonest impact possible and it will probably be flashing so fast you can read it.
-
Well theres many ways you could do this.
Like have the impact time within the planet stats when you put your mouse over the planet.
Really the big problem is calculating this.
EDIT: The only way I can think of is to have every body have a trace line going out in the direction it's going. (Kind of like the line "Projected Path" makes. If that line hits another body then a "Estimated" (Since Dynamic it will most likely not be perfect on the Impact time) time will start.
It could also be warped by gravity like projected path lines are. (When it curves around bodies).
Though I'm not sure it's possible or if it is it would take a lot of calculating for bodies that have a Impacts that are not direct.
Example would be...Let's say 2 bodies crossed orbits with each other every so often and eventually they will meet and intersect at some point. I'm not sure this would be easy to calculate especially since US doesn't know if an impact will or will not happen. It only knows when an impact is happening or not happening.
Even if this was possible and done...I imagine it would take a lot of CPU power to calculate all this unless it was made to happen. (Not sure the word for it..) umm like it was set up to do it and technically it's not Dynamic then.
If you don't know what I am talking about...sorry.
If I have no clue about what I am talking about Dan...sorry. lol
-
Anything is possible... It's always a matter of weighing the time required to implement the feature verses the payoff.
The big problem is that in order to calculate a collision the simulation must run ahead of what it's displaying (which could be really slow).
Like Chaotic Cow said the feature could be similar to the way projected paths work, but there are issues with that too.
Sell me on this feature... Why do you want it? What would you use it for?
-
One good use for it is that you will be able to more easily time your collisions in a system.
Though I can't think of anything else off the top of my head.
-
Anything is possible... It's always a matter of weighing the time required to implement the feature verses the payoff.
The big problem is that in order to calculate a collision the simulation must run ahead of what it's displaying (which could be really slow).
Like Chaotic Cow said the feature could be similar to the way projected paths work, but there are issues with that too.
Sell me on this feature... Why do you want it? What would you use it for?
let me explane:
Object A will cross the path of Object B, remember, Object A must be smaller Object b
the program calculates the time and speed of Object B and changes the speed and stats to the stats that impact the object B
i made a Timid impact with paint
(http://i29.tinypic.com/1h9qp2.png)
first the paths circle around the star
(http://i30.tinypic.com/2n06iqe.png)
then (after the ''TimImpact'' button is clicked) he begins calculating
(http://i29.tinypic.com/6i739v.png)
after the calculating is finished, the stats are changed to the impact stats
well....that's much it
-
isn't that Possible?
-
Hmmm
This might be possible if you could get X Y Z Coordinates.
That way US can check if the 2 objects will ever be in the same X Y Z Coordinate in the future then give an estimate of when it will happen.
One problem is that US will have to predict the future by running a Invisable simulation ahead of the regular one to see if it is possible.
Of course you could limit this into cycles or "Years ahead"
Example: Let's say the Cycles for Our Solar System is 10 Cycles meaning the "Invisable" simulation ahead of the regular one will only go up to 10 years ahead. (Not real time obviously)
So it knows what will happen within a 10 year limit.
If any impact happens it then adds a timer.
That's one way to to do Non-Direct impacts.
-
Hmmm
This might be possible if you could get X Y Z Coordinates.
That way US can check if the 2 objects will ever be in the same X Y Z Coordinate in the future then give an estimate of when it will happen.
One problem is that US will have to predict the future by running a Invisable simulation ahead of the regular one to see if it is possible.
Of course you could limit this into cycles or "Years ahead"
Example: Let's say the Cycles for Our Solar System is 10 Cycles meaning the "Invisable" simulation ahead of the regular one will only go up to 10 years ahead. (Not real time obviously)
So it knows what will happen within a 10 year limit.
If any impact happens it then adds a timer.
That's one way to to do Non-Direct impacts.
thanks, but i realy mean a Impact Timed
if 2 planets are 500 M KM away, they still will impact
now, its impossible to make 2 objects collide from 500 M KM away. but with the TimImp (timid Impact) it can
and Dan what do you think?
-
Hmmm
This might be possible if you could get X Y Z Coordinates.
That way US can check if the 2 objects will ever be in the same X Y Z Coordinate in the future then give an estimate of when it will happen.
One problem is that US will have to predict the future by running a Invisable simulation ahead of the regular one to see if it is possible.
Of course you could limit this into cycles or "Years ahead"
Example: Let's say the Cycles for Our Solar System is 10 Cycles meaning the "Invisable" simulation ahead of the regular one will only go up to 10 years ahead. (Not real time obviously)
So it knows what will happen within a 10 year limit.
If any impact happens it then adds a timer.
That's one way to to do Non-Direct impacts.
Got an idea!
This require I think from tens to thousends MB (so even GB) of RAM memory (I don't believe that the normal one is suitable):
Start the simulation: nothing appear. Data are stored. After some time (set by the user), maybe 10 years, maybe 1 year (obviously of the simulation), maybe dipending on the time step, the stored data are converted in images. The result? A simulation that runs with data of the future really accurate. The problem is the performance that this buffering sucks: data must be stored, which can take A LOT of RAM, especially in big systems, and then reloaded and calculated (the coordinates of every object must be stored frame by frame. For a ten-object simulation, which is really small, there are 30 numbers with lots of decimal digits every frame). I think that it could be optimized for dual/quad cores (with a dedicated core the RAM occupied will be smaller).
Moreover if you sacrifice your knowledge of the future you can make the simulation with auto time step more "linear" using the buffer.
Pros: A future more accurate
- Projected paths way more accurate
- Collisions predictable for sure within some seconds/minutes (according to your CPU/will)
- Auto Time step more fluid (scarifying your knowledge of the future).
Cons: Performance needs increasing exponentially with the number of objects.
EDIT: It also permit to move the simulation forward or backward maybe with a time bar. E.g:
-3 -2 -1 0 1 2 3
|----------|----------|----------|-O-------|----------|----------|
-
@SuperNova - I know what you mean. We are just taking the idea a step further.
A Direct Impact is much easier to calculate then one that is not apparent.
@FGFG - Are you saying the US should take images and then use the images to detect impacts?
-
No. It calculates, for example, the data for 100 frames, then in starts to convert the datas in images (what you normally see on screen), after the calculation.
-
dan, do you think this can be in US sandbox and do you like the idea?
-
Speaking for myself I think such a feature would be sure interesting, but thinking about the small amount of new possibilities it opens, I don't think it would be worth being added currently. It also sounds very brutal when I start to think of describing such a mode with formulas. Since these will also have to consider thousand of different influences.
-
imagine if it was some kind of minigame.
Where the player has to launch an object from Earth and hit it using the gravity of other planets and knock it out of the way before time was up.
-
Just like Cassini, it goes to Venus gravity, then again, then the Earth, then the Moon, then Jupiter, and at last Saturn.
-
I challenge you to do it without any formula...
-
lol. I been trying to shoot objects out of Earths hillsphere into venus and back. It's hard and I've yet to do it.
All it does is as soon as it 'leaves the hillsphere it oribits the sun.
-
Well, to do things accurately you have to learn some rocket science, and it's isn't the most simple subject to learn... ;)
-
lol. I been trying to shoot objects out of Earths hillsphere into venus and back. It's hard and I've yet to do it.
All it does is as soon as it 'leaves the hillsphere it oribits the sun.
What have you been doing? I want to try to get some random asteroids in to orbit around the sun.
-
Slow the time step. Target Earth..zoom in close and aim (Not AT Earth) but in same direction.
Choose your object you want to fire.
Aim...and press F..
If it's at the right speed it will follow Earths Orbit for a bit and when / if it leaves the hillsphere the sun will have the hold and drag it in until it settles into an orbit.
-
dan, do you think this can be in US sandbox and do you like the idea?
So your idea is for a button or command that will automatically alter the velocity of a body to collide with another?
It's a good idea, but I'm going to have to think some more about how the math would work. I think this could be lots of fun.
I will probably improve projected paths to show potential collisions. In version 2 when using the move tool projected paths turn on for that body which will make it easier to visually line up collisions.
-
In version 2 when using the move tool projected paths turn on for that body which will make it easier to visually line up collisions.
COOOL!!!!!!!!!!!!!!!!!
-
That is awesome Dan!
-
dan, do you think this can be in US sandbox and do you like the idea?
So your idea is for a button or command that will automatically alter the velocity of a body to collide with another?
that my idea, but there's a dropdown list near it so you can chose where it impacts
I will probably improve projected paths to show potential collisions. In version 2 when using the move tool projected paths turn on for that body which will make it easier to visually line up collisions.
but about that, is it possible that you can let a object on a object from 300 AU away impact?
-
I don't think the distance will be a problem - I think the biggest problem will be the fact that other bodies, especially if there are many, will interact with the body that's being set to impact another.
(Ofc the distance will be a problem if you want to set a body so far away from another that the line can't reach... But it might be editable...)
-
Like I said before. It can be Dynamic like projected paths. Changing based on whats in it's path.
-
You said you need the interacitions with all the bodies...
Got an idea!
This require I think from tens to thousends MB (so even GB) of RAM memory (I don't believe that the normal one is suitable):
Start the simulation: nothing appear. Data are stored. After some time (set by the user), maybe 10 years, maybe 1 year (obviously of the simulation), maybe dipending on the time step, the stored data are converted in images. The result? A simulation that runs with data of the future really accurate. The problem is the performance that this buffering sucks: data must be stored, which can take A LOT of RAM, especially in big systems, and then reloaded and calculated (the coordinates of every object must be stored frame by frame. For a ten-object simulation, which is really small, there are 30 numbers with lots of decimal digits every frame). I think that it could be optimized for dual/quad cores (with a dedicated core the RAM occupied will be smaller).
Moreover if you sacrifice your knowledge of the future you can make the simulation with auto time step more "linear" using the buffer.
Pros: A future more accurate
- Projected paths way more accurate
- Collisions predictable for sure within some seconds/minutes (according to your CPU/will)
- Auto Time step more fluid (scarifying your knowledge of the future).
Cons: Performance needs increasing exponentially with the number of objects.
EDIT: It also permit to move the simulation forward or backward maybe with a time bar. E.g:
-3 -2 -1 0 1 2 3
|----------|----------|----------|-O-------|----------|----------|