dbostream

75 reputation

bio	website
	location
	age
visits	member for	1 year
	seen	Apr 21 at 8:35
stats	profile views	11

	75 reputation	bio	website		visits	member for	1 year
5 badges		location			seen	Apr 21 at 8:35

summary answers questions tags badges favorites bounties reputation activity

45 Actions

suggestions reviews revisions comments badges posts accepts all

Nov 30	comment	Arrive steering behavior Ok I will see if I can come up with something to make use of more force if there is any available, because as I mentioned earlier right now if max speed is for example 16 no more force than 16 is used even if I set max force to 100 it seems. One thing I noticed when I was playing around a bit was that when I remove `deltaTime` from the velocity calculation the object decelerates nice and sharp with my original code. However this change also means my objects reach max speed after just one Update() call which is bad. I think `deltaTime` should be there I just found the result interesting.
Nov 30	comment	Arrive steering behavior What do you mean when you say the "difference" is the pseudo code only uses speed not acceleration? different oppose to what? my current code uses speed as well. I tried to implement code in my arrive method from the pseudo code and objects do stop if I use a very big slowing_distance, but then it takes ages to reach the target. Looking at your second link, that animation shows pretty much how I want it but I cannot get anywhere near that sharp deceleration. If my object would max the use of force when decelerating I believe I could start slowing down much closer to the target like I want.
Nov 25	comment	Arrive steering behavior Thank you that makes sense, since I want to be able to stop pretty hard I thought I would increase the max force. I set max force to 100 and max speed to 16, when logging the force used each time step it never seems to use more than what max speed is set to and this happens when the object is standing still and is about to start moving for the first time -> desiredVelocity.Length() = max speed and Entity.Velocity = 0. If increasing max force doesn't make the object use all the force it needs to stop in time what should I do?
Nov 25	comment	Arrive steering behavior I still get the oscillating behavior, I tried many different threshold values. Also the book uses the distance > 0 comparison and apparently it works. Thanks for trying to help.
Nov 25	asked	Arrive steering behavior
Jun 9	accepted	Sliding collision response
Jun 9	comment	Sliding collision response Took a break from this but I guess using -1.8 will have to do. Thanks for all the help.
Jun 3	awarded	Supporter
Jun 3	awarded	Scholar
Jun 3	accepted	Circle inside circle collision
Jun 3	comment	Circle inside circle collision I have found what appears to be a solution for my special case now. As far as I can tell everything works great now, just need some more testing but I will mark this question as solved. Thanks for all the help.
May 31	comment	Circle inside circle collision Thanks seems to work perfectly. When finding `k` variables like for example `d` and `k` sometimes get values like `-0.000000000000033750779948604759` and `-0.00000000000013500311979441904` etc., to prevent the algorithm from failing I simply set values very close to `0` to `0`, from what I can tell this is ok to do. Now I just have to find a solution for the special case when the small circle starts outside the big one and moves inside it.
May 30	comment	Circle inside circle collision It works pretty well now to rotate the position and velocity by an angle. One problem though is right now it always rotates counter-clockwise no matter the direction of speed. Using `Vt/(R-r)` the `Vt` I use is `double arcLength = (destinationPosition - collisionPoint).Length();` i.e. the distance left to travel after the collision, I believe this is the problem since this will always give me a positive angle. How can I solve this?
May 30	comment	Circle inside circle collision Ok thanks will check it out.
May 30	comment	Sliding collision response Thanks a lot, looks good. If I remember correctly I implemented something similar as your picture shows and that's when I got that oscillating behavior, especially noticeable at low speeds. A value of -1.8 made it a lot better though, any lower like for example -1.7 made the puck travel almost the entire corner before it failed. But the closer to -1.0 I go the earlier it fails. Will try your new suggestion tomorrow.
May 30	comment	Sliding collision response Says the page does not exist when I click the link.
May 29	comment	Circle inside circle collision Sure we can use the chat, not sure how it works here though but I will be around tonight. I think my collision detection code is not working properly and I don't completely follow the part about finding `k`.
May 29	comment	Circle inside circle collision @Darkwings, ok thanks anyway. Just a question about the collision scenario in my drawing, would it be (completely) wrong to move circle B to point D as a sliding response? It would keep circle B inside circle A but I guess there is some error involved since it seems too easy. :)
May 28	comment	Sliding collision response That's ok, some new ideas would be much appreciated since I am out of them. :)
May 28	comment	Circle inside circle collision @Darkwings Thanks, I drew a picture, this is how I interpreted your explanation. As you can see in the picture, B is outside so it has to be moved inside. The collision point seems to be at F which will position circle B at point E. But you say the collision is at point C (...starting to suspect my drawing is wrong) which would position circle B at point D. Assuming E is correct the remaining distance (BE) to travel will be in the form of sliding towards the point D. Comments?