GRW's ideas.

[GroudonRobotWars]

Could it be possible to make a little combat zone with a pit and a pile of dirt next to it into a version of the stunt arena making a new file of it in death match called RIWS1 Buried Alive Match. And modding the combat box to put robots in the box, and the combat box walls to be big enough to not throw OOTA. And you win when putting the bot in the pit and droping the dirt on it. The box will be over the pit in the final arena, the Robot Wars arena. Because I want this for a little thing after the main fights in Robotic International Wars used as one special vengance match that will happen.

[Trovaner]

Its possible but I can't guarantee that it would be easy. With no precedent, making the dirt look realistic wouldn't be easy. You would also need to make it out of one object. Otherwise, you need to figure out a way of keeping the lag down while being able to track if all the dirt was in the pit.

I'm not sure exactly where you're thinking the pit and dirt should go but I might understand the rest. You want a combat arena without the glass walls and a pit with dirt somewhere...

[Vertigo]

Sounds a lot to ask for just one match IMO

[GroudonRobotWars]

If it is not possible than it is ok. I just thought it would look cool.

[Clickbeetle]

Not to totally shoot down the idea or anything, but how about you focus on getting RIWS1 voted into signups on the next poll first, and THEN focus on getting the main tournament done, and THEN, if you still have time/interest, think about doing a bonus match.     I can guarantee you everyone who enters the tournament cares more about the main event than special bonus matches.

[Naryar]

Not to totally shoot down the idea or anything, but how about you focus on getting RIWS1 voted into signups on the next poll first, and THEN focus on getting the main tournament done, and THEN, if you still have time/interest, think about doing a bonus match.     I can guarantee you everyone who enters the tournament cares more about the main event than special bonus matches.

Oooh, manipulation...

Can realistic dirt can even be done in such a physics engine anyway without LAGGING THE GAME TO DEATH ?

[NFX]

My guess is no. My comp starts lagging whenever I knock the bricks down in the Obstacle Course. And then when I get to the second set...

[GroudonRobotWars]

I am actually not going to use it for RIWS1. But it could be used for other stuff or if other people want to use it.

[JoeBlo]

nobody else brought it up so I may as well...

how do you bury a robot alive when it is technically never living?

how about you focus on getting RIWS1 voted into signups on the next poll first, and THEN focus on getting the main tournament done, and THEN, if you still have time/interest, think about doing a bonus match.

I second this 

[Clickbeetle]

Can realistic dirt can even be done in such a physics engine anyway without LAGGING THE GAME TO DEATH ?

Nope, you would need to use a cheap substitute, like a big block that falls down and plugs the pit, or possibly even a small pile of bricks. But not actual dirt.

[GroudonRobotWars]

BUMP

I would say bricks would work best.

Anyways this will probably not be for RIW, it would probablly used for another tournament I create in the future.

[Trovaner]

Well, I could show you how to make your arena in RA2 but I don't think anyone has the time to make it themselves. The problem is that your instructions are not clear yet and it doesn't seem like there is much reason for having it.

After the opponent drops into the pit, there is no rush to bury them which sort of defeats the purpose of burying them in the first place. To remedy things, you could try adding an 'escape timer' which would launch the bot out of the pit after a certain amount of time.

Another problem is that you would need to prevent both bots from being immobilized/defeated by other means. You could prevent being immobilized by turning off the timer but you would run into problems when the bots are too low on power to push each other. They could have unlimited power but that isn't fair for bots with more power to begin with. To prevent both bots from defeating each other, you could add health to the chassis of both bots. You may also run into problems if both bots are missing a wheel (or drive motor). You could fix this problem by having the game check to see if both bots are down to one wheel but there are no ties in RA2 so ending the game would show that one of them won.
 
The biggest problem that you will run into is making sure that the bots are buried alive. If you didn't keep track of the 'dirt', you would run into problems when it got dropped in before the opponent or when the AI has to guide it into the pit after it has been tossed across the arena. Keeping track of more than one object (such as a few cement blocks) is no easy task, either, because the lag gets to be way to high. Making the AI drop ONE big object into a pit would require much of the same coding that I used for my AI Soccer Arena (which would be done if it weren't for the fact that I am making the AI more strategic).

[G.K.]

(which would be done if it weren't for the fact that I am making the AI more strategic).

Do you have seperate AI's for different positions?

[Trovaner]

(which would be done if it weren't for the fact that I am making the AI more strategic).

Do you have seperate AI's for different positions?

I currently have the Arena guiding the AI bots (that way you don't need a special AI to play) but I made it possible for specialized soccer AI's to override the Arena's coding. Right now, the AI is capable of scoring but it lacks enough strategy to win. Although its been awhile, I've been trying to make the AI check for obstacles, act as a goalie occasionally, and use the walls to their advantage.

[GroudonRobotWars]

So a problem would be getting the AI bots to get the opponent in the pit and making sure all parts of the object gets ito the pit.

[Trovaner]

So a problem would be getting the AI bots to get the opponent in the pit and making sure all parts of the object gets ito the pit.

It is possible but I don't think it would be efficient. There is no python code that gives the location of a GMF object so you need to perform multiple c using the only function that may be influenced by them. Using self.RayTest will tell you if something is between two points but it is so precise that you need to perform several of them in order to figure out a rough estimate of an object's location. It also wouldn't be able to differentiate between two objects so if two blocks were next to each other, you would need much more coding to figure this out.

Guiding the AI is no real challenge but having the AI figure out where the blocks are would be. That is why I suggested using only one object. Keeping track of one big object is way easier than keeping track of many little ones (especially since there would be less scanning involved to find a big object as opposed to a small one). You could probably use nearly the same coding as my AI Soccer Arena but instead of aiming for a goal, you would be aiming for a pit.

One alternative would be to have the 'dirt' appear after a bot has fallen into the pit. That way you can make it appear at a specific location every time a bot needs to be buried. The problem with this is that you wouldn't want to figure out if ALL the objects have fallen into the pit (for the same reason that you wouldn't want to keep track of that many objects).

Another alternative would be to lock the GMF objects to a specific location until a bot has fallen into the pit. Unfortunately, this would suffer the same problem as the above alternative and you wouldn't be able to have an "escape timer" (because the objects would have been moved after the first time).