A query on combat possibilities

Hello everyone,

First off, I apologize if I missed the answer to this question elsewhere in the site, or if this question is in the wrong place. I looked around a bit but didn’t really find what I was looking for.

I’m helping to develop an independant MMOG, Genesis, and currently I’m searching for an engine for us to use to playtest combat system prototypes. We have a basic 2D engine for basic combat, but we’re going to want to test out some more complex controls for phase 2.

Being familiar with Blender, as I took an independant study last semester to make art assets, I wanted to ask you guys first and foremost about your opinions on combat systems using the GameEngine. We’d prefer a system that can use hit locations, so if that’s something that the GameEngine could do relatively easily, I’d be… well, psyched.

Thank you for reading, I hope to hear something soon.

By “hit locations” do you mean, for example, that a character would take more damage if struck to the heart, as opposed to the arm? This might be (somewhat) simple to hack together, possibly by using collision-sensitive cubes parented to various parts of the player.

I’ll warn you right now though, the BGE gets very slow when it has to process large scenes, especially if they contain a lot of complex logic.

That is exactly the kind of response I was hoping for, thank you so much. Yes, I do mean hit locations such as your example, but I’d also hope that the engine can be reactive as well.

For instance, your attack doesn’t go through your opponant, it connects with them and they react to it. So less like WoW, more like a fighter.

I’m not too worried about big scenes, this is mostly for one-on-one fights that we’re prototyping for.

Depth makes things considerably more complicated.

What exactly do you mean by depth?

I think what PlantPerson probably means is highly detailed, for example, if the combat had no depth, it would be the whole character model taking the same damage, whereas combat with depth would be the whole more damage to head and chest areas than leg and arm areas. :smiley:

Ah, I see. Thank you for clearing that up. So, even if it is complicated, would 14 hit boxes (7 for each model) be too much to ask?

With regards to depth. You would have to allow points per the available depth. Then based on each entities available points for sending a blow from that item and decide how many would be taken off the opponents available points.

If you follow me? An example:

A has a hurt left arm from a bullet, so A can not hold a gun in left hand, but can punch, with a blow of 50% of its available 100% blow points for the left arm.

B is 100% all over(except a 50% ankle damage that cause him to move slower now, and no jumping, lol). B gets hit in his 100% face by A’s left punch and looses 5% of face points. Face=bigger bone then knuckles.

Sorry if this isnt what you were talking about. I just love brainstorming.

I interpreted your second query as asking, is it possible to make, say, a sword that goes in two inches cause less damage than one that goes in eight inches. If that’s not what you meant, I apologize.

Ah okay, that makes more sense now. No, I didn’t mean that, thankfully. That’s what random variation in damage is for. Egad, we are not ready for nitpicky details like that in fighting games. o.O

For me, this kinda reopens the issue on when the parent-to-bone system comes back. When it does, you can just parent all the hitboxes to the character bones, so if you animate a dodge, the hitboxes will move automatically. Plus it’d make it easier to carry weapons.

Just a bit of a rant there. I guess the point I’m trying to make is that, as of now, your hitboxes won’t easily move in conjuction with armature animation. So that automatically makes it a lot more annoying.

Oh, no, what would you have to do instead, and when do you think it’s going to come back?

All you can really do is have some sort of IPO system, causing hitboxes to animate themselves…but obviously it’ll be impossible to have them run smoothly, and they’ll easily break sync when another armature action suddenly starts. And I have no idea when parent-to-bone will come back. It does work okay outside the GE, so it may not be long.

EDIT: It seems that two other topics about parenting to bones have come up in the last little white, so it may help draw attention.

1 Like

That is a great damper indeed, I don’t think we can do what we need without a reliable hit box structure. Bummer. Thank you so much.

Hm. Does anyone know any open source game engines that work well with Blender models, that could substitute for the time being?

It’s true that you won’t be able to do this without parent-to-bone, but if you are only prototyping ideas, you could use Blender Publisher (Blender 2.25). The parent-to-bone feature was still active back then (not sure exactly what version we lost it at). Blender 2.25 is still a rather capable game engine in its own right. It just has the older Sumo physics (as opposed to the new Bullet physics), it doesn’t have the new materials options, and it doesn’t have native joystick support (which may be another setback to you).

I haven’t used other game engines enough to suggest an alternative.

OGRE and CrystalSpace are both supposed to be pretty good, but I haven’t tried either one.

Crystalspace might work well, and you can easily use Blender created models in it. OGRE is only a rendering engine, though, so you’d still have to code the physics and logic system for the game yourself (or use it on top of another game engine).

I will look into Crystalspace as well, then, and I’ll ask one of our few coders to look into OGRE. Thank you all so much, you’ve been very helpful. Also, I invite you all to take a look at Genesis, and drop a line if you feel like it.

Arcadian, I don’t know what your development schedule is, but it looks like Monster submitted a fix for the parent to bone thing which will be in the next version of Blender. snailrose said that it has been submitted to the svn, so if you (or one of your coders) know how to compile Blender from source, you might look into it. Also, I’d recommend PM’ing Monster if you have trouble getting parent to bone to work.

It’s very laid back, at the moment. We don’t really have anyone working on it full-time yet, but now that most of us are out of school or graduated altogether, we can devote a lot more time to it.

I’ll see if I can’t get someone to compile it. Thanks so much for the information!