Ogre Forums

I have an open question regarding how people play RTS style games. The current design that alot of games go by (SupCom, Warcraft, etc..) involves Team A killing Team B and the annihilation of the other team being the primary goal. Diplomacy seems tacked on as an afterthought.

I suppose that if organic limits on the amount of combat were imposed, that new strategies for winning the game would be required.

A scenario (space rts):
Player A has awesome ships, and Player B has a cloak. Neither can destroy the other, but you have to have a winner. So B waits for A to overextend themselves and then strikes.

Right now in most games, as soon as A amasses such a large fleet, B is pretty much dead because there is no negatives for having such a large fleet. Having a resource drain for having such a large fleet is not an option because its artificial. (i.e. your ship explodes if you run low on supplies?)

My fix would be to make the ships break part by part until they fall into disrepair and cannot be used to engage the enemy if that enemy surgically targets their supply routes (w/ that perfect cloak )

Any thoughts?

-I

incidenta5 wrote:Having a resource drain for having such a large fleet is not an option because its artificial. (i.e. your ship explodes if you run low on supplies?)

Why is something not an option because it is artificial? There are tons of other abstractions in games that are neither logical nor realistic. And that is definitly fine

The basic RTS concept is based on a positive feedback loop (the player who gathers resources faster and can build more units can in return conquer even more resources, build more units and so on..). Tons of ways exist to weaken that positive feedback. Making resource gathering more ineffective as a player controls more units is just one way. Which is what Warcraft does (plus a few more subtle things).

I don't really see what that point has to do with how diplomacy is incorparated into the game design For diplomacy I'd try to think about changes that make players cooperate and give them more tools to do so (most RTS only offer sharing of resources and unit control, plenty of design space left in that regard I guess).

It seems you are on to a very nice idea in my opinion. One time I was thinking about making a game, where in you need a certain number of survivors to say open a gate to the enemies base after all of the enemy had been killed. This would add a little something to the game, but it was never a very good idea alone.

Maybe if the team with more people moved at a slower pace, as in real armies the more numbers the harder to move the army from one location to another.

What got me onto this train of thought was the idea that games should last longer than 3-4 hours if you want to consider them 'strategic'.

Right now, especially in SupCom or Spring, once you get your mex/reactor farm going you can dominate the map and will have no shortage of units. This just seems a bit unnatural to me.

For example, if you mine an asteroid for metal the actual facility that does it should take awhile to construct and you shouldn't get the resources instantly.

This allows for new tactics that don't involve destroying anything. I.e blockading it to get your enemies attention long enough to do something sinister in another sector on the map.

Just an idea.

-I

Well, right off the bat these things come to my mind:

- Have a meaningful unit limit (that has a lot of other positive side effects on the technical side and design usually.. first of all no esoteric hardware requirements like SupCom)
- Make larger armies harder to control: e.g. only allow a player to select a limited number of units at a time
- Map design - larger armies have less of an advantage at choke points for example
- Tweak the balance parameters. Adjust costs and build time for expansions so it takes a few minutes until they turn into a resource advantage
...

Wouldn't it also work to model a real command structure (for making things difficult to manage). i.e. Introducing delays in communication and orders.

That way large groups of units would still be possible but micromanaging them in realtime would be a bit more difficult.

I thought about that, too. But it might be difficult to implement this in a way that doesn't make the game feel slow and tedious.

I diagrammed this out a while back. It may only be useful under certain conditions, which will limit how it can be applied. For example, you might start out this way and have the delay be proportional to the distance from your starting area. That way you can't expand too fast until you research the appropriate techs.

I've posted parts 1 and 2 of a series of posts about this type of idea on my website at: http://www.0x1337.com/?q=node/6

There is also some information regarding the RTS I'm building with these ideas in mind, if anyone is interested. I'm still playing with the idea of making it open source, although that really depends on the interest it generates. There's something like 10,000+ lines of code written for it already last time I checked.

Comments are welcome

incidenta5 wrote:For example, you might start out this way and have the delay be proportional to the distance from your starting area. That way you can't expand too fast until you research the appropriate techs.

With that you haven't done anything against massing, you just slow the whole game down. After all a delay for commands doesn't hinder me from building expansions. It does however make attacking early less attractive as the defender will have the advantage that at his base he can command units without delay and for combat that does matter.

I think the only solution is to put in some negative feedback loops, i.e. associating risks and costs with expansions and making it so that each further expansion yields less of an advantage.

About the post on your site: I strongly advise against making an RTS with "thousands of units". Look at Supreme Commander. It can get unplayable on high end quad core machines. And even taken performance aside you'll have endless pain with unit movement behaviour and AI in general

jjp wrote: About the post on your site: I strongly advise against making an RTS with "thousands of units". Look at Supreme Commander. It can get unplayable on high end quad core machines. And even taken performance aside you'll have endless pain with unit movement behaviour and AI in general

I wanted to make every object in the game a unit. I.e. all resources, missiles etc.. The player should only have a couple of battle units (<400 perhaps?) and then dozens of weapons per unit.

Although I haven't yet run into the hard cap in supcom w/ my Q6600 [I have on my laptop though, heh]