Sign in to follow this  

Strategy Game: C3 Interfaces

This topic is 4809 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I played a demo of Relic's Warhammer: Dawn of War and came away with mixed feelings. On the one hand, I liked their abstraction of resources by gaining control of resource points over harvesting (though it's still a rather artificial way of allowing a player to create more units), and I also liked the squad-based combat instead of having to control every single little unit. A morale bar indicating the psychological ability to fight was a nice inclusion too. Several options for "stances" or priorities of attack and defense were also nice. But I still noticed several glaring faults. Firstly, it was way too fast. There simply was no time to click on any of the special abilities like throwing grenades...something like this should have been at the discretion of the AI. But something I really would have liked to see is a command, control and communications interface (C3 Interface). I don't know why virtually every strategy game out there makes you scroll and pan all over the map trying to find your units or your building resources. And although the mini-maps on such games help, there's no labeling whatsoever....just a friend or foe identification, or if the game is nice, it'll tell you the difference between units and buildings. Instead, why not create an interface system that allows you to connect directly to your known resource systems and the battlefield commanders in charge of your units? For example, if you want to create some more units of type X, then instead of hunting and trying to remember where FactoryX is, you simply go to the C3 Interface, click on the Resources tree, and directly create your units from there? If your game considers something called Unit Integrity, then you can have all your units grouped together with a relative distance to one another. For example, in the above mentioned Dawn of War example, the basic unit is a squad. My idea is that you can link the squads together under another leader type to create a platoon (in my own system, I call squads, clusters, and groups of clusters BattleGroups). Unit Integrity enforces the principle that you can't have squad 1, 2nd platoon of Foxtrot company, halfway across the map (unless they were elite units who have the ability to act independently of the chain of command). The ability to access units via a simplified interface seems simple enough, so I wonder why no one implements it. The only reason I can think of is because no game has an established chain of command. In other words, you only have your basic units, but no way to build larger forces under the command of a leader (other than hitting Shift+FX to lump together a bunch of disparate unit types with no regard for combined arms capabilities). In my opinion, the map should mostly be used as an observation tool...to enjoy watching the carnage, and to develop your tactics and strategies. I think that directly clicking on units on the map should be optional, and not required in order to access and interface with your troops.

Share this post


Link to post
Share on other sites
I can't personally play turn based strategy games but many of the concepts involved in tbs's compared to standard rts's could solve this little problem.

Instead of simple repedative tactics thrown together as fast as posible (think 'twitch strategy', the counter-strike of rts's so to speak), or the slow highly specialized tactics of turn based strategy games, what I'd like to see is a real time strategy game that involves more restricted units (not necessarily less of them) that act smarter.

The military doesn't tell each infantry unit what exactly to attack, hell most high ranking officers don't even know who their indirectly commanding, so why should we have to? (or for another analogy, why should we be forced to micro-manage, a tack much more suited to less inteligent beings!).

I agree that being able to control squads is definately more appropriate, but it would be nice if one told the squads what they objectives are and give the ai some freedom of movement, maybe some more complex gameplay could be created using this! Would be strange if we were able to see units retreating because of being overwhealmed, and then back them up with another squad (telling them where to go still, so not being not in control, just possibly a different kind of control).

One thing that I think is important is to be able to have an influence on the graphical representation, as in when a squad moves you know how they're going to move. Since I'm into sci-fi, I think something super-modern/slightly futuristic would benifit from this greatly. You grab a unit and tell it to cover a corner of a wall, and queue a command that has them tear around the corner placing covering fire for the squads anti-tank infantry and having them take out whatever they can, retreating back behind the wall if they take too much damage (retreating tactic attribute would help, probably a slider/option control on gui).

Hail to a new forms of RTS gaming, because they're already long overdue! There can only be so many clones.

Share this post


Link to post
Share on other sites
Hmm...seems I was a little off topic, got carried away.

As for C3 interfaces, I seem to remmember C&C doing something similar to this except that you had to remember where your units were going to appear from. Ground Control 2 does this as well, you use dropships to purchase reinforcements based on points allocated by achieving various objectives. Just tell the dropship what drop location to use and every time you purchase some units they appear excatly where you knew they were going to and there is no clunky interfacing issues (well sorta, it does work though).

There is a downside to this type of interface, if you remember Total Annihilation the ability to use multiple factories, a common tactic (think flash rushes or aircraft only players). It would be possible to implement something but not nessecarily the simplest thing to build. You would have to do something like telling it what factories you wished to use at the current time, and then tell it to build the units. The two systems combined could be very useful however! (inteligent build menus!)

Share this post


Link to post
Share on other sites
Finally, another one of Dauntless' famous RTS threads. No more of that namby-pamby religion-in-gaming stuff again, I hope.

The closest analogues I can think of are from 4X games, which often have lists of your empires resources. There's a limited similarity to Earth 2150's Platoon system, which was not well received. Problems in interface and AI limited the utility of these groups.

The interface for Earth 2150 consisted of a panel at the bottom that had tabs to switch between Structures, Platoons, and Units. The problem was that these tabs hid the Orders screen (which had the special orders a unit could perform), which meant that it was not available when it was needed. The AI was problematic in that repair units (which, like Starcraft: BW medics, healed things automatically) did not repair things automatically when in a platoon. Nor could they be ordered to do so manually because individual units could not be selected in a platoon (or not easily anyway). The lack of a command structure made the Units tab useless as well. Clicking through an entire armada is not practical.

I've been thinking about making a game where you set up a battle plan and then simply watch it play out. Obviously, this would require a rather different interface than a typical RTS, as the orders one can give to a unit are not conducive to autonomy. You would need to be able to set up formations, tactics, contingency plans, SOP, acceptable levels of casualties, and so on.

Share this post


Link to post
Share on other sites
Hey thats cool Dauntlaess.
I had an idea like that back in the summer, for almost the same reason as you (and I have the GUI to prove it as well :D), I told my (older) brother about it who said it wasn't a good idea (he had just got CC generals), but then again he isn't a game programmer anyway..
In my idea I thought of having the whole thing kind of data driven, you'd have a panel with maybe buttons on it for each sector of the battle field, when you clicked a button you'd get a drop down list with tanks & different units on it, which you could look up for that sector. I thought maybe you could have a separate screen for each sector as well or something.

Share this post


Link to post
Share on other sites
Thumbs up to dauntless. I can't see the point of not providing such an interface, except that it might make the RTS a little too transparent as it were. I've often lamented that I can't issue more precise or detailed orders to units, and that's because I'm scrabbling about looking for the units. In short, this sort of traditional design needs you to fight the interface of the game, and that always unconditionally sucks.

So, I'm with you brother.

Share this post


Link to post
Share on other sites
I doubt there's some sort of prejudice against this sort of design; it just happens to not be obvious. I mean, I had no idea what C3 meant until I read this, so I'm guessing most developers have no idea either.

I would still use the map to guide units for most operations. I prefer seeing the information at a glance in a 2D format and acting directly upon it.

Share this post


Link to post
Share on other sites
Oluseyi-
Screenshots of my game?? I wish :) I've got most of the basic unit classes nailed down, some of the templated container classes to represent the clusters, and a real basic implementation of a Fctory Manager to produce unit types, but that's about it. I'm learning clanlib right now, since the syntax seems really clean to me (at least after looking at SDL and allegro). Although I really wish I had the time to look at Ogre3d....perhaps after I get the core prototype 2d game working, I'll move what I can over to a 3d engine. Hopefully within the next few months I'll have a basic logical pipeline for the game and work on some kind of "World_database" that'll keep track of everything.

School's killing me though...in fact, I should be studying right now....4 midterms in one week....AND a 68k processor assembly project...sigh.

Share this post


Link to post
Share on other sites
Flarelocke-
Didn't like my RPG-type posts? :) What can I say, I like PnPRPG's, and like my strategy games, I want to make them different. Hence, I wanted to create something that was more deep and profound.

But I haven't forgotten about my strategy game. I'm still trying to decide even what exactly the GUI portion of the interface would look like. Obviously the interface can't take up too much screen real estate, but since it's the most critical aspect of control for the game, I'm envisioning something that's somewhat large.

My initial vision has mini-icons or banners representing the top-level chain of command and another set that represents your resources. Now, in my game, you don't have factories per se, but instead have allocation request pools, and you have Distribution Hubs to worry about. So in a nutshell, you've got Commanders, Allocation Requests, and Logistics.

Because communication links as well as transportation links are vital in my game, you can lose contact with any basic entity (I define an entity as any game object which can be given an order either directly by the player, or indirectly by an AI Commander), up to and including top-level Commanders. Jamming equipment, poor terrain or weather may cause an inability for some part of your armed forces to be cut-off from your control. Just ask the British 1st Airborne during the battle for Arnhem bridge how sucky it is when an entire division can't communicate with headquarters, nor the regimental or battalion commanders between each other.

To model this, if communication is cut off, they simply become "greyed out" on your interface screen, and on the main map itself, you lose any information that only those entities would have been privy too. In other words, although those units are still alive, they're now no longer under your control, and hence, the re-establishment of communications should be a top priority. Obviously, the AI has to be robust enough to recognize and handle this, and it should be a "hard encoded" rule that if any of the C3 links are broken, it should make it a top priority to fix them.

So communications is relatively easy to implement. The Command and Control aspect is a bit more tricky though, mostly since it relies on very flexible and adaptable AI algorithms. The Command and Control aspect is also less influenced by the actual Interface screen, since the Interface screen is used more to set certain Order parameters, which in turn are used by the AI Commanders to set the Control scheme. For example, determining aggression levels or standing orders would be done through the interface and used as guidelines for the AI. As an example, you might order a Company Commander to hold a certain position and set the priority to high (maybe this company is on the flank edge, and if it folds, then it will threaten the rest of the battalion, hence the high priority). You can set these through the interface, but the actual Command of the Company falls to the AI Commander who implements the defense given the priorities that you input.

I'm still working on how to implement Orders vs. Motivations. Orders are the set of actions along with their conditions and parameters that you give to a Commander (and hence to all the basic unit types under his chain of command, and ergo, this means that depending on the Rank of the Commander, different Order types will be available). Motivations on the other hand are the messy real world conditions that people in battle must face. It is a blend of psychology of the troops, tactical considerations, and morale.

Share this post


Link to post
Share on other sites
After playing a shitload of starcraft (and the Dawn of War demo), I think I know why that interface has become the genre standard. While it may not be all that great for implementing strategy and tactics, it does mean there's never a dull moment in the game. You're always doing something, and when big battles get going, it gets pretty frantic.

However, as you say, it really isn't all that great for the strategy side of things. Player time is probably the most important resource in the game, and the interface makes it rather difficult to make the most efficient use of it.

It might be quite interesting to experiment with some sort of 'briefing' system, whereby you don't control your units directly, but you give them a series of objectives and waypoints - much like you might be given in a first person game - and let them get on with it themselves. You can watch the events as they unfold, even issue new objectives and throw in reinforcements, but you don't have direct control over the unit's movements, and you certainly aren't bothered with little details like whether to throw a grenade or not.

Share this post


Link to post
Share on other sites
Quote:

I've often lamented that I can't issue more precise or detailed orders to units, and that's because I'm scrabbling about looking for the units. In short, this sort of traditional design needs you to fight the interface of the game, and that always unconditionally sucks.


This C3 type of interface certainly sounds like an interesting idea, and one I'd definately like to play. However, I think you'd lose some fans of traditional RTS - I've noticed that those who tend to really excel at the type of game don't have the issue mentioned here - I actually got some friends to play a quick game with me (C&C Generals) for testing purposes before posting, with a non-playing friend testing us, and in almost all cases, any of us could have a specific requested unit displayed center of screen and selected ready to issue an order within a second.

I think that although the C3 interface is an excellent idea, some players will feel that it costs them control - and indeed, I'm sure some people could still be far more efficient through micro-management. Also in favour of C3 though, is the fact that it will probably more closely represent reality.

Share this post


Link to post
Share on other sites
Quote:
Original post by Kazgoroth
I think that although the C3 interface is an excellent idea, some players will feel that it costs them control - and indeed, I'm sure some people could still be far more efficient through micro-management.


A lot depends on scale, I think. The standard paradigm works extremely well for relatively small numbers of units. With careful micromanagement, you can pull off some fairly cool tricks with a small number of units. However, it doesn't scale well at all - as soon as you start getting large groups of units with lots of different abilities, the standard control mechanism falls apart completely.

The C3 approach sounds like it might well work better with large numbers of units, but you lack the fine control to be able to do as much with a small number. Perhaps some kind of combination of interfaces might be a good idea - this way you have better scalability, and a slightly more approachable game, since people are already familiar with the small scale interface.

Share this post


Link to post
Share on other sites
Quote:
Original post by Sandman
Quote:
Original post by Kazgoroth
I think that although the C3 interface is an excellent idea, some players will feel that it costs them control - and indeed, I'm sure some people could still be far more efficient through micro-management.


A lot depends on scale, I think. The standard paradigm works extremely well for relatively small numbers of units. With careful micromanagement, you can pull off some fairly cool tricks with a small number of units. However, it doesn't scale well at all - as soon as you start getting large groups of units with lots of different abilities, the standard control mechanism falls apart completely.

The C3 approach sounds like it might well work better with large numbers of units, but you lack the fine control to be able to do as much with a small number. Perhaps some kind of combination of interfaces might be a good idea - this way you have better scalability, and a slightly more approachable game, since people are already familiar with the small scale interface.


Ah yes, with more units, perhaps with additional abilities available, the C3 approach would indeed begin to offer notable advantages.

Share this post


Link to post
Share on other sites
Quote:
original quote by Sandman
I think I know why that interface has become the genre standard. While it may not be all that great for implementing strategy and tactics, it does mean there's never a dull moment in the game. You're always doing something, and when big battles get going, it gets pretty frantic.


Well, that's precisely the point (or problem, depending on how you want to look at it). Strategy games shouldn't be a fast paced frenetic click-fest. Imagine if Chess was played in the same fashion as most RTS games are nowadays. That's why I second the notion of Krysole's TBS games solving some of the issues that crop up with how RTS games play. I'd daresay that most RTS games aren't really strategy per se, but algorithmic formulations of what unit production facilities to build which units for your particular tactic...all under a vicious time constraint. Indeed, the time constraint is so great, that sometimes RTS games feel more like races than about strategy.

So I think the whole RTS genre needs to focus back on its core element....strategy. This in turn requires the ability to plan. I personally want to create a system that not only factors in tactical choices, but to recreate the elements that a battlefield commander must face.

I was in another forum about PnPRPG design, and because of my computer background and my own philosophical bent, I tend to be very precise, rigorous and simulationist in my design methodology. I had to explain to some people there that when you try to model a system, the very inputs you chose not only dictate what the output will be (domain -> range), but by your very choice of including those inputs, it makes those outputs (or results if you want to think of them that way) as important.

For example, let's say I design my game in which communication is a control factor. By making that decision, communication now plays a vital role in the gameplay itself, because communication inputs (and initial conditions or states) affects the output state. As Sandman pointed out, time is the most critical resource we have, though not just from a system architecture standpoint but from a user standpoint as well. Hence, time itself must be seen as a (hidden) input which affects all the outputs of the game.

Share this post


Link to post
Share on other sites
Dauntless,

As someone else has stated here, I think what you're suggesting would bring RTS closer to TBS. So, I wonder if your suggestion would also make it easier to have a strategy game that allows both options. I think it would.

- Rob

Share this post


Link to post
Share on other sites
Oluseyi-
Any help is greatly appreciated. I wish I had more time to devote to the game, but between school, work, and working out....I'm lucky if I can spend an hour or two a week on my game design.

Tell you what, what's your area of expertise? To be honest, I'm not even sure where I should be prioritizing what software modules to build first. Sometime after mid-terms are over, I'm going to put up a user requirements specification that will detail how the game will appear to the player. Later, I'm going to draw up a developer requirements specification that will list most of the software components that I could think of. I developed a very basic UML class diagram that I can put up, but no Use-cases or sequence diagrams or anything like that.

Share this post


Link to post
Share on other sites
RobAU78-
I was thinking of doing something similar to what Combat Mission did, or Battle Isle, in that the game would be a hybrid TBS and RTS system. In essence, the real time part only lasts a minute or two, and every RT phase is preceded by a turn based system. They are in effect, "pauses" which you use to give orders, rally troops, organize, and properly manage your forces. The key here is the insightful application of forethought, because the orders you give here are what is going to be carried out in the RT phase.

I've been toying with the idea of having "Command Points". The idea is that you "pay" for your orders. This will simulate the notion that as a commander, you can't micromanage everyone all the time, and you have to focus your attention at any one moment. So if the caca hits the fan, you pay your points to order whichever Commander's force you wthink is the most important. This will also further reinforce the Chain of Command purpose. IF you have a limited amount of Communication points to "spend" your command points on, then if you issue them to the right leader, that Commander can in turn tell his subordinates what to do.....exactly how the chain of command is supposed to work.

Of course, so much of this is riding on good AI implementation. It really is the Achilles heel of my game design. So anyone good at AI paradigms would be a great help for me (I'm pretty much ignorant of AI solutions....which sucks considering how important it is).

Share this post


Link to post
Share on other sites
Quote:
Original post by Dauntless
Flarelocke-
Didn't like my RPG-type posts? :) What can I say, I like PnPRPG's, and like my strategy games, I want to make them different. Hence, I wanted to create something that was more deep and profound.
Well the honor thread was interesting, but the rest of it seemed like the usual, "Will you guys still like me if my art is meaningful" stuff that plagues artists in every medium.

Quote:
But I haven't forgotten about my strategy game. I'm still trying to decide even what exactly the GUI portion of the interface would look like. Obviously the interface can't take up too much screen real estate, but since it's the most critical aspect of control for the game, I'm envisioning something that's somewhat large.
I'm suddenly reminded of Pax Imperia. The map was accessible by hitting F2, research by hitting F3, a list of fleets by hitting F4, a list of planets by hitting F5, espionage by hitting F6, and a closeup of a planet with F7. The interesting part for this discussion, though, is what happens when you go from one to another. If you select a planet in the planet list (F5) and press F2, the map shows the system the planet is in. If you select a planet and hit F7, it zooms in on the selected planet. Same kinda thing with the fleets. Select the fleet in the fleet list (F4), press F2 (system map), and it displays the location of the fleet, with the fleet selected. Finally, the galaxy map (which systems connect with which others) was visible on those screens where you needed it, such as the fleet screen, so you could order ships to move to a specific system (although nowhere specific within that system) without using the map. I guess I forgot to mention that when you selected a ship or planet in the lists, it highlighted your location with a lovely pulsating blue circle, and that when you selected a system in the galaxy map to visit, it highlighted your path on the way there, both of which are important to an effective user interface.

The only downside to this interface was that it's not intuitive, and that the galaxy map isn't shown on every screen. They made a token effort to fix this by implementing a lobby (F1) where you can get to each of the screens by clicking (and learn the hotkeys with mouseovers), but in practice, this is not ever used. On the screens without the galaxy map, you're afraid to do much work for fear of losing touch with the situation.

Now, in a ground-based RTS, you'd need to replace the galaxy map with a minimap, of course, and you'd probably want a way of bringing up more than one list or hierarchy at a time, so that you can order one to do something to the other, but I think the principles are sound.

Quote:
My initial vision has mini-icons or banners representing the top-level chain of command and another set that represents your resources.
You're going to have to be more specific. In what context are these mini-icons or banners displayed? I'm picturing something like an org chart -- how about you?

Quote:
Now, in my game, you don't have factories per se, but instead have allocation request pools, and you have Distribution Hubs to worry about. So in a nutshell, you've got Commanders, Allocation Requests, and Logistics.

Because communication links as well as transportation links are vital in my game, you can lose contact with any basic entity (I define an entity as any game object which can be given an order either directly by the player, or indirectly by an AI Commander), up to and including top-level Commanders. Jamming equipment, poor terrain or weather may cause an inability for some part of your armed forces to be cut-off from your control. Just ask the British 1st Airborne during the battle for Arnhem bridge how sucky it is when an entire division can't communicate with headquarters, nor the regimental or battalion commanders between each other.

To model this, if communication is cut off, they simply become "greyed out" on your interface screen, and on the main map itself, you lose any information that only those entities would have been privy too. In other words, although those units are still alive, they're now no longer under your control, and hence, the re-establishment of communications should be a top priority.
Increasing priority as mission objectives are resolved, but I get the point.
Quote:
Obviously, the AI has to be robust enough to recognize and handle this, and it should be a "hard encoded" rule that if any of the C3 links are broken, it should make it a top priority to fix them.
Different battle plans call for different priorities. If you're holding a strategic location, communication links are not as important as when you're part of a coordinated assault.

Quote:
So communications is relatively easy to implement. The Command and Control aspect is a bit more tricky though, mostly since it relies on very flexible and adaptable AI algorithms. The Command and Control aspect is also less influenced by the actual Interface screen, since the Interface screen is used more to set certain Order parameters, which in turn are used by the AI Commanders to set the Control scheme. For example, determining aggression levels or standing orders would be done through the interface and used as guidelines for the AI. As an example, you might order a Company Commander to hold a certain position and set the priority to high (maybe this company is on the flank edge, and if it folds, then it will threaten the rest of the battalion, hence the high priority). You can set these through the interface, but the actual Command of the Company falls to the AI Commander who implements the defense given the priorities that you input.

I'm still working on how to implement Orders vs. Motivations. Orders are the set of actions along with their conditions and parameters that you give to a Commander (and hence to all the basic unit types under his chain of command, and ergo, this means that depending on the Rank of the Commander, different Order types will be available). Motivations on the other hand are the messy real world conditions that people in battle must face. It is a blend of psychology of the troops, tactical considerations, and morale.
Wanting to follow orders is just one of many motivations...

Share this post


Link to post
Share on other sites
I hadn't thought of an Org chart, but that'd definitely work for the interface.

Now that you bring it up, prioritizing certain key actions is very important and should be a part of the C3 interface. As you brought up, depending on the battleplan (the standing orders that are initially given at the beginning of the game), then certain priorities need to be established.

The most fundamental priority is of course self-preservation. In effect, any order that is given (whether by the player or the AI Commander) must compete against this priority. High morale, good Commanders, and high motivations will allow the actions to be carried out. So being able to modify or influence internal priorities and psychological states (which are motivations...related to the Morale/Motivation aspect I mention in my other current thread) along with prioritizing standing orders and battlefield orders should be a key component of the interface.


Hmmm, man, I think my interface screen is going to be huge....

Share this post


Link to post
Share on other sites
Quote:
Original post by Dauntless
Oluseyi-
Any help is greatly appreciated. I wish I had more time to devote to the game, but between school, work, and working out....I'm lucky if I can spend an hour or two a week on my game design.
Believe me, I feel your pain.

Quote:
Tell you what, what's your area of expertise?
System architecture, integration, logic and algorithms. I've fallen off graphics in the last half-decade.

Email me.

Share this post


Link to post
Share on other sites

This topic is 4809 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this