This post is split into two sections, each one divided into a common statement, a selection of text, and a question. It’s a bit long, and my first post, so please forgive me if it’s a little rough. Good reading. ------------------------- "If you want to get good at RTS games, you should really learn all of the hotkeys." It is my belief (and I'm sure others share this belief) that RTS games are a little too much twitch and not quite enough tactics. Too much of the player's time is spent either with a ridiculous amount of micromanaging, or having to handle half a dozen different things. The result is a very clear cut "skill level" that relies more on knowing which buttons to press and manual dexterity than well thought out planning and strategic management. It could just be my opinion, but I don't think this is a particularly good thing. I think it may be a good idea to free up the player’s brain a little bit. A real commander shouldn’t have to worry about each and every unit’s position and tasks, and neither should the player. The player should be able to play the game with a level of involvement that they chose, not a vastly unnecessary array of incredibly tedious tasks which would normally handled by officers two levels below him. He CAN control everything, but he doesn’t HAVE to control everything. Would a high level of automation in a RTS be a good thing, or a bad thing? (Obviously, the player would be able to choose what is and isn’t automated.) And now, the second matter. ------------------------- "If you want to get good at RTS games, one of the most vital things is speed." The rush is the bane of newbies. A player will get overrun by a small strike force at the beginning of the game, and they’ll be more or less screwed over. This is highly enjoyable for the winner, for they get the great feeling of victory for very little effort. This is also a very bad thing for the newbie, because they never get a chance to be "strategic" until they learn to be lightning fast with their opening routine. And even when they do, speed of creation is usually a more important thing than anything else. Ask any RTS player what one of their best games was. You’ll usually get a response describing a game with a lot of involving combat. Usually a long game. From what I can tell, the majority of RTS players fit into three basic categories, in a sense; Players who like to build, players who like to fight, and players who like to win. Now obviously there’s more to players than that, and everybody likes to win, but you’ll usually find players who prefer to build, prefer to fight, or prefer to win, and don’t care how they have to go about it. Here’s my concept; Leave the "Real Time" part out of base construction. Give the player a base creation mode. They’re given a set amount of resources, and all the time in the world they want to create a solid "Home Base" fortification, complete with a barracks, a solid HQ, walls, tanks, aircraft, defensive turrets, and anti-air installments. Include an "Auto-create base" function for people who don't want to have to mess around, and you’ve got the best of both worlds. The builders are happy, because they have all the time in the world to build their defenses up. The fighters are happy, because they can jump straight into good combat. The people who like to win probably don’t care either way. Everybody’s happy. Of course, there are obvious problems which require balance; you shouldn’t be able to create nothing but 500 units and a small shack for a base, and you can’t make 500 feet of solid brick walls as the only thing between the enemy and certain death. But otherwise, it’s a solution which fixes the problems of rush tactics, and satisfies the majority. Would the pre-built base concept be a good base for a RTS, as opposed to the current "worker and home base" routine? ------------------------- Now I’m sure that both the idea of high level automation and a pre-built base aren’t new ones. The question is, are they ultimately beneficial to add as core game concepts? Any input is greatly appreciated. Thank you for reading. Edit 1: Fixed a couple of grammatical errors. Edit 2: I've made a couple of comments which help clarify the issues discussed here; it might be easier to understand the concepts if those are read before posting. [Edited by - Dexter Omega on October 19, 2005 1:35:49 AM]

I think the second idea is pretty good. The fact of the matter is that for nearly all RTS games, the first 10 minutes is the same. Follow the 'best' build pattern or you lose.

The first idea, perhaps not so good. If the AI is worse than a player [and it pretty much will be] then someone cannot just automate it, because they'll lose to every player who's good enough to do it themselves. I'd just automate it for everyone. Make the decision be the tactical one rather than control.

Though I'm [yes, personal opinion forthwith!] skeptical that RTSes can be made interestingly tactical. More often than not it's just who's got the bigger force, since tactical decisions often offset when there's so few of them due to time constraints.

I agree totally with the build problem. How about this:

The much-reviled "moneymap" players in the StarCraft Community often institute a 'No Rush" rule in their games. It allows them to build at their leisure, using the limitless resources of the custom map to build robust defenses and mighty armies before actually engaging their adversaries.

In five minutes, for instance, I can get my base established with some good unit production and enough static defense to keep me from getting overrun before my army responds.

How about if a player can "record" a starting build? You go into "build mode" and have five minutes, using the resources on the map and your starting units, to make the base of your choice to the best of your ability. When you join a multiplayer game with the "five-minute base" rule enabled, you get to choose from your library of five-minute builds. It skips the first five minutes of the game, assuming that each player did exactly what was recorded, and you swoop into the game with that base, those units, etc.

Five minutes may be too much, and you might have to record openings for every map, but in principle, I'd like to be able to deplay "Lost Temple Vult Rush Build" rather than have so much depend on cloning my SCVs, getting the build order just right, and pumping those first two vultures right on time every time I play. It's too easy to bone myself in the first three minutes.

I also concur about the second point. The ability to have a little prebuild time would definately have its advantages. I don't think that someone should be given an unlimited amount of time, but a set amount of time, as to keep things moving. Even an option on the server to do both; prebuild time: with the option of unlimited (until everyone clicks ready), or a set amount of time, or to start in the traditional *craft-ish way of a couple of builders and the home base.

Thanks for the feedback guys, it means a lot. I'll solidify a couple of concepts more clearly.

For the AI idea, the idea is to have routines that can be turned on or off at the commander's will. If the player can do better than the AI, he should be free to try by all means, but he should have something to fall back on if he so choses. It's to give the newer players support, not to replace other tasks completely. As for AI efficiency, heck, why not take it a step further; give the player access to a simple scripting language or AI "interface" to pre-set what their AI will do? The key is to let the player do the planning, not just the thinking. Preperation is half the battle, after all.

As for the base-building.. The idea is that you're given a set amount of resources, and unlimited time, to build a base. This building isn't done online, it's done completely offline. When the game starts up online (or in a skirmish offline, or whatever), you basically select from a list of your prebuilt bases, and the base is placed down at the begining of play in its entirety.. Each map, as opposed to just having a "starting position" as in most games, would have starting "plots", which would be squares of flat land allocated for base building. Obviously this is a tad problematic with the issues of multiple base sizes and such, but I imagine it's something that can be worked around.

Just thought I'd clear up the concept, and again, it's great to have feedback.

Quote:

Original post by Dexter Omega Would the pre-built base concept be a good base for a RTS, as opposed to the current "worker and home base" routine?

I'm gonna try to answer this question by pimping a RTS game idea that's been floating around in my head for a while. It's not really unique by any means, but I haven't seen it implemented all that well yet.

I'm the kind of person that generally prefers pure combat RTS games. Games like Ground Control or any of the Total War series, for example. In reality, I really like base building, but you pretty much hit the nail on the head by saying that base building is what spawns ugly strategies like early rushes. I've always thought an excellent way to alleviate the problem is to completely separate each player's base from the main battlefield.

Basically, imagine you have three islands (in reality you could implement it however you want, islands are just a good way to explain the concept). The central island is the largest and it contains the resources that you'd want to fight over. On each side of the central island (the 'battlefield') are two smaller islands. Each of the smaller islands is the main base for one of the players.

Units are primarily constructed on the main base and then transported over to the battlefield to fight. Likewise, resources are gathered on the battlefield and transported back to the main base where they can be used to construct more units. You could, of course, use whatever method you want to justify how units move around - teleportation, transports, whatever. What's important is that neither player can reach the other's main base for the majority of the game. You can say that each main base is protected by a force field that requires advanced technology to penetrate or anything, just so long as each main base is completely separated from the action until the endgame.

What I like so much about this idea is that it preserves the real-time aspect of the game while preventing an early rush win. Since the main bases are separated from the action the players are fighting over resource nodes instead. I'd probably allow a limited number of structures to be constructed on the battlefield - defenses, collection buildings, things of that nature, but none of the primary construction buildings. Gameplay would have to revolve around the fact that neither player can rush the other since they're limited in how many units they can bring to the battlefield.

I could go on a lot more about it, but you probably get the idea. [smile]

Quote:

Original post by Dexter Omega Would a high level of automation in a RTS be a good thing, or a bad thing?

Not automation. Delegation. You assign a duty to a subordinate, and the subordinate carries out you orders to the best of its ability.

As for resource management: abstraction. By which I mean, "suppy lines." I always found it funny that RTSes expected me to take them seriously when set in the modern era but still requiring players to set up camps, factories, research labs... in the middle of a battlefield. It's one thing if I'm building a kingdom from nothing (still, why are the surrounding areas empty? why am I not uniting the existing populace?), but it's completely different when I represent the "Allies" against the "Axis" and clearly have been sent into a huge war theater with no supplies, no logistics, no equipment and no intelligence support.

I'm going to draw an arbitrary line in the sand and divide all RTSes according to their relation to that line. To one side are "Historical" RTSes, set in the era of feudal kingdoms when political borders were highly fluid and nation-states were nascent. If your RTS falls into this category, come to terms with the fact that rarely is your land completely empty, nor do you need to set up a "Birthing Center" to yield woodsmen, villagers and soldiers. A more intelligent model is to require me - a small feudal lord - to shrewdly grow my holdings by protecting the residents and thus earning their fealty, forcibly subjugating them, or negotiating fiefdoms with other lords. As I amass power, I come into conflict with larger opposing entities in regions further out, until I eventually rule the entire kingdom. I can then enter into trade or wage war against other nations, and I have to quell internal intrigues and threats of civil war.

In this model, it is reasonable to commission castles to be built, mine resources out of the earth and so forth, as I am essentially engaging in nation building, starting from the grassroots level. I still require delegation, and I require that your game helps me with managing my growing retinue of advisors and underlings by allowing me to tier them and promote and demote them between tiers - and execute and replace them, too!

If, on the other hand, your game is a "Modern" or "Contemporary" RTS, then we're looking at a significantly different model. Unless you have me landing on an alien planet, I shouldn't be building a lot of bases. Instead, especially if my theater is close to a city or large town, I should commandeer use of an existing structure. If my theater is in the jungle/desert/icy wilderness, then I should have camp materials that I can pack up and take with me, so I have a mobile basecamp. Further, I expect my country to have a military industrial complex that is producing needed equipment at home and transporting them to me; my only challenge is creating and protecting avenues for resupply, from commandeering an airport to creating a makeshift one, to finding and protecting a truck route from a neighboring and friendly country.

Remember, in modern times, no country goes to war without a supply chain plan.

Furthermore, politics and statemanship have been divorced from combat to some extent in contemporary statecraft, with a higher degree of role specialization. This means that my objectives may go as far as installing a puppet regime and securing the appearance of legitimacy by policing the first election, but generally no further. It's not my job to build an economic infrastructure; I came in to kick ass and take names! Hoo-rah!


Of course, In My Opinion.

One Word: Z

A few thoughts:

"If you want to get good at RTS games, you should really learn all of the hotkeys."

It's true that current RTS games involve a lot of twitch--and I'll mention Warcraft III specifically since that is about as twitchy a RTS game as you can get right now--but maybe not as much as you'd think.

In Warcraft III, the base building IS the strategy part. But before I go on we need to define terms: I'm using the generally accepted definition of strategy as being the planning before the battle and tactics being the methods by which put than plan into execution.

When you get into a game, you need to consider a lot before you even place your first building: what races are my enemies, what races are my allies, what is map, and from that what is the best strategy to defeat my enemies knowing their races and that of my allies. Furthermore, even once you've decided all that, there is also the placement of your buildings to consider, which will also vary depending on the conditions above. Are you going to lay out your buildings such that they are well protected, or such that they offer good protection for your army and that of your allies? Are you going to make an early expansion? Are you going to place scout buildings to anticipate enemy attacks? Fast-tech or no?

Now with all that said, yes, you could very well provice a mechanism to automatically construct your base or save your own build order, but there are typically so many variables that go into the organization of a base at a high-level of play that such a feature would go unused except for newbies, and could potentially stop them from ever getting better because you are avoiding for them the need to make such strategic considerations.

edit: Another point I'd like to make is that I disagree with the notion that fast games are necessarily lacking in strategy. From my experience, which I'm afraid to say is quite extensive, the best games are those which are evenly matched. A "rush" simply begins the contest of skill--and by skill I refer not just to "twitch" but also stategy and tactics--earlier.

[Edited by - Silex on October 18, 2005 2:21:59 PM]

Now this is a cool idea!

I need to do some schoolwork but I'll be back in a bit with more feedback.

(Nice idea, keep it alive)