next »

When one is making some sort of "artistic" project (and an RTS is something of an art), one must look at other examples to see what they did to succeed (or fail). One of the end results from looking at lots of different RTS games is that you will begin to see a structure among skirmishes, and further study deviates from there. I have developed a very simple SSS for the "big three" RTS games which are most often cloned (AOE, C&C, 'Craft). Analysis to figure out which traits we want will be vital, especially since I think the plan is to have an extremely unique SSS, particularly because of faction diversity.

Common Skirmish Structure:
Start: Each player starts with the bare essentials to start a camp. This is either just a stockpile of resources (C&C) with the means to build new workers, or a very small stockpile of resources, some workers, and the means to create new workers.
Attack: The start of the game primarily based on setting up for the first attack. Any development from there is generally to counter a new attack, or prepare for a greater attack. The start is generally used to prepare for an early attack, either defending or attacking. In a more C&C style, this is 100% accurate, as players construct their turrets and small attack forces. In a craft style, this attack is usually rare and always a rush. In AOE, the start is very clearly seperated from the First Attack by the fact that the player must advance an age.
Development: After the first attack, development is key. There are a lot of variables to take into account, but foremost in importance is following up. In the case of the attacker, he is doing multiple things: A, he could be constructing a defense in case of counter attack, B, following up his attack, or C, developing his forces for a stronger attack. The player chooses depending on which he feels is most effective. The defender is also doing one of three things: A, preparing a counter attack, B, Developing his forces, or C, Repairing his defenses and making them more effective. C is probably the least likely in the average RTS, though in BFME1, it was what Good side was often doing.

This is the standard model. After development is finished, players will use their development to defend, expand, or attack. It continues in a cycle until one player makes the wrong choice and is overwhelmed. This is a very open model, and is not as in depth as my next one (which I'll spend more time on) will be.

I recommend, Darvin specifically, that you guys write a detailed model with each of the factions showing how they A: Start, B: Develop, C: Attack, and D: How the cycle continues.

I HIGHLY recommend a battle report for factions prior to even doing balance. Just having faciton strategy is not enough; having a 1v1 using the cycle between each faction will give me a good idea as to how the numbers should add up. The result is that we will have utter control over gameplay and strategy.

Skirmish start and development is a huge issue.  I've got a good idea of how Balkurn and Varnost should start and develop, but I've got to develop on the other three.  Needless to say, it's not a bad idea that this be my next focus.

Certainly, I think that the start should be the most unique place for the factions. Even Starcraft didn't have a unique start for each faction, and when there were deviations they were very small. Here are some ideas that might be helpful:

In a technical way of looking at things, Varnost (in my opinion) should solely be preparing for an attack. New military camps, more slums etc should be the main focus of the player. Attacks should be repeated, and almost non-stop. As a final note, a Varnost player who is defending should be considered a losing player; A varnost player's defense is his offense, which means that intercepting an incoming attack would be best. Perhaps when attackers reach the Varnost base, some sort of thing is triggered or the base is simply vulnerable, the end result being that the Varnost are hurting.

I don't have a clear picture of the Balkurn, but I think (to be unique) Balkurn should focus it's Attack phase (First attack phase, or the one directly following the start) on defense and build up. In essense, it should have a defensive building. You may say "well, they should attack," (okay, you probably wouldn't say that but in case you do...) to which I reply that Balkurn will have launched an attack force at around the same time as it's opponent. However, it will move so slowly that it won't arrive until after Balkurn has likely been attacked.

[Balkurn]

Starting Units / structures:

Balkurn:  starts with a rudimentry base with some basic defenders.  Their economy takes a while to get rolling, but their army is at full tenacity from the earliest stage of the game, since their strength is in rock hard discipline and well equipped basic soldiers.

Varnost:  starts with a force of warriors (particularly scouts) which can establish camps and settlements about the map, but no structures.  Essentially, Varnost starts with a military rather than an economy, in contrast to the other factions.

Caeluin:  starts with a minor hamlet and a decent force of defenders; starts with higher levels of defences than other factions (accomodating for the fact that Caeluin needs to set up a defensive perimeter and multiple settlements to meet its defensive needs).

Nisyrra: starts with a slave settlement with a military compound at the center, as well as some rudimentary guards.  Their basic economy gets rolling early, but building up the requisit high level structures takes time and effort, forcing them to deal with early skirmishes, or take the fight to the enemy prematurely.

Tharwain: starts with a moderate settlement with an established economy, but no military weapons built up.  Using their draft function, they can defend themselves until they've built up a proper military, but rushing will require either some economic sacrifices or a well executed use of a few good units.



Development Choices:

Balkurn:  essentially, the Balkurn base takes quite some time to fully develop, and choosing which sectors to develop early on will define how you play that part of the game.  Essentially, defence, economy, and military are their focusses.  Even with an established military, Balkurn cannot necessarily attack.  A smaller Balkurn settlement can lose key operations for the duration of a battle - which can cripple it.  Having walls to take the brunt of the offense is critical, both to successful survival of enemy rushes, and in freeing up soldiers for your own rush.  Essentially, the powerful but slow Balkurn army is kept in check by the need for defenses before they can head out.  There's one wild card left: economy.  How you develop your economy while pursuing military and defense will affect how you enter the mid-game. 

Since Balkurn cannot make resource camps, maximizing their mining capacity is critical to keeping up to rival ecnomies, and establishing secondary citadels is crucial to future development; but that requires a strong military to hold the position until walls are up.  Balkurn's ability to continually place fortresses upon the map will make their late game a brutal one if they're successful, allowing them to reinforce besieged settlements from others.  However, their weakness is that the loss of a citadel is a brutal one, since the economic investments are massive.

Varnost:  the exact opposite of Balkurn, Varnost doesn't need to consider what structures they build.  While placement is an issue, their settlements grow rampantly.  Instead, their concern is in keeping their enemies in check.  Establishing offensive staging posts, while their hordes build up is critical.  Varnost has one critical weakness, however; their development is unabating, and it is unsustainable.  They are eventually forced to relieve this pressure by capturing territory or getting men killed.  The two factors keep what is otherwise an unstoppable killing machine balanced; their weaker military forces and settlements can be defeated individually before becoming powerful, and they are eventually forced to play their hands whether or not they're ready to do so.  Essentially, their development is the inverse of the other factions, without checks on economic and military development, but instead they are checked by those things.

Caeluin:  one of Caeluin's unavoidable priorities is setting up multiple settlements.  Their need for metal isn't as great as other factions, and as such instead of simple mining camps they will establish full bases around all their resource points.  They use a network of settlements to control their territory, and their defensive plan revolves around making an enemy's fights fruitless.  A battle for any one Caeluin settlements will result in great losses as the Caeluin forces harass the sieging force, while the other settlements muster their strength to damage the enemy's position while they are occupied.  In this way, despite being outnumbered, Caeluin's flexible economy allows them to survive and prosper on the battlefield.

Caeluin's development is less about economy as it is on their strategic position.  Their military is small, and they rely on establishing a complex network of offensive and defensive staging posts.  They also need to constantly be aware of enemy troop positions and keep their mobile army on the tail of them.  Essentially, they must establish a military position, not an economic one, to compete against a foe which has more power than they do.  Military preparation is their strength, not industry or commerce.

Nisyrra:  development for Nisyrra involves using their slaves to build a powerful forward city to serve as a holy retreat for their warriors.  Their strength is drawn from the magic of their homeland, and building temples and shrines - which takes a great deal of effort beyond mere resource collection - to channel that energy is required to assemble their full forces.  Increasing their economic capacity and supporting more slaves, while protecting them, is Nisyrra's difficulty.

Nisyrra is required to expand upon their military and economic position tactically in order to protect their growing strength while pressuring the enemy.  A defeat on a distant battlefield is meaningless to Nisyrra, as their numbers can be drawn anew.  The loss of temples and shrines will prevent them from fighting effectively in the area, and can reduce their strength substantially.  Essentially, Nisyrra's military is small and powerful, but disposable.  This is their critical advantage; so long as they can replace it in time, the loss of their powerful units is not a great one.

This is VERY well done, Darvin. Your talent at making these things is phenomenal. Now, the only real thing that you need is a prelimnary battle report for each of the factions, preferably with a control. I recommend that the Tharwain faction becomes the control, and a report is written for the other factions vs. Tharwain on the same map under the exact same conditions. The map in mind should be perfectly balanced (ala Tournament Snow). Of course, this is all mental visualization, but it will still come in handy.

A suggestion is that a game that is very easy to mod, such as Warcraft 3, could be used to make a crude "Balance Model" for the game. I think most of the things in game could be implemented, though it may not be worth the challenge to get battalions working. It's just an idea if you are interested, as pretty much anybody with a little time and patience can mod WC3. Regardless, a report is a good place to start, so that I can begin thinking of balancing stats (such as armor types, whether morale could be implemented etc).

The danger with the WC3 editor is that as you're stuck with Jass, which is procedural code not a lot better than the C language in terms of the amount of time you need to put into it (and much more limited in terms of what you can do).  That's a LOT of work effort as you start to make something complex, and even a few relatively minor differences between the final engine and the WC3 interpretation will make balance results in the test useless.  I just don't think it's practical for this.

As for battle reports, I think that's jumping a little ahead.  I'd prefer to give factions their strengths and weaknesses, then see how they interact when we hit the alpha testing stage, which will be some time in the future.

I agree with you on what you are saying; my suggestions are little more than projections for future ideas. Yeah, I was looking at the World Editor again, and I realized that my idea was rather crappy. Of course, I haven't broken out world editor for 2 years now, hehe. I think the first step to success (before strengths and weaknesses) is to have a development tree for each faction in a 1v1 skirmish, and to then convert that to a battle report. Like this (using a generic faction template, though):

Generic Good Empire Faction Skirmish Tree:
Start:
Opt.:
1. A farm and Barracks for an early Generic Footman (Attack)
2. Multiple farms and large numbers of Generic Workers for early economy

Development:
2. (Note that 2 is before 1. The reason is that, if using 2, development comes earlier)
 a. Swordsmen if Rush is suspected
 b. Ranger Hut if rush is not suspected
1.
 a. If rush is succesful:
  i. Boost economy
  ii. Follow up Rush
 b. If rush is unsuccessful
  i. Rebuild defenses for counter attack
  ii. Use time to tech

Etc. The goal of this "graph" is to list common tactics that you want to be faction goals. In the case of the mighty Generic Empire, they have early game flexibility. Obviously, you'll want much more detailed graphs for all of your highly complex and differing factions. Note that the goal of the graph is not to pidgeon hole the player; it's goal is to show how a battle might happen. You may be asking the purpose, of course. Well, in my opinion, the goal is to have an extremely clear idea of how the factions fight and operate. BFME2 suffered from factions who claimed to be one way (Dwarves...) and really were something else (Dwarves...). Also note that I do not necessarily say that you should start working on this straight away. I am simply talking about things that I highly recommend are done at some point prior to the "final version" or the version which has all of the wanted game features nicely packaged.

1.
 a. If rush is succesful:
  i. Boost economy
  ii. Follow up Rush
 b. If rush is unsuccessful
  i. Rebuild defenses for counter attack
  ii. Use time to tech

if you're going to make 'strategy' like that, you're actually removing the strategy. any bot can do that stuff, but the human mind is much more complex and predicting

The point is that this is how "generic faction" works in a skirmish. These are two extremely simple tactics that "generic faction" is good at, and that a human player should be able to do without trouble. This isn't to say that generic faction isn't capable of other things. Like I said, the other factions should have more complex systems. It was an example, not to mention that "generic faction" is extremely boring. Also, the graph was not a complete one.

The reason I'm writing all of this stuff is because I think a very organized structure can be extremely useful, and ensures that the factions do not deviate from their core theme. A good example where themes went to the dark side would be BFME2: A good example within BFME2 would be Dwarves. The theme of slow and strong, as well as highly defensive, was changed into agressive rush with slow (but powerful) fast training units. If you have an idea as to how the faction will operate, than you can better prepare for the future.

If a faction deviates from its core theme, we can evaluate why that happens.  In the case of the dwarves in BFME2, it was because they had the strongest early game units AND they were built the fastest AND they had a way to get into the oppponent's base faster than anyone else.  Is it any wonder they were the ideal rushing faction?

I think you're rushing ahead.  We need to spend more time on the engine.  It's good fun developing ideas like this, but they're not going any further than paper and imagination until that engine is well on its way.

No doubt. Is work on the engine actually starting yet?

Let me put it this way: don't expect anything remotely playable in 2007.

No surprise there. Well, no point in talking about storyline or game play until the engine is playable. We should really knuckle down SOON.

We really need more people.  I've been flipping through some old manuals, and two dozen people dedicated to programming seems to be the norm - and that's in addition to other staff.  Plus we're talking about professionals here, not kids with half (or less) of the training they need and none of the experience.

that's true. i think that making and releasing the pre-project (which is easy enough) will attract some more people

If you're really serious about a pre-project, then I'd recommend finding a revolutionary concept and idea to really attract attention, even if some parts of the game are rushed or lacking.  An original idea and fun gameplay can make people forget many other faults of a game.

ok, i'm open to ideas

what i think would really attract people is a big action fest, with bullets, particles and destroyed pieces flying everywhere

Doing something graphically intensive, or with lots of physics like that would probably be hard.  Not because it's not doable, but because it would likely be laggy.

If you'd indulge me, I could go through my library and pull out half a dozen game concepts I've thrown on to paper in the past.

never thought i'd ask the meaning of an english word, but what's 'indulge'?

anyway, those concepts would be nice. remember that it's likely going to be a scrolling-shooter (cause that is easiest)

Literally, to indulge is like to treat.  So, buying a child a chocolate bar is to indulge him.  "If you'll indulge me" implies that I'm quite happy to list out my ideas.

ok well, i'll be quite happy to hear them, so...

ok  well, i'll be quite happy to hear them, so...

...so he'll indulge you

Anyway, we really need to make a pre-project, to catch people's eyes. And we musn't polish this pre-project off, but make it seriously, and if you (darvin) can give us a really good game concept it can only be better!

I've got five (or rather, four and a half, since one of the concepts is just an extension of the other, but I'm wary of what the work-load may entail), which I will now post in their own thread in the appropriate section.