Hey, guys! I thought I'd make this little thread here, the purpose of which is to teach you guys some of the principles I used to create Crimson Moon and Amber Sun. I'll cover as much as possible (feel free to contribute ideas), spanning everything from terrain, cinematics, story-telling, AI, etc. There won't be much structure, though I'll try to stay on aspect at a time, starting with terrain.
I see a lot of bad terrain. Some of it is acceptable, as in regular Arcade maps, where the purpose of terrain is little to non-existant (though better terrain almost always improves any map), but in campaigns, it is not! Campaigns require the player to be immersed in the story and its world, and bad terrain will not accomplish that.
Now, I'll admit: terrain is hard to get right! I struggle a lot with it, to be honest, and I've learned my share of tricks throughout the many missions I've made. It requires a sense of aesthetics that is hard to learn, and I find it no surprising to learn that the very people who master the editor (usually people with programming talent) doesn't necessarily know how to do it. But I think most people can learn, and I'm happy to be here to teach. I'll start at the basic, showing how you can create the following terrain:
That is very easy to create. There are only three tiles, and only a few doodads are used. But it highlights some of the principles I follow, in a manner of which I will demonstrate now.
The diagonal principle
If you've played the original SC and tried to make maps with it, you might have noticed how vertical and horisontal cliffs simply looked horrible, effectively forcing the mapmaker to create diagonal cliffs. Just look at the bottom right in the following picture, just to the right of the vulture:
That is the closest a SC1 map-maper could get to a vertical cliff. Notice how it's basically just a bunch of corners. Now, take a look at the cliff besides the reaver. It is diagonal, and flows smoothly.
By the way, when I say "horizontal", "vertical" and "diagonal" cliffs, I'm not talking about how cliffs reach up from the ground, but how they look from above. Allow me to demonstrate with pictures from SC2.
Those are horisontal/vertical cliffs. And they look horrible. It looks a chess board or Tetris. See an example of diagonal cliffs below:
It looks a whole lot better, in my opinion. Yes, it is the same cliffs from another angle, but they look a whole lot more immersive. The 3D feel is enhanced, and we've taken another step towards creating campaign terran. Yes, it really is that simple. Which is why it annoys me to see the opposite.
However, we're still not quite where we want to be. Nature is random, and so must we be. The previous two pictures were very stale. By applying a more random approach, we can create something that is more realistic:
Notice how there is very little trace of a pattern (although the diagonal principle is followed). It is all very random, and it feels unplanned. When I create cliffs like that, I don't go in with a plan. I'll brush a little there, raise a little there, lower some over there, until it all looks good. Notice how some of the cliffs are horizontal. That is actually okay. We don't want to remove too many weapons from our arsenal, and horizontal cliffs can work to, though they look best when used in combination with diagonal cliffs.
Using tiles
Frankly, just telling people to use more than one or two tiles is actually a good start for this section. I see too much barren wastelands, which is sad, because, as with cliffs, the way to fix it is really quite simple. When creating out-doors terrain, I usually follow what I call the "layer principle", which is that you have one tile at the bottom, then fill up the middle with another tile, and then alternatively a third one on top of that. Let's look at a prime example of that:
I started out with the darkest ground tile, then added one lighter above that, and the lightest on top. Notice how it forms a path that is aligned with the cliff (though still with a bit of randomness tossed in).
Next, I do the same with the cliffs above the path. Notice how I use the second-lighest tile right beside the cliff. Generally, that is a clever thing to do, because the cliff becomes easier to spot when there are two different tiles between the two levels. This is very important on tilesets like Char, where the tiles are darker and some of them are almost the same color as the cliffs themselves.
I've only used three tiles so far, but they look good. And they flow off each other very well!
Placing doodads
Doodads are a bit trickier than both cliffs and tiles, which are generally very easy to do right once you know how. Doodads take more time, more precision and more planning. I'll primarily use only three doodads to lighten up this image: trees, rocks and brambles.
I start out with placing rocks, starting with rock large. Then I place some trees amongst them (because I know beforehand that I'm making trees, I leave room for them).
In-between, I place some brambles to make it look nicer. I find that if you place a lot of rocks together, it often looks a bit dry. Using brambles is an easy way to mix it up a bit. Let's look at an example.
Without brambles:
With brambles:
See the difference? Good! An easy way to lighted up the map! Place some trees there, some crates, maybe some cracks and a dead body, and it's a real party!
Final result
I continue creating rocks, trees and brambles on the upper cliffs as well. On the top cliff, I isolate them with a fence, though I leave some doodads on the outside. Nature is uneven, and so must our maps be. I place some floodlights behind the fence, because that always looks good.
Notice how the fence work as both an additional cliff, and an alternative one. It is easy to create too many cliffs. Sometimes, using fences, walls and things like that can be a good alternative. Why? Because you generally can see beyond them. Placing stuff like a forest or a collection of corpses behind them (maybe even prisoners, in form of units) can work really well.
To top it off, I place a crane in the distance. The rocks in the doodad looks the same as the regular rocks, which is good for us, because it means it'll flow well with the rest of the map. Don't be afraid of putting a few brambles there as well. I also added a propaganda billboard, a few small rocks on the road, and some cracks.
This will be awesome and should be on the tutorial wiki when it is more complete. I chat with a lot of people who have some interest in making there own campaign. It is pretty overwhelming to come to sc2mapster and find a good jumping off point, not to mention the immensity of the editor. I ended up spending a lot of time watching tutorial videos that really have no usefulness for a beginner level campaigner, eg learning about dialogs. I would say having a programming background is helpful, but like myself, if you just sit down and code, make mistakes and push yourself you will get it. However, if a newbie campaigner had a nice structured list of advice/tutorials, common pitfalls, story production, dedicated alpha testers, then maybe we could boost the scene!
That's about some decent tutorial you have, detailing how you managed to do it and such. One thing that's lacking though, IMO, is a more deeper explanation in the concepts of terraining, of picking terrain, how's it like, how terrain actually influences on gameplay and stuff. I could write something like that, eventually, or I could help you with that.
I’m gonna show you a few pictures now, and I want you to do one thing: take an honest guess of which WOL campaign maps you think I’ve drawn. Don’t spend too much time thinking about what the arrows represent. Just look at the drawings and see if you recognize these missions.
1.
2.
3.
4.
5.
Did you recognize the maps? Well, take a look now:
1. The Evacuation
2. The Dig
3. Smash and Grab
4. Breakout
5. Engine of Destruction
Are you starting to get it? Allow me to demonstrate my point. The idea of the drawings is to strip away all the details of the maps, until all you are left with is the pure basics. (The arrows describe movement, whilst the colors denote the player. It’s really that simple.)
I chose these five maps because their basic structure is so distinct, and so in sync with the mission objectives. Let’s look at The Evacuation again.
The mission essentially consists of two parts, which is reflected in the arrows. First, you are employed with a small team that has to make it to the colonist base. Then the main part begins. Just look how simple it is. Yes, sure, the map is very detailed when considered as a whole, and looks absolutely stunning when you play it, but when you strip all that away, you are essentially left with an arrow facing west, and one facing north-east.
But what can we learn from this? Well, for starters, we can learn to plan ahead and save time. By asking yourself what the purpose of your map is, you can jot down a few arrows of your own and essentially outline the map right then and there on the paper. This is economical, because you won’t waste time creating a map structure that is dissonant with its primary purpose.
But let’s not stay theoretical. Let’s say you wanted to make an escort mission of your own. You sit down with your pen and paper, and maybe you come up with something like this:
What we see here is a simple, snake-like path that you have to traverse in order to safeguard an escort. But, as you might have noticed, I’ve added a little twist, more specifically the smaller, dotted arrows. They represent alternate routes that are blocked by the enemy. With only a few strokes, I’ve given the player the chance to employ an alternate strategy, namely to cut through a more dangerous, but shorter path. And it only took ten seconds to create! By that estimate, we can say that if you had two full minutes at your disposal, you could jot down 12 scenarios.
Imagine how much better your map would be if you took the time to consider the many different ways you could structure it, and think about how little time it would take! Being aware of time is crucial, because creating a map takes a long time, and you don’t want to be making one that is subpar to the one you were only five seconds away from starting to create. Again, and I can’t stress this enough: a map time takes a loooong time to create. Make sure you are creating the one you want to create!
While I will admit to using this strategy far too rarely myself, I almost always outline my maps in some way with just pen and paper before I begin creating them. Because time and space is limited, I find it very useful. In addition, I find that the terrain gets better when I do it too. When I don’t have a plan, I find that the terrain often gets a very random feel, which is not good. You see, there are a lot of components that is influenced by structure, and if we create bad structure, it has a trickle-down effect.
One of the most prominent components is difficulty. Think about how much more difficult you could make a map just by reducing the enemy’s attack path in half, for instance. Or how you could place just a few cliffs by an enemy base and essentially give the player a location to unload his siege tanks without the possibility of repercussions.
Another component is space. If your structure is bad, you might end up wasting good space on nothing, or perhaps realize to late that you haven’t made enough room for an enemy base or a second expansion (I assure that I speak from experience here).
A third component is strategy. If you make an island map, for instance, you are forcing the player to use air units and dropships. One of my Amber Sun maps, Desert Dogs (mission 07) underwent that exact change. The objective of the map is to protect three generators that are positioned away from the main base. At first, the map had no low cliffs, so the player could just walk his reinforcements to the generators, as well as between them, and back to the base. By creating low cliffs, I changed the map completely. Now, the player had to transport his units to the generators. He had to change they way he strategized and how he played.
Imagine that. Just by creating some cliffs, the map suddenly became a lot more intense. I would also say it fixed it. You see, before the change, the map was boring. Afterwards, it became a test of APM endurance. If that wasn’t enough, the improvement of the map actually inspired me to keep working on the map until I reached a point of satisfaction. Inspiration is precious. Don’t waste it.
So find a pen and paper. Think about what mission you want to create. Is it a survival mission? A base destroying mission? An installation mission? This might sound trivial, but these three different scenarios require three different kinds of map structures. Once you’ve realize that, you’re not only well on your way with creating a great map, but you are also far ahead of the guy who is halfway through his “I-make-it-up-as-I-go-along”-map, because he is not far away from realizing that he perhaps made a mistake.
Nice post. Very good to point out time efficiency. Chances are if you spend much idle time working on a map you'll lose interest.
This reminds me of another advice I usually follow: Insert as much of the gameplay as possible before you begin to do terrain to make you sure you won't make drastic changes and do alot of terrain work in vain.
The method I have used during the last couple of missions I've worked on is to paint a gameplay sketch with a 255 increment texture brush around the map to indicate map structure, similar to what you do with pen and paper but only inside the editor. Then I have a small corner on the side of the map where I experiment with doodads, textures and cliffs to create the atmosphere and "feel" of the map. This way I can work on gameplay and terrain at the whim of inspiration without having to do alot of risky work that might have to be redone anyway.
But I have to admit, I am the “I-make-it-up-as-I-go-along”- type so I usually expand the map or make radical adjustments to the gameplay midway. Nevertheless, having these advice's in mind make me more structured and time-efficient then I would have otherwise.
I should say there is value in creating things as you go along too, actually. I mean, you can't plan every detail! It's sort of like playing an instrument. If you improvise occasionally, you might come up with some cool stuff.
I agree that it would probably be best to create most of the gameplay first, but I actually finish most of my terrain first. Making terrain inspires me, and when it starts to get detailed, I get a rush and want to keep going. That's why I often finish certain sections of the map (down to the smallest detail) before I even start on the rest. I guess it has to do with the fact that I find it a bit hard to imagine what the final product will look like when I just place cliffs or something, so I have to create it to see it. I also create the opening cinematic pretty early, just so I can test the map and get a feel of how this all starts out. However, that is something that is good to postpone too, otherwise you're limited to using only a certain part of the map in the cinematic. One of my own favorite openings (AS07) uses the entire map, and I think that's one of the reasons I like it so much.
I've been doing a hybrid of both. For my first map, I just threw down the game play and then built up the terrain and added in cinematics...sorta shooting from the hip most of the time. I think doing it this way for newbies is very important. The editor is just too overwhelming unless you have the perseverance of protoss Templar :)! Laying out the gameplay and just getting something you and others can play through can give you a sense of accomplishment and help you push forward...especially if you are new to programming. However, I only recommend this for your first map, once you got the basics then it is really important to plan.
I think these tutorials should go on the Custom Campaign Initiative Home Page. As well as some of the basic/beginner tutorials from the wiki. What do you guys think?
I place great value on good terrain, as I feel that campaigns, more than any other type of custom map, greatly depends on it for immersion (therefore, few things annoy me more in custom campaigns than bad terrain). When you’re telling a story, good terrain helps the player immerse himself in the worlds you’ve created. In this tutorial I’ll be showing some examples of how you can use terrain to augment your story.
But before I get so far, I’d like to share a little anecdote. If you’ve seen the movie The Shawshank Redemption, you might remember Morgan Freeman playing a convicted felon, who has repeated parole hearings throughout the story that all end in the same way (“denied”). Supposedly, director Frank Darabont would admit that, on set, it hardly seemed like Freeman was doing anything at all. Only when he saw Freeman on screen, where he could bask in the subtle mannerisms of his performance, did he notice how much he actually was doing.
I tell this anecdote to illustrate a point: when you are a mapmaker, you don’t have the privilege of Morgan Freeman. You can’t zoom in on a character and tell the story through his face (be it eyes, ears or mouth). Our characters are small, lifeless creatures. If you don’t have voice acting, you’re even worse off. But that doesn’t mean all is hopeless. You just need to find other ways to tell the story. And that’s where the terrain comes in.
You see, because we do not have the option of subtlety, we are free to go over the top, where we have the opportunity to indulge in all the clichés known to man. So if you are writing a sad or very serious scene, you can just place a rain doodad and make it pour like hell. In movies, using weather to reflect how the characters feel would be considered an eye-roll moment, but in a custom map, it’s a perfect fit. Trust me.
Look no further than the last scene in Crimson Moon for an example (I won’t spoil it for those who haven’t seen it). It is a very serious scene, one where a harsh truth finally finds it way to the surface, and to augment that, I used very obvious tools: really dark lighting, broken buildings, blood splats, and yes, heavy rain. I didn’t go as overboard as that when I first made the scene, and you know what? The scene didn’t work. Only when I went completely overboard did I achieve the feel I wanted.
That doesn’t mean we should go so far, all the time. But it is wise to always consider how you can use terrain, doodads, lighting, and even music to instruct the player on how to feel. Another example would be Roverville, the location that the Amina miners have traveled to in the beginning of Crimson Moon. Theirs is not a successful endeavor, and although I could have easily conveyed that with a tileset such as Mar Sara (which lends itself very easily to a dig site look), I ultimately found that Port Zion was just perfect. Why? Because Port Zion looks like the “rectal end of the universe”, and that’s exactly where I wanted the miners to feel like they were. They are so badly out of their element, and the terrain compliments that perfectly. It’s bad enough that they don’t find the riches they were looking for and have to fight a lot of Zerg, but to have to drink from green water too? Now that really sucks!
On the opposite end of the spectrum, we have two missions where the main characters are coming home. The first is The Shortest Straw from Crimson Moon, and the other is Homecoming from Amber Sun. In both examples, I used tilesets with a much warmer feel (Mar Sara and Zerus, respectively), because I wanted to mirror that “home, sweet home”-feeling. Both tilesets just lend themselves to that in such an effortless way. The same thing can be said of the first Amber Sun mission, The Overlord Hunt, which is set in beautiful shallow water grounds. Here, I wanted to convey a feeling of peace, as the war presumably was over, and the time had come for celebrations. Think of how lost that feeling had been if I had used other tilesets. Imagine if I had used something like Char, or perhaps even Tarsonis! Use the right tileset!
There is a point in the story in Amber Sun where things begin to get a whole lot darker, and what did I do to show that? I set fire to Amina (technically, it wasn’t me, but I am guilty of placing the doodads). What’s great about that is that you don’t have to say that things are down (when you can show something instead of telling it, you do that), because the characters’ homeworld burning says it for you.
These are all some very obvious examples so far. Luckily, there are many other ways to use terrain to augment the story as well, and many of them have a more micro-based focus. For instance, I often decorate certain rooms with blood and corpses (even animal corpses) whenever I want to convey a really dark feel to the story. It’s not subtle, and yet it works, particularly because it’s so suggestive. I mean, if none of the characters say what went down with all these dead people, you are left to speculate exactly what did happen.
For a more mysterious feel, using fog works perfectly, especially if it has the right color. Below is a picture from AS07, where I used blue fog to add a very creepy touch to the lab. It really goes well with the general lighting of the map as well (not to forget the doodads, which consists of stasis tubes with blue water).
And so we can continue. There are so many options. Want to create a big boss battle? Use the Char platforms. One to make a calmer mission? Use Agria. Want to convey a happy feeling? Insert a Mar Sara skybox and point the camera upwards. Place some white birds in the air as well. Whatever you want, think long and hard about what exactly it is that you want to convey with your mission, and try and find the best tileset to go with that, as well as creating the terrain in the fashion that best augments the story. You can also go backwards. Maybe you have a certain image in your head (or even in the editor) that you think look great. You can write a moment, scene or even an entire story out of that image. Hell, I only got the idea for Alzadhar after I saw someone create a map with a nuke-bombed dead city.
Boss battles was one of the cool additions HOTS brought to the SC2 table (although it could be argued that WOL was first with Safe Haven). I had wanted to include bosses in Crimson Moon, but I lacked the necessary skills to do it (not to mention templates to learn from). When creating Amber Sun, however, I finally succeeded in creating not just one, but three boss battles, and I am very pleased with the results. Now, the time has come for me to teach you how to do it, and I am happy to say that it is actually very easy to do!
Contents
A few reminders
Boss context
Creating dodgeable attacks
Aesthetics
Programming the fight
A few reminders
There are many ways to create boss battles, and because they often work better the more they stand out, it is impossible to create a comprehensive guide. So consider this an entry-level tutorial. It is up to you take your boss battle to the next level from here.
Much of what I will teach you here are the results of wild (and frustrating, I might add) experimentation. My forte does not lie in this area of the editor. So although you will be able to replicate the stuff I do here, be aware that I might not always offer the most optimal solution, and I might not always be able to explain in detail what I have done. I might also have done more than I have needed to. Regardless, I have done my best to make sure everything here is as understandable as possible, and I will answer any question you might have as well as I can. Just know that what I present here works (unless I have left something out).
It would probably be best to consider this as much a “making of” as a tutorial. . However, I have to generalize as well as I can. Just be aware that I can’t be holding your hand all the way. I can teach you how to create a good boss battle, but it is up to you to make it great.
This is a very long tutorial. If there is only a specific aspect you want to learn about, feel free to jump ahead.
Boss context
Before we go on to the technical aspects, I want to talk a little bit about the importance of context in boss battles. A boss needs more than just buffing and resizing to make him great; he also needs a standout look and an interesting background. The more he’s a part of the story, the better. This is not the same as saying he needs his own dossier, but rather that he should be given a proper context.
The first of my bosses, Phantom Locust, appears in AS06 – A Rose With Thorns. He comes a bit out of nowhere, which is half the point. We first encounter him (or rather, his voice) when we enter a building and are spotted by a locust, which turns out to be one of his underlings. My intention here was to create tension and to build up to his fight. When we encounter him, we learn that he mans a large siege tank and uses locusts as spotters to compensate for his near-sightedness. In addition, he also resides in quite a dark room that shaped like a cross and heavily fogged. All these details help the player get immersed in the fight. They are not essential, but a good custom campaigner should recognize their importance nonetheless (if you can, always consider giving your boss a name). When the fight is finished, we see what remains of his body lying on the ground, and we learn his background and are actually given the option to sympathize for him (in what I personally consider to be one of the best scenes in the campaign).
Later, in the same mission, we stumble upon the series’ main villain, Ghost Caine, who mans the Geraldus Rose, a humongous battle tank capable of firing several nukes at once. The player has to dodge Rosie’s attacks in the hangar, without the possibility of fighting back. The reason I made it impossible to defeat Rose in this fight was to build her up for the final mission, where she appears again, now as the final boss. By leaving both Caine and his weapon undefeated, I hoped to make the player realize that another fight was inevitable, consequently hinting of what would come later, as well as building tension.
A funny fact: the reason that Phantom Locust is near-sighted and that Rosie only fires nukes is that I hadn’t yet figured out how to create dodgeable missiles. Considering the importance of being able to dodge a boss’ attacks, you could safely say that I had painted myself in a corner, and consequently getting creative when trying to get out. I must say I am pleased with the results. So let that be another lesson: sometimes the best ideas come from unfortunate situations. You should always indulge your creative side when creating bosses.
If you have played AS10, you might have noticed that you have to walk for a short while before facing up with Caine and Rosie again. This was done, once again, to build up tension (as well as give the player ample opportunity and encouragement to save). Do not underestimate the impact such a small decision can have on your boss fight. After all, it is not a far cry from Link climbing the many stairs that leads up to Ganondorf at the end of Ocarina of Time.
I cannot stress the importance of creating a good context for your boss. Think about how he looks, what music you will use, where the fight takes place, what you can make to make it more special, etc. Look to other games for inspiration, or even to the mechanics of StarCraft 2 itself. After all, there are so many units in SC2 that make up for great bosses even if you don’t modify them! How about using a reaper or a colossus, as they can scale cliffs? Or maybe a roach or infestor, as they can move while burrowed? Perhaps give your boss a shield that needs to be disabled before you can attack him? Perhaps have two bosses at once? The possibilities of boss fights are endless, and their greatness is limited only by your imagination.
Creating dodgeable attacks
In this part of the tutorial, I will show you how I created the dodgeable missiles that Geraldus Rose fires at you in AS10. This is actually very easy, but you should be aware that replicating them might not entirely work for you. For instance, the player’s ability to dodge the attacks will depend on the movement of his unit(s). In my case, I set Janus’ speed to 2.75, which is very fast. Had I used his regular speed, he would not have been able to dodge the attacks. Alas, I still don’t know how to slow down missiles (I have to remind you that this part of the editor is out of my comfort zone), though I am happy to learn.
The first thing I did was to change Rose’s attack weapon, which I did by looking up the unit in the editor, and adding the Banshee – Backlash Rockets instead of the regular ground attack. I also added the Colossus - Thermal Lance, although I won’t write much about that here, except to say that by setting its max range to 9 and the minimum range of the rockets to 10, I was able to give Rose the ability to switch effortlessly between the two weapons based on the player’s position without the use of programming. My reasoning behind using the Banshee weapon was that it utilized missiles, so that the player can see what it is he is trying to dodge.
Next, I looked up the weapon’s effects, of which there are three of importance. First is Banshee – Backlash Rockets (Create Persistent), where I changed the Location – Value to Target Point. Secondly is Banshee – Backlash Rockets (Launch Missile), where I changed Impact Location – Value to Target Unit/Point. Now, we are nearly there now, but the most important thing remains. Now we go to Banshee – Backlash Rockets (Damage). I changed the Amount to 24. The Impact Location – Value was already set to Target Unit/Point, so I didn’t need to change it (I stress this fact because it might not be the case if you use another attack).
Lastly, we go to Search. We are still under Damage, although I will be presenting this with Detail View, not Table View. What we will do now is to specify the impact point of the attack, so that it does full damage to the unit if it doesn’t move, and so that it does reduced damage if the unit has removed itself from the impact point, but not far enough to dodge the missile entirely. If we don’t do this, the missile will simply fire into the ground.
First, we will create three values for Area – Arc, all with 360 degrees. This makes sure the unit takes damage no matter what side of the impact it stands. Secondly, we will create three values for Area – Fraction, which specifies how much damage the unit takes, based on its position relative to the impact of the missile. We set the first value to 1, which is full damage (2 would be double), and then the next two to 0.5 and 0.25. Note that you don’t need three values. You can have just two, or even just one (even four), just don’t make it zero. Also note that each value in each Table corresponds to the same ranked value in the other tables. You’ll realize the importance of this when faced with the final table, Area – Radius, where we specify the size of each value’s impact location. I’ve set the first value to 1, and this corresponds to the 1 in Area – Fraction. Then, the next one is 1.5, which is 0.5 in Fraction. The logic here is that the Fraction of the damage is reduced the further away we are from the impact location. If this is hard to understand, look at the Nuke Damage effect, which works the same way. Lastly, I’ve set the final value to 2. Note that by changing the Fraction and Radius, we can indirectly compensate for the player’s unit speed. So a slower unit might need a smaller radius, and a faster one might need a bigger one. All the other values will insert themselves, so this is all we need to do!
Aesthetics
Here, I will show how I changed the aesthetics regarding the boss. First, I found the Banshee Attack Impact under Models, and changed its Model to the Siege Tank Impact (which has a more impressive explosion), and changed its size to 2.0. I also changed the size of Backlash Rockets LMWeapon to 2.0. Lastly, I found Banshee_AttackImpact and changed its sounds to SiegeTank_ImpactMedium7.wav (up to 9). Because the Geraldus Rose is so big, I found that its attack needed to have more beef than the regular banshee attack.
I also sized up the model of the unit itself. I used the Blackhammer Thor, as I found it to look cooler than the regular Thor, but also because it made it stand out from the regular unit. I also added Caine’s portrait (technically, Tosh’s). Lastly, I gave the unit the Build Nuke ability, which makes it look like the nuke is being fired from the unit and not the silo (if you want to do the same, remember to give the unit sufficient nukes by using the Add Ammo To Magazine trigger).
Programming the fight
Now that we have created our boss, it’s time to program his actions, which we do by triggers. This is a bit more complicated, but it’s actually very easy to do (though the difficulty will increase with your ambition). Here, we need an integer variable, which I have called “Geraldus Rose Fight Variants” and set to its default 0. We will be using it to randomize Rosie’s actions.
Next, we create a trigger for the fight itself. Because the fight has only phase, we will only need one. If you want to create several phases for your fight, you should create a trigger for each one, and have an If condition check the boss’ life (or whatever it is you want to separate the phases). I did this for Phantom Locust, but won’t be showing it here. It’s not too hard to do. However, if there is demand for it, I will add it later.
Now, what we need first in the trigger is to make sure it repeats forever. Then, we check if the boss is alive (otherwise we’d be ordering him around post-mortem, which is a bit of information overload for the editor). Next, we randomize the Fight Variants integer. I’ve chosen 1 through 12, as I have 12 different variants for him (a lot of them are almost the same action, except that they order Rosie to different locations). Each of the variants has the same chance to run, which can be used to your advantage. If, for instance, you want your boss to have two attacks, but you want one of them to run more often than the other, you can assign three variants, and have the two first be the same (which means the variant you want to use the most run 2/3 of the time).
I have copied the trigger below. As can be seen, it's only a fraction of the real thing (I included only variant 1), as it is really, really big (see the entire thing in the map file). Note the Wait action at the end of each If action, which separates the variants. Also note that I order Rosie to move and not attack when walking towards locations, as the latter would have made her stop to attack you before reaching her destination. I also used Wait for Condition actions to make sure she didn’t fire her nukes too early.
This was a very long guide, and there are a lot of other things I could have included, but I hoped that you found use for it nonetheless. I left out a few things (like how to create a boss bar), but there are other tutorials for that. Now I hope to see your boss fight sometime in the future!
Contains spoilers for Crimson Moon (up to mission 5)
In this tutorial, I will be looking at a storytelling device that is so common that chances are you have seen it used hundreds of times already without knowing what it’s called. The device is called “Chekhov’s gun”, after what Anton Chekhov said:
“If you show a gun in the first act, you must fire it by the third.”
What he meant by that was that any element of a story should not be included unless it serves a specific purpose. It doesn’t have to be a gun; it could be a character, an event or... well, whatever the hell you want it to be. Today, the term has taken on a new meaning, more specifically “an object or other element that is first introduced as unimportant, but later turns out to be very important to the main narrative”.
Examples
One of my favorite examples is from the movie The Departed, more specifically an envelope labeled “citizens”. It first appears as a largely irrelevant objective, but it then reappears near the end, and when it does, it marks the moment when Leonardo DiCaprio realizes that Matt Damon is the undercover criminal in the police department he has been looking. The important scene here is not the realization itself, but rather the one where the envelope was introduced. Because the audience (like DiCaprio) has seen the envelope before, we can share his shock at seeing it. We already knew that Damon worked undercover, but until now, we didn’t realize how the envelope was important. It is a very chilling moment, and I can still remember my surprised reaction to it when watching the film in the cinema.
Breaking Bad also made good use of Chekhov’s gun at numerous occasions. The primary example would perhaps be the episode “Box Cutter”, where the camera focuses on a box cutter early in the episode, clearly telling us it will be important later, creating tension before it’s finally used (albeit not in the fashion we may have guessed). An even better example would be the ricin cigarette that is introduced in the first season, where the use of it is teased every so often, but then it is actually not used until the fifth season.
Why use it?
Using Chekhov’s gun is a good way to keep your story economical. Stories can get fat pretty quickly, and although it’s not forbidden to have an extraneous element, it is always wise to ask yourself whether something actually should be cut or not. Perhaps you have a character that provides neither plot importance nor any form of entertainment value, or maybe you’ve set up a storyline that ultimately goes nowhere. Your audience is going to be disappointed if you set them up for something you can’t deliver on.
On the flip side, Chekhov’s gun can have a lot of advantages as well. For one, it will make your stories flow better, as the resolutions you create for the third act will have a basis in the first. Try to consider a final showdown between a hero and a villain you’ve never been introduced to. That seems rather odd, doesn’t it? There certainly wouldn’t be a lot of reason to be emotionally invested in such a fight, would there?
How to use it
From my experience, it is best to have a loose perspective on Chekhov’s gun. Don’t interpret it too literally (it doesn’t have to be a gun), and don’t feel pressured to use it in the same way as other people do. There are tons of ways to utilize it, and if you manage to put your own spin on it, you manage to make it your own. Also, be careful that you don’t become a zealot to the rule. Not all elements in a story need to be equally important, and some can be important in their own way. For instance, if you introduce a back story to a character that is not important to the main plot, you don’t have to remove it. After all, the back-story could serve a purpose of its own: maybe it helps us invest in the character, or maybe it rationalizes the choices he commits.
I’ve learned that a good story is sometimes written in reverse. That doesn’t mean you should begin at the ending and end at the beginning, but rather that it would be wise to consider what the direction the story is headed in, and try to anticipate this in some way. Maybe use foreshadowing to hint at a coming event, or perhaps create a character that is wholly different to how you want him to end up, essentially setting him up for a sudden or gradual change.
Did you know that Chekhov’s gun is actually the reason why Crimson Moon has ten missions instead of five? At first, the campaign had half as many missions as it has now, focusing primarily on Janus instead of the miners, but when I decided to give the miners a bigger role, the story got twice as long. The reason why I wanted to give more room for the miners was that I had written a scene I was particular scene I was quite proud of: the death scene of Brother Tyrone. It's one that plays heavily on Chekhov's gun.
Brother Tyrone was, as you might or might not remember, the leader of the Roverville miners. He was constantly shown berating sergeant Henderson, and the idea behind this was that, at one point, Henderson would finally have had enough of being belittled, and he would abandon Tyrone when Tyrone actually needed him the most. What happens in the story is that Tyrone first insults Henderson one time too many, then asks him to stand guard, which Henderson neglects doing, instead going to relieve himself. This leaves Tyrone unguarded when Quarinius comes knocking, and he is killed.
This particular scene wouldn’t have worked if I hadn’t laid out the groundwork first. It wouldn’t have made sense that Henderson chose to abandon his duties if I hadn’t established that he didn’t care much for Tyrone. However, because we had seen Henderson repeatedly being berated by Tyrone, we understand it completely when he decides to ignore Tyrone’s order. The scene also becomes much powerful because of this, not only because the story has secretly been leading up to it so far, but also because it marks the moment where Henderson finally realizes he needs to take charge.
How to play with it
Today, Chekhov’s gun is so common that a seasoned viewer will have no trouble spotting it a mile away. “The camera focuses on a particular object? Why, it must be important!” This is unfortunate if the point if we were to realize the importance of the “gun” in retrospect only. If the writers don’t want us to know where the story is headed, but we guess it anyway, that is unfortunate. Luckily, knowing this gives us the ability to play with the device.
One way of doing that is to actually break the rule, or actually realize that it’s not a rule at all (because there are no rules in storytelling). This might contradict what I said before, which is why I find it important to remind you that I am offering advice, not telling you what to do. If you want to break the rule, what you might want to do is set up a gun and imply that it will be important to the audience, and then have it turn out to be insignificant. This can act as a set up for some nice plot twists, and it can simply be used to throw the audience off-balance. Perhaps you had a second, better hidden gun, or perhaps you wanted to lead the audience down one path only so they wouldn't realize they were heading in another. Just be careful, though. Even if you can get away with this, there are good and bad ways of doing it.
+1 We need more story-related tutorials like this.
Quote:
I’ve learned that a good story is sometimes written in reverse. That doesn’t mean you should begin at the ending and end at the beginning, but rather that it would be wise to consider what the direction the story is headed in, and try to anticipate this in some way. Maybe use foreshadowing to hint at a coming event, or perhaps create a character that is wholly different to how you want him to end up, essentially setting him up for a sudden or gradual change.
Did you know that Chekhov’s gun is actually the reason why Crimson Moon has ten missions instead of five? At first, the campaign had half as many missions as it has now, focusing primarily on Janus instead of the miners, but when I decided to give the miners a bigger role, the story got twice as long. The reason why I wanted to give more room for the miners was that I had written a scene I was particular scene I was quite proud of: the death scene of Brother Tyrone. It's one that plays heavily on Chekhov's gun.
Brother Tyrone was, as you might or might not remember, the leader of the Roverville miners. He was constantly shown berating sergeant Henderson, and the idea behind this was that, at one point, Henderson would finally have had enough of being belittled, and he would abandon Tyrone when Tyrone actually needed him the most. What happens in the story is that Tyrone first insults Henderson one time too many, then asks him to stand guard, which Henderson neglects doing, instead going to relieve himself. This leaves Tyrone unguarded when Quarinius comes knocking, and he is killed.
To nitpick, this isn't really an example of Chekov's gun. Lots of people interpret Chekov's gun as just mere foreshadowing, but that's not really true. It's not about what should be included in a story, it's about what shouldn't be included. It's about keeping the story economical like you said. That's why the phrase is "If in the first act you have hung a pistol on the wall, then in the following one it should be fired. Otherwise don’t put it there" not "If a pistol will be fired in the second act, it ought to be hung on the wall in the first act."
Likewise, you didn't necessarily have Tyrone killed as a direct result of pestering Henderson; it would be more accurate to say that you had Tyrone berate Henderson because you knew you wanted him to die. In fact, laying the necessary groundwork only made the story larger by your admission, so it's not really an example Chekov's gun.
Quote:
Not all elements in a story need to be equally important, and some can be important in their own way. For instance, if you introduce a back story to a character that is not important to the main plot, you don’t have to remove it. After all, the back-story could serve a purpose of its own: maybe it helps us invest in the character, or maybe it rationalizes the choices he commits.
This is why I'm not the biggest proponent of Chekov's gun. I'm all for minimalistic storytelling, but sometimes you just want to have something in a story for its own intrinsic sake and the aesthetic of realism.
But it doesn't matter what came first in writing, because to the audience, the order is the same. B follows A, even if A originated from B. What I wanted to emphasize was that the Henderson-Tyrone conflict has a point to the story before Tyrone's death, just not in the way the audience is supposed to be aware of. I get what you're saying, though.
And I agree that you shouldn't always follow Chekhov's gun, which is why I said you shouldn't be a zealot. In fact, I am a big fan of the way David Chase played with anti-climaxes in The Sopranos and often left plot threads hanging (the russian, the ending), but while it made watching the show a lot more interesting than its counterparts, it also made it a frustrating experience to watch sometimes.
Anyway, I'll try to include more story-telling tutorials. Any particular you want seen? I thought about writing about plot twists.
The problem is that it's a storytelling technique, yet people use it as a plot device ie. this object or that object is an example of Chekov's gun. People have deviated from Chekov's original advice of merely not including into the story anything that serves no purpose.
Good storyline tutorial. You bring up some interesting points and it is very inspiring. This gave me some useful ideas as to how I can progress the story of my own campaign in the later missions, and it turns out that I have actually used this method already without even knowing it ^_^
I like how you bring us some insight in how these were applied through your own stories and other ones as well. This helps us have a better understanding of how it works.
Btw, I agree on the plot-twist tutorial. That would be incredibly useful for a lot of people. Me included, even though I think I have the basics of it down, and I also have a couple planned, it's always nice to get ideas :D
Rollback Post to RevisionRollBack
StarCraft II: Annihilation Campaign - A 24 mission single-player campaign featuring unique heroes, custom tracks, boss battles, a varied set of mission scenarios and three intersecting storylines.
Nice to see a storytelling tutorial, should prove useful. :) I got a grasp on the concept, should probably look into it deeper though. Looking forward to more story-related tutorials!
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Hey, guys! I thought I'd make this little thread here, the purpose of which is to teach you guys some of the principles I used to create Crimson Moon and Amber Sun. I'll cover as much as possible (feel free to contribute ideas), spanning everything from terrain, cinematics, story-telling, AI, etc. There won't be much structure, though I'll try to stay on aspect at a time, starting with terrain.
Overview
Terrain
AI
Story
Terraining principles
I see a lot of bad terrain. Some of it is acceptable, as in regular Arcade maps, where the purpose of terrain is little to non-existant (though better terrain almost always improves any map), but in campaigns, it is not! Campaigns require the player to be immersed in the story and its world, and bad terrain will not accomplish that.
Now, I'll admit: terrain is hard to get right! I struggle a lot with it, to be honest, and I've learned my share of tricks throughout the many missions I've made. It requires a sense of aesthetics that is hard to learn, and I find it no surprising to learn that the very people who master the editor (usually people with programming talent) doesn't necessarily know how to do it. But I think most people can learn, and I'm happy to be here to teach. I'll start at the basic, showing how you can create the following terrain:
That is very easy to create. There are only three tiles, and only a few doodads are used. But it highlights some of the principles I follow, in a manner of which I will demonstrate now.
The diagonal principle
If you've played the original SC and tried to make maps with it, you might have noticed how vertical and horisontal cliffs simply looked horrible, effectively forcing the mapmaker to create diagonal cliffs. Just look at the bottom right in the following picture, just to the right of the vulture:
That is the closest a SC1 map-maper could get to a vertical cliff. Notice how it's basically just a bunch of corners. Now, take a look at the cliff besides the reaver. It is diagonal, and flows smoothly.
By the way, when I say "horizontal", "vertical" and "diagonal" cliffs, I'm not talking about how cliffs reach up from the ground, but how they look from above. Allow me to demonstrate with pictures from SC2.
Those are horisontal/vertical cliffs. And they look horrible. It looks a chess board or Tetris. See an example of diagonal cliffs below:
It looks a whole lot better, in my opinion. Yes, it is the same cliffs from another angle, but they look a whole lot more immersive. The 3D feel is enhanced, and we've taken another step towards creating campaign terran. Yes, it really is that simple. Which is why it annoys me to see the opposite.
However, we're still not quite where we want to be. Nature is random, and so must we be. The previous two pictures were very stale. By applying a more random approach, we can create something that is more realistic:
Notice how there is very little trace of a pattern (although the diagonal principle is followed). It is all very random, and it feels unplanned. When I create cliffs like that, I don't go in with a plan. I'll brush a little there, raise a little there, lower some over there, until it all looks good. Notice how some of the cliffs are horizontal. That is actually okay. We don't want to remove too many weapons from our arsenal, and horizontal cliffs can work to, though they look best when used in combination with diagonal cliffs.
Using tiles
Frankly, just telling people to use more than one or two tiles is actually a good start for this section. I see too much barren wastelands, which is sad, because, as with cliffs, the way to fix it is really quite simple. When creating out-doors terrain, I usually follow what I call the "layer principle", which is that you have one tile at the bottom, then fill up the middle with another tile, and then alternatively a third one on top of that. Let's look at a prime example of that:
I started out with the darkest ground tile, then added one lighter above that, and the lightest on top. Notice how it forms a path that is aligned with the cliff (though still with a bit of randomness tossed in).
Next, I do the same with the cliffs above the path. Notice how I use the second-lighest tile right beside the cliff. Generally, that is a clever thing to do, because the cliff becomes easier to spot when there are two different tiles between the two levels. This is very important on tilesets like Char, where the tiles are darker and some of them are almost the same color as the cliffs themselves.
I've only used three tiles so far, but they look good. And they flow off each other very well!
Placing doodads
Doodads are a bit trickier than both cliffs and tiles, which are generally very easy to do right once you know how. Doodads take more time, more precision and more planning. I'll primarily use only three doodads to lighten up this image: trees, rocks and brambles.
I start out with placing rocks, starting with rock large. Then I place some trees amongst them (because I know beforehand that I'm making trees, I leave room for them).
In-between, I place some brambles to make it look nicer. I find that if you place a lot of rocks together, it often looks a bit dry. Using brambles is an easy way to mix it up a bit. Let's look at an example.
Without brambles:
With brambles:
See the difference? Good! An easy way to lighted up the map! Place some trees there, some crates, maybe some cracks and a dead body, and it's a real party!
Final result
I continue creating rocks, trees and brambles on the upper cliffs as well. On the top cliff, I isolate them with a fence, though I leave some doodads on the outside. Nature is uneven, and so must our maps be. I place some floodlights behind the fence, because that always looks good.
Notice how the fence work as both an additional cliff, and an alternative one. It is easy to create too many cliffs. Sometimes, using fences, walls and things like that can be a good alternative. Why? Because you generally can see beyond them. Placing stuff like a forest or a collection of corpses behind them (maybe even prisoners, in form of units) can work really well.
To top it off, I place a crane in the distance. The rocks in the doodad looks the same as the regular rocks, which is good for us, because it means it'll flow well with the rest of the map. Don't be afraid of putting a few brambles there as well. I also added a propaganda billboard, a few small rocks on the road, and some cracks.
And that's it!
Wow really cool job! I can't wait to see what you do With all of this. Looks like it will be very useful to everyone.
Good job! Looking forward to more from you.
@EivindL: Go
This will be awesome and should be on the tutorial wiki when it is more complete. I chat with a lot of people who have some interest in making there own campaign. It is pretty overwhelming to come to sc2mapster and find a good jumping off point, not to mention the immensity of the editor. I ended up spending a lot of time watching tutorial videos that really have no usefulness for a beginner level campaigner, eg learning about dialogs. I would say having a programming background is helpful, but like myself, if you just sit down and code, make mistakes and push yourself you will get it. However, if a newbie campaigner had a nice structured list of advice/tutorials, common pitfalls, story production, dedicated alpha testers, then maybe we could boost the scene!
That's about some decent tutorial you have, detailing how you managed to do it and such. One thing that's lacking though, IMO, is a more deeper explanation in the concepts of terraining, of picking terrain, how's it like, how terrain actually influences on gameplay and stuff. I could write something like that, eventually, or I could help you with that.
Also, I've just started something concerning AI that you can read here, if you'd like: http:www.sc2mapster.com/forums/resources/tutorials/62267-trigger-ai-making-your-own-sky-net/#p1
Yeah, sure. We'll call it a "guest tutorial". It'll be awesome! :D
I have more terrain tutorials coming up, though. I just wanted to explain some principles and basic stuff to get it started.
Map structure
I’m gonna show you a few pictures now, and I want you to do one thing: take an honest guess of which WOL campaign maps you think I’ve drawn. Don’t spend too much time thinking about what the arrows represent. Just look at the drawings and see if you recognize these missions.
1.
2.
3.
4.
5.
Did you recognize the maps? Well, take a look now:
1. The Evacuation
2. The Dig
3. Smash and Grab
4. Breakout
5. Engine of Destruction
Are you starting to get it? Allow me to demonstrate my point. The idea of the drawings is to strip away all the details of the maps, until all you are left with is the pure basics. (The arrows describe movement, whilst the colors denote the player. It’s really that simple.)
I chose these five maps because their basic structure is so distinct, and so in sync with the mission objectives. Let’s look at The Evacuation again.
The mission essentially consists of two parts, which is reflected in the arrows. First, you are employed with a small team that has to make it to the colonist base. Then the main part begins. Just look how simple it is. Yes, sure, the map is very detailed when considered as a whole, and looks absolutely stunning when you play it, but when you strip all that away, you are essentially left with an arrow facing west, and one facing north-east.
But what can we learn from this? Well, for starters, we can learn to plan ahead and save time. By asking yourself what the purpose of your map is, you can jot down a few arrows of your own and essentially outline the map right then and there on the paper. This is economical, because you won’t waste time creating a map structure that is dissonant with its primary purpose.
But let’s not stay theoretical. Let’s say you wanted to make an escort mission of your own. You sit down with your pen and paper, and maybe you come up with something like this:
What we see here is a simple, snake-like path that you have to traverse in order to safeguard an escort. But, as you might have noticed, I’ve added a little twist, more specifically the smaller, dotted arrows. They represent alternate routes that are blocked by the enemy. With only a few strokes, I’ve given the player the chance to employ an alternate strategy, namely to cut through a more dangerous, but shorter path. And it only took ten seconds to create! By that estimate, we can say that if you had two full minutes at your disposal, you could jot down 12 scenarios. Imagine how much better your map would be if you took the time to consider the many different ways you could structure it, and think about how little time it would take! Being aware of time is crucial, because creating a map takes a long time, and you don’t want to be making one that is subpar to the one you were only five seconds away from starting to create. Again, and I can’t stress this enough: a map time takes a loooong time to create. Make sure you are creating the one you want to create!
While I will admit to using this strategy far too rarely myself, I almost always outline my maps in some way with just pen and paper before I begin creating them. Because time and space is limited, I find it very useful. In addition, I find that the terrain gets better when I do it too. When I don’t have a plan, I find that the terrain often gets a very random feel, which is not good. You see, there are a lot of components that is influenced by structure, and if we create bad structure, it has a trickle-down effect.
One of the most prominent components is difficulty. Think about how much more difficult you could make a map just by reducing the enemy’s attack path in half, for instance. Or how you could place just a few cliffs by an enemy base and essentially give the player a location to unload his siege tanks without the possibility of repercussions.
Another component is space. If your structure is bad, you might end up wasting good space on nothing, or perhaps realize to late that you haven’t made enough room for an enemy base or a second expansion (I assure that I speak from experience here).
A third component is strategy. If you make an island map, for instance, you are forcing the player to use air units and dropships. One of my Amber Sun maps, Desert Dogs (mission 07) underwent that exact change. The objective of the map is to protect three generators that are positioned away from the main base. At first, the map had no low cliffs, so the player could just walk his reinforcements to the generators, as well as between them, and back to the base. By creating low cliffs, I changed the map completely. Now, the player had to transport his units to the generators. He had to change they way he strategized and how he played.
Imagine that. Just by creating some cliffs, the map suddenly became a lot more intense. I would also say it fixed it. You see, before the change, the map was boring. Afterwards, it became a test of APM endurance. If that wasn’t enough, the improvement of the map actually inspired me to keep working on the map until I reached a point of satisfaction. Inspiration is precious. Don’t waste it.
So find a pen and paper. Think about what mission you want to create. Is it a survival mission? A base destroying mission? An installation mission? This might sound trivial, but these three different scenarios require three different kinds of map structures. Once you’ve realize that, you’re not only well on your way with creating a great map, but you are also far ahead of the guy who is halfway through his “I-make-it-up-as-I-go-along”-map, because he is not far away from realizing that he perhaps made a mistake.
Nice post. Very good to point out time efficiency. Chances are if you spend much idle time working on a map you'll lose interest.
This reminds me of another advice I usually follow: Insert as much of the gameplay as possible before you begin to do terrain to make you sure you won't make drastic changes and do alot of terrain work in vain.
The method I have used during the last couple of missions I've worked on is to paint a gameplay sketch with a 255 increment texture brush around the map to indicate map structure, similar to what you do with pen and paper but only inside the editor. Then I have a small corner on the side of the map where I experiment with doodads, textures and cliffs to create the atmosphere and "feel" of the map. This way I can work on gameplay and terrain at the whim of inspiration without having to do alot of risky work that might have to be redone anyway.
But I have to admit, I am the “I-make-it-up-as-I-go-along”- type so I usually expand the map or make radical adjustments to the gameplay midway. Nevertheless, having these advice's in mind make me more structured and time-efficient then I would have otherwise.
I should say there is value in creating things as you go along too, actually. I mean, you can't plan every detail! It's sort of like playing an instrument. If you improvise occasionally, you might come up with some cool stuff.
I agree that it would probably be best to create most of the gameplay first, but I actually finish most of my terrain first. Making terrain inspires me, and when it starts to get detailed, I get a rush and want to keep going. That's why I often finish certain sections of the map (down to the smallest detail) before I even start on the rest. I guess it has to do with the fact that I find it a bit hard to imagine what the final product will look like when I just place cliffs or something, so I have to create it to see it. I also create the opening cinematic pretty early, just so I can test the map and get a feel of how this all starts out. However, that is something that is good to postpone too, otherwise you're limited to using only a certain part of the map in the cinematic. One of my own favorite openings (AS07) uses the entire map, and I think that's one of the reasons I like it so much.
I've been doing a hybrid of both. For my first map, I just threw down the game play and then built up the terrain and added in cinematics...sorta shooting from the hip most of the time. I think doing it this way for newbies is very important. The editor is just too overwhelming unless you have the perseverance of protoss Templar :)! Laying out the gameplay and just getting something you and others can play through can give you a sense of accomplishment and help you push forward...especially if you are new to programming. However, I only recommend this for your first map, once you got the basics then it is really important to plan.
I think these tutorials should go on the Custom Campaign Initiative Home Page. As well as some of the basic/beginner tutorials from the wiki. What do you guys think?
Using terrain to augment the story
I place great value on good terrain, as I feel that campaigns, more than any other type of custom map, greatly depends on it for immersion (therefore, few things annoy me more in custom campaigns than bad terrain). When you’re telling a story, good terrain helps the player immerse himself in the worlds you’ve created. In this tutorial I’ll be showing some examples of how you can use terrain to augment your story.
But before I get so far, I’d like to share a little anecdote. If you’ve seen the movie The Shawshank Redemption, you might remember Morgan Freeman playing a convicted felon, who has repeated parole hearings throughout the story that all end in the same way (“denied”). Supposedly, director Frank Darabont would admit that, on set, it hardly seemed like Freeman was doing anything at all. Only when he saw Freeman on screen, where he could bask in the subtle mannerisms of his performance, did he notice how much he actually was doing.
I tell this anecdote to illustrate a point: when you are a mapmaker, you don’t have the privilege of Morgan Freeman. You can’t zoom in on a character and tell the story through his face (be it eyes, ears or mouth). Our characters are small, lifeless creatures. If you don’t have voice acting, you’re even worse off. But that doesn’t mean all is hopeless. You just need to find other ways to tell the story. And that’s where the terrain comes in.
You see, because we do not have the option of subtlety, we are free to go over the top, where we have the opportunity to indulge in all the clichés known to man. So if you are writing a sad or very serious scene, you can just place a rain doodad and make it pour like hell. In movies, using weather to reflect how the characters feel would be considered an eye-roll moment, but in a custom map, it’s a perfect fit. Trust me.
Look no further than the last scene in Crimson Moon for an example (I won’t spoil it for those who haven’t seen it). It is a very serious scene, one where a harsh truth finally finds it way to the surface, and to augment that, I used very obvious tools: really dark lighting, broken buildings, blood splats, and yes, heavy rain. I didn’t go as overboard as that when I first made the scene, and you know what? The scene didn’t work. Only when I went completely overboard did I achieve the feel I wanted.
That doesn’t mean we should go so far, all the time. But it is wise to always consider how you can use terrain, doodads, lighting, and even music to instruct the player on how to feel. Another example would be Roverville, the location that the Amina miners have traveled to in the beginning of Crimson Moon. Theirs is not a successful endeavor, and although I could have easily conveyed that with a tileset such as Mar Sara (which lends itself very easily to a dig site look), I ultimately found that Port Zion was just perfect. Why? Because Port Zion looks like the “rectal end of the universe”, and that’s exactly where I wanted the miners to feel like they were. They are so badly out of their element, and the terrain compliments that perfectly. It’s bad enough that they don’t find the riches they were looking for and have to fight a lot of Zerg, but to have to drink from green water too? Now that really sucks!
On the opposite end of the spectrum, we have two missions where the main characters are coming home. The first is The Shortest Straw from Crimson Moon, and the other is Homecoming from Amber Sun. In both examples, I used tilesets with a much warmer feel (Mar Sara and Zerus, respectively), because I wanted to mirror that “home, sweet home”-feeling. Both tilesets just lend themselves to that in such an effortless way. The same thing can be said of the first Amber Sun mission, The Overlord Hunt, which is set in beautiful shallow water grounds. Here, I wanted to convey a feeling of peace, as the war presumably was over, and the time had come for celebrations. Think of how lost that feeling had been if I had used other tilesets. Imagine if I had used something like Char, or perhaps even Tarsonis! Use the right tileset!
There is a point in the story in Amber Sun where things begin to get a whole lot darker, and what did I do to show that? I set fire to Amina (technically, it wasn’t me, but I am guilty of placing the doodads). What’s great about that is that you don’t have to say that things are down (when you can show something instead of telling it, you do that), because the characters’ homeworld burning says it for you.
These are all some very obvious examples so far. Luckily, there are many other ways to use terrain to augment the story as well, and many of them have a more micro-based focus. For instance, I often decorate certain rooms with blood and corpses (even animal corpses) whenever I want to convey a really dark feel to the story. It’s not subtle, and yet it works, particularly because it’s so suggestive. I mean, if none of the characters say what went down with all these dead people, you are left to speculate exactly what did happen.
For a more mysterious feel, using fog works perfectly, especially if it has the right color. Below is a picture from AS07, where I used blue fog to add a very creepy touch to the lab. It really goes well with the general lighting of the map as well (not to forget the doodads, which consists of stasis tubes with blue water).
And so we can continue. There are so many options. Want to create a big boss battle? Use the Char platforms. One to make a calmer mission? Use Agria. Want to convey a happy feeling? Insert a Mar Sara skybox and point the camera upwards. Place some white birds in the air as well. Whatever you want, think long and hard about what exactly it is that you want to convey with your mission, and try and find the best tileset to go with that, as well as creating the terrain in the fashion that best augments the story. You can also go backwards. Maybe you have a certain image in your head (or even in the editor) that you think look great. You can write a moment, scene or even an entire story out of that image. Hell, I only got the idea for Alzadhar after I saw someone create a map with a nuke-bombed dead city.
These are beautiful tutorials and very effective, I am glad you do them.
Creating boss battles
Boss battles was one of the cool additions HOTS brought to the SC2 table (although it could be argued that WOL was first with Safe Haven). I had wanted to include bosses in Crimson Moon, but I lacked the necessary skills to do it (not to mention templates to learn from). When creating Amber Sun, however, I finally succeeded in creating not just one, but three boss battles, and I am very pleased with the results. Now, the time has come for me to teach you how to do it, and I am happy to say that it is actually very easy to do!
Contents
A few reminders
Boss context
Before we go on to the technical aspects, I want to talk a little bit about the importance of context in boss battles. A boss needs more than just buffing and resizing to make him great; he also needs a standout look and an interesting background. The more he’s a part of the story, the better. This is not the same as saying he needs his own dossier, but rather that he should be given a proper context.
The first of my bosses, Phantom Locust, appears in AS06 – A Rose With Thorns. He comes a bit out of nowhere, which is half the point. We first encounter him (or rather, his voice) when we enter a building and are spotted by a locust, which turns out to be one of his underlings. My intention here was to create tension and to build up to his fight. When we encounter him, we learn that he mans a large siege tank and uses locusts as spotters to compensate for his near-sightedness. In addition, he also resides in quite a dark room that shaped like a cross and heavily fogged. All these details help the player get immersed in the fight. They are not essential, but a good custom campaigner should recognize their importance nonetheless (if you can, always consider giving your boss a name). When the fight is finished, we see what remains of his body lying on the ground, and we learn his background and are actually given the option to sympathize for him (in what I personally consider to be one of the best scenes in the campaign).
Later, in the same mission, we stumble upon the series’ main villain, Ghost Caine, who mans the Geraldus Rose, a humongous battle tank capable of firing several nukes at once. The player has to dodge Rosie’s attacks in the hangar, without the possibility of fighting back. The reason I made it impossible to defeat Rose in this fight was to build her up for the final mission, where she appears again, now as the final boss. By leaving both Caine and his weapon undefeated, I hoped to make the player realize that another fight was inevitable, consequently hinting of what would come later, as well as building tension.
A funny fact: the reason that Phantom Locust is near-sighted and that Rosie only fires nukes is that I hadn’t yet figured out how to create dodgeable missiles. Considering the importance of being able to dodge a boss’ attacks, you could safely say that I had painted myself in a corner, and consequently getting creative when trying to get out. I must say I am pleased with the results. So let that be another lesson: sometimes the best ideas come from unfortunate situations. You should always indulge your creative side when creating bosses.
If you have played AS10, you might have noticed that you have to walk for a short while before facing up with Caine and Rosie again. This was done, once again, to build up tension (as well as give the player ample opportunity and encouragement to save). Do not underestimate the impact such a small decision can have on your boss fight. After all, it is not a far cry from Link climbing the many stairs that leads up to Ganondorf at the end of Ocarina of Time.
I cannot stress the importance of creating a good context for your boss. Think about how he looks, what music you will use, where the fight takes place, what you can make to make it more special, etc. Look to other games for inspiration, or even to the mechanics of StarCraft 2 itself. After all, there are so many units in SC2 that make up for great bosses even if you don’t modify them! How about using a reaper or a colossus, as they can scale cliffs? Or maybe a roach or infestor, as they can move while burrowed? Perhaps give your boss a shield that needs to be disabled before you can attack him? Perhaps have two bosses at once? The possibilities of boss fights are endless, and their greatness is limited only by your imagination.
Creating dodgeable attacks
In this part of the tutorial, I will show you how I created the dodgeable missiles that Geraldus Rose fires at you in AS10. This is actually very easy, but you should be aware that replicating them might not entirely work for you. For instance, the player’s ability to dodge the attacks will depend on the movement of his unit(s). In my case, I set Janus’ speed to 2.75, which is very fast. Had I used his regular speed, he would not have been able to dodge the attacks. Alas, I still don’t know how to slow down missiles (I have to remind you that this part of the editor is out of my comfort zone), though I am happy to learn.
The first thing I did was to change Rose’s attack weapon, which I did by looking up the unit in the editor, and adding the Banshee – Backlash Rockets instead of the regular ground attack. I also added the Colossus - Thermal Lance, although I won’t write much about that here, except to say that by setting its max range to 9 and the minimum range of the rockets to 10, I was able to give Rose the ability to switch effortlessly between the two weapons based on the player’s position without the use of programming. My reasoning behind using the Banshee weapon was that it utilized missiles, so that the player can see what it is he is trying to dodge.
Next, I looked up the weapon’s effects, of which there are three of importance. First is Banshee – Backlash Rockets (Create Persistent), where I changed the Location – Value to Target Point. Secondly is Banshee – Backlash Rockets (Launch Missile), where I changed Impact Location – Value to Target Unit/Point. Now, we are nearly there now, but the most important thing remains. Now we go to Banshee – Backlash Rockets (Damage). I changed the Amount to 24. The Impact Location – Value was already set to Target Unit/Point, so I didn’t need to change it (I stress this fact because it might not be the case if you use another attack).
Lastly, we go to Search. We are still under Damage, although I will be presenting this with Detail View, not Table View. What we will do now is to specify the impact point of the attack, so that it does full damage to the unit if it doesn’t move, and so that it does reduced damage if the unit has removed itself from the impact point, but not far enough to dodge the missile entirely. If we don’t do this, the missile will simply fire into the ground.
First, we will create three values for Area – Arc, all with 360 degrees. This makes sure the unit takes damage no matter what side of the impact it stands. Secondly, we will create three values for Area – Fraction, which specifies how much damage the unit takes, based on its position relative to the impact of the missile. We set the first value to 1, which is full damage (2 would be double), and then the next two to 0.5 and 0.25. Note that you don’t need three values. You can have just two, or even just one (even four), just don’t make it zero. Also note that each value in each Table corresponds to the same ranked value in the other tables. You’ll realize the importance of this when faced with the final table, Area – Radius, where we specify the size of each value’s impact location. I’ve set the first value to 1, and this corresponds to the 1 in Area – Fraction. Then, the next one is 1.5, which is 0.5 in Fraction. The logic here is that the Fraction of the damage is reduced the further away we are from the impact location. If this is hard to understand, look at the Nuke Damage effect, which works the same way. Lastly, I’ve set the final value to 2. Note that by changing the Fraction and Radius, we can indirectly compensate for the player’s unit speed. So a slower unit might need a smaller radius, and a faster one might need a bigger one. All the other values will insert themselves, so this is all we need to do!
Aesthetics
Here, I will show how I changed the aesthetics regarding the boss. First, I found the Banshee Attack Impact under Models, and changed its Model to the Siege Tank Impact (which has a more impressive explosion), and changed its size to 2.0. I also changed the size of Backlash Rockets LMWeapon to 2.0. Lastly, I found Banshee_AttackImpact and changed its sounds to SiegeTank_ImpactMedium7.wav (up to 9). Because the Geraldus Rose is so big, I found that its attack needed to have more beef than the regular banshee attack.
I also sized up the model of the unit itself. I used the Blackhammer Thor, as I found it to look cooler than the regular Thor, but also because it made it stand out from the regular unit. I also added Caine’s portrait (technically, Tosh’s). Lastly, I gave the unit the Build Nuke ability, which makes it look like the nuke is being fired from the unit and not the silo (if you want to do the same, remember to give the unit sufficient nukes by using the Add Ammo To Magazine trigger).
Programming the fight
Now that we have created our boss, it’s time to program his actions, which we do by triggers. This is a bit more complicated, but it’s actually very easy to do (though the difficulty will increase with your ambition). Here, we need an integer variable, which I have called “Geraldus Rose Fight Variants” and set to its default 0. We will be using it to randomize Rosie’s actions.
Next, we create a trigger for the fight itself. Because the fight has only phase, we will only need one. If you want to create several phases for your fight, you should create a trigger for each one, and have an If condition check the boss’ life (or whatever it is you want to separate the phases). I did this for Phantom Locust, but won’t be showing it here. It’s not too hard to do. However, if there is demand for it, I will add it later.
Now, what we need first in the trigger is to make sure it repeats forever. Then, we check if the boss is alive (otherwise we’d be ordering him around post-mortem, which is a bit of information overload for the editor). Next, we randomize the Fight Variants integer. I’ve chosen 1 through 12, as I have 12 different variants for him (a lot of them are almost the same action, except that they order Rosie to different locations). Each of the variants has the same chance to run, which can be used to your advantage. If, for instance, you want your boss to have two attacks, but you want one of them to run more often than the other, you can assign three variants, and have the two first be the same (which means the variant you want to use the most run 2/3 of the time).
I have copied the trigger below. As can be seen, it's only a fraction of the real thing (I included only variant 1), as it is really, really big (see the entire thing in the map file). Note the Wait action at the end of each If action, which separates the variants. Also note that I order Rosie to move and not attack when walking towards locations, as the latter would have made her stop to attack you before reaching her destination. I also used Wait for Condition actions to make sure she didn’t fire her nukes too early.
Final words
This was a very long guide, and there are a lot of other things I could have included, but I hoped that you found use for it nonetheless. I left out a few things (like how to create a boss bar), but there are other tutorials for that. Now I hope to see your boss fight sometime in the future!
Chekhov’s gun
Contains spoilers for Crimson Moon (up to mission 5)
In this tutorial, I will be looking at a storytelling device that is so common that chances are you have seen it used hundreds of times already without knowing what it’s called. The device is called “Chekhov’s gun”, after what Anton Chekhov said:
“If you show a gun in the first act, you must fire it by the third.”
What he meant by that was that any element of a story should not be included unless it serves a specific purpose. It doesn’t have to be a gun; it could be a character, an event or... well, whatever the hell you want it to be. Today, the term has taken on a new meaning, more specifically “an object or other element that is first introduced as unimportant, but later turns out to be very important to the main narrative”.
Examples
One of my favorite examples is from the movie The Departed, more specifically an envelope labeled “citizens”. It first appears as a largely irrelevant objective, but it then reappears near the end, and when it does, it marks the moment when Leonardo DiCaprio realizes that Matt Damon is the undercover criminal in the police department he has been looking. The important scene here is not the realization itself, but rather the one where the envelope was introduced. Because the audience (like DiCaprio) has seen the envelope before, we can share his shock at seeing it. We already knew that Damon worked undercover, but until now, we didn’t realize how the envelope was important. It is a very chilling moment, and I can still remember my surprised reaction to it when watching the film in the cinema.
Breaking Bad also made good use of Chekhov’s gun at numerous occasions. The primary example would perhaps be the episode “Box Cutter”, where the camera focuses on a box cutter early in the episode, clearly telling us it will be important later, creating tension before it’s finally used (albeit not in the fashion we may have guessed). An even better example would be the ricin cigarette that is introduced in the first season, where the use of it is teased every so often, but then it is actually not used until the fifth season.
Why use it?
Using Chekhov’s gun is a good way to keep your story economical. Stories can get fat pretty quickly, and although it’s not forbidden to have an extraneous element, it is always wise to ask yourself whether something actually should be cut or not. Perhaps you have a character that provides neither plot importance nor any form of entertainment value, or maybe you’ve set up a storyline that ultimately goes nowhere. Your audience is going to be disappointed if you set them up for something you can’t deliver on.
On the flip side, Chekhov’s gun can have a lot of advantages as well. For one, it will make your stories flow better, as the resolutions you create for the third act will have a basis in the first. Try to consider a final showdown between a hero and a villain you’ve never been introduced to. That seems rather odd, doesn’t it? There certainly wouldn’t be a lot of reason to be emotionally invested in such a fight, would there?
How to use it
From my experience, it is best to have a loose perspective on Chekhov’s gun. Don’t interpret it too literally (it doesn’t have to be a gun), and don’t feel pressured to use it in the same way as other people do. There are tons of ways to utilize it, and if you manage to put your own spin on it, you manage to make it your own. Also, be careful that you don’t become a zealot to the rule. Not all elements in a story need to be equally important, and some can be important in their own way. For instance, if you introduce a back story to a character that is not important to the main plot, you don’t have to remove it. After all, the back-story could serve a purpose of its own: maybe it helps us invest in the character, or maybe it rationalizes the choices he commits.
I’ve learned that a good story is sometimes written in reverse. That doesn’t mean you should begin at the ending and end at the beginning, but rather that it would be wise to consider what the direction the story is headed in, and try to anticipate this in some way. Maybe use foreshadowing to hint at a coming event, or perhaps create a character that is wholly different to how you want him to end up, essentially setting him up for a sudden or gradual change.
Did you know that Chekhov’s gun is actually the reason why Crimson Moon has ten missions instead of five? At first, the campaign had half as many missions as it has now, focusing primarily on Janus instead of the miners, but when I decided to give the miners a bigger role, the story got twice as long. The reason why I wanted to give more room for the miners was that I had written a scene I was particular scene I was quite proud of: the death scene of Brother Tyrone. It's one that plays heavily on Chekhov's gun.
Brother Tyrone was, as you might or might not remember, the leader of the Roverville miners. He was constantly shown berating sergeant Henderson, and the idea behind this was that, at one point, Henderson would finally have had enough of being belittled, and he would abandon Tyrone when Tyrone actually needed him the most. What happens in the story is that Tyrone first insults Henderson one time too many, then asks him to stand guard, which Henderson neglects doing, instead going to relieve himself. This leaves Tyrone unguarded when Quarinius comes knocking, and he is killed.
This particular scene wouldn’t have worked if I hadn’t laid out the groundwork first. It wouldn’t have made sense that Henderson chose to abandon his duties if I hadn’t established that he didn’t care much for Tyrone. However, because we had seen Henderson repeatedly being berated by Tyrone, we understand it completely when he decides to ignore Tyrone’s order. The scene also becomes much powerful because of this, not only because the story has secretly been leading up to it so far, but also because it marks the moment where Henderson finally realizes he needs to take charge.
How to play with it
Today, Chekhov’s gun is so common that a seasoned viewer will have no trouble spotting it a mile away. “The camera focuses on a particular object? Why, it must be important!” This is unfortunate if the point if we were to realize the importance of the “gun” in retrospect only. If the writers don’t want us to know where the story is headed, but we guess it anyway, that is unfortunate. Luckily, knowing this gives us the ability to play with the device.
One way of doing that is to actually break the rule, or actually realize that it’s not a rule at all (because there are no rules in storytelling). This might contradict what I said before, which is why I find it important to remind you that I am offering advice, not telling you what to do. If you want to break the rule, what you might want to do is set up a gun and imply that it will be important to the audience, and then have it turn out to be insignificant. This can act as a set up for some nice plot twists, and it can simply be used to throw the audience off-balance. Perhaps you had a second, better hidden gun, or perhaps you wanted to lead the audience down one path only so they wouldn't realize they were heading in another. Just be careful, though. Even if you can get away with this, there are good and bad ways of doing it.
+1 We need more story-related tutorials like this.
To nitpick, this isn't really an example of Chekov's gun. Lots of people interpret Chekov's gun as just mere foreshadowing, but that's not really true. It's not about what should be included in a story, it's about what shouldn't be included. It's about keeping the story economical like you said. That's why the phrase is "If in the first act you have hung a pistol on the wall, then in the following one it should be fired. Otherwise don’t put it there" not "If a pistol will be fired in the second act, it ought to be hung on the wall in the first act."
Likewise, you didn't necessarily have Tyrone killed as a direct result of pestering Henderson; it would be more accurate to say that you had Tyrone berate Henderson because you knew you wanted him to die. In fact, laying the necessary groundwork only made the story larger by your admission, so it's not really an example Chekov's gun.
This is why I'm not the biggest proponent of Chekov's gun. I'm all for minimalistic storytelling, but sometimes you just want to have something in a story for its own intrinsic sake and the aesthetic of realism.
But it doesn't matter what came first in writing, because to the audience, the order is the same. B follows A, even if A originated from B. What I wanted to emphasize was that the Henderson-Tyrone conflict has a point to the story before Tyrone's death, just not in the way the audience is supposed to be aware of. I get what you're saying, though.
And I agree that you shouldn't always follow Chekhov's gun, which is why I said you shouldn't be a zealot. In fact, I am a big fan of the way David Chase played with anti-climaxes in The Sopranos and often left plot threads hanging (the russian, the ending), but while it made watching the show a lot more interesting than its counterparts, it also made it a frustrating experience to watch sometimes.
Anyway, I'll try to include more story-telling tutorials. Any particular you want seen? I thought about writing about plot twists.
The problem is that it's a storytelling technique, yet people use it as a plot device ie. this object or that object is an example of Chekov's gun. People have deviated from Chekov's original advice of merely not including into the story anything that serves no purpose.
@EivindL: Go
Good storyline tutorial. You bring up some interesting points and it is very inspiring. This gave me some useful ideas as to how I can progress the story of my own campaign in the later missions, and it turns out that I have actually used this method already without even knowing it ^_^ I like how you bring us some insight in how these were applied through your own stories and other ones as well. This helps us have a better understanding of how it works.
Btw, I agree on the plot-twist tutorial. That would be incredibly useful for a lot of people. Me included, even though I think I have the basics of it down, and I also have a couple planned, it's always nice to get ideas :D
StarCraft II: Annihilation Campaign - A 24 mission single-player campaign featuring unique heroes, custom tracks, boss battles, a varied set of mission scenarios and three intersecting storylines.
Starcraft II: A War Story Campaign - A multi-act campaign with custom factions. Follow my discord for updates: https://discord.gg/Ztu44gZ
If you want to see a list of other awesome campaigns, go to: Custom Campaign Initiative
Nice to see a storytelling tutorial, should prove useful. :) I got a grasp on the concept, should probably look into it deeper though. Looking forward to more story-related tutorials!