[D] Base Vulnerabilities and You
PREFACE
Have you ever seen a map that you didn't like because it seemed too... plain?
I don't mean aesthetically plain. I mean strategically plain. It might have been well-proportioned. It might have been well-balanced. It might have even featured some semi-revolutionary, unique feature. But in the end it just wasn't interesting.
When I you are done reading this, I want you to look at that map again (it may have even been your own map), and I want you to think about what could have been changed to make it more interesting.
(By the way I start off talking about simple things but that's just to get everyone on the same page. I end up talking a little about some pretty advanced stuff. If you are sincere about becoming a better mapmaker it's probably to your benefit to pay attention the whole way through.)
INTRODUCTION
Turtling. An infamous word. Why? Because it's boring.
Nobody likes to watch two players macro for 15 minutes with little to no engagements and throw maxxed armies at each other in the middle of the map. Either one side simply rolls over the other and the game is over instantly, or both sides lose a lot and meet in the middle again 5 minutes later. This is obviously uninteresting to an "intellectual", but even a "simpleton" doesn't want to wait around so long to see the pretty lights.
So as mapmakers we try not to make our maps favor a turtling strategy. Arguably, one single map in an entire map pool should be a very large map that favors turtling. But we already have one. We call it Terminus (more on that later). We don't need another one (yet).
So wait a second... isn't "turtling" a lot like "macroing"? Well, yes. But no. While turtling forgoes all offense until after having an army advantage, macroing often involves asserting pressure and/or map control to secure additional bases. That's another big difference. A turtling player is content with 2, sometimes 3 bases depending on the map. A macroing player is always looking to expand.
Let's take another look at that word "expand".
Expand - [ik-spand] (verb): to spread or stretch out.
So when people expand they are stretching themselves out. The resources are their goal and an important factor, but the real key here is: just how much are they stretching themselves out? This is something I think too few people are paying enough attention to. StarCraft is extremely complicated, and there is no shame in not knowing what to look for. I am writing this to give insight as to what I have learned to pay attention to.
Specifically, I will be explaining some things I call "base vulnerabilities" and their role in "attack path layout density". Perhaps more important than that, I want to show how maps can encourage both expanding and attacking while rewarding the player with good scouting and multitasking skills.
THE FOUR TYPES OF BASE VULNERABILITY
[1] Ground Pathways / Chokes
- Most vulnerable, can be attacked by: All units.
- Blocked / Unblocked (Destructible Rocks)
- A choke/pathway that is blocked with destructible rocks is usually easier to defend than one without destructible rocks.
- Width
- + Show Spoiler +
- Wider = Defender is more vulnerable.
- + Show Spoiler +
To calculate the width of a diagonal choke/pathway, we use the Pythagorean theorem (a² + b² = c²).
a² + b² = c²
7² + 9² = c²
49 + 81 = c²
130 = c²
v(130) = c
~11.4 = c
+ Show Spoiler +
The "standard" width (imo) is 9 to ~11.4 squares long (just enough to block with 3 barracks or gateways).
- + Show Spoiler +
- Length
- + Show Spoiler +
- Shorter = Defender is more vulnerable.
- + Show Spoiler +
- Concave
- + Show Spoiler +
- More area/openness can be both a good thing and a bad thing for either side, depending on the units used. Generally, Zerg can overwhelm a Terran or Protoss that spreads themselves too far in an open area (though all races have the capacity for this). By the same token, more room to spread out can make splash units (like banelings) less effective).
- Pay close attention to how big the concave is on both sides in relation to how wide the choke/pathway is.
- Also note that this is closely related to how easy it is to "contain" someone.
- + Show Spoiler +
- Ramps
- Having troops on the high ground always gives you an advantage over troops on lower ground. Whether they have to risk air units to give them vision or use a scan (aka 270 minerals) or simply take the hit of not knowing what they're moving into.
- Example: Consider how zerglings can defend a high ground natural ramp against hellions (because the hellions can't shoot the zerglings until they're in range of the zerglings). This allows the zerglings to perform much better. Things tend to matter more in the early game though (more on that later).
- Having troops on the high ground always gives you an advantage over troops on lower ground. Whether they have to risk air units to give them vision or use a scan (aka 270 minerals) or simply take the hit of not knowing what they're moving into.
[2] Cliffs
- Second most vulnerable, can be attacked by: Air/Transport, Cliffwalking, and (to a lesser extent) Ranged units.
- Much like ramps, the person on the high ground has the advantage.
- Siege tanks are the first thing that comes to mind for most people when they think of a ranged unit (not including collosus) that can take advantage of cliffs. But even a short ranged unit like the roach can take advantage of a cliff. Ask yourself: Is there a mineral line/geyser near the cliff? Is there a reason to build near the cliff? Is there a reason for the enemy to position troops near the lower half of the cliff?
- It is possible that cliffwalking units cannot actually walk over certain cliffs. To me, this is very closely related to how sometimes there can be strips of parallel terrain that are separated by only a small amount of air space. Note that in these areas transport units can be particularly effective.
[3] Air / Space
- Third most vulnerable, can be attacked by: Air/Transport units.
- Consider how much room there is between the edge of the map and the base. Also consider how the air space is restricted by other terrain. Pay attention to whether or not air units can be "trapped" in certain areas (or caught on their way out).
- Even though Air is only the third most vulnerable type, do not underestimate what a good air force can achieve given enough room to operate.
[4] Edge of Map
- Invulnerable; cannot be attacked from this angle.
- + Show Spoiler +
Notice how bases in a corner of the map have less vulnerable surface area than anywhere else on the map.
OTHER THINGS THAT CAN AFFECT BASE VULNERABILITY
Xel'Naga Watch Towers
- Has the potential to help either defender or attacker. As defense, it can help you see threats quicker giving you more time to respond to them. However, the XWT can also be used against you.
Line of Sight Blockers
- LoS Blockers have the potential to help either defender or attacker. Usually the favor goes to the side with air units, but being able to set up ambushes is good too.
Raised, Unpathable Terrain
- Similar to LoS blockers, this has the potential to help either defender or attacker. Usually the favor goes to the side with air units.
Ingenuity
- I am definitely not claiming that all possibilities are within the scope of what I am explaining here. The thing that keeps me mapmaking is that there is always room for new ways of doing things. Maybe you can combine some of these elements I'm talking about to make something that will someday be standard in most maps. Maybe there are different ways of thinking about these things that can make something entirely new.
PROGRESSIVE VULNERABILITY
As a general rule, the closer a base is to your main the easier it should be to defend. Actually that tends to happen anyway simply because of that virtue, but it should also be coaxed and controlled by manipulating the four types of vulnerabilities.
Another important thing to understand is that things tend to be more important earlier in the game. This is a difficult concept to explain, but it's true. I guess what I'm saying is that you should pay closer attention to earlier bases' vulnerabilities.
A prime example of this is how easy it is to take your natural expansion. When you think about it, the decision of when to expand (or not to expand) is really the first really big decision you have to make in a game. If you don't expand fast enough and don't put pressure on your opponent if he is, then you might not have enough map presence to be able to claim a third base when it comes time to. Therefore we should pay close attention to just how easy it is to expand. Perhaps the most important factor in this (that I'm not sure most people are aware of) is something I will call the "ramp -> cc/nexus relation".
+ Show Spoiler +
I'm certainly not saying that every map should have the same or even similar ramp -> cc/nexus relation, but I don't think enough people are aware of just how important it is for an FE'ing Terran or Protoss. The more buildings it takes to close your ramp and connect it to your cc/nexus, the harder that natural is to FE into. I would almost argue that this is so pivotal that maps should be balanced around it.
By the way it's perfectly fine to throw in an easy-to-defend base later on in an expansion pattern (creativity is the heart of mapmaking after all), but you will generally want to give this type of base less resources (there is a strong connection between this "progressive base vulnerability" concept and how many resources each of those bases should have).
Also, it's important to take an entire expansion and attack path layout pattern as a whole. Each base has an effect on every other base. Always try to take in the big picture without forgetting all the little details.
BASES COVERING BASES
The true key to encouraging both expanding and attacking is utilizing a technique I call "bases covering bases". Basically, you can use later, more vulnerable bases to help earlier bases become less vulnerable by placing them in such a way that they cover certain attack paths.
+ Show Spoiler +
Take a look at Crossfire SE. The main base has a backdoor entrance with destructible rocks blocking it (path 1). However, when you take (base a), (path 1) essentially closes while (path 2) and (path 3) open up. Furthermore, when you take (base b), (path 3) closes while (path 4) and (path 5) open up. That's a total of 5 paths, but when you actually take two of those bases you are down to 3 paths. Very cool.
There's actually a more extreme way of doing this (sorry for no pictures to explain it), but basically there are ways to make a base cover many paths. Usually this is done by placing it towards the center of a map... and it doesn't necessarily have to "cover" any paths as much as it simply becomes a much easier target (so when they're busy attacking it they're not attacking more tucked-in bases).
Let me sorta squeeze this in here: + Show Spoiler +
CASE STUDY: TERMINUS
+ Show Spoiler +
VULNERABILITY TYPE:
Red = Ground
Cyan = Cliff
Green = Air
Yellow = Edge of Map
Red = Ground
Cyan = Cliff
Green = Air
Yellow = Edge of Map
The first thing you should notice when analyzing Terminus for vulnerabilities is that the first three bases all share a single choke point! I chose this example specifically to show what a map that favors turtling could look like. If you can defend all ground paths to your first three bases with a single choke point (especially since it happens to not be a very wide choke point), then that definitely encourages a player to decide to turtle.
It is interesting to consider the turtle-ness of the first three bases in stark contrast to the massive vulnerability of the fourth base in the middle of the map.
Let me be clear, however, and say that this is actually quite an excellent map. I said earlier that it's okay if there is a single map in a map pool that favors turtling, and I meant it. This map happens to fill that role extremely well.
Those three bases may not be very vulnerable to ground, but they're actually quite vulnerable in other ways. The third base is very vulnerable to air, and the main is somewhat vulnerable to air. The main is also rather vulnerable to being dropped (notice how there is a small strip of air terrain between the main and the fourth... as mentioned earlier this makes an excellent staging ground for drop play). Also the main is somewhat vulnerable to cliffwalking units. (I could be mistaken but I think the main ramp is double-wide like scrap station, which makes it more vulnerable in the early game).
Furthermore, turtling on this map is risky simply because doing so makes you lose map control, and there are a lot of bases on this map (16... the most any map should have), so your enemy could just take another 3 bases while you're turtling, so that it's 6-7 bases vs your turtled 3.
All things considered, this is actually a pretty damn good turtle map. But it's still a turtle map, and we don't want more than one of those in a map pool. Let me emphasize once again the high density of ground attack paths on this map, and now let's look at how it contrasts with an attack path layout with lower density.
CASE STUDY: XEL'NAGA CAVERNS
+ Show Spoiler +
VULNERABILITY TYPE:
Red = Ground
Cyan = Cliff
Green = Air
Yellow = Edge of Map
Red = Ground
Cyan = Cliff
Green = Air
Yellow = Edge of Map
Oh my! Look at all that red going all over the place. Plenty of cyan too. This is a great example of a lower density attack path layout. + Show Spoiler +
Do not be confused and think that more red/cyan means more density... what it really means is how those attack paths are in relation to the bases themselves... That is, the attack paths are surrounding/all over the bases instead of concentrated into a few areas.
Let us not forget that Xel'Naga Caverns was, for a long time, considered the most balanced map around (and arguably the most interesting). You might think that high vulnerability of all the bases discourages expanding. This looks good on paper, but when it comes down to it most games you see on this map actually go into the late game with plenty of back-and-forth. Why is that?
Well first it starts with a good (low building requirement) ramp -> natural cc/nexus relation. That is, taking your natural expansion isn't actually that hard even though it is quite vulnerable/open. From there, something interesting (and really important) happens. Basically, skill becomes a huge factor.
This particular map with all of the base vulnerabilities and low attack path density rewards the player who can scout and split up their forces (both offensively and defensively). This concept combined with using the "bases covering bases" technique are really the keys to making a map encourage both expanding and attacking.
UNIT COMPOSITIONS, BASE VULNERABILITIES, AND STATIC DEFENSES
The last important concept I want to talk about is also an enormous metagame step that will inevitably be made some time in the future.
Most people who play this game understand the concept of trying to figure out what your enemy is making and then making the correct units in response. This is only about half of the picture.
You should also consider how unit countering (and capabilities) relates to the vulnerabilities of your bases, and how that works into a long-term expansion plan. You want to think about how well the units your enemy is making can take advantage of each vulnerability in your bases. You want to think about which vulnerabilities your own units can take advantage of, and how, when, and where you should protect your own base(s) (with static defenses or not) in relation to the enemy's forces. You'll also want to consider which bases you should be taking (and when to take them) based off of the capabilities of your own forces and your enemy's, and also based off of what you'll be making after you take said bases.
I could go on and on about this, but basically base vulnerabilities and their positioning and the timings that you can take advantage of them (or defend them) will be (and is) at least as important as the unit timings and positionings themselves.
Perhaps one perspective is to think of the bases themselves as just really big units, and the units are just the base's weapons or defense/hitpoints. Base vulnerabilities are a key component of the base's characteristics. You might not be able to change unit characteristics (in a melee map at least...), but you can definitely change base characteristics.
CONCLUSION
I'm surprised that I am the first person to outline this overall concept (am I?), because to me these things are essentially the bread and butter of balanced mapmaking. These are the bare fundamentals. To some this is old news (but I'm sure even they could use some thought reinforcing). To some it is critical to learn well in order to breaking into the next levels of mapmaking.
So what was my true goal in writing this? Too often I see maps that feature a slew of bases with extremely boring base vulnerability characteristics (which leads to boring overall layouts). Spice that shit up people! ^_^_^
In general, this is what I think about how much each type of vulnerability is used.
Type 1 (Ground) - Multiple base entrances are not used enough.
Type 2 (Cliff) - Not used enough.
Type 3 (Air) - Not used enough.
Type 4 (Edge of map) - Used too much (on bases outside of main).
I'm Back <3
- Barrin