Player versus Developer posted the latest in a series on RMTs today. You should read it. I haven’t really talked about RMT, but the posts highlight why I find the concept of designing around RMTs distasteful. I’m neutral on whether to do them, but I prefer as few barriers in the way of my design goals as possible. (Note: subscriptions can also be a barrier.)
I gave my lecture on Balance Constraints: Lessons from Pirates this past Wednesday. Today I threw together a hybrid of my slides with my notes that are somewhere between notes and a full paper in an attempts to record what I said. Hopefully it parses fine. 
I devoured the original Puzzle Quest: Challenge of the Warlords. It’s a great game, and it’s rather well known for being challenging. I didn’t find it terribly difficult, but I did need to think to make my way through the game. Part of that may be because I played a Wizard, and I found they were quite strong. Challenge of the Warlords got a lot of buzz through word of mouth and good publicity.
Puzzle Quest: Galactrix’s biggest accomplishment is the way it managed to bounce me right out of the game despite being a Sci Fi hungry Puzzle Quest addict. I was forcing myself to play, but I never found the game I hoped was buried underneath the endless luck-driven timed puzzles.
Then Puzzle Kingdoms showed up. It completely slipped under my radar - I only heard about the game a few days before it shipped. That was unusual, because Galactrix got quite a bit of hype. Infinite Interactive’s site doesn’t even list the game as released, and I haven’t seen a single review despite it releasing two weeks ago. And yet it’s a much better game than Galactrix. I’ll describe the basic mechanics since it’s not as well known.
As another color-matching game, Puzzle Kingdoms relies on acquiring mana (power) to charge units in your army so they can attack the enemy army. Armies consist of four units. Each unit requires a specific kind of mana, and the first unit of that in your army receives that mana. Anytime you acquire mana that doesn’t go to one of your units, it goes into the hero’s generic mana pool.
Initially, attacking is strategic. If you attack with multiple units at once, they all gain a damage bonus. Sometimes you want to hold off to give yourself a little bit more damage to finish off the enemy in fewer attacks, but you make yourself vulnerable to the enemy killing off your charged units. As you level up, your hero stats increase the attack and defense of your units. The number scale is much smaller than Puzzle Quest, so the +1-3 damage you get from combos should always be useful.
Unfortunately, Puzzle Kingdoms bears too much resemblance to an area where the original Puzzle Quest was weak: item balance. About midway through Puzzle Quest, I found a few items that gave my character insane amounts of power. There were still some difficult fights, but I’d have to self handicap to make the average fight mildly interesting. Puzzle Kingdoms has the same problem, but it is much more egregious in its flaw. After an intriguing early experience, I found two items early that basically ensure I win any fight after a couple rounds. Every fight is trivial. It’s pretty common to kill the entire army of a boss in a single attack. Enemies are so un-threatening that I routinely give them damage block combines just so I can match the kind of mana I want. There are a few factors that cause this:
1) Items. Early in the game, I found an item that increases my unit hitpoints by 3. Suddenly, even cheap 1 hitpoint glass cannons became effective front line units. Shortly after that, I found a weapon that gives me +2 red power everytime I combine red blocks. Combine that with a goblin (and later a wolfrider) who only needs 2 red power to attack, and I get 1.5 attacks for combining a simple block. Without that, you need 2-5 combines to charge a unit. Right now I use 3 wolf riders, and I charge all three by making two basic red combines (or a single 4 combine charges two instantly).
2) Heroes & Scaling. Your hero quickly gains power, but past the first few zones, the enemies you fight don’t scale up for a very long time. I’ve conquered a good 12 kingdoms, about half the game from what I can tell, and the creaures are rarely harder than the ones I saw by the second or third kingdom. Meanwhile, my hero increases the attack/defense of my units by around 5/3. That addition is stronger than most of the creatures I fight. And I outnumber them. And I have spells. And I have items. My cheap wolf riders have enough base damage (10) to one-hit everything I’ve fought short of some dragons, and they attack almost instantly. Meanwhile, I keep fighting creatures that have 2-4 hitpoints and insufficient damage to kill my “frail” wolfriders.
I’ve thought about self handicapping to play the game, but even when I tried that it wasn’t a challenge with an unequipped level 1 hero and basic units. My own skill has increased disproportionately to the difficulty curve in the game, which means there’s no real way for me to get a challenge unless I field less than four units. The heroes level up fast enough that I quickly gain strength on new characters and once again outmatch the enemy units.
I’m curious whether it’s just my experiences or if the game’s difficulty curve is ridiculously, horribly broken. It’s disappointing because I like the game so much otherwise.
As Login ‘09 approaches, most of my weekend time is going to be dedicated to getting my lecture on balance constraints together. This morning, I was gradually converting my many pages of notes into a lecture. At the same time, I discovered Warren Spector’s blog and started reading it beginning with the oldest posts.
I was working on a note about creativity flourishing within constraints, and how I want to be given a box so that I can find a way to break out. Coincidentally, this happened at the same time as I was reading Spector’s post on Clean Slate vs Reactive creativity, and I realized he had written a much more elegant and thorough take on the subject. The post is a year and a half old, but it’s definitely worth a read if you haven’t seen it before. I also recommend reading the two posts that lead up to that one.
I just noticed I didn’t post once in March. Surprisingly, that led to a minor crash in my website traffic! What, is nobody satisfied re-reading the same dull posts?
There were no posts in March because I was moving, and a good chunk of that left me without an easily accessed home computer. I probably won’t be too active in April either, because most of my blog time will be going to writing my lecture for Login ‘09. It’s coming along pretty well. The abreviated (note only, not full text) version is about 7 pages currently. I’ll probably post a preview up here sometime, but if not I will at least post the lecture material when I’m done.
In other news, I finally finished Stalker, and I was quite pleased with the ending. There were many available, and I saw a number of them. Stalker impressed me with its non-narrative driven story. I left with a vivid picture of the game world in my brain.
Solid new game designers typically have the game research side of things covered, but it’s common for them to miss out on other disciplines. While game design is relatively new, design itself is not. There’s a plethora of subjects that impart useful game design lessons, ranging from psychology and architecture to traffic. Today I’m going to talk a little about traffic.
Freeway design is a science, right down to proper sign placement and how frequently and early the driver should be notified about exits. The purpose of a road is to get the users from point A to point B. A well designed road system does that as quickly, efficiently and hassle-free as possible. These goals are often mirrored in our game designs. Often, we want to funnel players to specific experiences in the game without giving them the feeling that they’re being led by the nose. We want intuitive paths. We don’t want players to get lost, yet they need to be able to explore and find their bearings. I’m sure you’ve all experienced poorly designed roads that led you astray or caused huge traffic jams. Those are the problems we want to eliminate in our games.
I’m going to use an example from a local road to demonstrate how simple flaws can compromise the integrity of the entire design. Note: the lessons used to solve this problem come from psychology.
First, let’s take a look at the road layout. Cars are trying to reach the freeway (X) and they come from points A and B.
During normal work hours, there was usually a two-cycle delay for each car to get through a street light. The morning and evening commutes backed up across this entire stretch of road, leaving cars trapped for up to 30 minutes on this ~1 mile stretch (depending on the point of origin).
On the way to the destination, there’s a longer side path. This side path simply requires paralleling the traffic jam on a minor street and then merging before the freeway entrance. It’s an appealing option to drivers who are stuck in traffic. They bypass the logjam and only have to deal with a small amount of traffic before entering the freeway. Now, with that in mind: where does the traffic originate?
The bulk of the traffic comes from the cars on the side path merging with the main road.
The commuters who used the side path were taking a common self-serving approach to the problem. They knew they could bypass most of the traffic for themselves, but they did not evaluate the problem from a wider perspective to realize that they were responsible for the problem they’re trying to avoid. This is expected behavior. Of course, you may be thinking this is a chicken-egg scenario, because the traffic had to exist for drivers to attempt to bypass it to create traffic. On this point, we can reasonably assume an origin C coming from the north, resulting in the merges before the freeway entrance.
The city/county/whoever eventually solved this problem by removing the option to merge onto the main road via the side path. Traffic immediately plummeted, as it is all about flow. Each time cars cross they create a small delay. Multiply that delay and you get traffic jams. In order to deal with traffic jams, we need to guide people down the paths with as few collisions as possible.
I simplified the problem a bit in an attempt to avoid cluttering my point… and because it was a lot easier to draw! All design is interconnected, just like roads. We can learn lessons from studying traffic patterns, just as this traffic problem found its answer in psychology. This problem also demonstrates how critical it is to spend time gathering data and analyzing the problem to find the right solution. The kneejerk response is to add another lane to reduce the traffic, but that would have ignored the cause of the traffic. Design cannot administer random solutions any more than doctors can prescribe random medications.
Lately, I’ve been pouring lots of time into S.T.A.L.K.E.R. Shadow of Chernobyl. I haven’t played anything else quite like it, and I found it drew me in a lot faster than Fallout 3. I’ve been meaning to write about it for awhile, especially after seeing a good Rock Paper Shotgun post on the game a few weeks back.
For those of you who don’t know anything about Stalker, it’s an open-world FPS set in a post-nuclear Chernobyl chock full of RPG influences. It manages to successfully pull off tricks that I’ve hated in other games, because it does them right. What really impresses me about Stalker is how many stories I can tell based on my experiences. Unfortunately, there’s only one story I ever tell because I don’t want to spoil some of the cool elements of the game. Even so, this post contains a few low level spoilers. Here’s that story:
Walking through an underground lab
I returned to the first area of the game after exploring a big chunk of the world to find a group of bandits had taken a military outpost that I wiped out a few days before. I was able to fight my way through the camp by sneaking around and shotgunning guys around corners. One of the bandits had a secret on him, which creates treasure at a marker on my map. I headed towards the marker, and I found a large band of hostiles fighting off a huge swarm of dogs in a lightly wooded area. There were probably 10 guys and 20+ dogs. They were trying to pick off the dogs as they’d swarm in and attack one guy to try to take him down then flee. I used that as an excuse to start sniping the hostiles. This drew all their attention to me and the dogs began to scatter. I couldn’t get reliable shots with so many guys firing at me, so I swapped to an assault rifle and began darting from tree to tree, poking my head out to fire short bursts at any exposed hostiles. After a long firefight and a couple close calls where I had to use a pistol, I wiped out the band and took all their gear.
I can't see it yet, but I can hear a firefight in the distance.
I continued towards my destination, but there was still that huge swarm of dogs. As I ran down the road, I could hear them howling/barking all around me. I kept taking snapshots at them as they ran by the trees and bushes, trying to keep them away from me. Periodically, I’d get swarmed by 4-5 of them and spray bullets everywhere to get them to back off. I killed several of them and ended up arriving at a friendly outpost who took up position to help me fight off the remaining ~10 dogs. What’s cool is that was totally random, and it was a more unique experience than most games ever manage to provide.
Moving on: death is an interesting subject in Stalker. It manages to simultaneously make death completely meaningless and yet moving. NPCs are dying everywhere. There’s no real impact on the world when one of a host of random people dies, and it’s supposed to be that way. It’s a hostile world and most of the people in it are meaningless. And yet, every once in awhile, I’ll come across a body of a loner. When I examine it, I find the loner labeled as a friend. Loners become friends when you do things to help them - protect them from bandits or wildlife, fix them up when they’re wounded, etc. There aren’t many friends in this world, and it’s sad to see them die. It makes me think back to what happened before. There was one specific guy who I’d protected several times before. I’d bandaged him when he was wounded and traded him medikits. When I found his body, it was actually sad.
I stagger and my vision blurs after taking a painful attack.
Even though I’ve never heard it called a survival horror game, it’s a great example of the genre and it stands out from the usual experiences. I’m fighting to survive instead of fighting for the sake of reaching the next fight. Stalker plays mind games. There are strange things in the world, but they aren’t fully explained or presented. They’re introduced gradually. There isn’t always something lurking around the corner. Enemies are everywhere, but it’s less scripted and more a generally dangerous environment. As you travel through the world, your character’s perception of the world changes depending on the environment and the surroundings. They employ a variety of camera filters, visions and sounds. Seeing your vision change plays a critical role in creating the mood. It’s just one element of the detailed and compelling atmosphere that keeps drawing me in.
Colors change and wash out in certain environments.
When it released, Stalker was incredibly buggy. It’s smooth sailing several patches and two mods later. If you play Stalker, I highly recommend the Zone Reclamation Project (bug fix mod) and an endurance mod to raise the weight limit. I’m one of those people who can’t stand small inventories in games full of loot, and I find I carry 600 weight in supplies when the original cap is 50. Stalker isn’t a mass market game, but it’s gold for the right audience.
Eric Heimburg’s post on the MMO industry is a good read, and I’m going to use it as a lead in for the thoughts I’ve been wanting to share on the subject. If the ideal business world as a whole is a ship construction yard, then the MMO industry is a group of castaways building a raft to escape a deserted island.
Building a Raft…
Of course, these characteristics are not isolated to MMO development, and they are not representative of any single developer. Neither are they indicative of the people working on the projects.
This post isn’t intended to be bleak. Growth always comes from learning the lessons of the past.
My lecture proposal was approved, so I’ll be speaking at Login again this May. The session title is Balance Constraints: Lessons from Pirates. It’s similar in format to a postmortem, except it’s specifically aimed at the process of balancing an MMORPG. There’s a mix of balance philosophy and lessons from Pirates. I’m going to talk about what caused the various problems and how we could have solved them earlier.
Raph Koster has an interesting post up on socialization. I’m posting to highlight this part:
But it bears repeating, because there are many voices out there that will decry anything that resembles “social engineering.” But the shape of your map is social engineering; the layout of your buildings, the colors on your HUD, the placement of your text box — it’s all social engineering. We need to get past the simplistic observation that we shouldn’t treat users like ants in an art farm, and get real about the fact that we do have significant impacts on what our users do and how and when they do it.
It’s difficult to overstate how important social engineering is in online environments. Ignoring it achieves nothing. Depending on how you utilize it, you can create a game where newbies are valued (City of Heroes shortly after superbases and newbies generating the base equivalent of money) or abused (Shadowbane allowing you to train players and loot their corpses in the protected newbie lands). Social engineering is how you setup your game environment, and it has a huge impact on retention. Choosing to ignore social engineering is the equivalent of ignoring character advancement in an achievement-driven RPG, except the results are more insidious.
