Click "Sleep" for a dark background.
Click "sleep" again if text isn't dark.

 

Entries in RTS (11)

Thursday
Jun182009

Critical-Gaming Pikmin Course: Final Presentations

It was the last day of the class, and as scheduled we gave our presentations designed to advance our understanding of Pikmin 2's battle mode by analyzing some part of the game. The other option was to deepen our understanding of the RTS genre as a whole through some kind of detailed investigation. Since there weren't many of us in the class, we could comfortably fit all of our presentations in one class period. I took notes on parts of the presentations that I found to be useful or interesting.

 

SPRAYS AND GELS:

  • The effectiveness of the Ultra-spicy spray increases the more Pikmin you have (mid-late game) because you can powerup more units per spray.
  • 1 spicy Pikmin can take out 3 normal Pikmin out of a group of 50. 5 can take out 21. And 10 can take out all 50. Perhaps this is too great of an advantage for the spicy gels especially considering that they last 40 seconds.
  • The effectiveness of the Ultra-bitter spray decreases the more Pikmin your opponent has (mid-late game) and the more they multitask. Unless a valuable target is unguarded by the opponent's Avatar unit, chances are any forces you plant into the ground can be reinforced very easily. Also, the bitter sprayed Pikmin only stay planted for 10 seconds. This seems very short compared to the highly effective 40 second bitter spray time. Perhaps the Pikmin should stay planted until plucked. 
  • It's best to hold on to all of your Ultra-spicy sprays until later in a match. This means, when you use a spicy spray early on or on a relatively small number of Pikmin and the opponent uses a bitter spray to counter you, you are losing out in the long run.
  • Be careful when drinking gels from the field. Just like greedy Pikmin and nectar, both Avatar units can drink from the same spicy/bitter gel thus adding one spray to both of their inventory. If you find a spray in the field, drink it quickly.
  • Gel Gambling: Whenever you use a bitter spray on an enemy of any size and kill that enemy while it's petrified/frozen, instead of leaving behind a carcass they will drop a nectar/spicy/bitter gel. The chances of an enemy dropping a spicy/bitter gel is about 10%. If you can use a bitter spray on a lot of small/weak enemies, it's possible gamble and win a more valuable spicy gel or better yet more than one spicy/bitter gel. 
  • Cryo Freezing: Using a bitter spray on a large, dangerous enemy is a great way to attack safely. If you want to harvest the carcass (for non Cherry spawned enemies), then you can weaken the enemy while frozen, wait for it to break free, and then finish it off easily.

 

CHERRIES

  • Cherries are the most influential and valuable resource in the game. They can give a player enemies to deploy, Pikmin seeds to pluck, gels to turn the tide of battle, colored marble security, and more.
  • A single Pikmin can carry a Cherry so swiftly that it can maneuver through dangerous enemies will a high degree of success without aid.
  • A set amount of Cherries must be in the field. After harvesting a Cherry, another one will drop into the field approximately 10 seconds later in the general area where Cherries were previously found. It's possible to leave behind Pikmin in these areas so that spawned Cherries are automatically picked up and carried back to your base. This is called cherry picking.
  • Cherry hoarding: To prevent your opponent from getting and using Cherries against you, it's possible to store all of the Cherries in your base by stopping your Pikmin before they deposit a cherry into the Onion. Though this is a very laborious process, it's also an impractical one on some levels. As many as 7 Cherries can be spawned into a single level. Managing all 7 would be too difficult especially considering that they would be scattered across a very large area of the field.
  • The Cherry roulette is influenced by the changing game state between the two players. For example, when one's colored marble is stolen, there is a very high chance that the next Cherry harvested will be a "return to base" power. This kind of design is very common in Nintendo games (Mario Kart, Mario Strikers Charged, & Super Smash Brothers)
  • In normal game conditions the percentage breakdown of the roulette options are:
    • 1% Recover Stolen Marble
    • 4% Blowhog fire/water
    • 5% Vanish
    • 8% Withering Bloghog
    • 8% All Pikmin Bloom
    • 9% Ultra-Spicy Spray
    • 9% Ultra-Bitter Spray
    • 9% Snitchbug
    • 10% +5 flower
    • 11% +10 flower
    • 11% Volatile Dweevil
    • 15% Boulder Drop
  • While Cherries simplify the resource gathering and harassment in Pikmin, their random yet tide turning effect on a match can be too much for some gamers. Perhaps, the number of Cherries should be toned down to a maximum of 3 instead of 7. Another interesting option would be to design the roulette to escalate in useful items when multiple Cherries are  harvested in a row. In this way, a single Cherry won't do much good/damage. But by collecting many over time, not only will you gain access to more powerful items, but you'll have a way of reducing the random range of the Cherry roulette.


CONTINUOUS CHOICES WITH PIKMIN POPULATIONS

Two types of continuous choices that are most common in a Pikmin match are deciding how many Pikmin to assign to undig a marble and how many Pikmin to assign to carry back an object. For every Pikmin you give up for a task you lose offensive and defensive power. At the same time, you gain speed. The more Pikmin working on a task, the faster it'll be accomplished. The faster the job gets done, the most quickly you can get your Pikmin back to do something else. Because speed is the main factor we're considering, we need to look at some data to help us understand exactly how much each Pikmin speeds things along.

It takes:

  • 132.7 seconds to undig a yellow Marble using 1 leaf/bud/or flower Pikmin.
  • 43.9 seconds to undig a yellow Marble using 3 flower Pikmin
  • 33.2 seconds to undig a yellow Marble using 3 flower boosted Pikmin

According to the data, the time it takes to undig a yellow marble can be expressed using this equation: t = 132/x - 3 3.b where x is the number of Pikmin undigging the marble and b is the number of boosted (Ultra-spicy spray) Pikmin undigging the marble. Using this knowledge, you can better keep track of the timing of such events. So, just in case Shearwigs are unearthed along with the yellow marble, you can swing back around just in time to save your Pikmin from being eaten up.

When carrying an object back to your base, how many Pikmin should you assign to the task? Each object has a different minimum and maximum Pikmin carrying count. A marble requires at least 1 and at most 8 Pikmin while a Red Bulborb requires 10-20. On top of this, leaf/bud/flower Pikmin all move at different speeds. First we looked at some data showing how quickly a leaf/bud/flower Pikmin could carry a Cherry a set distance when unboosted and then boosed. 

  • LEAF: 9.8s       BOOSTED 9.6s
  • BUD: 9.5s        BOOSTED 9.4s
  • FLOWER: 9.1s BOOSTED 9.1s

The data for carrying a marble a set distance using all flower Pikmin was presented.

  • 1 Pikmin 42.3 seconds
  • 5 Pikmin 24.6 seconds
  • 8 Pikmin 18.8 seconds
  • 8 Pikmin boosted 18.8 seconds

From this limited amount of data I managed to copy into my notes, the boosted Pikmin don't help  at all when carrying a marble compared to a flower Pikmin. With marbles, you can double your speed with around 6 Pikmin compared to the travel time of a single Pikmin. The more Pikmin you assign to carry a marble, the safer it'll be from attacks because of the increased mobility and Pikmin count. If an opponent puts a single Pikmin in the way of a marble carried by 1 of your Pikmin. There's a chance your Pikmin will be killed and the opponent's Pikmin will take the marble over. But if you have 2 of your Pikmin carrying the marble in this case, the 2 Pikmin will most likely win and your marble will continue traveling to your base.

 

 

MAP VARIABILITY

One presentation was on map variability. This student made a map of every stage in the game and diagramed the different variations of elements within the maps. He charted out where the Cherries are likely to drop and where to find the yellow marbles. This was all very useful information that's hard to reproduce here in my summary.

 

PIKMIN RUSH

My friend Chang did an interesting presentation about rushing your opponent's base in the early game. I couldn't help but think that he was inspired by the classic Zergling rush strategies. Basically, if you can't find a way to kill your opponent's Pikmin outside of a PvP battle, extending yourself that far away from your base puts you, the rusher, at a great disadvantage. If there are enemies in or very close to your opponent's base, a rush strategy may be viable. Also, for this strategy, the closer you are to your opponents base the better. With such low numbers of Pikmin at the start of the match, planning ahead and using each spray wisely is crucial.

There are two main ways to rush the opponent: initial rush and an early game rush. The initial rush involves using your starting Pikmin right off the bat. If you can make it to the opponent's base when they're still harvesting flower pellets/fighting enemies, you can fight without having your travel time significantly affect your strategy. Otherwise, if you have really good micro skills, you can spend the first minute or so growing a larger army and then use that to overcome your opponent. With this strategy it's even possible to leave a few of your Pikmin behind to cherry pick and play the odds for additional advantages.

If your rush fails, it's important to know when to pull out. Retreating early and using your temporary map control to take Cherries back with you is not a bad way to start a match. There's even a chance that you can gain more Pikmin and gels from the Cherries you carry back with you than you lost pushing such a risky strategy. Overall, rushing is a strategy that requires a lot of micro dexterity, knowledge, and adaptation.

 

WAITeRTS

Another student did a very interesting presentation drawing a parallel between being a waiter in a restaurant and playing a RTS game. Like tech trees or Pikmin orders, from the moment a customer walks into the door there's a step by step process for dealing with them. From the top the customer must be greeted, seated, brought menus and napkins/silverware, you must take drink orders, fill drinks, get food orders, punch in the orders, take out the orders when the food is ready, check up on the table, check for drink refills throughout, deliver check, process payment,  get to-go boxes/cups if necessary, clean off table, and secure tip. This order is like an assembly line that, for the most part, must be done in this order.

The units in a restaurant are the manager(s), waiter(s), cook(s), and bar tender(s). Interestingly, this student explained that where he works, he's the only waiter/manager/bar tender on duty and he works with only one cook. So, aside from cooking the food, this guy had to do everything. At this point, he discussed the types of actions he could do. The incompressible (must do) actions include seating, taking drink/food orders, and delivering the food/drink/check. Compressible actions include things like getting refills/condiments, answering the phone, and getting to go boxes/cups. Extraordinary actions include talking with customers or looking up extraneous information for them.

He explained that one table is easy to take care of. But handling any more than one table at a time begins the juggling act. Figuring out how the shuffle the independent orders of different tables together requires understanding where the gaps of time are between one action and the next and the priority (incompressibility/compressibility) of each action. In general, after you take a drink order, you have a small amount of time to return with the drinks before the customers get antsy. As long as you return before or about the same time as they decide on their food, everything is good. In this time, it's possible to greet, seat, and take the drink orders of a few other tables without missing your "clicks" so to speak.

Because being a waiter means moving around the world of the real bound by real time and space, being a waiter moving customers from one end of the order to the end requires physically moving throughout the restaurant. Between the table, the drink machine, the register, and the food window, you'll be moving back and forth throughout the restaurant space like the Avatar unit does in Pikmin. In a waiter's world, space is folded and can be folded so intricately that the "level design" classification goes beyond folded level design. The presenter didn't know how to describe it, so I'll add here that everything in this real world of ours probably falls under the pure organic classification which can be thought of as the most advanced and most organic extension of folded level design.

For a waiter and any RTS, time is precious and using resources is essential. Instead of Pikmin, minerals, gas, or resources like that, a waiter has resources like rolled napkins, individual food items, and clean tables. Without clean tables (or enough clean tables), there's no place for customers to sit. This can easily create an incompressible action (clean a table) for the waiter that must be addressed. Without napkins already rolled, a waiter must find time between seating customers and delivering food to roll some. Hickups with resources are annoying, but they pale in comparison to real harassments.

When you're the only worker responsible for the entire front house (everything that's not the kitchen and the office) there are a multitude of ways your timing and scheduling can be thrown off. When customers walk in wanting to place to-go orders, they take up a waiter's immediate attention. The same goes for any customer that walks up to the front counter. Likewise, spills, telephone calls, and customers requesting your attention for whatever reason can really throw a wrench in things. (This is not even mentioning the obvious rude customers that purposefully harass you.) Because all of these elements are unpredictable, it's best to work in a manner that saves as much time as possible all the time. This is where micro comes in.

The functions of working or the goal of being a waiter can vary from person to person. Some only do it for the money/tips. Some do it to serve others. Others do it to support the company/business. Whatever combination of goals you might have is what you keep in mind as the bigger picture. Regardless, optimising the smaller processes will give you more time and control over your work which is a good thing to have no matter what your goals are.  Everything from filling drinks, carrying out food, taking orders, running credit cards, cleaning a table, to answering the phone can be optimised to save time. One way is to carry multiple items for multiple tasks at once thus compressing two or more round trips into one. Other technique is to memorize food and drink orders so you don't have to waste time writing them down.

And then there's scouting. In a typical customer order, a table needs less and less attention as the order moves closer to completion. It's bad to be late on a step. It's good to be early to a step. And it's best to be right in time as if you're some kind of magical server who simply appears whenever needed. ("Poof! What do you need?") In general, a waiter must always keep watch around the room. Things to notice include body language for people upset with their food or requiring additional condiments, folded menus for people ready to order, low drinks, people picking at their plates when they're ready for their check, and people still reading their menus that need more time. A good waiter must keep an eye out for all of these signs. Recognizing that a table isn't ready to order yet can give you the go ahead to attend to another table or get another task done. Like scouting in StarCraft, it's best to look out for information on customers that can influence you to make a time saving discrete decision.

The student ended the presentation by proposing that waiters who utterly fall apart when the pressure is on would probably be no good at games like StarCraft or Pikmin. Likewise, gamers well versed in different RTSs will find themselves well equipped to handle hectic waitering situations.

 

The professor commented on how this presentation was exactly what he had hoped one of us to do.  If there was a game centered around being a waiter, it would be a very interesting RTS in that it would feature zero fighting or battling whatsoever. Pro.K shared with us his hopes that we would now look at all RTS games and even activities from our own lives using the language and the concepts from real time strategy games. Though when you break it down, each concept/step is simple enough. Being able to understand, comment on, and explain how all the parts come together and interact is a level of education that can best be achieve through study like with this class.

The professor shared with us that he felt that there were most likely several elements to Pikmin 2's battle mode design that would hold it back from being a very successful highly competitive game. Like StarCraft to Brood War, little tweaks here and there would go a long way if anyone was willing to make those tweaks. The new play control version of Pikmin 2 would have been a great opportunity to make any changes, but already that opportunity is gone. Pro.K explained that everyone who made it through the course (and now all of you readers that have made it to this last summary) are part of the small group of Pikmin players that are the best players in the world, if not some of the best gamers suited to analyze its matches.

"The only good Pikmin is a working Pikmin. Resources are potential. ie only good when used. Knowledge is power. So put your knowledge to good use." The professor wished us well in life and class was dismissed.

 

Oh, and if you're wondering what I did for my presentation, I must say that I didn't get around to doing one. The reason I didn't bother with a presentation is because I had planned on challenging Pro.K to a Pikmin best out of 3 match since he announced the option weeks ago. After class was dismissed, we pulled out the game station and I started my final exam. I beat him the first round. He beat me the second. And I passed the class for the 3rd round. I only wish we could have recorded the video.