7 Wonders is a classic Eurogame. It requires resource production to build upgrades which give the player points towards winning the game. In addition there are is a trading mechanism so you can utilize resources of other players. The interesting thing about 7 Wonders' resources is they act more like public goods than any other Eurogame I have played. This can result in massive underproduction of resources. Let's break down the mechanics and player incentives to see why this happens.
In 7 Wonders players each receive a hand of cards, they choose one to build, and then pass the remainder to the player next to them. This process repeats until the hand is exhausted, then a new hand of more expensive cards is dealt and the drafting continues. The game consists of 3 ages or rounds of drafting. At the end of the game points are totaled to determine the winner.
The catch with building cards is the player must be able to pay the cost listed on the card, usually some number of resources. Resources are gained in one of two ways. The first is through the cards a player has already built or as an ability of the Wonder the player is playing. This method of producing resources is permanent and usable in every build phase. So if the player can produce 1 Wood due to having that resource card, every time they build they can use that 1 Wood. But if the player has a card that costs 2 Wood, they would need access to two different sources of Wood. In later ages, cards can get very expensive requiring access to many types and large quantities of resources. This is the main incentive to drafting cards the produce resources as none of the cards give players points directly.
The second way to gain resources it to purchase them using gold from a neighbor. The twist in 7 Wonders over usual trading mechanics is this one is both non-exclusionary and non-rivalrous. This means that the player receiving the gold can't reject it and prevent the other player from using his or her resources. Further even if a player is using a resource, a neighbor can still purchase it from them. So if a player has 1 Wood that they are using to build this turn, their neighbor can also pay gold to use that Wood on that turn too.
It these interesting properties of trade that can result in a game being completely resource starved in the last age, where no new cards produce resources. In Ages I and II there is a disincentive to taking resources early because that guarantees a higher quality of cards which actually affect points at the end of the game though having more options. Plus a player following this strategy can simply buy from those that do draft the resource cards early gaining the double benefit of more resources they can access and a better quality of card choices through the resource cards being removed. And if people are at a table were everyone else is taking non-resource cards first, then it is optimal to do that too in order to maintain card quality parity and not feeding others the best cards when you are taking only resources.
When all players follow this take resources late strategy it often ends up that there are fewer resources on the field than needed for the more expensive building because the last card of an Age is discarded instead of drafted. Which leads to relatively depressing Third Ages due to everyone being resource starved. The interesting thing is this is actually optimal! It is clearly better to buy resources from others to get more scoring cards than buy resources. And any player building resources is clearly the sucker in the game as the other players use his production and snap up all the good cards. The only thing that might temper such a trading strategy is running out of gold, which is possible if the other players are not also trading due to the lower velocity of gold in such a game.
This is the only example I know of where the existence of trade and a reliance on it leads to the players getting locked out of upper tier parts of the game. It is definitely worth digging deeper into the mechanics to determine if trade having this harmful effect generalizes or what the key component of the system that makes trade harmful is.
Tuesday, June 25, 2013
Monday, May 6, 2013
Pandemic: The difference a rule makes
Recently I got to try Pandemic. In this game four diseases threaten to wipe out mankind. The players must work together to keep the plagues from spreading while racing to find a cure before time runs out. Each player can be one of five roles (7 roles in the new version) which grant them special abilities. Like the Researcher who can give any card in any location, instead of just the card corresponding to that location. This is important because a set of 5 (or 4 for the Scientist) cards of the same color are needed to cure a disease. It is a very good game and I recommend that you give it a try when you have the chance.
When I was playing with some friends, they were playing that you could only give cards on the giving player's turn and were finding the game quite challenging. Most of that first game with them was spent figuring out how the receiver could end up in the correct city to be given the card they needed. This made the game more an exercise in creating board moving than actually about stopping pandemics and it wasted a lot of turns, which are really the limiting factor in Pandemic.
After some rules discussion, we played the next game by relaxing that giving cards could be done on either the giver's or receiver's turns. The difference in difficulty was astounding. We won easy without a sweat, whereas before it was more like a coin flip. The other big difference to me was the type of discussions we were having. Instead of trying to go through weird gyrations to pass cards, we talked more about stopping diseases. Now we debated the trade-off of heading off the spread of a plague or getting a cure for another sooner. The focus had returned to what the designers had in mind, instead of trying to go through the gyrations of a small rule.
This was a very nice example of how a very simple rule change could have a dramatic impact on how the game is played and really the quality of the game. The lesson I took away from this is when most of the discussions in a game are about how to work within one rule, unless that rule is central to the concept of the game, there is probably something wrong. And then letting the players do what they want to, give cards on any turn in this case, can make a more enjoyable experience as then the players' focus is not distracted by obtuse rules.
When I was playing with some friends, they were playing that you could only give cards on the giving player's turn and were finding the game quite challenging. Most of that first game with them was spent figuring out how the receiver could end up in the correct city to be given the card they needed. This made the game more an exercise in creating board moving than actually about stopping pandemics and it wasted a lot of turns, which are really the limiting factor in Pandemic.
After some rules discussion, we played the next game by relaxing that giving cards could be done on either the giver's or receiver's turns. The difference in difficulty was astounding. We won easy without a sweat, whereas before it was more like a coin flip. The other big difference to me was the type of discussions we were having. Instead of trying to go through weird gyrations to pass cards, we talked more about stopping diseases. Now we debated the trade-off of heading off the spread of a plague or getting a cure for another sooner. The focus had returned to what the designers had in mind, instead of trying to go through the gyrations of a small rule.
This was a very nice example of how a very simple rule change could have a dramatic impact on how the game is played and really the quality of the game. The lesson I took away from this is when most of the discussions in a game are about how to work within one rule, unless that rule is central to the concept of the game, there is probably something wrong. And then letting the players do what they want to, give cards on any turn in this case, can make a more enjoyable experience as then the players' focus is not distracted by obtuse rules.
Sunday, April 7, 2013
The CSM Memebers live in EVE's Hinterland
[Null Sec] importing goods from Jita and exporting harvested materials to Jita, Elise argued, was a bad design. [CCP] Grayscale agreed...This line in the CSM minutes really frustrated me because it show a lack of understanding in both why EVE Online is an interesting game and how economies work. I believe that what Elise is observing is evidence of good design. The shear volume of trade passing through Jita, the main trade hub in EVE, shows that EVE has a deep and highly connected economy. Further the exporting of raw materials and importing of finished goods indicates a large degree of specialization, which is evidence of a highly developed economic system.
Elise & CCP Grayscale don't see that they would have to turn the entire EVE world on its head to achieve a higher level of industry in Null Security Space and less reliance on High Security Space. The full exchange on page 45 discusses changing industry incentives and resource distribution to make Null Sec look more desirable misses the logic completely. Higher production efficiency and more abundant resources would have lots of trouble off setting the inherent locational and security advantage of High Sec.
High Sec is the only region of EVE where NPCs police behavior. Any crime, like blowing up another ship, will result in speed retribution. To the extent that it is impossible for a player to keep his current ship once he has committed a crime. This makes it really easy to move items and materials around. Especially low value to volume items which can auto-piloted to the destination, leaving the player free to do other things too. On top of the level of safety, High Sec is also the central region of space. Meaning it is easier to move good where ever they are needed as opposed to an operation on the far side of EVE's galaxy.
Null Sec is space that can actually be owned by players. It forms the outer ring of the galaxy. Also there is no police response of any kind, so your ship is not safe. Granted a player that is part of the alliance which own the space will be very secure there. There are also certain logistics structures which makes moving things around fairly easy for those that inhabit the space. The issue is any independent player will not be able to gain a market foothold without becoming part of the alliance and subject to its rules & taxes. Further any moving requires an active player, auto-pilot is certain death in Null Sec.
The real issue is Null Sec would not be Null Sec without the features mentioned above. But it is these very features which make the space undesirable to industrialists. So I see no way to improve Null Sec industry without turning the game on its head, or providing very extreme incentives to make players to what they don't want to. Which is bad design because player's don't enjoy being forced to do things.
To make a historical analogy, High Security is the center of a strong empire. People are free to come and go as they please and are relatively safe. Null Security is the hinterland or frontier of the empire. It may be ruled by local strong men demanding tribute or even be completely lawless. Often this hinterland contains materials which the central empire needs, but rarely is any actual production of finished goods done there. It is too easy for the local strong my to extort away all profits and too much output is lost in transport to thieves.
In this light the truly strange thing is the CSM is composed mostly of these frontier strong men with very few representative of the strong central empire. This seems like a case of the fox guarding the hen house.
Friday, April 5, 2013
The Experience of Design
I started reading The Art of Game Design: A Book of Lenses by Jesse Schell recently. So far it is a pretty interesting book. Some of the information is stuff I have come across in other readings, which is a good thing as this is supposed to be an introductory textbook for game design.
Chapter 2 of the book focus on how a designer is not creating a game but an experience. The game is just a medium for the experience. At the core this seems approximately correct for what a game is. Most video games seek to let you experience some story. Even board and card games often strive to create an experience. For example Magic attempts to simulate a wizards duel. This lens certainly provides focus for a designer as she can constantly evaluate the components of the game against the criterion of how they contribute to the experience she is trying to evoke.
On the other hand focusing on just the experience is limiting and would seem to reduce the depth of the game. Evaluating the concept that a designer creates an experience through the lens of Mark Rosewater's Psycographic Profiles, this focus leads to just designing for Timmy. Johnny and Spike are not drawn in by raw experiences, they have other base needs that drive them. A game that simply provides an experience has little replay value for them because Johnny can't express himself and Spike can't prove himself just through experiencing what the designer had in mind.
I think that Schell's focus on the experience because that is an actual starting place. Starting with either of the other two's desires would be extremely tough. The need to express oneself or prove something makes more sense in the context of experience or existing game, but not much on the blank slate that is the start of design. While I think it is necessary to have Johnny & Spike elements in any game to give it replay value these add to the Timmy experience which starts the creative process.
Chapter 2 of the book focus on how a designer is not creating a game but an experience. The game is just a medium for the experience. At the core this seems approximately correct for what a game is. Most video games seek to let you experience some story. Even board and card games often strive to create an experience. For example Magic attempts to simulate a wizards duel. This lens certainly provides focus for a designer as she can constantly evaluate the components of the game against the criterion of how they contribute to the experience she is trying to evoke.
On the other hand focusing on just the experience is limiting and would seem to reduce the depth of the game. Evaluating the concept that a designer creates an experience through the lens of Mark Rosewater's Psycographic Profiles, this focus leads to just designing for Timmy. Johnny and Spike are not drawn in by raw experiences, they have other base needs that drive them. A game that simply provides an experience has little replay value for them because Johnny can't express himself and Spike can't prove himself just through experiencing what the designer had in mind.
I think that Schell's focus on the experience because that is an actual starting place. Starting with either of the other two's desires would be extremely tough. The need to express oneself or prove something makes more sense in the context of experience or existing game, but not much on the blank slate that is the start of design. While I think it is necessary to have Johnny & Spike elements in any game to give it replay value these add to the Timmy experience which starts the creative process.
Monday, April 1, 2013
Measuring Meat lost to Pirates
A friend of mine pointed out this article to me. It is written by Tommy Refenes, one of the developers of Super Meat Boy, on the joint topics of losses due to piracy and consumer confidence in a game studio. His points on the value of consumer confidence are spot on. Having a good reputation with your customers is key for them trying new games you produce & continuing to support your current offerings. Tommy considers keeping consumer trust as far more valuable than worrying about pirating. I completely agree with him on this. My disagreement with the article stems from his reasoning on why piracy is not a problem.
Essentially Tommy believes that piracy is not a problem because we can't precisely calculate the loss. As we can't pin it down exactly ever, we shouldn't even try to estimate it. An implication of this logic would be we should not come up with theories to try and figure out any counter factual. The problem is that coming up with theories, working out their testable implications, & then testing them is the basis of the modern scientific method. That is why to me his main argument can be summed up as giving up.
Really anyone can come up with rough bounds on the loss due to piracy, [0,Price*Pirated Copies]. The power of theory is then tightening these bounds, especially the lower one, as that is more relevant to the DRM worst case cost-benefit analysis. Granted the theory is only as good as you think the assumptions are. But we can compare games with different pricing strategies & types of DRM to try and separate competing assumptions. For example, comparing a game with a fixed price to one that has periodic, but randomly timed sales, which should allow us to understand how sensitive piracy is to price. I am sure with the data most companies collect quite a bit could be said about the true costs of piracy.
Returning to why piracy is less important than consumer trust, essentially piracy is a good problem to have. If people want to pirate your game, they must find it fun enough to want. So if people want to pirate your game, you clearly have some player confidence that you are delivering a quality experience. Granted you can have cases where the reason to pirate is the DRM is invasive and it is easier to play without the DRM on. But these are cases of companies throwing the baby out with the bathwater due to not understanding user issues and being overzealous in piracy elimination. So barring MMOs, which almost have perfect DRM by definition, there must be an optimal level of piracy & DRM where one balances increasing trust with providing just enough barriers to piracy to deter some people.
Essentially Tommy believes that piracy is not a problem because we can't precisely calculate the loss. As we can't pin it down exactly ever, we shouldn't even try to estimate it. An implication of this logic would be we should not come up with theories to try and figure out any counter factual. The problem is that coming up with theories, working out their testable implications, & then testing them is the basis of the modern scientific method. That is why to me his main argument can be summed up as giving up.
Really anyone can come up with rough bounds on the loss due to piracy, [0,Price*Pirated Copies]. The power of theory is then tightening these bounds, especially the lower one, as that is more relevant to the DRM worst case cost-benefit analysis. Granted the theory is only as good as you think the assumptions are. But we can compare games with different pricing strategies & types of DRM to try and separate competing assumptions. For example, comparing a game with a fixed price to one that has periodic, but randomly timed sales, which should allow us to understand how sensitive piracy is to price. I am sure with the data most companies collect quite a bit could be said about the true costs of piracy.
Returning to why piracy is less important than consumer trust, essentially piracy is a good problem to have. If people want to pirate your game, they must find it fun enough to want. So if people want to pirate your game, you clearly have some player confidence that you are delivering a quality experience. Granted you can have cases where the reason to pirate is the DRM is invasive and it is easier to play without the DRM on. But these are cases of companies throwing the baby out with the bathwater due to not understanding user issues and being overzealous in piracy elimination. So barring MMOs, which almost have perfect DRM by definition, there must be an optimal level of piracy & DRM where one balances increasing trust with providing just enough barriers to piracy to deter some people.
Thursday, March 28, 2013
Seagull Droppings
I have been doing some catchup reading on the CSM minutes. (The Council of Stellar Management is a player elected group which CCP consults on design focus & decisions for EVE Online.) One of the most interesting sections, and the section that has made me the most hopeful for the direction of EVE Online, was the session titled EVE Online - The Next Decade. This section was lead by CCP Unifex and CCP Seagull.
I was extremely impressed with CCP Seagull's comments on how the company is starting to analyze new features for EVE. The biggest contribution seen in the CSM minutes is her additions to the EVE developer lexicon. She identifies 5 major types of players:
To me the fact that people in the company are starting to definite types of players and design specifically for those types of players shows that CCP is maturing. This shows a better understanding of their game and more focus in improving EVE. In reading the minutes it is clear that CCP Seagull it talking on a level above the CSM and even most of the other developers in the meetings. This terminology and way of thinking is clearly very new to many people are still adjusting to it.
There is a similar evolution in vocabulary and way of thinking that happened in Magic: the Gathering, most of which was driven by Mark Rosewater. I have blogged a bit about the Psycographic Profiles before. It is interesting to see that both CCP and Magic created terminology to talk about different types of players after about 10 years of existence. This seems to be a natural evolution of game companies as their products age they develop a lexicon to facilitate thinking and discussion of feature.
The last thing worth pointing out is the 5 player types CCP Seagull defines are categorized by what role they play in a group. The Psycographic Profiles of Rosewater are about why people play. These two different systems don't compete with each other as they define different parts of the player. In theory one could use both systems to construct a more detailed profile of a player. Though that is an exercise for another blog post.
I was extremely impressed with CCP Seagull's comments on how the company is starting to analyze new features for EVE. The biggest contribution seen in the CSM minutes is her additions to the EVE developer lexicon. She identifies 5 major types of players:
- Instigator: Players with ambitions who inspire groups to act.
- Enabler: These players run the logistical side of large groups so that things function smoothly.
- Lurker: Solo players who have little contact with others.
- Group Player: Players that want to engage socially with other though joining activities.
- Small Scale Leader: These players put together activities for others to do. They differ from instigators in not having as large a scale, more an evening's activity verses conquering an empire.
To me the fact that people in the company are starting to definite types of players and design specifically for those types of players shows that CCP is maturing. This shows a better understanding of their game and more focus in improving EVE. In reading the minutes it is clear that CCP Seagull it talking on a level above the CSM and even most of the other developers in the meetings. This terminology and way of thinking is clearly very new to many people are still adjusting to it.
There is a similar evolution in vocabulary and way of thinking that happened in Magic: the Gathering, most of which was driven by Mark Rosewater. I have blogged a bit about the Psycographic Profiles before. It is interesting to see that both CCP and Magic created terminology to talk about different types of players after about 10 years of existence. This seems to be a natural evolution of game companies as their products age they develop a lexicon to facilitate thinking and discussion of feature.
The last thing worth pointing out is the 5 player types CCP Seagull defines are categorized by what role they play in a group. The Psycographic Profiles of Rosewater are about why people play. These two different systems don't compete with each other as they define different parts of the player. In theory one could use both systems to construct a more detailed profile of a player. Though that is an exercise for another blog post.
Thursday, March 14, 2013
Social Games are not designed for Spike
I just read this blog over at Gamasutra and my impression is that the author is a type of Spike. He wants to demonstrate mastery through goals in game and doesn't understand how games like FarmVille can be appealing. Essentially these games are not designed for him but a very different market segment.
The way these social games work is people have their space, in FarmVille a farm, which they can customize & use to produce things. These products can be used for further customization or traded with other players. Also each player's space can be view & discussed by their friends. In truth, this is not much different from having a flower garden in one's yard. It provides a topic of conversation among friends and complementing the garden is a compliment to the gardener. In essence the garden is an expression of the gardener's self, just like the farm in FarmVille. Which is what Johnnies are all about, expressing themselves.
This isn't to say that Spikes and Timmies can't enjoy the game. Timmies love the experience of hanging out with their friends and bonding over a shared interest. Some Spikes will relish mastering the production method in FarmVille to create the most productive farm. Granted once a Spike masters the game and proves that he has mastered it he will probably get bored. But this does not mean social games suffer from poor game design, just very targeted design.
The way these social games work is people have their space, in FarmVille a farm, which they can customize & use to produce things. These products can be used for further customization or traded with other players. Also each player's space can be view & discussed by their friends. In truth, this is not much different from having a flower garden in one's yard. It provides a topic of conversation among friends and complementing the garden is a compliment to the gardener. In essence the garden is an expression of the gardener's self, just like the farm in FarmVille. Which is what Johnnies are all about, expressing themselves.
This isn't to say that Spikes and Timmies can't enjoy the game. Timmies love the experience of hanging out with their friends and bonding over a shared interest. Some Spikes will relish mastering the production method in FarmVille to create the most productive farm. Granted once a Spike masters the game and proves that he has mastered it he will probably get bored. But this does not mean social games suffer from poor game design, just very targeted design.
Subscribe to:
Posts (Atom)