Thoughts on score-based game design
by Ian Omroth Hardingham · 05/26/2009 (5:54 am) · 14 comments
(This is a copy-and-paste job from our podcast's blog, Visiting The Village, which I thought you guys might find interesting here)
A lot has been made of how bad death is in games, and “modern” design often points in the direction of trying to remove death. I believe that this does not apply for heavily generative and score-based games. In a heavily architectured game, death is bad because it means you have to repeat things you have done successfully before - this is not fun. In a heavily generative game, death is a clean slate - another world is created for you to play in, you see new things straight away, and you have a new opportunity to succeed. We don’t want games of pinball or Spelunky to routinely go on for hours - that would almost imply that the first wedge of time you spend on each game is not meaningful. This leads to one of my core tenets as a designer:
In a highscore-based game, the risks and rewards at game start should not be exponentially smaller than those x minutes in.
This is violated all the time. Look at Tetris’ classic mode, where blocks start off falling slowly and gradually speed up. At the start the game is basically easy, but you can’t score many points. This is not interesting for the experienced player - why go through a period where your play effectively doesn’t matter to get to the real thing? As for the inexperienced players, that’s what easy mode is for.
You should be able to die in the first thirty seconds, and your likelyhood of dying shouldn’t be that much lower than it is ten minutes in. We can look at this mathematically. In games which violate the above tennet, a “great” playing time is not that much longer than the average - because most of the playing time is comparitavely meaningless.
Average Tetris playing time: 10 minutes. “Great” Tetris playing time: 13 minutes.
Average Theatre Of Magic (Pinball) playing time: 3 minutes: “Great” Time: 15 minutes.
This leads up to another tenet of mine:
In a highscore-based game, the average playing time should be low - 3 minutes is a good start.
When you start playing a highscore-based game, you want to know you could be having a “great” game soon. You want to be saying “wow I’m on FIRE today” after a minute or two, you don’t want to have to grind for ten minutes before you can even hope to sniff a good performance. I had this particular problem with Lumines, where an average game seemed to take about half an hour. A three-minute game is an easy thing to get into. A half hour game seems daunting to even start.
It seems ingrained into games designers that difficulty should ramp up over time. I call this as being outmoded in a generative context. A game should give skilled players the opportunity to do riskier things, not treat them with kid-gloves for the first x levels. If a game is different every time you play it, death is good-frustrating: it compels you to play again.
Finally:
You should only have one life in a score-based game.
Let’s look at Spelunky as a bit of a criminal of this. In Spelunky, you have health. Health is difficult to find and if you lose even one heart on the first couple of levels you’ll want to restart the game. This is bad. Pinball is also a culprit - if you lose your first ball without getting many points, you may be less motivated to play the game through. (I recently turned the number of balls per game on my pinball table down to one).
About the author
Designer and lead programmer on Frozen Synapse, Frozen Endzone, and Determinance. Co-owner of Mode 7 Games.
#2
Death has to be fair and always the player's fault though, and the feedback of why you died and how you can improve has to be very clear.
05/26/2009 (7:30 am)
Not if they had fun in those two minutes. In fact, I think it's much more likely that people will come back if they feel they haven't got the measure of a game the first time around. People stop playing because "oh no, I have to do that AGAIN?!". People come back when they don't get enough of something they like.Death has to be fair and always the player's fault though, and the feedback of why you died and how you can improve has to be very clear.
#3
That's probably true, but if they come back 100 times for a game that lasted 3 minutes each time, that's 5 hours of total playing time. By this time, they've probably gotten the measure of the game, and are tired of dying. Meanwhile, the guy sitting on the couch played one game of Super Mario Brothers during that time, and he'll probably do the same again tomorrow.
Only if "that again" isn't something worth doing again.
I can agree with you in the context of games like Tetris. There's really not much to it, and you're basically doing the same thing with the same pieces in the early stages as you'll be doing in the late levels, they'll just be going faster.
But with games that give you a variety of content and features to boost your score, and obstacles to kill you, there's a lot to be gained through a slower progression. If you just throw random content at them every time they play, it won't take long until they've seen everything, and then it'll just be the same thing over and over again at 3 minutes a pop.
05/26/2009 (8:04 am)
Quote:
In fact, I think it's much more likely that people will come back if they feel they haven't got the measure of a game the first time around.
That's probably true, but if they come back 100 times for a game that lasted 3 minutes each time, that's 5 hours of total playing time. By this time, they've probably gotten the measure of the game, and are tired of dying. Meanwhile, the guy sitting on the couch played one game of Super Mario Brothers during that time, and he'll probably do the same again tomorrow.
Quote:
People stop playing because "oh no, I have to do that AGAIN?!".
Only if "that again" isn't something worth doing again.
I can agree with you in the context of games like Tetris. There's really not much to it, and you're basically doing the same thing with the same pieces in the early stages as you'll be doing in the late levels, they'll just be going faster.
But with games that give you a variety of content and features to boost your score, and obstacles to kill you, there's a lot to be gained through a slower progression. If you just throw random content at them every time they play, it won't take long until they've seen everything, and then it'll just be the same thing over and over again at 3 minutes a pop.
#4
A core gamer would not have a problem diving into a game that was going to take several hours and probably wouldn't enjoy a game that only lasted a few minutes.
Although you have some good ideas for casual games, as a game designer, you should generally not have "tenets". Every aspect of a game design needs to be re-evaluated specifically in the context of the game you are designing.
This does not mean a designer cannot reuse elements from previous games, nor does it mean that a tenet will ever be wrong, but they all still need to be re-evaluated.
This is the path to innovative game designs.
Slaughter those sacred cows! :P
05/26/2009 (8:07 am)
Since I'm a core gamer and not a casual gamer, I immediately rejected all of your tenets, but after reflecting on it for a few minutes I realized you were probably talking about casual games.A core gamer would not have a problem diving into a game that was going to take several hours and probably wouldn't enjoy a game that only lasted a few minutes.
Although you have some good ideas for casual games, as a game designer, you should generally not have "tenets". Every aspect of a game design needs to be re-evaluated specifically in the context of the game you are designing.
This does not mean a designer cannot reuse elements from previous games, nor does it mean that a tenet will ever be wrong, but they all still need to be re-evaluated.
This is the path to innovative game designs.
Slaughter those sacred cows! :P
#5
05/26/2009 (8:10 am)
Certainly, I'm specifically talking about score-attack games. I love long FPS and RPG games as a core gamer - but for games which are supposed to be pick-up-and-play, I find that having to grind for minutes is completely contrary to the idea. 
#6
This does not mean pick-up-and-play. There are lots of pick-up-and-play games that are not score-attack and lots of score-attack games that are not pick-up-and-play.
This is a hearken back to arcade games. I would be pissed if I went to an arcade, dropped in a quarter to play my favorite game, and got to play for only 3 minutes.
How "dynamic" could a 3 minute game be? If not dynamic, how stale would it become? The problem with such a short game life is that you have to micro balance things even more than a full length game.
You should re title your tenets. Instead of "Tenets of Score Based Games" it should be "Tenets of the Highest Diffuculty, speed run, super pro replays of score based games" where the player blazes through all of the content and never dies once.
For us meager mortals (an overwhelming majority of the players,) we would like a little meat and some room for error while discovering more about the game.
While your views seem correct, I think they are distilling what is a rich gameplay experience into a condensed "shot" of a perfect high score game. I used to agree with you, as I used to play competitive games at tournaments etc. but you really are alienating the quiet masses who otherwise completely support the sales of your game. Not everyone hates dying and getting to retry at the cost of their score. Some people just want to see the end of the game. And if it only lasts 3 minutes I would always feel cheated, no matter how "rewarding" the game play is.
I think between 10-30 minutes is a good "pick up and play" game. If it's a free web based game, then feel free to make it 3 minutes, that way people don't feel cheated.
05/26/2009 (8:52 am)
Score attack games are meant to be unilateral competition. That is, you compete indirectly with people who aren't playing with you at the same time using a common weighting system called your "score".This does not mean pick-up-and-play. There are lots of pick-up-and-play games that are not score-attack and lots of score-attack games that are not pick-up-and-play.
This is a hearken back to arcade games. I would be pissed if I went to an arcade, dropped in a quarter to play my favorite game, and got to play for only 3 minutes.
How "dynamic" could a 3 minute game be? If not dynamic, how stale would it become? The problem with such a short game life is that you have to micro balance things even more than a full length game.
You should re title your tenets. Instead of "Tenets of Score Based Games" it should be "Tenets of the Highest Diffuculty, speed run, super pro replays of score based games" where the player blazes through all of the content and never dies once.
For us meager mortals (an overwhelming majority of the players,) we would like a little meat and some room for error while discovering more about the game.
While your views seem correct, I think they are distilling what is a rich gameplay experience into a condensed "shot" of a perfect high score game. I used to agree with you, as I used to play competitive games at tournaments etc. but you really are alienating the quiet masses who otherwise completely support the sales of your game. Not everyone hates dying and getting to retry at the cost of their score. Some people just want to see the end of the game. And if it only lasts 3 minutes I would always feel cheated, no matter how "rewarding" the game play is.
I think between 10-30 minutes is a good "pick up and play" game. If it's a free web based game, then feel free to make it 3 minutes, that way people don't feel cheated.
#7
Which arcades did you go to? 3 minutes is absolute norm for an arcade game, pinball game, quiz game... you name it.
3 minutes may be too low... fair enough. But if not 3, then personally I would say 10ish - this is just for me, not saying that this is the "right" length.
I'm amazed by all the strong opinions - many thanks for them, this is educating me a lot.
05/26/2009 (8:55 am)
Quote:This is a hearken back to arcade games. I would be pissed if I went to an arcade, dropped in a quarter to play my favorite game, and got to play for only 3 minutes.
Which arcades did you go to? 3 minutes is absolute norm for an arcade game, pinball game, quiz game... you name it.
3 minutes may be too low... fair enough. But if not 3, then personally I would say 10ish - this is just for me, not saying that this is the "right" length.
I'm amazed by all the strong opinions - many thanks for them, this is educating me a lot.
#8
I remember many a day of dropping three quarters for a game of T-Mek at a cinema, and so long as someone else would set down in the other pod(s) I could reign supreme and play for free until someone killed me -- one time I missed a whole entire movie because I was so "in the zone".
Any game, casual or otherwise, that only takes three minutes to play is quickly forgotten about by me.
05/26/2009 (11:14 am)
Quote:You got that right!
I would be pissed if I went to an arcade, dropped in a quarter to play my favorite game, and got to play for only 3 minutes.
Quote:Then you must not have been any good ;) I used to be able to walk into an arcade, plunk in a quarter on just about any scrolling shooter and still be playing 15 minutes or even an hour later (rarely) from that initial quarter. My initials were associated with so many high-scores it was crazy, and yet I never spent more than a dollar or two in several hours at the arcade.
Which arcades did you go to? 3 minutes is absolute norm for an arcade game, pinball game, quiz game... you name it.
I remember many a day of dropping three quarters for a game of T-Mek at a cinema, and so long as someone else would set down in the other pod(s) I could reign supreme and play for free until someone killed me -- one time I missed a whole entire movie because I was so "in the zone".
Any game, casual or otherwise, that only takes three minutes to play is quickly forgotten about by me.
#9
Harkening back to the arcade days is part of what I was thinking about too. Then even for the high-scoreathons, it wasn't even just about the high score, it was also about how long you stayed on the game. If you could keep the arcade open for an hour or two after their normal closing time while they watch you finish your game, then you were really legit :P
Of course there were some that were just obsessed with high scores... I had a friend who rigged an electric sander up to the controls on Track & Field at the local bowling alley that he worked at, so that he could post an unbeatable high score. I think he may have done that in about 3 minutes :o
05/26/2009 (11:49 am)
Yeah I think I was thinking in somewhat broader terms than the OP in relation to "score based games".Harkening back to the arcade days is part of what I was thinking about too. Then even for the high-scoreathons, it wasn't even just about the high score, it was also about how long you stayed on the game. If you could keep the arcade open for an hour or two after their normal closing time while they watch you finish your game, then you were really legit :P
Of course there were some that were just obsessed with high scores... I had a friend who rigged an electric sander up to the controls on Track & Field at the local bowling alley that he worked at, so that he could post an unbeatable high score. I think he may have done that in about 3 minutes :o
#10
I would like to play devil's advocate a little though...
If Tetris fails so much in this area, why is it still one of the most played games of all time? Does Tetris succeed *despite* this weakness because of its other mechanics that shine or there something about the slow ramp up that appeals to a majority of people? It seems to me that if this was a true failure on Tetris' part then some other game would have eclipsed it long ago (it is easy enough to build a Tetris game tuned around 3 min perfect sessions with little to no difficulty ramping).
One thing that strikes me about your "give the player everything from the beginning so they can finish the perfect game quickly" philosophy is that it seems to run counter to the "easy to learn, hard to master" tenet of pretty much every fun, accessible, and successful game out there.
05/26/2009 (2:37 pm)
Definitely some good thoughts here =)I would like to play devil's advocate a little though...
If Tetris fails so much in this area, why is it still one of the most played games of all time? Does Tetris succeed *despite* this weakness because of its other mechanics that shine or there something about the slow ramp up that appeals to a majority of people? It seems to me that if this was a true failure on Tetris' part then some other game would have eclipsed it long ago (it is easy enough to build a Tetris game tuned around 3 min perfect sessions with little to no difficulty ramping).
One thing that strikes me about your "give the player everything from the beginning so they can finish the perfect game quickly" philosophy is that it seems to run counter to the "easy to learn, hard to master" tenet of pretty much every fun, accessible, and successful game out there.
#11
Firstly, I don't think you should give the user all content straight away. I also don't think that all content should have the same risk/reward. What I'm saying is that risk/reward should not be a simple increasing function of time-into-the-game.
Second - and this comes from the arcade stuff - I still think that losing quickly to start with adds meaning to the game. So maybe I should modify my second tenet - it's not the average playtime that should be three minutes, but the first few playthroughs.
I imagine you guys won't agree with that, possibly saying that it will put users off if they lose too quickly. But I'll stand by this - I think it's worse to not lose quickly enough - I like to feel my play is meaningful, and games without loss are just toys. I like competing, and surely that requires losing, especially as you learn? Losing does not have to be un-fun if it's handled properly.
I understand that this is all vastly controversial, and a lot of it quite personal - but I'm definitely enjoying the conversation.
05/27/2009 (4:19 am)
Ok, I'm just digesting some of the responses here.Firstly, I don't think you should give the user all content straight away. I also don't think that all content should have the same risk/reward. What I'm saying is that risk/reward should not be a simple increasing function of time-into-the-game.
Second - and this comes from the arcade stuff - I still think that losing quickly to start with adds meaning to the game. So maybe I should modify my second tenet - it's not the average playtime that should be three minutes, but the first few playthroughs.
I imagine you guys won't agree with that, possibly saying that it will put users off if they lose too quickly. But I'll stand by this - I think it's worse to not lose quickly enough - I like to feel my play is meaningful, and games without loss are just toys. I like competing, and surely that requires losing, especially as you learn? Losing does not have to be un-fun if it's handled properly.
I understand that this is all vastly controversial, and a lot of it quite personal - but I'm definitely enjoying the conversation.
#12
The inicial slower paced part of the game has a very important part to play - it helps the player adjust to the gameplay - this is specially important for games that have different mechanics from others like it.
For exemple, the game Wanted: Weapons of Fate has several things not common in most FPS / TPS games out there (curve bullets, time distortion, etc.) so when you get a new skill they put in a level where you can't loose and you can learn to use that skill.
I think this is important for non hard-core gamers (gets them a chance to learn the skill in a controlled arena where they can't feel frustrated if the first few times they don't get the hang of the skill) and the hard-core gamers get to be introduced to the skill when they play the first time and get really fast back into action since that level is so short and easy
05/27/2009 (5:52 am)
I see things in a different wayThe inicial slower paced part of the game has a very important part to play - it helps the player adjust to the gameplay - this is specially important for games that have different mechanics from others like it.
For exemple, the game Wanted: Weapons of Fate has several things not common in most FPS / TPS games out there (curve bullets, time distortion, etc.) so when you get a new skill they put in a level where you can't loose and you can learn to use that skill.
I think this is important for non hard-core gamers (gets them a chance to learn the skill in a controlled arena where they can't feel frustrated if the first few times they don't get the hang of the skill) and the hard-core gamers get to be introduced to the skill when they play the first time and get really fast back into action since that level is so short and easy
#13
05/27/2009 (5:54 am)
Also - it's those first 20 - 30 min lower paced action that gets the non hard-cored gamer involved enough in the game to want to endure the other harder faster paced hours of the game
#14
I believe a lot of newer tetris versions have super death mode; the pieces drop instantly. There is no ramp up in difficulty.
There is some competitive following for that style of gameplay. However, it's not what people think of when they picture tetris.
@Ian -
Losing quickly will add meaning to a game, but it will piss off people if they payed any money for it. The game "I Wanna Be the Guy" (free downloaded 2d sprite game) is one of the hardest games ever, but I would have asked for my money back if I payed for it.
This is an important thing. When money is involved, people expect a different experience. If they are just playing around on the internet and come across a really fast paced hard game that's fun, they'll play it for a long while, but then you have to go through advertising to gain revenue.
Games are a business, and your tenets describe a charity. That's not to say that charities or your tenets are bad, but they have a different place in the economy.
05/27/2009 (8:56 am)
@Matt - [Re:Tetris]I believe a lot of newer tetris versions have super death mode; the pieces drop instantly. There is no ramp up in difficulty.
There is some competitive following for that style of gameplay. However, it's not what people think of when they picture tetris.
@Ian -
Losing quickly will add meaning to a game, but it will piss off people if they payed any money for it. The game "I Wanna Be the Guy" (free downloaded 2d sprite game) is one of the hardest games ever, but I would have asked for my money back if I payed for it.
This is an important thing. When money is involved, people expect a different experience. If they are just playing around on the internet and come across a really fast paced hard game that's fun, they'll play it for a long while, but then you have to go through advertising to gain revenue.
Games are a business, and your tenets describe a charity. That's not to say that charities or your tenets are bad, but they have a different place in the economy.
Torque Owner Gerald Fishel
Development Ninja
Not everybody is playing a score based game just because they want a high score. The high score is nice, but playing the game is still supposed to be fun and diverting, and there aren't all that many fun and diverting things that last 3 minutes. (This is a family oriented site, so no comments from the peanut gallery.)
There are some instances where it makes sense. For instance, in a game where you'll take turns with other people, a shorter time with a more aggressive level of difficulty makes sense. Nobody wants to wait 30 minutes for their turn, so 3-5 minutes would probably be acceptable. But when people are playing games by themselves, they prefer to play longer games. If they keep dying in the first couple of minutes, they'll turn it off and find something else to do.