Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 23, 2014
arrowPress Releases
October 23, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Player Controlled Progression
by Josh Bycer on 07/03/13 02:00:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Reprinted From my Site: Game-Wisdom

In the past I've talked about progression in game design by defining it as short-term and long-term. Where short-term represented minor or constant changes like new equipment. And long-term were major changes that affected how the player continues to play such as runes in Diablo 3.

However we can further break down progression based on the amount of control the player has over it and as always, there are specific advantages and disadvantages to consider.

Diablo 3

Foundation Building:

In many games with RPG elements these days, the player is given control of how their character is developed. This could mean everything from what skills to have active at one time, or in what order upgrades become unlocked.

Basically the player knows at any given time a good idea of what they're going to unlock next and how to get it. Even in a RPG with randomized loot like Diablo 3 or Din's Curse, the majority of the player's abilities are based on what skills they have access to, with said skills based on leveling up.

Giving the player a time-frame for when they are given new toys to play with is a popular motivation mechanic to keep people playing a game. Who here while playing a MMO or RPG has ever stared at their experience bar after a fight to see how much of it filled up?

With open world games built around upgrades, the time-frame is relative, based on how quick the player pursues the required elements. But the player still knows that they can become stronger if they kill X enemies or gather Y items.

Diablo 3
All equipment in Dark Souls was set and finely balanced, allowing players to personalize their gear and attack style.

The advantage of giving the player control of the progression is that it can make the character become personal to the player. As the player is in control of how their character develops.

As mentioned, it is easier to motivate players with set goals or a check-list, as opposed to telling them that they "may" find something useful while playing.

From a design point of view, knowing exactly the upgrade paths and choices the player can make over the course of playing, makes it easier to balance the game out. For example: If you know that by the first boss fight that it is only possible to acquire more health, then you don't want to make that fight require additional upgrades.

Many designers don't sweat the details with an upgrade system and instead balance the game around a loose play-style: someone who plays the game and not focusing on getting every upgrade. That way the people who do try to get them all will notice that the upgrades paid off without punishing those that didn't.

While having set upgrades makes it easier to view progress, there are several issues to be aware of. By having a completely rigid list of upgrades for the player to find, means that there are specific actions and limits to what the player can accomplish. Meaning that anything not relating to progression can be seen as wasted time.

If the only way for the player to become stronger is to kill 10 red enemies, then spending the next two hours fighting nothing but blue enemies will seem like a waste.

What's worse is if you have tiered progression that requires the player to complete X before Y begins to count. What ends up happening is that the player will likely perform both X and Y actions at the same time, but only the X actions will count. This was frustratingly seen in Red Dead Revolution with the hunting side quests that required you to kill specific animal types in a set order.

Diablo 3
Even completely open world titles like Just Cause 2 featured set progression in the form of upgrade crates. But this created a limit on character improvement.

The other problem with set progression is right in the name: there is a set number of ways for the player to improve. Once the player has reached max level, or found all the upgrades, there are no other ways to show meaningful improvement.

In Diablo 3, so much of the character progression and abilities were based on unlocking new skills and runes in contrast to equipment. After the player unlocked the last rune, progression grinds to a halt with the only area of progression left are random upgrades. There was also the fact that set upgrades also meant a linear path through the game, making repeat plays less interesting.

When there is nothing more to look forward to or the goal post is placed too far away, most people will see this as the time to stop playing. This is where providing variable upgrades comes in.

Rolling the Dice:

Knowing when and how upgrades will happen can provide steady motivation for playing, but random upgrades can be a good surprise.

As we mentioned up above, randomized upgrades can co-exist with set upgrades, but for this category we're going to talk about games where they are the primary means of improvement.

Not counting Diablo 3, most ARPGs give the player a wide variety of randomized equipment to build their characters from. This equipment can have a noticeable affect on the player's play style from that point on. Such as giving a warrior life steal, or creating a build around using ice weapons to freeze enemies.

Diablo 3
Drox Operative featured a variety of randomized equipment designed to alter your play style based on what you have equipped.

While Borderlands 2 did feature set upgrades in the form of skills for each class, the main form of progression came from finding new guns with wildly varying differences in abilities. Changing guns could have a huge impact on your style of play: such as switching from an assault rifle to a shotgun.

The fact that you never know when a new item will appear can be as motivating (if not more so) as knowing how much experience you need for the next level. With randomized progression, there is a lot more replay ability as you know that there isn't a set limit for how powerful you could become.

In terms of design, games built around randomized progression are more open compared to set progression, both in terms of upgrades and the general path through the game. The player is usually given a wide variety of weapons and possible strategies, as the game has just as many different enemies and situations.

If the game has enough variety in terms of modifiers and different equipment, with enough stat tweaking you could have a game that goes on for a very long time. If not infinitely with a consistently rising curve of equipment quality and danger.

However it's time to talk about the problems and there are several big issues that can happen. First is that it’s hard to personalize a character when everything that makes up their utility is random. If you like using one specific weapon or style, it can be disheartening to be forced to use something else on account of only finding upgrades of a different type.

While having random upgrades can be motivating by the fact that you never know, it also means that you have no idea if you'll get any upgrades. Equipment upgrades are meant to be short-term as you want to be consistently getting better gear the more you play. But if you're going upwards of several hours using the same equipment, then there is a problem.

Another problem is if the power curve increases too slowly over the course of playing. In this case the player will find a lot of side-grades, but not a lot in terms of major upgrades to their character. I had this issue with the Torchlight series, as I had to keep stopping and starting after each fight to check the minute differences between equipment.

Finally the big issue is that game balance can be very tricky to pull off when the designer has no idea what the general power level of the player is at anytime. As with Diablo 3, the designers went overboard with the initial inferno mode settings and they weren't balanced around the relative range of equipment stats that were available to the player.

Or in Borderlands 2's case: you may come to a section that was easier with a specific weapon or damage type, but you may have not found the specific type relative to the current threat level.

Diablo 3
Many action titles these days feature RPG mechanics but when not balanced properly, can lead to frustration.

What's worse is if the game features set battles or situations, but with randomized equipment. What usually ends up happening is that either the fight will be so easy because the player found stronger equipment, or it will be frustratingly difficult due to lack of upgrades.

Examining both sides of progression there isn't a clear-cut winner. Both types of progression can be used to make an amazing game. Deciding what kind of progression you are going to use is a pivotal decision and one that will affect everything about the design of your game from that point on. It's better to fully commit to either one or the other, or a combination of both early on and stay the course.

Defining all areas of progression in a title is like building the foundation of a house: if you alter it once the building begins to go up, the whole thing could come crashing down.

Related Jobs

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer
Bohemia Interactive Simulations
Bohemia Interactive Simulations — ORLANDO, Florida, United States

Game Designer


Yiannis Koumoutzelis
profile image
Random upgrades is a simple and meaningful way to do it, but still does not really set you free from level/equipment requirements in order to progress. I think skill based flow complimented with random gear bonus could be a better solution opposed to the classic level/equipment based gameplay in order to achieve that.

You could have a world that lives in a different pace than the player, but that would require a tremendous effort and cost in order to constantly provide relevant hooks for a player's ascension and opportunities for them to get involved with what is happening in that world at different levels. It could be covered with live development and "events" at various stages of the game i.e. certain main quests being available in a particular location only for a fixed amount of time and then they move on and the story evolves. this would really cost a lot of money to set up it could be a huge challenge and it would constitute a huge risk imo. If i am not mistaken URU was meant to be something like that but in a more limited puzzle adventure/lore based paradigm.

Ramin Shokrizade
profile image
The focus here seems to be on single player experiences, and from my experience the time a consumer will play one of these is a lot shorter than for an MMO. Thus when focusing on a single player game it is my opinion that dopamine effects are primary and as such randomized loot is very helpful, avoiding grinds is helpful, and as always equity losses should be used carefully. Players tend to need a break from intense play every 100 to 120 minutes, so distributing your content in "chunks" of about this length makes it easy for players to take breaks and refresh.

In an MMO environment dopamine is still important, but now you have peer to peer interactions that take place in other parts of the brain that can ultimately trump dopamine effects. Since MMOs are my thing I focus a lot of attention on prestige and equity effects which are not helped as much by randomization as they are by skill building and display. Both mechanisms are important, but these two approaches are so different that you are likely going to need separate groups building both systems at once.

For more insights my Virtual Achievement chapter from 2009 on Gameful is a useful resource. If you want to read about how the D3 RMAH disrupted the reward mechanisms in that game, there is my 2011 Smedley's Dream... paper also on Gameful.

Kaitlyn Kaid
profile image
I very much agree that for mmo-style games, pride is often as large a driving factor for reward as actual statistical advantage. I can not count the hundreds of inventory slots I've used over the years to hold a cool looking weapon, or set of armor, simply because of the appearance or to show off my completion of difficult content. Random rewards really fail in this since there is no way to know how much effort/skill was required to acquire the reward (oh the number of times I had to run X content to get a cool looking item to the point where I could complete it blindfolded, vs the number of times I got the cool item as a fluke on my first try).

This is one field where I think that a lot of improvement is left to be made in the single player, non-mmo multi-player games.

David Serrano
profile image
I've found many games in recent years feature a type of progression - level up paradox. To unlock reward B, players must first obtain reward A. But realistically... reward A can't be obtained unless players have reward B equipped. Or the drop rate for reward A is set so low that obtaining it is reduced to pure luck. Final Fantasy XIII was a great example of this. Some rewards needed to craft more powerful or rare gear had 1 percent drop rates. And there were no alternative options for obtaining those items. So you could literally re-fight the same battles hundreds to thousands of times and still not earn the rewards.

While there's no published data to confirm this, it's a pretty good bet these types of obstacles will cause a majority of players to stop playing.

Darren Tomlyn
profile image
I wish I could give you a really long, informative reply to this, but I'm not really in a position to fully back it up - all of which is exactly what my blog is for and about.

The fact that I'm having to start with, and base everything upon, a post I'm currently writing called 'On The Functionality And Identity Of Language', however, should tell you that there's more to this than you are (and everyone else is) currently focused upon.

One of most basic symptoms of the problems I'm describing in my post atm, is how we use the language to describe itself - which is an (very) indirect part of the problem here, because of how we describe the information we use the word game to represent.

The problem with your post is three-fold:

1) Your description of the behaviour/things that happen we're concerned with as being 'progression' is too broad and not specific enough.

2) Your descriptions of how this is applied and enabled is too specific and limiting - (and some of the terms used are inconsistent for the elements involved, though this is not currently recognised (i.e. 'RPG')).

3) The player being given a random 'reward' is not consistent with player-control unless they have influence over the randomness itself, (such as the act of directly rolling a pair of dice) - which your post has not specified.

It's because of 3) above that the bigger, more fundamental, picture is needed to fully explain and understand the specifics of what you're talking about, especially in relation to each other and games themselves, which is something we currently do not have.

So, the bigger picture is that we're talking about games, specifically - which 'player controlled progression' doesn't limit itself to. More importantly, we're talking about the behaviour of the player when playing a game itself - gameplay.

However, progression isn't precise enough, because it doesn't best describe what we're talking about - change - and how and why it happens over time as the game is played.

In short, what we're talking about here, is:

Gameplay development.

Now, of course, this is a little problematic because we call the act of creating a game itself, development, which can be confusing, but I feel it's the only term that really describes what is happening, (and what we want to happen), within a game being played.

Unfortunately, we now have a problem - a MASSIVE, FUNDAMENTAL problem - and answering the question of how and why such a problem can even exist, has lead me to figuring out problems with our current perception, recognition and understanding of language itself...

We need to recognise and understand the difference between gameplay development on behalf of the player, and gameplay development on behalf of the game/the developers/creators.

The OP has failed to do just that - and therefore demonstrated a basic symptom of the problems I'm looking at:

Of confusing something a person DOES, with something that happens TO them.

Do you understand why they can NEVER be the same thing?

If I don't roll the dice, it's not part of my behaviour, and therefore the gameplay as a player.

ALL types of basic gameplay can be developed on behalf of either party as a game is being played - to limit the systems and mechanics (rules) governing such development in ANY way - as and by the playing pieces involved, for example - is to confuse its application with its definition.

As such, the term 'Role-Playing Game' has absolutely NOTHING WHATSOEVER to do with such gameplay mechanics, since they have absolutely nothing to do with 'role-playing' at all. (Merely controlling a character (as a playing piece) is not role-playing, and neither is changing their attributes etc. as the game is played.)

Since games are (or should be!) DEFINED as and by the behaviour of the player(s), ONLY the gameplay development that the player has any power over, (as part of playing the game), can be used to define it as a (type of) game.

Note: there is a lot more that needs to be derived from/based upon this, but it'd take far too long to go through it all.

Jonathan Jou
profile image
Oh, it's you again! How have you been, Darren?

I'm reading over your comment here, and I'm sorry to say that the tone you take and the clarity of your message still leave me feeling defensive rather than interested. Have you developed your linguistic doctrine into something the world might be willing to listen to? I feel like you have a lot to say and it's too bad that people aren't likely to be receptive of your philosophy. How's the linguistics degree coming?

I feel like your point in that randomness can be a choice, but isn't always a choice is interesting. On the other hand, I think more than enough people have discussed why Role-Playing Games are somewhat ill-defined and misleading as a genre.

And most importantly, have you tried finding people who are genuinely interested in the ambiguity of language to discuss your true passion with? I worry that your linguistic ambitions fall on the ears of game developers, who might be happy to call anything which walks and talks like a duck exactly that.

Best of luck!

[Edit: I have no intention of derailing a perfectly good article by continuing this conversation here, and honestly am probably neither problem nor solution since I spend my time on proofs and algorithms these days. Best success in your crusade!]

Darren Tomlyn
profile image

After talking the basics over with a friend involved in linguistics at Cambridge University (UK), he told me to write it up for him, which I'm trying to do as we speak - (it's taking a while, because it's hard to describe problems when the whole reason for their existence is the lack of the correct/consistent/accurate labels/words to describe/represent them - and I have no academic/linguistics background to help).

If I'm right about what I've realised/found - (and it's so basic and obvious that there's no way I can be wrong - (my friend at Cambridge didn't argue with anything at all, he just asked me to clarify a few things then told me to write it up for him)) - then I'm going to be helping to re-write the understanding such degrees are based upon - (studying a particular language for a degree is always going to suffer if our very perception, recognition and understanding of language in general is inconsistent and problematic!)

Choice is completely the wrong word to describe what is happening, how and why. The act of choosing to do something (e.g. roll a pair of dice), is separate from (exercising the power and influence over) the act of actually rolling them.

If I tell someone to roll a pair of dice for me, what has THEIR behaviour got to do with my own, accept as an effect, that is defined and labelled separately? If someone else rolls the dice and moves the playing piece, even at my request, how can I then be the one truly playing the game of Snakes and Ladders, let alone be a game of chance if I don't roll the dice?

Getting confused between the player's behaviour and the game's (on behalf of its creator(s)) is an extremely fundamental mistake.

(Note: There are ways to have a game of chance without directly involving the equivalent of 'rolling the dice' on behalf of the player, but it still involves more input from the player than merely asking someone to roll a dice for you - (such as in draw-based poker.))

A slot machine is NOT a game - it is a competition. A game is an activity in which people compete by doing something for themselves, and a competition is an activity in which people compete to be told whether or not they have won - mainly by a 'judge's' opinion or a RANDOM DRAW - do you people not understand and recognise how much certain people (a particular industry in particular) have to gain by convincing people that something that happens TO them, is exactly the same as something they DO?

Yes, it's called propaganda - and lots of people have fallen for it.

So, are you part of the problem, or the solution?

Eric Schwarz
profile image
A few thoughts.

1) I think that it is worth looking at progression in the context of a full game rather than under a microscope. Progression, specifically player character progression, is only as important as the wider structure of gamplay makes it out to be.

2) This article seems to imply that character progression for its own sake is a good thing. You need to analyze why progression works in a game and why it might not work in a game. For instance, in Tetris there is no need for progression because the core gameplay is centered around mastery of a fixed set of mechanics; progression is a natural result of the player getting better and better at working with the game mechanics.

3) A better question to ask, summarily, is "what does character progression enable for gameplay? is this meaningful to the game experience I am trying to create?"

Broadly, I look at a game as having a definitive beginning, middle and end - specifically an introduction phase, a learning phase, and a mastery phase. The introduction phase introduces the core concepts of gameplay to the player; the learning phase allows the player to experiment with these concepts and grow in skill; the mastery phase represents the peak of the game before the finale and puts all the player has learned to the test. Games themselves are broken up into larger and smaller phases that all follow similar structures. Pacing is the frequency these peaks and valleys occur and how long they last for.

In this interpretation, the idea of an "unlimited game" makes no sense. There is no such thing. All games are going to run out of content. All games are going to get boring. If you look at a typical MMORPG, for example, eventually the player will reach the level cap and even exhaust endgame content. Because progress is defined almost solely using character levels, when those are depleted by the player there is no more progress to make and the game must naturally end as part of its course. Yet developers and publishers seem to expect players to keep going at it forever. It's no wonder many MMOs peak a few months in only to slowly decline as players run out of things to do and novelty wears off.

It's much smarter, therefore, to design a game to follow one of two models:

1) The game has a definite beginning, middle and end, as described above. The game is created as a finite experience from the start and all design decisions are informed by this. Example: typical story-driven single-player action game.

2) The game has the same beginning, middle and end structure, but it is divided into much smaller and repeatable gameplay sessions. Long-term progression can be maintained near-indefinitely by introducing an extremely high skill ceiling through the use of complex mechanics over simplistic and shallow ones. Even once mastery is achieved, the natural rhythm and short session length still ensure gameplay is always interesting. Example: any MOBA-style game like DOTA2 or League of Legends.

Last, I don't think randomness really exists as far as player progression goes. Most games don't feature true randomness; loot-driven action games have strong bias for instance, which will adjust based on the player's level and character build. In many of these games progression is still very fixed because there is still a "predefined" arc to it; it's simply "fuzzier" than what you'd get in other games. The only games that I'd say would have truly open and player-defined progression are games with heavy focus on exploration, research and similar, namely strategy and 4X games.

Josh Bycer
profile image
1. There is definitely merit in examining progression on a game by game basis and we could probably have multiple articles for each game.

Regarding your models, where do you think games built around casual sandbox play (Minecraft) or titles where the player is given everything and challenged to survive or accomplish something (Don't Starve, Dwarf Fortress, the current state of Kerbal Space Program) fit in?

With randomization, this is something that we largely see in games built around Rogue-like elements: Binding of Isaac, FTL, Dungeons of Dredmor and so on. In these cases all it takes is one item or piece of equipment to radically change your playstyle or survival rate. With BOI, I can't count the number of times that one item meant the difference between winning the entire game and losing at the final fight.