The Pacing Discussion

[Charles]

So there was some talk of pacing in the latest podcast. I think it's an interesting discussion because there's a lot of problems with pacing and what it means for games. Devs try and control it somewhat, but ultimately, there's two cases:

1. Dev controls pacing, game is completely linear.
2. Game isn't linear, or incorporates side missions, dev tries to control pacing, can't.

Number one gives us Modern Warfare. There's nothing for the player to do but move forward and follow the pacing provided. If you stop moving, nothing happens. Move forward, game takes over.

Number two gives us any open world game. There's pacing there, if you go mission to mission, but it's completely in the player's hands. If you stop to do any kind of side mission, you inherently break the pacing. Or more to the point, you inherently take control of the pacing, so it becomes your own fault if you break it (assuming, of course, that the main mission arcs are well paced on their own).

Granted, this doesn't mean the problem isn't solvable. I think despite Tom's statement that pacing is a big issue, the reality is that most devs, at least in my experience, don't really think about it enough. Or more to the point, the writer will think about it and everyone else will ignore him. People talk about it online but it's really only people who can appreciate writing and crafting of story who really care, and sadly, a lot of game devs can't.

I think for linear games there's not much to be done, but for open games, there's a lot that can be done if devs think about it up front.

1. Place "breather" points in the story. Mini-wrap-ups where the story effectively tells the player that it's cool to do side missions now without breaking the pacing.
2. Tweak side missions so that there are different types based on where you currently are in the pacing. So if you are in a rising point in the pacing, any side missions you take will reflect that feel of the pacing.
3. Lock out side missions when you want the player to speed through a part of the story.
4. Toss story out completely and let the player make his own.

I guess 4 is wishful thinking at this point, but maybe one day.

[tromik]

You're limiting this to overall action or adventure games, right? Number four applies to a lot of strategy and simulation games.

I think there are some ways to communicate a story and control pacing while allowing the player to deviate from a linear path that haven't really been explored yet. What if you were to take concepts from Portal and Bioshock in terms of delivery, and combine it with a tonne of various non-external event triggers? Let's say you stop chasing the campaign to start a side mission in GTA4. It feels like you're completely separated from that linear storyline. If the world did a better job responding to previous events, and new events occurred regardless of what you were doing, it could feel like you were still moving forward. Little queues based on event, temporal, and state triggers - billboards changing, the city degrading, weather changing - could communicate a lot if done right. GTA4 had those news bulletins, but then you'd go right back into your regular music loop. Taking these types of things further might work.

[Zylon]

Is this thread about plot pacing or gameplay pacing? That needs to be made clear or the discussion will devolve fast.

[Charles]

Is this thread about plot pacing or gameplay pacing? That needs to be made clear or the discussion will devolve fast.

The two are intrinsically linked.

[Charles]

You're limiting this to overall action or adventure games, right? Number four applies to a lot of strategy and simulation games.

I think there are some ways to communicate a story and control pacing while allowing the player to deviate from a linear path that haven't really been explored yet. What if you were to take concepts from Portal and Bioshock in terms of delivery, and combine it with a tonne of various non-external event triggers? Let's say you stop chasing the campaign to start a side mission in GTA4. It feels like you're completely separated from that linear storyline. If the world did a better job responding to previous events, and new events occurred regardless of what you were doing, it could feel like you were still moving forward. Little queues based on event, temporal, and state triggers - billboards changing, the city degrading, weather changing - could communicate a lot if done right. GTA4 had those news bulletins, but then you'd go right back into your regular music loop. Taking these types of things further might work.

True but part of what you said "responding to previous events" is not a realistic solution. At the moment our only choices is to hardcode scenarios, as there's no way of generating this stuff. Devs have to foresee all situations and add content accordingly.

I think part of what you are saying though, is like in Oblivion, instead of having the gates sitting there waiting, have the monsters slowly encroach on the world making it difficult to avoid. It's a good solution but it's expensive at the very least, and that's before we get in to complexity.

I think there's a lot to be said for having the game's story continue without the player but a lot of people flip out at the thought of that happening.

[Vesper]

While a lot of people didn't like it, I think Fallout's approach to this is a step in the correct direction. Largely the game was 'open world', but there was an ongoing timer you had to beat in able to properly finish the game. I like the idea of a hybrid of forced pacing mixed with open world possibilities by utilizing time as a resource. Fallout's problem is that it was largely invisible and it was hard to judge how much time you had left.

[Charles]

While a lot of people didn't like it, I think Fallout's approach to this is a step in the correct direction. Largely the game was 'open world', but there was an ongoing timer you had to beat in able to properly finish the game. I like the idea of a hybrid of forced pacing mixed with open world possibilities by utilizing time as a resource. Fallout's problem is that it was largely invisible and it was hard to judge how much time you had left.

At that point it's just poor design. There should be something in the world, always, which can let the player know how close he is to failing.

Long long ago, when I was a pie in the sky game designer, I had this dream of a game where you started as either the hero or the villain, and the first half of the game was a race to find an artifact which gave one power, and the second half of the game was overcoming your opponent, either once you acquired it, or once they did.

Whichever side you picked, the other side would be run by an AI which would be competing against you on a time line, from a different part of the world but working towards the common center.

[Zylon]

The two are intrinsically linked.
I'm going to disagree. Look at the Thief games. Completely linear plot, but highly non-linear gameplay. Ditto for Deus Ex, or really any "open map" but not open-world game.

[Fugitive]

There's also the issue of pacing and difficulty, especially with RPGs. The pacing is generally more relaxed to begin with, so they don't feel too out-of-place, but if you let them have meaningful rewards then the player might become too powerful and the main quest loses its challenge (e.g., in FFX the final areas and boss fight goes from ball-bustingly hard to trivial if you do some extra questing and get the right summons).

But then maybe some players will get frustrated if they feel the main quest is too difficult for them without the ability to compensate for it by building themselves up a bit more... Balancing the two approaches is the hard part here.

[Charles]

I'm going to disagree. Look at the Thief games. Completely linear plot, but highly non-linear gameplay. Ditto for Deus Ex, or really any "open map" but not open-world game.

If they are completely separate then they are working against the medium rather than exploiting the strengths. It's the same as saying "Look, I can fit a book and a movie together, because I have parts where you read in between the parts where there's a movie."

[tromik]

True but part of what you said "responding to previous events" is not a realistic solution. At the moment our only choices is to hardcode scenarios, as there's no way of generating this stuff. Devs have to foresee all situations and add content accordingly.
I meant respond to previous user-driven events that have already occurred. I used the GTA4 example of the news bulletins on the radio which comment on what you just did and provide some story information. However, after that you go right back into the same radio cycle you were previously. It's a blip at a point in time rather than a change for the duration.


I think part of what you are saying though, is like in Oblivion, instead of having the gates sitting there waiting, have the monsters slowly encroach on the world making it difficult to avoid. It's a good solution but it's expensive at the very least, and that's before we get in to complexity.
I didn't play enough Oblivion, but I like the sound of that. Temporal event triggers combined with user-driven event triggers could be pretty powerful. I'm not a developer, so I have no idea how hard that would be. I'll take your word that we're still quite a ways off of seeing it done properly.


I think there's a lot to be said for having the game's story continue without the player but a lot of people flip out at the thought of that happening.
Yea, take it too far and you end up with the time-limitation as in Fallout, which I agree, is a pretty poor and unimaginative contrivance to keep you moving.

I keep thinking of games like Bioshock or Portal though. What if you moved through an area (One you'd been through previously) after some event or some amount of time and you saw messages scribbled on the wall? The game would have to place them where you'd be sure to see them and judge the amount of time since you left the critical path to make it effective.

[Shadarr]

True but part of what you said "responding to previous events" is not a realistic solution. At the moment our only choices is to hardcode scenarios, as there's no way of generating this stuff. Devs have to foresee all situations and add content accordingly.

I think it's doable, because games have done it. I'm thinking specifically about the point in GTA:SA where the city is engulfed in riots. You might not be able to do something like that for every mission, but you could definitely have triggerpoints that affect the gameworld and give some indication that what you're doing is having an impact.

Saboteur did it too, both with various missions triggering massive changes like the riots, and more subtly with their control mechanic where there would be fewer Nazis and the world would change to colour in areas you'd liberated. And obviously RF:G had a similar system that allowed you to modify the gameworld with a hammer.

I don't think you can do a full, living world like Space Rangers 2 with a real time 3D environment on today's hardware, but that should be the goal. The more you can do to make it seem like your actions matter, the better. In SR2, there was no scripting and nothing compelling you to push the story forward, aside from the fact that there was a war going on and if you did nothing, the enemy would win.

The equivilent in a sandbox game might be to have something a little more granular than Saboteur's system, where depending on which missions you've done and which side stuff, the enemy has more or less of a presence in the area, which makes it harder to do missions. And to make it really engrossing, make the enemy slowly rebuild the stuff you've destroyed if you bugger off and play poker for two hours. That would give the sense that you need to keep working toward the goal to avoid backsliding. And then you can have breaks periodically where you go to a new section that is completely enemy controlled, and at that point there's no penalty for "wasting" time.

[peacedog]

I think part of what you are saying though, is like in Oblivion, instead of having the gates sitting there waiting, have the monsters slowly encroach on the world making it difficult to avoid. It's a good solution but it's expensive at the very least, and that's before we get in to complexity.


Magic Candle 3 immediately sprung to mind, where you watched the world get slowly covered in that plague-thingy. Random combats fought in infested areas were different than out, IIRC.

Exile/Avernum 3 had monster plagues that slowly encroached into "civlized" lands. It was a pretty slow delay. The actual counter was invisible but the game did a decent job pestering you (with the added benefit that you are advised that taking out the plague sources now might help you later, and it makes the end game easier).

Prototype did this but in a completely restricted/linear way. The infection spread after each story mission. It was a bummer.

[Charles]

I think it's doable, because games have done it. I'm thinking specifically about the point in GTA:SA where the city is engulfed in riots. You might not be able to do something like that for every mission, but you could definitely have triggerpoints that affect the gameworld and give some indication that what you're doing is having an impact.

Saboteur did it too, both with various missions triggering massive changes like the riots, and more subtly with their control mechanic where there would be fewer Nazis and the world would change to colour in areas you'd liberated. And obviously RF:G had a similar system that allowed you to modify the gameworld with a hammer.

I don't think you can do a full, living world like Space Rangers 2 with a real time 3D environment on today's hardware, but that should be the goal. The more you can do to make it seem like your actions matter, the better. In SR2, there was no scripting and nothing compelling you to push the story forward, aside from the fact that there was a war going on and if you did nothing, the enemy would win.

The equivilent in a sandbox game might be to have something a little more granular than Saboteur's system, where depending on which missions you've done and which side stuff, the enemy has more or less of a presence in the area, which makes it harder to do missions. And to make it really engrossing, make the enemy slowly rebuild the stuff you've destroyed if you bugger off and play poker for two hours. That would give the sense that you need to keep working toward the goal to avoid backsliding. And then you can have breaks periodically where you go to a new section that is completely enemy controlled, and at that point there's no penalty for "wasting" time.

Yeah there are ways to do it, but there isn't a way to recognize everything the player did without it being prohibitively expensive. Not until we have some really advanced AI with natural language processing, anyway.

[Jason McCullough]

Like most things with gaming, you don't have to do it perfectly, you just have to fake it well enough. Thief is a perfect example of that.

[Marcin]

True but part of what you said "responding to previous events" is not a realistic solution. At the moment our only choices is to hardcode scenarios, as there's no way of generating this stuff. Devs have to foresee all situations and add content accordingly.

If the user's interaction is limited to combat, you could actually generate quite a bit from simple stats, such as we see in GTA. Peds ran over, miles of terrain covered, terrain explored, various gun stats ... etc. If you make the world respond to your actions, you could also record NPC stats and create events out of those as well. Seems pretty straightforward, actually.

Din's Curse does a lot with both player and NPC activity. There's a pool of various events - wars, treaties, migrations, deaths, traveling merchants - and all of it on a timer. It's quite a convincing "open world" storytelling attempt, as long as you accept that the stories are restricted to those events. :)

[Shadarr]

You don't have to track everything individually, though, you can track a couple of variables that are modified by player action. Like you could track the enemy presence and threat level. When the presence is high, there are more soldiers and when the threat level is high, they behave more aggressively. Sort of like the notoriety level in AC2 only without the annoying mechanic of pulling down posters. And maybe a third variable that modifies the behaviour of civilians, so that sometimes the city is full of regular people, sometimes the streets are deserted and sometimes they fight back (or even rat you out).

Really, anything you can do to make the NPCs react like people instead of just being objects is good. SR2 did a pretty good job of this, GTA4 and Saboteur not so much. A perfect example was the the way people reacted to your car in Saboteur. They would walk right in front of a moving car, and wait way too long to dive out of the way as if they had no idea you were there. You could click the left stick to honk the horn, but aside from making noise it didn't accomplish anything, the NPCs didn't react at all.

Something that might be fun would be to combine the dress-up aspect of SR2 with notoriety. Have posters and radio broadcasts saying the police are on the lookout for someone matching your description, whatever that happens to be. Maybe it's a man with blond hair and a goatee wearing a blue t-shirt, maybe it's a woman in a bikini and a fireman hat. Rather than just tearing down posters or sitting on a park bench, it would give you a reason to change your appearance. You could choose to change regularly to keep a low profile, or you could always put on the chicken suit when committing crimes, raising your notoriety on purpose like Zoro.

[Charles]

If the user's interaction is limited to combat, you could actually generate quite a bit from simple stats, such as we see in GTA. Peds ran over, miles of terrain covered, terrain explored, various gun stats ... etc. If you make the world respond to your actions, you could also record NPC stats and create events out of those as well. Seems pretty straightforward, actually.

Din's Curse does a lot with both player and NPC activity. There's a pool of various events - wars, treaties, migrations, deaths, traveling merchants - and all of it on a timer. It's quite a convincing "open world" storytelling attempt, as long as you accept that the stories are restricted to those events. :)

I didn't say it was hard, I said it was expensive.

[Zylon]

If they are completely separate then they are working against the medium rather than exploiting the strengths. It's the same as saying "Look, I can fit a book and a movie together, because I have parts where you read in between the parts where there's a movie."
Ummm... separating story from gameplay *is* working with the medium. If blending story and gameplay was an inherent strength of the medium, game devs wouldn't have so much trouble with it, and you wouldn't have people claiming that great gameplay and great storytelling are mutually exclusive.

[Charles]

Ummm... separating story from gameplay *is* working with the medium. If blending story and gameplay was an inherent strength of the medium, game devs certainly wouldn't have so much trouble with it.

No, separating it is ignoring the strengths of the medium, because the medium is not completely matured, and instead of facing and overcoming the new requirements of this medium the industry in general chooses to partition it in to old parts and new parts.

[Zylon]

Maturity doesn't enter into it. Games are inherently interactive. Absorbing a story is inherently passive. Both games and storytelling have been around for millenia, so if there was an effective way to integrate the two, I'm pretty sure someone would have stumbled on it by now.

[Charles]

Maturity doesn't enter into it. Games are inherently interactive. Absorbing a story is inherently passive. Both games and storytelling have been around for millenia, so if there was an effective way to integrate the two, I'm pretty sure someone would have stumbled on it by now.

Plenty of people have. It's the concept of a player story. You can have player stories and provide a narrative, but it's difficult when shoehorning in an old style story is easy. That's why you don't see much of it. Also it takes a good amount of adventurous development to have even a chance of integrating the two.

[sinnick]

I think for linear games there's not much to be done, but for open games, there's a lot that can be done if devs think about it up front.

1. Place "breather" points in the story. Mini-wrap-ups where the story effectively tells the player that it's cool to do side missions now without breaking the pacing.
2. Tweak side missions so that there are different types based on where you currently are in the pacing. So if you are in a rising point in the pacing, any side missions you take will reflect that feel of the pacing.
3. Lock out side missions when you want the player to speed through a part of the story.

When thinking about Sandbox games like GTA, I think all of these are good ideas. In particular, for #3, an alternate I can see is having the main mission continue whether you are there to take part in it or not. I think it would be great in GTA if, during lulls in the main story you are free to fuck around in the world, but when the story starts amping up, you have to keep up with it. You have to be where the drug dealer left the car bomb ... you have to sneak into the back room of a house to overhear what's said in a secret meeting. I think there would be ways to have these little bundles of non-linear missions happen, and having them continue without the player's consent would add to the sense of urgency -- a key thing in pacing.

As for completely story-driven, linear games, the podcast mentioned Dragon Age as an example of bad pacing, and I agree: the mage's tower, and the dwarven deepmine missions were way too long, for too much of the same thing. Having lots of little quests in the same area isn't the solution because I think the player feels a need to "complete" an area before they've really moved on. I think Dragon Age (and others like it) needs to break its areas up into two or more sub-areas which are quite different from each other, and let the player feel like they've "completed" one area before they hit the next one.

When I think of linear, story-driven adventures with great pacing, the game that springs to mind is Chrono Trigger. That game introduced three areas early on, with a number of side quests in each. You feel like you are making progress in the main story, plot details are gradually revealed, you fight some bosses, you think you know what to expect and then BAM! They introduce a whole new area that looked dramatically different from anything you'd seen before, with a whole new set of story to take on. The pacing in that game was exceptionally good. And it had multiple endings too, so even then it wasn't 100% linear.

[belgerog]

Number one gives us Modern Warfare. There's nothing for the player to do but move forward and follow the pacing provided. If you stop moving, nothing happens. Move forward, game takes over.



What I like about the Half-Life games is that, even though they are linear, they give you the illusion that you are exploring the environment. That's something CoD in general doesn't do as well, I think. In fact, when thinking about pacing in linear games, HL has been one of the most effective for me. It's obviously directed, but I don't feel that way when I'm playing. The progression is very coherent, it makes sense when you change the type of place you're in, it's not disconnected.

This was one of my biggest disappointments with MGS4. MGS3 gave you a great sense of progression through the environment, because it was one big place and you knew how you were going through it. In MGS4, when a place was getting interesting, the act suddenly changed and threw you somewhere else.



Number two gives us any open world game. There's pacing there, if you go mission to mission, but it's completely in the player's hands. If you stop to do any kind of side mission, you inherently break the pacing. Or more to the point, you inherently take control of the pacing, so it becomes your own fault if you break it (assuming, of course, that the main mission arcs are well paced on their own).


I think that in many open world games, there is some structure. The game is not directed, but that structure allows your specific progression to be well paced. I remember it being mentioned in a podcast that the Fallout 3 devs made an effort to always put enough points of interest around you, wherever you were, so that you always had something to do.

The later Bioware games have put even more structure in their RPGs, and I think too much to an extent, or at least it doesn't work as well. The intro/4-places/ending scheme feels too forced to me. It worked better in KOTOR than in Mass Effect and Dragon Age, and Baldur's Gate 2 felt much more natural too, despite being less open-world than Fallout 3 or Oblivion.


I'm going to disagree. Look at the Thief games. Completely linear plot, but highly non-linear gameplay. Ditto for Deus Ex, or really any "open map" but not open-world game.

I imagine it must be easier to handle pacing in "open-map" games, since they are usually a series of missions, so progression through the environment is less important. On the other hand the gameplay on each mission has to be solid enough (as it is in the Thief, Deus Ex, or even Swat 4) so that they remain interesting. But it's also cool when some event in the mission changes the gameplay or advances the story. I've been playing Freespace 2, and that happens, part of the story advances through radio when you're flying your ship, or a new capital ship jumps in from subspace.

[sinnick]

What I like about the Half-Life games is that, even though they are linear, they give you the illusion that you are exploring the environment. That's something CoD in general doesn't do as well, I think. In fact, when thinking about pacing in linear games, HL has been one of the most effective for me. It's obviously directed, but I don't feel that way when I'm playing. The progression is very coherent, it makes sense when you change the type of place you're in, it's not disconnected.

Yeah, I couldn't agree more. From the very first moments in HL2 where you are chased through a stairwell by guards, you feel as though you are running blindly in a random, panicked direction when in reality you are going exactly where they want you to go. The level design and the pacing are intrinsically linked in that game.

[Mordrak]

So there was some talk of pacing in the latest podcast. I think it's an interesting discussion because there's a lot of problems with pacing and what it means for games. Devs try and control it somewhat, but ultimately, there's two cases:

1. Dev controls pacing, game is completely linear.
2. Game isn't linear, or incorporates side missions, dev tries to control pacing, can't.


This is an interesting issue and I found myself at odds with many people regarding pacing in discussions of Dragon Age and Mass Effect 2.

In Dragon Age, many people felt the Deep Roads was way too long, but for me.. I really enjoyed the length of the Deep Roads because it was one of my favorite sections of the game. I think they could have improved it though like adding areas to camp down there. However, even though I got into the Mage Tower quest and enjoyed it, it felt a little overly long.

Mass Effect 2 though, the mission lengths and objectives were so evenly spaced that it bothered me. They were so regular, that my experience felt mechanical and boring. To me, ME2's pacing was over polished, so I felt like I was trapped by the gameplay rhythms decided upon by a data mining algorithm.

What interesting though, this is less of an issue when I watch episodic TV, but I think that may have to do with the linearity. I'm unsure what it is about the format that changes my experience.

[Mordrak]

Yeah, I couldn't agree more. From the very first moments in HL2 where you are chased through a stairwell by guards, you feel as though you are running blindly in a random, panicked direction when in reality you are going exactly where they want you to go. The level design and the pacing are intrinsically linked in that game.

Going back to HL2, I realize how much game developers have learned since then. In those early hours of the game, I found myself trying to go places I couldn't which ruins that sense of reality that you are really exploring on your own.

Developers have gotten really good at using light sources and other seemingly subtle clues to direct the player, but again... once you learn those it can feel like you're chasing the little nodes rather than exploring a world. Portal felt a bit like that for me, but with regard to the hints for the puzzles. Pretty quickly, you couldn't help but notice the hints and start mechanically following them.

At that point, it felt like Valve was solving them for me.

[belgerog]

Mass Effect 2 though, the mission lengths and objectives were so evenly spaced that it bothered me. They were so regular, that my experience felt mechanical and boring. To me, ME2's pacing was over polished, so I felt like I was trapped by the gameplay rhythms decided upon by a data mining algorithm.


Yes, I felt a similar way. I think missions in ME2 always follow the same template, and that takes out from the immersion and the illusion that you are in the game. Also, you can always see the limitations and "invisible walls", you rarely get the feeling that "there's more out there". I too liked the Deep Roads, and Dragon Age was better than ME at making you feel that each part was a small journey, but not as good as Baldur's Gate 2. Also, even in Dragon Age, the way the different parts are separated feels somewhat disconnected. It felt "gamey" in that you know that you have to go through 4 parts and complete each one. I prefer when it's a bit fuzzier.

[Zylon]

Going back to HL2, I realize how much game developers have learned since then. In those early hours of the game, I found myself trying to go places I couldn't which ruins that sense of reality that you are really exploring on your own.
It feels to me like Valve has learned a little too much in this regard. Ever since running through Lost Coast in commentary mode, I've become uncomfortably aware of their espoused design formula when playing their other games. Here's a combat area. Here's a rest area. Here's a rest area that will later become a combat area. Here's a minor obstacle designed to slow you down. Here's a point of no return. Etc. It's all just so sterile and formulaic. Creating a grammar for level design certainly has advantages for a company like Valve, but it seems to be not a particularly rich grammar, and thus is limiting their level design in fairly obvious ways.

[Damien Neil]

If they are completely separate then they are working against the medium rather than exploiting the strengths. It's the same as saying "Look, I can fit a book and a movie together, because I have parts where you read in between the parts where there's a movie."

Strong disagreement, on the grounds that Thief clearly did handle both story and gameplay well, despite keeping them almost entirely separate.

[Mordrak]

It feels to me like Valve has learned a little too much in this regard. Ever since running through Lost Coast in commentary mode, I've become uncomfortably aware of their espoused design formula when playing their other games. Here's a combat area. Here's a rest area. Here's a rest area that will later become a combat area. Here's a minor obstacle designed to slow you down. Here's a point of no return. Etc. It's all just so sterile and formulaic. Creating a grammar for level design certainly has advantages for a company like Valve, but it seems to be not a particularly rich grammar, and thus is limiting their level design in fairly obvious ways.

Well, the language of game design is still developing. Basically in an any art (and any student of the arts) the progression basically goes through three stages:

A) Learning the Rules.
B) Mastering the Rules.
C) Breaking the Rules.

It's not wholly linear, you don't need to learn all the rules before you master some and then break them. It's essentially about manipulation and as Valve continues to design to players expectations and patterns, they'll learn how to effectively break them as well to create a powerful experience.

I think Portal is an awesome little narrative game for instance, even if I think they undermine the puzzle solving a bit to do it.

[roguefrog]

Strong disagreement, on the grounds that Thief clearly did handle both story and gameplay well, despite keeping them almost entirely separate.

Yeah especially with Thief 2, where they created the levels first and then added in the story afterwards...

[Zylon]

Is that "yeah" serious or sarcastic? Metal Age is generally considered to have a much weaker plot than Dark Project.

[Wobbo]

If they are completely separate then they are working against the medium rather than exploiting the strengths.
I disagree, many of the best games have no story at all. Regardless, plot pacing is much different than gameplay pacing, and the two often have nothing to do with each other.

Even in linear games, the player tends to set the gameplay pacing, but how quickly events happen within the fiction of the plot is an entirely different matter. In general, I'd argue that WAY too much of entertainment in American society is super fast-paced. The reasons why have a lot to do with advertising, shorter attention spans and other topics beyond this discussion, but suffice to say that book writers, television producers, movie directors and so on could gain a lot by slowing things down. There's rarely any subtlety, nuance, or tension in modern entertainment, and games are no exception. I'd like to see that change.

[Charles]

Strong disagreement, on the grounds that Thief clearly did handle both story and gameplay well, despite keeping them almost entirely separate.

I didn't say that doing them separate made anything bad. You can do them just fine, separate. But that doesn't mean it's using the medium properly.

[CLWheeljack]

My favorite example of excellent pacing in a game is Skies of Arcadia. Obviously this kind of thing varies based on personal preference, but I felt that it expanded your access to the world at exactly the right rate to keep you moving. The free-roaming sidequest sections came just when you were getting tired of the enforced storyline linearity, and the sidequests were varied enough and optional enough that they allowed you to move along just when you were itching to get back into the game.

I've never played a JRPG that doled out the rewards quite as well as Skies of Arcadia.

[Wisbechlad]

X-COM is still a great example of pacing for me, and your in game actions did have observable impact (countries withdrawing funding)

[Damien Neil]

I didn't say that doing them separate made anything bad. You can do them just fine, separate. But that doesn't mean it's using the medium properly.

I think my argument is that if Thief is using the medium improperly, I don't want to be proper.

[Equis]

I think pacing is a matter of doling out the player feedback in an appropiate manner, whether it be story and narrative elements, in-game rewards, or variation of game play challenges. Half Life and Half Life 2 certainly were very polished, pacing wise and their commentaries revealed a lot of thought that goes into the presentation of the game. Another game that had absolutely stellar pacing was Uncharted 2.

Personally, the two best paced game in the past half year have been Bioshock 2 and Mass Effect 2. It's harder to paced an Open World game, (and Rockstar is absolutely terrible at this) but not impossible with the promise of the next reward, or the next reveal.

I want to write more, but work beckons.

[Charles]

I think my argument is that if Thief is using the medium improperly, I don't want to be proper.

I'd rather have a story that involves me rather than being told to me.

[Zylon]

I'd rather have a story that involves me rather than being told to me.
That's fine, as long as you're willing to accept that it won't be as good of a story. As I noted earlier, there's a reason the best stories... or really any decent stories at all... don't involve audience participation. Your insistence that "proper" use of the gaming medium involves strong storytelling integration simply doesn't hold water. Sure it's possible, and when pulled off successfully the results are impressive, but so is building a house of cards from a poker deck. In both cases it's not something the medium naturally lends itself to.

[Charles]

That's fine, as long as you're willing to accept that it won't be as good of a story.

You should be happy that I don't accept that. It means someone in the game industry is trying to improve rather than just saying "Well, we can't do it!"



As I noted earlier, there's a reason the best stories... or really any decent stories at all... don't involve audience participation. Your insistence that "proper" use of the gaming medium involves strong storytelling integration simply doesn't hold water. Sure it's possible, and when pulled off successfully the results are impressive, but so is building a house of cards from a poker deck. In both cases it's not something the medium naturally lends itself to.

That's not really true. You are commenting more on the fact that A) No one tries, and B) people usually fail.

[Rock8man]

This whole discussion seems to have started with the notion that gameplay pacing and story pacing are different, and deliberation on how interconnected they are.

I disagree with Zylon that they are not tightly connected in Thief. Story pacing has a HUGE effect on gameplay pacing in Thief. One of my biggest complaints with the game is that when the story (as told to us in the cutscenes) gets really exciting and starts hurtling towards a conclusion, the gameplay pacing completely buggers it up by giving you a huge level called "Escape" in which the player is on edge wanting to know what happens next, only to be confronted with a long uninteresting level with no story advancement. That was the only misstep in an otherwise great game for me. So even though gameplay pacing and story pacing are different, I agree with Charles that they are very dependent on another in a game, even a game like Thief where they try to keep the two separated.

Luckily, the last level of Thief did not have this problem. At a time where the story reaches a fever pitch, and I really wanted to know what happened next, the last level is not a long ponderous level about stealing things from someone's house, or anything inappropriate like that. It ties in wonderfully with the story and the theme of the whole game, and it is very short, and brings the player quickly to the story conclusion. And that's one case where gameplay pacing and story pacing are brought back into harmony at the end.

[RickH]

That's fine, as long as you're willing to accept that it won't be as good of a story. As I noted earlier, there's a reason the best stories... or really any decent stories at all... don't involve audience participation.

I suppose that would be why the "choose you own adventure" books never took over a section of the bookstore.

The two (opposing?) philosophies here seem to be linearity (ex. MW2) versus openness (GTA4). But my immediate reaction was that linearity works for a shooter, that's its language. When I play a shooter, the question is always "where is my next target?" I control progress through the game by physically advancing through the game, bringing me into contact with my enemies. Too much openness leads to frustration and delay. So if the shooter doesn't answer the question "where is my next target?" it is failing as a shooter, despite any successes it was experiencing in the storytelling department. In that situation, context- or position-triggered events can take care of all the storytelling I need. Pehaps even more storytelling than I need in the case of Cortana taking over my screen every 5 minutes.

In contrast, there are more than one question(s) at work in a open-world game. "Who am I?" "What do I want to do next?" "How can I become more powerful?" "What will happen if I do THIS?" All of a sudden context triggers lose effectiveness because there's no guarantee you will ever trigger the event in question, or that you will trigger it in the right order. All the designer can do, short of taking away options (ex. last set of missions in Saints Row) is suggest actions or try to entice the player.

This means that in an open-world game (or any sufficiently open game structure), linear storytelling can not be intertwined with gameplay. Story elements cannot be structured 1-2-3-4 if the player can activate the story triggers 4-2-1-3. In response, Bioshock made its deep storytelling optional and non-linear, since there was no guarantee a player would discover or listen to any particular audio log. Big story events required trapping the player in a room and guaranteeing that story could occur. Rockstar approached this in GTA3 by locking off sections of the play area until certain conditions were met, in Bully by imposing a day/night cycle within a seasonal school calendar that would not progress until certain conditions were met (my favorite method thus far). In Saints Row Volition used "core" missions and side missions, requiring the player to decide whether they wanted the bonuses from activities or whether it was time to change the world (this worked well because it maintained the player as the center of the game's universe).

Throughout all of these structures, pretty much any narrative can be unfolded. The question is who is in charge, the designer or the player? Am I unwrapping a present for me, or am I a minor player in a world that's moving on without me? My personal preference is for any story elements to wait for me, rather than the other way around.

[Shadarr]

The two (opposing?) philosophies here seem to be linearity (ex. MW2) versus openness (Crackdown).

GTA4 had an open world, but was completely linear. It's my go-to example of how not to present a story, because instead of guiding the player to the next objective like a well-designed linear game, it simply made deviation from the story a failure state without telling the player what was expected.

Crackdown gave you objectives rather than scripted missions. There was no cutscene telling you to get on top of a specific building with a sniper rifle — and scripting that prevented anything else from happening until you did that — you were given a target and if sniping from a near-by building seemed like a good idea to you, you could do that.

[Zylon]

That's not really true. You are commenting more on the fact that A) No one tries, and B) people usually fail.
If as you claim gaming naturally lends itself to storytelling, then why do people usually fail?


I disagree with Zylon that they are not tightly connected in Thief. Story pacing has a HUGE effect on gameplay pacing in Thief.
I think we're interpreting my statement differently. In Thief, you can successfully play every mission without watching a single cutscene, and you can get 99% of the story from the cutscenes without playing a single mission. That's what I mean when I say the story and the gameplay aren't tightly coupled. The player is presented with story and gameplay in alternating fashion, only rarely combining them.

Contrast with any of the Half-Life games (but mostly the earlier ones), where the gameplay *is* the story.

[RickH]

GTA4 had an open world, but was completely linear.

That must be why I gave up on it after a couple of hours. Worst of both worlds.

[Rock8man]

I think we're interpreting my statement differently. In Thief, you can successfully play every mission without watching a single cutscene, and you can get 99% of the story from the cutscenes without playing a single mission. That's what I mean when I say the story and the gameplay aren't tightly coupled.

Contrast with any of the Half-Life games (but mostly the earlier ones), where the gameplay *is* the story.

While this is true, I was partly just pointing out that this point was a tangent. The original contention was that gameplay pacing and story pacing are linked. That's what Charles was originally saying. This is, after all, the pacing discussion thread. Now, the fact that Charles agreed to go off on this tangent with you doesn't mean that his original contention should be in question. He was right. They are linked. And I think Thief is an excellent example that they are indeed linked, even in a game where gameplay and story are have a wide separation between them.

[Charles]

If as you claim gaming naturally lends itself to storytelling, then why do people usually fail?


Because you can't tell the story in the same way that you do in established linear media. The story in a game is as much about the player's actions as what the game itself tells the player.

[Zylon]

Because you can't tell the story in the same way that you do in established linear media. The story in a game is as much about the player's actions as what the game itself tells the player.
I thought you were talking about narrative that adapts itself to player actions, but now it sounds like you're actually redefining what "storytelling" means.

On a related note, Dan "Chinese Room" Pinchbeck has just released his doctoral thesis, Story as a function of gameplay in First Person Shooters: An analysis of FPS diegetic content 1998-2007 (http://www.thechineseroom.co.uk/thesis.pdf).

[Charles]

I thought you were talking about narrative that adapts itself to player actions, but now it sounds like you're actually redefining what "storytelling" means.

Narrative that adapts is a part of it. But long before we get there, we should at least manage a narrative that respects player actions.

One of the big problems with pacing is that it is ultimately under the player's control in a game. If the player decides to go explore a secondary hallway during a sequence that's supposed to be rushed, well, the pacing is now broken cause the player decided to take a breather.

One thing that can be done that I didn't mention is dynamic situational content. This is something that would actually be really easy to do, if someone were so inclined to experiment. The idea is that the story calls for events, and the game delivers, regardless of where the player is. If the story calls for an ambush, then the game checks for the nearest ambush position, and drops it on the player, regardless of where he is. If the player is supposed to escape to a hideout, have the hideout be in the place the player is heading towards.

The trick is to incorporate the player's actions so that he can't ruin the story's pacing, and so the story doesn't lead to situations where the player is lost or confused.

[Shadarr]

The idea is that the story calls for events, and the game delivers, regardless of where the player is. If the story calls for an ambush, then the game checks for the nearest ambush position, and drops it on the player, regardless of where he is. If the player is supposed to escape to a hideout, have the hideout be in the place the player is heading towards.

I'd go the other way, and not try to be that granular in what the player has to do. Rather than deciding that the player must get ambushed, which leads to a lot of annoying gameplay where you know there's an ambush but you can't progress until you walk up and trigger it, simply put the ambush in place. Take a step back and realize that the point of this section of the story is the journey, not the individual aspects of the journey, and let the player control the pacing. If the player realizes there's an ambush, that aspect of the journey has already been achieved. It doesn't matter whether he walks into it on purpose, or tries to get around it, or ambushes the ambushers via stealth or sniping, so long as the general arc of meeting an obstacle and overcoming it is conveyed.

This is where I think the reliance on scripting and cutscenes to tell the story really holds back the medium. From the writer's perspective, the next part of the story isn't the player overcoming an obstacle, it's whatever happens in the cutscene, therefor that cutscene has to play, therefor the player has to be maneuvered into position to play the cutscene, even if that means placing artificial limitations on the player.

In your corridor example, my response is "so what?" If the player turns down a side hallway, that shouldn't matter as long as your enemies react in an intelligent way. If there is no other way through than the ambush hallway, then they will just wait. If the second hallway provides an alternate route, then they should leave their posts to try to surround and intercept the player. If it's done right, the sense of danger will be the same regardless of which way the player turns, and will be much greater than a scripted trigger system where the player can literally choose to just stand around and take a break before walking into the ambush they know is coming.

[Dufresne]

If it's done right, the sense of danger will be the same regardless of which way the player turns, and will be much greater than a scripted trigger system where the player can literally choose to just stand around and take a break before walking into the ambush they know is coming.

I'm reminded a lot of what Deus Ex did when you had to get captured to progress the story. It gave you an objective, (to get to the subway station and escape,) which, like everything else in the game, had a bunch of different ways in which you could accomplish it. In this case, though, it was really hard to get there. If you died at any point you'd wake up in a detention cell, and there were some serious overwhelming odds. It's pretty tough to just make it out of the hotel room where you're ambushed.

If, however, you were able to make it through the ambush, past the Men in Black in the lobby, past the soldiers and security mechs in the street, and past the additional soldiers in the subway station itself, only then would the game say, "Okay, enough already, this plot is progressing whether you like it or not," and put you into an unwinnable battle against an invincible Gunther Hermann. But no matter how far you get, the sense of danger and feeling of being hunted is there constantly.

[Zylon]

One of the big problems with pacing is that it is ultimately under the player's control in a game. If the player decides to go explore a secondary hallway during a sequence that's supposed to be rushed, well, the pacing is now broken cause the player decided to take a breather.
You've illustrated right here the fundamental tension between gameplay and storytelling-- good gameplay puts the player in control, while good storytelling puts the teller in control. If these truths aren't immutable, they are at the very least long-lived. As I keep pointing out, both play and the telling of stories have been around since human prehistory. Given that play and storytelling remain resolutely distinct (despite human storytellers being far more adaptive than any computer program short of a Strong AI) I have to wonder where your optimistic notions spring from. The move to computers isn't relevant, since play and storytelling are defined by human nature itself.

[Hanacker]

It sounds like Charles is talking more about a specific type of game he'd like to see than something that will change the entire storytelling paradigm in gaming. A lot of people seem to like tight, linear, stories that the player has little ability to majorly change. People like completing as many little sidequests and finding as many hidden treasures as possible in RPGs, despite the fact that doing so much sidetracking in the face of imminent doom is inherently ridiculous (and that it tends to totally screw with the pacing). I think Charles' vision sounds great, but I don't want every game to be like that.

[Charles]

I'm not saying there's one true way. I'm saying there's a way that hasn't been properly explored that actually leverages the strengths of the medium.

Games will always have stories told in the old style in the same way that there are still movies that are made like they were in the early days of movies. But right now, while games have probably progressed past the silent film era of movies, I'm convinced that games are still in the "filmed stage play" era, comparatively. Most developers are treating games like a platform for showing movies, just like early directors treated movies as if they were just stage plays to be shared with the world.

Just like they weren't using the medium to its full potential, we still aren't. As long as we continue to believe that telling static stories in games is the most games are capable of, not only will it continue to be true, but we won't actually find out what is possible.

It's not that I'm redefining storytelling, it's that this is a new kind of storytelling. Telling a story with a movie is radically different than telling one with a book, which is radically different than telling one in person. So why would you think that a story in a game can be done in the same way as other mediums?

Anyway, the point to this side discussions is that pacing is invariably part of both the story and the gameplay, and linking the two is necessary to cracking the problem of pacing as a whole. Until then pacing will continue to be very subjective and random.

[Tim James]

Why is it oddly difficult for me to think about story in games? I was looking through my backlog to try to relate to this thread, and can only pick out a handful where the story even registered as a blip. With games that had one I'm sure I was enjoying it at the time, but almost all of my fond gaming memories seem to be about gameplay or atmosphere.

[Charles]

Why is it oddly difficult for me to think about story in games? I was looking through my backlog to try to relate to this thread, and can only pick out a handful where the story even registered as a blip. With games that had one I'm sure I was enjoying it at the time, but almost all of my fond gaming memories seem to be about gameplay or atmosphere.

A good chunk of that is because there are very few writers of any quality working in the games industry.

[Tim James]

On a related note, Dan "Chinese Room" Pinchbeck has just released his doctoral thesis, Story as a function of gameplay in First Person Shooters: An analysis of FPS diegetic content 1998-2007 (http://www.thechineseroom.co.uk/thesis.pdf).This seems to reference STALKER quite a bit. I guess I'll be skimming through it now.

[Zylon]

Y'know, I'm still waiting for board games to mature enough that they can tell a good story.

[Togra]

This seems to reference STALKER quite a bit. I guess I'll be skimming through it now.
That's for a reason, especially the last game (Call of Pripyat) has excellent pacing and offers the best of both worlds (linear setpieces and freeform open world stuff). If Stalker 2 can stick to that but with a larger world then GSC has got it sorted.

I have a feeling Stalker doesn't get enough credit compared to the so-called open world FPSs US/UK devs come up with. Clear Sky was pure horror in terms of pacing (let's just say the pacing got replaced completely by difficult grinding and the biggest design mistake any FPS designer can make: an overload of illogical enemy respawns, it was GSC's Far Cry 2 moment) but Shadow of Chernobyl and Call of Pripyat should never be left out of these discussions.

[belgerog]

That's for a reason, especially the last game (Call of Pripyat) has excellent pacing and offers the best of both worlds (linear setpieces and freeform open world stuff). If Stalker 2 can stick to that but with a larger world then GSC has got it sorted.


I had the opposite reaction with CoP, I much preferred the pacing in the first game. The environments were great in CoP, but I found that I spent most of my time fetch questing and just traveling around.

For me part of the problem was that there were not enough human encounters. I loved that about the first stalker, when your were called to defend or attack an outpost. Since the maps were very open, you always planned your attacks in your head, and you had to be careful not to stay in the open too much. Doing night ops in Stalker Complete is even cooler, as even you can't see much.

Another reason the pacing was better in the first one is that you didn't get powerful weapons as fast as in CoP, where I found good rifles from the beginning, and there's less motivation to explore and look for equipment. The first game also provided some great combat situations in the very beginning, when all you had was your crappy makarov and the shotgun. In fact, I think it became a bit easier than it should once you got a scope for your AK, I guess because the enemy AI didn't adapt well to those situations.

In general, the balance between scary lab parts and human encounters with bandits and all was better in the first one. You fought a lot of bandits in the beginning, and then they changed the situation by putting you in the labs, then more soldiers, and another lab.

In CoP too much time was spent against mutants, and by the time there were larger human encounters (in Prypiat) you were already too powerful in terms of equipment. Also, the scarier monsters (bloodsuckers, snorks poltergeists) were too common in the overworld, it made them scarier when you only met them once in a while in a mostly empty lab.

Overall, I actually think they didn't handle well the lack of structure, compared to the first game, that was in CoP.

So I definitely hope Stalker 2 is closer to the first one.

[Charles]

Y'know, I'm still waiting for board games to mature enough that they can tell a good story.

It's a medium which is based on neither visuals nor prose. It's not the same thing at all, but nice try.

[Charles]

I had the opposite reaction with CoP, I much preferred the pacing in the first game. The environments were great in CoP, but I found that I spent most of my time fetch questing and just traveling around.


And this is exactly the core of the issue -- pacing is currently more dependent on how you play or approach the game, rather than what the devs do. Which is the very crux of the discussion. Until games incorporate what the player himself does, pacing will continue to be something which is made or broken by the player himself.

[RickH]

It's not that I'm redefining storytelling, it's that this is a new kind of storytelling. Telling a story with a movie is radically different than telling one with a book, which is radically different than telling one in person. So why would you think that a story in a game can be done in the same way as other mediums?

The difficulty is that the act of storytelling depends on control by the storyteller versus freedom of the listener (for lack of a better term). Good storytelling requires specific elements of character, conflict, tension, foreshadowing, etc. If you allow the listener to wander away from these elements, the storytelling fails. If you force the listener to experience these elements, you run the risk of the listener resenting the intrusion on his play experience.

I consider Saints Row to be an excellent example of storytelling in the context of a game. The player can play significant parts of the game without experiencing any story. However, when story elements are experienced, the player consents to them by activating the particular missions. Each of the three factions in the city have missions that must be played sequentially, although player can freely alternate which faction's track they are playing. Once all 3 factions have been completed, a fourth track opens for sequential play. The elements of stortytelling are fulfilled within the context of the player's broader freedom.

True freedom in the part of the player, however, will kill any storytelling other than an emergent experience based on the scripted behaviors of non-player actors within the game world. At that point, there are no events that must occur, there are simply events that may occur based on the player's actions. I'm not convinced that the game industry is capable of making something that sophisticated. Especially when development assets tend to go towards visually appealling "on the screen" elements rather than background tasks such as AI.

[Charles]

The difficulty is that the act of storytelling depends on control by the storyteller versus freedom of the listener (for lack of a better term). Good storytelling requires specific elements of character, conflict, tension, foreshadowing, etc. If you allow the listener to wander away from these elements, the storytelling fails. If you force the listener to experience these elements, you run the risk of the listener resenting the intrusion on his play experience.


As I said before though, there are ways we can make these happen with the player, without removing his freedom. You don't "force" him to see the elements, you simply present them to him no matter what he does.



True freedom in the part of the player, however, will kill any storytelling other than an emergent experience based on the scripted behaviors of non-player actors within the game world. At that point, there are no events that must occur, there are simply events that may occur based on the player's actions. I'm not convinced that the game industry is capable of making something that sophisticated. Especially when development assets tend to go towards visually appealling "on the screen" elements rather than background tasks such as AI.

We're not talking about true freedom. True freedom implies a toy which implies there is no pacing, so that is outside the bounds of this discussion.

Also, the AI programming team on AC2 was over 20 programmers. So I have a feeling your information about how development assets are used is out of date.

[Tim James]

And this is exactly the core of the issue -- pacing is currently more dependent on how you play or approach the game, rather than what the devs do. Which is the very crux of the discussion. Until games incorporate what the player himself does, pacing will continue to be something which is made or broken by the player himself.
I don't really know enough to comment, but the interesting thing about the original STALKER game is the developers had an internal struggle about whether to make it a linear shooter or do lots of cutting-edge living world emergent AI stuff. What they ended up with is a mostly open world FPS with a few very compelling linear levels that act as gates for the story. (Sometimes they acted as gates for the world as well, but other times there were minor events that limited travel until later in the game.)

It didn't feel like the game was trying to tell the player he was in a breather point, rather that the breather points were the entire game except for a few high paced linear sequences where it wasn't possible to jump out and do side quests. I suppose during the lead up to those sections (travelling to them, learning more about them) the player could decide to go off and do something else for a few hours, but it never felt like breaking something.

During the final two or three world maps, side missions were locked out to encourage the player to finish the story quickly. Mods later removed this restriction to allow free travel between these zones. It's been speculated this was more of a release date constraint than a conscious design decision, although it's somewhat supported by the story (Pripyat is now open, and all the factions are trying to rush through it to see what is in Chernobyl).

I'm not sure how much of this design was sheer luck and how much of it was brilliant planning. In any case, it works for just about everyone who plays the game -- the open world part is fun to waste time in, and the linear part is intense. The main problem is the initial learning curve and maintaining enough interest during that time to accept the game's numerous flaws; they probably could have done a better job with that.

To relate back to Togra's point, the third game in the series didn't do as well on the high paced story gates, although I thought it was compensated slightly by a more enjoyable open world part.

[RickH]

As I said before though, there are ways we can make these happen with the player, without removing his freedom. You don't "force" him to see the elements, you simply present them to him no matter what he does.

It seems we aren't having a discussion as much as launching clay pigeons up for you to shoot down.

That said, let's look at how games have solved the "providing story elements" problem via the force/involuntary presentation method:

Bioshock: You begin floating in water. There is no where to go other than the artificial island, within which there is no where to go other than the elevator down. As you pass through, you are presented with Ryan's image and motto. Then you are locked in the elevator and talked to by Atlas. You see combat, but the door stays locked until it's over. then you are released onto a linear path, guided by the voice of Atlas.

Mass Effect 2: Long, explanatory cutscene. Your character awakes in a medical bay, and you are talked to by Jacob, who feeds you information as you proceed through a linear path and engage in introductory-level combat. After a great deal of partially interactive exposition, you progress to a mission hub with restricted choices.

Fallout 3: Your character is created over intervals of time with user-selected options. Then, you are turned loose in a linear area of the Vault to accomplish certain goals (lots of locked doors). Some NPCs may be interacted with, others will approach you, some are "gatekeepers" who will not permit you to advance until their requirements are met. Player choice is limited to a few instances where you can decide whether to kill or save certain NPCs as you escape from the Vault. You emerge from the Vault into the world, as your pip-boy detects various things for you to investigate.

All establish the story of the game by limiting the player's freedom. This is what I meant by stating that storytelling is the opposite of freedom.


We're not talking about true freedom. True freedom implies a toy which implies there is no pacing, so that is outside the bounds of this discussion.

Absolutely, the question is how much freedom must be taken away from the character for storytelling purposes. Do I understand correctly that you believe there is a previously undiscovered method that will permit no more than a minimal reduction of freedom within a game environment?

I can see a game story's "pacing" impletemened by scheduling world events to occur at certain playtime intervals, similar to what we see in an MMO. However, in the MMO context the player can choose to ignore these events without consequense, which could not be allowed in a player-centered game. If a world event occurs it would have to be localized to the player, for example by having an NPC approach the player and engage in exposition.


Also, the AI programming team on AC2 was over 20 programmers. So I have a feeling your information about how development assets are used is out of date.

I notice you stated a certain amount of assets was applied to AI rather than offering a comparison of development asset allocation. I certainly don't have any concrete numbers to show allocations one way or the other, I am speculating based on the priorities I see identified by corporate bigwigs, marketers, and reviewers as being important to a game, and visual appeal appears to me to be at the top of the heap.

[Charles]

All establish the story of the game by limiting the player's freedom. This is what I meant by stating that storytelling is the opposite of freedom.


Actually, while the freedom is often limited at the beginning of the game, it's as much in service of teaching the player the game mechanics than anything else. Bioshock was probably more for story, but if you think about how ME2 and Fallout 3 started, you'll realize that they were there as much as glorified tutorials as anything else.



Absolutely, the question is how much freedom must be taken away from the character for storytelling purposes. Do I understand correctly that you believe there is a previously undiscovered method that will permit no more than a minimal reduction of freedom within a game environment?


I doubt it's undiscovered. But certainly I haven't seen anyone try it yet. The problem is that devs need to overcome their obsession with controlling every facet of the experience before they can approach it from a fresh perspective. The irony is that the only way they can make it better is to let go.



I notice you stated a certain amount of assets was applied to AI rather than offering a comparison of development asset allocation. I certainly don't have any concrete numbers to show allocations one way or the other, I am speculating based on the priorities I see identified by corporate bigwigs, marketers, and reviewers as being important to a game, and visual appeal appears to me to be at the top of the heap.

It was a majority of the programming team. Comparing allocations of programmers to non-programmers is pointless, however, since the non-programmers can't be expected to start coding AI.

As for shooting down clay pigeons, it seems more to me like people are happy with the status quo and believe it good enough. There are solutions to make things way better, and I know some people are exploring them. In general though, it seems likely that teams get scared by the lack of control and bail.

[Zylon]

As I said before though, there are ways we can make these happen with the player, without removing his freedom. You don't "force" him to see the elements, you simply present them to him no matter what he does.
This is even worse, because it would slap the player in the face with their lack of agency. Which is to say, players would very quickly figure out that the game world doesn't "play by the rules". It would be like that cartoon where every door has Droopy Dog behind it.

[Charles]

This is even worse, because it would slap the player in the face with their lack of agency. Which is to say, players would very quickly figure out that the game world doesn't "play by the rules". It would be like that cartoon where every door has Droopy Dog behind it.

Ehh. That's just a matter of tuning and balancing, it doesn't invalidate the whole concept. My examples above were there to show what can be done, not as a rule of how it should work.

[Rock8man]

Charles, I still like the concept of letting the player set their own pace as the best option. As a player of games, I've learned over the course of playing games how to get the best enjoyment out of them for myself. So when an open world game lets me set my own pace, I still like that the best, because as a player, I know what feels like the best pace for me.

That's one of the reasons why Assassin's Creed (the first one) is one of my all time favorites. Not only was the scenery and the combat seemingly tailor made for me, but also I loved that I was able to control the pacing of the story and the gameplay. I played the game over a series of 14 days, playing it for about two or three hours every day, setting my own pace so that each play session had some calm moments scoping out vantage points, a series of action sequences, and then usually a climax ending my session in a big fight. Everything I did fit well into the narrative and the world I was immersed in, so each session felt like a great self-contained experience but also part of a greater whole.

I still think that this is the best way to go about doing pacing that I've encountered so far. As an experienced gamer, it seems to me that I'm the one who is the best equipped to set the pacing myself in a game like that, because the designers gave me the tools to be able to set that pace at my leisure.

[Cl_Flushentityhero]

I think part of the pacing issue for shooters is the elephant in the room: most of the time you're just shooting stuff. If there are other gameplay elements, they're often half-baked or boring by comparison. Unfortunately, even shooting itself wears thin with enough repetition, so it gets boring anyway. What you often get in, say, a Rockstar game, is that "good pacing" is offsetting the greater boredom of repetitive shooting missions with the somewhat novel boredom of delivering pizzas or whatever.

Mediocre warfare realizes this and heaps scripted sequences on you to break up the shooting gallery. That's entertaining enough, though the experience is both homogenized and short.

I think ultimately (provided devs take the hint) we will see more games like Batman: Arkham Asylum that are hodgepodges of different elements to create a quality experience that doesn't overstay the welcome of any particular mechanic. It's not that I am ZOMG crazy about how good AA was, but I think it's an excellent design for maintaining interest among a gamer demographic who has 'seen it all' in action gaming. And, if we're talking about quality in the abstract, that's the audience developers should be targeting.

[sinnick]

I think part of the pacing issue for shooters is the elephant in the room: most of the time you're just shooting stuff. If there are other gameplay elements, they're often half-baked or boring by comparison. Unfortunately, even shooting itself wears thin with enough repetition, so it gets boring anyway. What you often get in, say, a Rockstar game, is that "good pacing" is offsetting the greater boredom of repetitive shooting missions with the somewhat novel boredom of delivering pizzas or whatever.

Again, though, Half Life 2 is a good example of pacing in a shooter. Like in film, they structure the action to rise and fall. Action beats occur regularly, followed by more slow-paced, thoughtful physics puzzles or obstacle puzzles, with large set-pieces less frequently. You never get bored of "just shooting stuff".

I think an interesting game to discuss pacing as it pertains to linear, first-person games is Mirror's Edge. Interesting because I thought it succeeded in some ways and failed in others. The story overall had a good setup, but kind of a cryptic progression and a letdown ending. But they made very good use of the gameplay mechanic (ie: thrilling rooftop chases) to structure story pacing: You have a chase which is exhilarating, followed by some quiet puzzle stuff, then some story, then some more chase -- it creates a good ebb and flow gameplay-wise, but also makes sense for the story.

The game wasn't perfect - it got frustrating because sometimes it wasn't obvious where to go next, or if you accidentally miss a jump or a turn you'd lose all your built up momentum which felt like a sudden jolt out of the game - but I think those imperfections are evidence that the pacing was working well. I wanted the game to keep going when those things happened.