Game Logic vs Choice & Consequence

Gameworlds have become ever-more lavish, but has there been a dark price paid for this? Craig Lager believes so. Production values are up but these worlds don't seem to react to players' actions as fulsomely as they once did, he worries - are we allowing games' strange logic to take us for granted? But there is yet hope. Frowned at: Deus Ex: Human Revolution, Dragon Age II, Skyrim. Smiled at: The Witcher 2, Dwarf Fortress, Outcast. Please note these are Craig's views, not necessarily those of RPS.

In my version of Human Revolution, the police station should be surrounded. There should be SWAT teams, negotiators, probably even an evacuation zone. Adam Jensen’s face should be being projected from every single screen that litters Detroit's streets as Eliza explains him as being a more-than-prime-suspect in a new, horiffic incident. An hour ago, she would explain, Jensen asked for access to the police morgue and was declined. Now the back door has been broken into, and a path of corpses and hacked computers lead to the morgue in which a body has been clearly tampered with. Instead, Jensen walks into the main lobby and is greeted with “Hello”.

In my version of Dragon Age II, Hawke should have been executed a hundred times over. Ignoring The Circle he wanders around Ferelden raining down fireballs on common thieves, all the while accompanied by (and probably going to be sleeping with) a Blood Mage - the most illegal type of already illegal gone-rogue mage there is. He flaunts his magical prowess and barely an eyebrow is raised as he walks into the city barracks, embers still practically tumbling from his hands.

In my version of Oblivion, the hero should be exiled from Cyrodill. He’s been locked up more times than anyone can count. He saunters up to random people in the middle of city streets and beats them half to death with a club. Then he stands still and waits to get arrested. Over and over, the same cycle - beating, arrest, jail, breaking out of jail. Then, when the time comes, he’s warmly accepted into The Blades and handed weapons so he can ‘save the world’, without so much of a discussion about the psycopathic idiot that’s frequented the Black Horse Courier’s pages so often.

It goes on. It’s game logic, or rather, a lack of. There is a clear disparity between what a player does or wants to do and what game environments or characters know how to handle. The above examples are extreme cases, of course, but it happens all the time - how often have you set off an explosion, killed a man, or even half flattened a city with barely any repercussion or consequence? How many times have you been forced to game a conversation tree to fit closest to what your character wants to happen; or been pushed down one path even though another makes much more sense?

Frustratingly, the gap is getting bigger. More and more our choices are restricted to fewer possible outcomes or ways in which an environment can handle what we’re doing, and to boil it right down, it’s because of the level of technology we’ve come to expect and the costs that come with it.

To pluck a couple of old titles out of the air - Planescape, Outcast, even the original Deus Ex - they were incredibly reactive to what a player could do and gave a multitude of options in what a player could say. They could do this because to code in options was cheap (or at least cheaper). Now, though, because of the level of detailing we expect from a title, a large amount of money has to be pushed at each new option that’s presented to a player. A new dialogue branch (never mind complete separate path for a story to take), for example, isn’t just someone typing in some text or getting some more recorded and bolted in - it’s a wealth of motion captured facial animation and figuring out details down to where a characters eyes should be looking. A new enemy type isn’t a simple model and textures - each limb is intricately detailed and animated, and the same goes for a new NPC and each new quest line.

With new ideas of elements to put in games there is a simple rule: as they get more detailed, they get more expensive, and there is only so much time and money that can go into making a game. It’s a choice that has to be made - detailing how something looks or sounds or detailing how it works with other things around it. This is why, along with simpler but exquisitely good looking titles, we see super lo-fi but expansive titles like Minecraft, Space Station 13, and, to push it to the extreme, Dwarf Fortress.

It’s a game where the simplest of mistakes can unravel into a chaos of problems as systems interact and bounce off each other, giving the player an incredible amount of choice of what to react to and how to react to it. For example, a solitary miner could be expanding a tunnel but ends up opening into a huge cavern. Inside, a Forgotten Beast has been waiting for an opportunity for years and bursts through the new opening and rampages through the fortress, killing dwarves indiscriminately. Eventually it is slain, but its blood somehow finds its way into the Fortress’s water reserve and poisons it. Inevitably, as Dwarves start to drink from the reserve they become sick and die, then, crushingly depressed by friends being ripped apart by a monster and another friend dying from poison, a blacksmith dwarf goes crazy at work and throws a masterwork of a goblet into a volcano. Faced with his proudest moment being thrown into a fiery end, the creator of the masterwork goes berserk, tracks down the blacksmiths wife and murders her for revenge, who then rises from the dead as a ghost and haunts the blacksmith into a deeper depression until he hurls himself into a void.

It goes on, and that is mapping one linear line through a set of events - each of which could be spawning a similarly complex line - but complexity like this comes at a cost. “It's partially a matter of prioritization and partially a matter of talent and resources.”, says Tarn Adams, developer of Dwarf Fortress. “If I have to choose between adding two game elements or adding one game element and its corresponding graphics and sound, then I'll add two game elements”” He works with the most lo-fi assets there are - ASCII characters, but it enables him to push how much the world and player can do, and he can expand on ideas super quickly without any emotional or financial investment.

“It's easy to add new things and it's easy to change my mind. It's also easier to do procedural generation - I just need to make a paragraph description of a randomized creature and the technical specs without needing to make a Spore-like model, and it's simple to add new detail there. I also don't have a team of people with personal investment in elements that might need to be cut or altered.” But then, of course, you’re left with ASCII characters and cube worlds, and while we all have a soft spot for Minecraft, it could easily have remained a niche indie offering because of it.

Making assets attractive and pushing graphics is obviously important. The quality of sound, graphical fidelity, particle effects and complexity in environments which all combine to make staggering sales figures and spectacular games (whether that be Battlefield 3 or Modern Warfare to look at some extremes) could never have happened without the lineage of good looking but fundamentally less complex games, for example. Basically, the games industry couldn’t have grown to be the biggest in all of entertainment while looking ugly, and stories - linear or otherwise - deserve to be told in the best way possible: mountains collapsing, spectacular explosions and breakthroughs in how to get a virtual character to emote and communicate are something that we couldn’t do without.

I wouldn’t ask for a world without the expressiveness of LA Noire, characterisation of Mass Effect, or amazing visuals of Crysis. I wouldn’t even ask for a world without the linear Michael Bay fests that crop up a couple of times a year as, clearly, people enjoy them and they make games companies money. At the same time, however, I would ask for a world where wonderfully reactive environments and AI, and stories with more greyscale and branching were something that were prioritised more.

Fortunately, there are a select few teams striving for everything - an amazing looking and sounding game with hugely diversifying story and worlds that properly acknowledge your existence. I can’t think of a better example than of CD Projeckt with The Witcher 2 as (much like the original) it’s a game that doesn’t shy away from presenting you with choices and forcing you to deal with the consequences forever.

“Fundamentally, we’re out to give players an epic tale” says Michał Platkow-Gilewski of CD Projekt “and the genre we’ve chosen is governed by its own narrative laws. These require integrating choice into the story we want to tell, even if it means our team will have to work that much harder to achieve this.”

Presenting choices on a scale of what to do with an individual prisoner to the entire plot structure of the game and making it as polished as The Witcher 2 has is impressive, but predicatbly makes for a lot of extra work: “As regards game size, making a game like The Witcher 2 requires at least as much effort as making three or four less complex titles. It’s a tough way of doing things, and one that only a few RPG developers choose.”

The Witcher 2 is split into 3 chapters, but the content of chapters 2, 3 and the epilogue that follows varies hugely depending on the choices that a player has made. (*spoilers*). Right at the end of chapter 1, the player has to decide who to assist: a rogue elf or an human spy working for the Realm. One choice results in a celebration in the nearby town, the other a slaughter. But it goes further too - in the next area the player goes to there is a fight between a King and a Rebel leader and who the party side with is decided not by the player, but by his new companion meaning that an unrelated choice earlier has just split the possible story of the game in two, because the choice wasn’t just of what skillset a companion should have, but what moral choices and allegiances they would tend to as well, or as Michal says: “Feeling the consequences of your actions can’t be reduced to a statistic. It can’t be about getting +10 karma for doing something ostensibly good and -10 karma for doing the opposite. If you want to generate real feelings about a game, you have to forget numbers and show the real consequences of the player’s choices and deeds in the game world.” (*spoilers end*).

There are smaller things too which give the world a very reactive, entwined feel. If Geralt so much as pulls out a sword inside a town, guards pounce on him, but instead of immediately starting a fight they aggressively ask that he sheathes his weapons. Or if he has offended or humiliated guards at some point earlier, a few guards might band together and ambush him in the streets to get even. All of the time people run for shelter in the rain, hide from monsters at night, and comment frequently and diversely on what Geralt has been doing.

So, while what CD Projekt do is fantastic, it’s also incredibly rare and certainly an expensive way to make a game, but really, it should be able to stop being either for everyone. Graphics are constantly getting pushed for more detail but they have to plateau at some point and personally, I’d argue that we really don’t need to revolutionize what graphics can do for a while - games look pretty damn good at the moment. And as with anything, methods for voice recording and good animation will inevitably become easier with time, so, with all this combined, if creating and plugging in assets can inch closer to trivial (just like the ASCII characters of Dwarf Fortress), the more resource it opens up for coding options, branches, and ways that a world can react.

In fact, if you look at Skyrim (a game I only need the smallest of excuses to talk about) you can see some sort of middleground there. People will comment on what you’re wearing, what type of character you’re playing, things that you’ve done, and even comment when you’re ill - all things that re-enforce your sense of place in the world. And then, mechanically, things like only getting a bounty on your head if a witness actually manages to report you to a guard make sensible player logic like “no witnesses” make sense. Of course, to counter balance you still get the stupid things like guards telling you to be careful around the mages guild even though you, in fact, run it, or that you can still be a murderous psychopath yet still invited to save the world. So, while still flawed and stories aren’t particularly branching - there’s progress there, especially when it’s part of such a massive, gorgeous, freeform (and admittedly buggy) world.

If the next five years of big-title gaming development can focus on closing that disparity of what we want to do in a world and what developers can afford to let us do, rather than pushing technology to add complexity to raw assets, we can end up with environments that are more interactive, manipulable, intelligent and alive, rather than just prettier. It’s important to empower us, as players, to do whatever we want in games and, essentially, to make the choices and options worthwhile by having the game react in the most interesting ways it can.