My past two blog posts have been focused on open world gaming. These posts have been continuations of an earlier post about the narrative "limbo" that many open world games create via their quest structures. In the first post in this second series, I pointed out what I perceive to be a problem with open world games that insist on turning their sandbox worlds into little more than convoluted mission-select screens and collectible checklists. In the following post, I described some games that I think managed to make successful open worlds by including features or mechanics that made traveling through the space (or knowledge of the space) into a meaningful mechanic. This time, I want to go back to some of the games that I singled-out in the first post in this series, and brainstorm some ways that they could have made better use of the large spaces that their maps offered so that traveling around the world wouldn't become so boring later in the game.

But before I do that, I want to re-emphasize that I don't hate these games. They're just not very good at using their space, and that's what I'm criticizing. Well, the newer Assassin's Creed games have been pretty terrible. Anyway, I pick on games like Skyrim and The Witcher III a lot, but I like them just fine - I bought the DLC for both. I pick on them, not because I hate them, but because I do like them and I want them to get better (or for their sequels to get better). Rather, my objective here is to find ways for these games to make better use of the large, open spaces that they provide the player, so that exploring the map feels more mechanically relevant, more interesting, or more rewarding; and to feel less like a time-sink.

Games like Skyrim and The Witcher III have massive worlds, but do a poor job of utilizing the space.

Bethesda's Skyrim and Fallout titles, as well as CD Projeckt Red's Witcher III and Rockstar's Grand Theft Auto V, already have open worlds that transcend being simple, convoluted mission-select screens like games like Assassin's Creed and Metal Gear Solid V. They populate their worlds with little narrative world-building details that make their worlds feel alive and lived-in (even though they may feel stagnant). So what could a game like Skyrim or The Witcher III have done to improve its open world?

I wrote a lengthy blog late last year about the stagnant, "limbo"-like feel of most open world games' narratives. I had written that blog mostly before I played Metal Gear Solid V, and so I wasn't able to incorporate my thoughts regarding that game into the blog. But I did come to a new realization about open world gaming while I was playing MGSV. In my review of that game, I noted that:

"Even the open world itself feels constrained, as sheer cliffs prevent you from travelling too far off of the roads and serve to functionally railroad the player towards the small set-piece outposts and villages."
    - from my Metal Gear Solid V: the Phantom Pain review

I realized while playing MGSV that the game had built this large, open world (well, two large, open worlds really, but I hadn't gotten that far yet), but it didn't really care to let the player actually traverse that space or use it in any meaningful way other than scavenger hunting for collectibles. At least those collectibles felt relevant to gameplay though! Roughly half the map is dead space that the player can't even access. There was also this strange focus on using the helicopter to drop in and drop out of missions, rather than actually living in the game world, as the character had to do in Snake Eater. The map started to feel less and less like a place, and more like a convoluted mission-select screen. At first, this seemed like a strange, isolated example of an open world game that really doesn't want the player actually exploring its world. But as I thought about it, I realized that this isn't really a new phenomenon; it's actually just a very extreme example of what has become a sort of defacto state in most open world games.

The Afghanistan map of Metal Gear Solid V feels heavily constrained by sheer cliffs.

Think about it this way: in a linear game with rooms and corridors, every hallway and room should serve some purpose or function. In most games, this function will be some kind of skill or system mastery test. An action game like Devil May Cry will throw enemies at you to fight; a puzzle game like Portal will have a puzzle (or a piece of a puzzle) in the room to solve; a stealth game like Metal Gear Solid 3 will have a sneaking challenge or obstacle to pass; and so on. In the best games, each of these challenges will also provide a unique or novel test of skill or system mastery: unique combinations of enemies, unique puzzles, or novel arrangements of enemies and obstacles. Other games can use those rooms for thematic or narrative purposes. A survival horror game like Resident Evil or Silent Hill will usually put enemies, puzzle items, or supplies in a room, but some rooms might instead contain a scripted scare. In some cases, a room might even be left completely empty in order to build some kind of tension or anticipation, or to delay the release of already-built tension or anticipation.

So what is the gameplay purpose of an open world map? ...

Normally, I try not to get excited about movie-tie in games. They have a very bad track record - with only a handful of exceptions. But this Mad Max game wasn't a direct movie adaptation, and it didn't release simultaneously with the movie, implying that it hopefully wasn't being rushed out the door to meet the movie's release. Warner Brothers Interactive had previously released Middle Earth: Shadow of Mordor, which was also sort of a tie-in to the Lord of the Rings and Hobbit movies, and that game was actually very exceptional! It had a novel and innovative concept around which the entire game revolved (making it very focused), and it was a very well-polished game that was immensely comfortable to control. So Warner Bros had earned some benefit of the doubt for its next game. I wasn't expecting Mad Max to match (let alone exceed) Shadow of Mordor, but I still had hopes that this one would turn out to be a well-realized game that could stand tall and proud as one of those rare, good movie tie-in games. After all, the concept of an open-world, post-apocalyptic action game about smashing spiky, nitrous-fueled cars into each certainly sounds like a solid premise for a game!

Well, not quite...

Wasteland chaos

Many actions are overloaded to the X button - the game even displays conflicting prompts at times!

Virtually every interaction that I had with the game was either naggingly uncomfortable in some way or was prone to glitches. Even the basics of moving around and interacting with objects in the game world was a constant chore. When one button does everything; it does nothing (see my Assassin's Creed III review). Fortunately, a couple really important functions (like getting in and out of cars) were mapped to different buttons, but virtually everything else uses the X button. So if you're standing in front of a ladder and holding a weapon, it's a crapshoot whether the game will decide to let you climb the ladder or make you drop the weapon, and then it'll be a crap shoot whether the game lets you pick up the weapon again. Oh there's button-prompts to tell you what you can and can't do, but sometimes they outright conflict with one another. Besides, when you're running or fighting, then you're reacting on impulse and muscle memory rather than reading screen prompts. It doesn't help that the character's movement is very fidgety, so it's hard to position yourself properly when trying to interact with objects. I think the developers recognized this, which is probably why they make you have to hold the button for a second in order to perform most actions - to give you time to ask yourself "are you sure this is the action you want to do?".

Not enough space for vehicular combat

Clunky movement isn't limited to walking on foot. Steering vehicles is also very fidgety and floaty, and I found it very difficult to perform any precision maneuvering in the cars. The cars all tend to understeer at high speeds, but then strangely oversteer or fish-tail whenever you let off the gas. Trying to hit a ramp or knock down an enemy scarecrow or ram a sniper tower would often require multiple passes in order to succeed, and doing slaloms through the canyons resulted in a lot of cheap impacts. The rough terrain also leads to a lot of spin-outs. The vehicles feel so weightless and floaty that they can park on nearly vertical slopes, and running over a pebble can send the car hurtling and flipping 20 feet in the air. On a more personal note, I prefer my driving games to have cameras very close to the action, and so Mad Max's driving camera feels like it's a mile away from the action, which makes it harder for me to get a feel for precisely where the car is in relation to the environment. Virtually none of the game's vehicular set pieces really worked all that well for me due to these nagging control and scaling issues. If the map were bigger to accommodate multiple vehicles running side-by-side on a road, then dealing with the low-traction sand or the unlevel rocks wouldn't be so much of a consistent problem. Even having the option to zoom in the camera (an option that I couldn't find) would go along way towards helping me make more precise maneuvers.

The primary gimmick of vehicular combat works fairly well in spite of the map not feeling big enough to support it.

This game really lives or dies based on how well the cars perform. The bulk of the game is played from within your car. You use the car to travel the world, and it's actually your primary weapon thanks to the game's novel vehicular combat. This vehicular combat would actually be really fun if the cars handled a bit better and were durable enough to actually take the beating that the combat entails...

Are you one of the poor suckers who paid $30-40 for Ground Zeroes and were ready for The Phantom Pain to make up for your disappointment with what was little more than a glorified (and over-priced) demo? I wasn't, because I got Ground Zeroes for free from my PSPlus subscription around the same time that Phantom Pain was released. I was obviously disappointed with the demo's short length, and I didn't bother doing any of the side missions. But since I didn't pay for it, I wasn't as enraged as some other players might have been.

I actually really liked what little gameplay Ground Zeroes had to offer. The Guantanamo Bay arena was well-designed and offered some good infiltration challenge that tested my Metal Gear capabilities. The A.I. was surprisingly competent and adaptive - not so much that I couldn't exploit them occasionally, but still good. The graphics, lighting, and weather effects all looked outstanding. It was a fun experience. Not "forty dollars fun", but pretty fun. At least part of the battle at Mother Base should have been playable, and I didn't like that large elements of the story were hidden away in collectible audio tapes, but whatever.

Ground Zeroes gave me flashbacks to the phenomenal classic Sons of Liberty demo that came packaged with Zone of the Enders on the PS2. At least that only cost me a $3 rental, and I got to play Zone of the Enders too. After Ground Zeroes, I was looking forward to getting my hands on the much bigger Phantom Pain, and was optimistic that it would provide an equally good experience that would be worth the purchase price. Phantom Pain is a very long, very complicated, and very uneven game. So buckle up, friend. This is going to be a long review.

Table of Contents

• Ground Zeroes and intro
• Building a mercenary army
  • Anyway, back to Mother Base
• Sometimes, sacrifices must be made for the greater good
  • More assertive stealth
  • Fluctuating mission quality
• And then chapter 2 starts...
  • ... And then chapter 2 ends ...
• Online play influenced by ... Demon's Souls?
  • Metal Gear Online ... without the "Metal Gear"
• Destined to be Kojima's magnum opus? Or his black sheep?

After having written a lengthy blog post about how open world, sandbox game design almost necessarily puts the game's narrative in a state of limbo, I was amazed to start up Metal Gear Solid V and see the very first mission took my criticisms to heart. Of course, the game had already been released by the time I had written that opinion piece, so I can't take credit for having influenced its development, but it was still refreshing and gratifying. Anyway, in the very first mission, Ocelot tells you that Miller has been captured by Soviets in Afghanistan, has been tortured for intel, and has three days - tops - to live. You must rescue him before that time.

Ocelot gives the player three days to rescue Miller...

At first, I didn't put much stock in Ocelot's claim. After all, sandbox games are notorious for saying that something needs to be done ASAP, but they never have the balls to actually walk the walk and enforce that objective. Until now. When checking my map, I noticed something in the corner that I hadn't noticed in other sandbox games before: an "elapsed time" counter. The game was plainly tracking how long it was taking me to complete the primary mission objective. I treated this timer with a certain degree of skepticism. But sure enough, failure to rescue Miller within the allotted time actually results in a "Game Over"!

This is exactly how I feel that priority objectives in open world games should be handled: make it apparent to the player (through dialogue and/or explicit notification) that an objective is being timed or that it is otherwise a priority, and make sure that there are reasonable, perceivable consequences for failure to achieve that objective within the expected conditions. Then design some early-game quests or objectives such that the player is put in a position in which they can (or must) fail; thus, teaching the player that when the game says "do x or else", the game actually means it. I put down the controller and gave Hideo Kojima a standing ovation. But would this opening mission set a precedent that priority missions must actually be prioritized, and would that precedent stand throughout the rest of the game? Or was this just a one-off occurrence that would not be representative of the rest of the game? Regardless, a tone was plainly set for the rest of the game, and the stakes had been raised.

... Failure to rescue Miller within the allotted time results in his death and a Game Over.

Would this refreshing precedent carry over into the rest of the game? Well, sort of...

"Pirate Lord Captain Gregle, Slayer of Ancients and World-Renowned Trapeeze Artist" sounds like a pretty legendary character, right? Well, he wasn't. In fact, he was a very lucky, over-achieving halfling rogue in a short-lived campaign of Dungeons & Dragons. I rolled for the character's initial stats, got fairly low constitution, and then rolled the minimum value for hit dice for the first few levels. The result was a sixth-level character with a pathetic sixteen max hit points! A single lucky shot from virtually any enemy could be an instant KO for that character, and getting engaged in melee would practically be a death sentence. While some power gamers may scoff at the idea, rage against their dice, and then remake their character with a standard array and average HP, I decided to run with it and role play the hell out of little Gregle.

Character sheet for "Pirate Lord Captain Gregle, Slayer of Ancients and World-Renowned Trapeeze Artist"

With the low initial constitution, I focused my character around stealth abilities, disengagement and evasion tactics, and ranged attacks, and continued to improve those skills as I leveled. Knowing that he's a pathetic weakling, Gregle overcompensated by being a very flamboyant braggart and narcissist (I took inspiration from Stephen Colbert), and thought that he was more charming than he actually was. He routinely hid in the shadows, taking pot shots at vulnerable enemies and racking up kill steals from afar while his two warrior companions did most of the heavy-lifting. He would occasionally disarm a trap or unlock a door, and once used a clever trick to pacify (and subjugate) an entire band of pirates. He then took credit for much of the party's achievements.

Despite having only slightly above average charisma, he leaned on his halfling luck to succeed on some charisma checks and make himself a bit of a celebrity with the local townies for his exaggerated heroics. He reveled in the unprecedented access to their community that the locals provided, and he reveled in the adulant gifts that they showered upon him, happily hoarding it all in his bag of holding. The other party members never called him out on it in public, since they were just happy to have the cooperation of the locals.

While the other players and DM enjoyed Gregle's antics, their characters only barely tolerated his presence. During the actual adventuring, he was constantly getting into trouble and needing to be bailed out by his fellow adventurers. He once falsely awakened the party during his night watch after mistaking a wyvern for a dragon. In another instance, he was KO'd while using spider-boots to walk up a ceiling to pursue an enemy that had climbed a rope to escape the conflict, and he became stuck on the ceiling, forcing the party to figure out a way to get him down. They reluctantly obliged to help him, since Gregle was the possessor of the party's bag of holding, and was actually good at sneaking around to perform recon, unlocking doors, disarming traps, coming up with clever plans to avoid direct conflict, and other appropriately roguish things.

Gregle was one of the most fun characters that I've ever played, and he provided me with one of my most entertaining gaming experiences. This is the power of role playing to a character's strengths and weaknesses. It's a power that Bethesda shows no interest in utilizing for Fallout 4.

Out of the vault and into the wastes

I have to give credit to Bethesda for making one really interesting decision with Fallout 4: the game starts in a time period prior to the Great War that triggered the nuclear holocaust, and so it explores as yet unseen elements of the series' backstory. Or at least, it does for all of fifteen minutes. Much like Fallout 3, the pre-war gameplay and time that you spend in the vault is really just an extended tutorial and character-creation process. But unlike Fallout 3, it doesn't give enough time and depth to those settings to make the player legitimately care about them or the characters in them.

You don't spend enough time in your pre-war home or vault to develop any attachment to the place or people.

After creating your character and setting your S.P.E.C.I.A.L. stats, you and your spouse immediately flee with your infant child to the neighborhood vault. Once inside, you're handed the trademark silly superhero pajama jumpsuit and then promptly cryogenically frozen. You awake to witness your spouse get murdered and child kidnapped by apparent raiders, but then get frozen again. Then you awake again to do the combat tutorial against radroaches before leaving the vault and starting the game proper.

You spend virtually no time in the pre-war time period; you don't bond at all with your spouse or child; you don't establish any connections with your home or neighbors. There is absolutely no emotional bond between the player and what is lost in the war. So when the game drops you in the wasteland with a dead spouse, a missing child, and a quest to track down the kidnapper/killer, it does so without creating any emotional connection or investment for the player. I could go to Concord and then to Diamond City and search for my son, or I could just wander off in any random direction fighting raiders and painstakingly building my own little settlement out in the middle of nowhere using salvaged car tires and scrapped raider armor. Fallout 4 doesn't waste any time taking a nose-dive into the open world limbo.

Fallout 3 simulated an entire childhood in the
vault, with friends, family, and even bullies.

Compare this against Fallout 3's prologue. It spent a considerably longer time developing your character and immersing you in the vault. Your dad (voiced by Liam Neeson) plays with you as a baby to teach movement and camera controls, he teaches you how to shoot, and throws a surprise birthday party for you. You interact with a childhood friend, other vault dwellers, and even a bully in order to tutorialize persuasion and speech checks and learn how to solve conflicts without violence. You even go to school and take a test to determine your default skills. In the short amount of time in the vault, you've lived a montage of an entire life.

With only a little bit of buy-in from the player, Fallout 3's Vault 101 becomes a living, breathing place populated with people who you can relate to and care about. You, as a player, have an investment in it and the characters that inhabit it. So when shit happens and you have to leave the vault, it's a monumental moment, and the events of your life, and the decisions that you've made, will shape your character's development over the rest of the game.

Even Skyrim gave the player interesting role-play decisions in its tutorial by required your character to make an immediate decision to follow the imperials or the Stormcloaks (though the scenario makes that decision a pretty one-sided one). And your initial choices of weapons and battle tactics would level up those specific skills; thus, starting the character down a path towards specializing in those skills as the game progressed (though you were completely free to change all that if you want).

And Fallout 4 has a perfect opportunity to take that father / son dynamic from Fallout 3, and invert it! The game could have opened with the birth of the baby. Since Bethesda had to record dialogue for multiple names for the player character, they could just as easily have done the same with the child's name. The doctor could hand you a paper with "This Year's Popular Baby Names", and you could chose one of those names that were explicitly recorded in dialogue. You could even be given the option to type your own name and replace the child's name with "my son" in dialogue. You could fill out the child's name and your own character's name on the birth certificate. Naming the baby would create a sense of ownership and connection to the child that might help encourage the player to pursue the main quest.

Fallout 4's vault serves only as a combat tutorial with no depth, emotional resonance, or meaningful decisions.

But it doesn't have to stop there...