Wednesday, November 23, 2011

Moral Ambiguity and Choices in Skyrim: All Setup, No Payoff

While black and white morality systems have been a staple of RPGs for years now - though less because of the inherent systems and more because of poor implementation by developers - it seems that over the last year or two it's become increasingly popular for RPGs to present morally ambiguous situations for the player, in the hopes of providing more compelling and intellectually stimulating decisions.  CD Projekt RED's The Witcher was perhaps the first game in this line, and shortly after, other RPG developers hopped on board, with Dragon Age, Fallout 3 and even Mass Effect 2 attempting to provide more difficult scenarios for the player that went beyond the question of "I want to be a bad guy/good guy."

With Skyrim, Bethesda have weaved a large and complex world, with rich lore that fits distinctly but unobtrusively into the existing Elder Scrolls fiction.  As part of realizing this world, they have, more than their previous games, attempted to do away with the binary good and evil dichotomy, by instead presenting a number of factions for the player to interact with, both with their given political agendas and their own more human elements.  However, while the setup plants the seeds for a game with interesting dilemmas with possible far-reaching consequences, due to either design limitations or oversights, instead Skyrim stumbles at providing interesting decision-making, even as it tries to provide a more realistic and morally ambiguous world.

Everyone's a Jerk but Me

The political landscape of Skyrim is defined largely by two factions, the Imperials (or more particularly, their military branch) and the Stormcloaks, a rebel army attempting to regain Skyrim's autonomy from the Empire due to perceived wrongdoings.  On the surface, it's easy to see what Bethesda was going for here.  While the Imperials are generally well-meaning and ensure equality for all under their laws, their philosophy also sees the individual concerns of smaller communities fall by the wayside in the name of the Empire, while the Stormcloaks, though fighting for a justifiable cause, do so often with cold-blooded murder, led by a man who seems more intent on proving his own superiority over others than truly saving his homeland.

In the mix of this are the Thalmor, a group of High Elves who control the Empire following the Empire's defeat in a prior war.  This on its own presents a far more interesting situation than just the two groups, because the Empire is largely being manipulated by the Thalmor, and many of the things the Stormcloaks fight the Empire over, such as the now-illegal worship of the god Talos, are in turn dictated by the Thalmor instead of the Empire, not out of malevolence, but because the Empire will be wiped out if they don't adhere to the terms of their treaties.  In many ways, the Empire is set up to take the brunt of aggression from the Nords of Skyrim, and they can do little about it without risking their own annihilation.  It's a definite "between a rock and a hard place" scenario.

Is there a "make my own faction that isn't full of idiots" option?
 The seeds here are sewn for interesting conflict and decision-making.  Choosing to support one of the factions is a big part of the game, and determines the player's role in the game world.  While both factions have good causes (the Thalmor are the only ones who are justifiably "bad guys"), their respective problems make the decision to join up a more difficult one.  Much like in Fallout: New Vegas, the goal is fixated on one outcome between the two groups - providing Skyrim with peace, safety and prosperity - but they have significantly different means of handling that.

Unfortunately, in attempting to build a morally ambiguous world, Bethesda actually go overboard and by and large overshadow most of the redeeming qualities the different factions have.  While the Stormcloaks are positioned as perhaps the "best" option due to their nature as freedom fighters, and the portrayal of their people as simple, hardy and well-meaning folk, it's clear that many are actually blatant racists and don't just want the Empire out of Skyrim, they want everyone who isn't a Nord to leave as well.  With the Empire, it becomes clear that while they might mean well, being such obvious puppets for a larger group makes it hard to support them as well.  Though many of their failings come more from what they don't do than what they do do, it's clear they have a lot of problems that simply defeating the Stormcloaks won't solve.

The human element isn't much better.  Without going into significant plot details, some background revealed about the Stormcloak leader, Ulfric, casts serious doubt upon his competence as a leader and his loyalties.  General Tullius on the Imperial side, on the other hand, is at least more immediately likeable, but doesn't really form any connection with the player; if I had one word to define his character, it would be "stern."  When you end up questioning the faction leaders themselves, and don't have much faith in their ideals or capabilities either, what else do you have?


In the end, I find myself not really caring for any of the factions available, and I've found many others who feel the same way.  While I made my choice, it felt like I was choosing the least incompetent party, not the one which best represented Skyrim and my character's beliefs.  Though it's great to have choices and better still for them to be interesting, the way the Imperials and Stormcloaks are set up, it's clear Bethesda went too far in giving the groups negative traits to counteract the positives, without actually spending time to build up those positives in the first place.  Good moral ambiguity in alliances makes for interesting choices, but in Skyrim, you're asked to choose between factions based not on which is better, but which one is least likely to screw things up.


You Want to do What?!

The second problem with Skyrim's moral decision-making has less to do with what it does and more to with what it doesn't do.  While it's important to provide the player with factions and characters that he or she can care about and become emotionally invested in, without constantly questioning their every move and motive, good moral ambiguity doesn't stop there.  A big part of that puzzle is giving the player logical and interesting choices to make that fit seamlessly into the situations the game presents, with the best decisions always being the most intuitive and clear-cut ones, rather than the most obscure.

Skyrim gets very close to this, but eventually trips over its own girth.  In the city of Markarth, a massive place built on the back of a Dwarven ruin cut into the mountainside, there are two factions, though more loosely defined than the Imperials and Stormcloaks.  The Forsworn are cultists (or at least religious extremists) who were pushed out of Markarth years ago by the Nords and placed under iron-fisted rule.  Understandably, they now hate Nords and the corruption they represent.  The Forsworn, despite being officially removed from the city, have many sleeper agents and other people on the inside working to gradually take the city back.  After investigating for some time, the player is eventually framed and arrested, only to break out of prison and either kill the Forsworn leader, or help the Forsworn escape and/or help as they butcher the entire city.

Markarth is one of the game's most interesting locations, but doesn't have the quest design to to back up the art design.
On the surface, this looks like your classic situation, and admittedly, the deaths of all those people is a pretty significant outcome which has long-term repercussions.  However, that's precisely what's wrong here.  There are only two outcomes: kill everyone with the Forsworn and let a group of violent, dangerous cultists take over, or kill the cultists.  It's pretty clear that these decisions are unreasonable and don't allow for much leeway.  Pacifism isn't an option - it's impossible to persuade the Forsworn leader to give up his crusade, and it's impossible to oust the corruption of the Nords without siding with the cultists.  In giving a radical binary decision to the player, Bethesda only draw even more attention to the fact that the options given are inadequate and, to be frank, highly unrealistic.

What's more, the player is rendered less an actor in the game world by such decisions, and more an observer.  Much of the fun in RPGs is getting a sense of influence over the game world and storyline, and yet in Skyrim, too many quests, including the Forsworn conspiracy quest, are happy to let the player stand on by without providing any real function save for actually speaking the dialogue that sets the quest in motion.  It's important not to build a game world that feels like it's just waiting for a hero to come along, but sometimes the lack of agency goes beyond the realm of plausibility and becomes irritating.

It's always easy to justify this sort of thing by saying that it'd be a lot of extra work to come up with so many different solutions and outcomes, but when your only options are "kill everyone" I really have to wonder what Bethesda's designers were thinking.  Such decisions aren't morally ambiguous or difficult; they're just stupid, and their violence and bluntness is so jarring precisely because so many other options are denied.  I'd be happy with those outcomes as options, but instead, senseless slaughter is mandatory, and ultimately the decision comes down to whether you want to kill cultists or civilians, not any sort of interesting battle of issues and ideals which can be reconciled without excessive bloodshed.

Status Quo is God

Skyrim loves to stick to the status quo.  Though the world it provides is massive and interesting, with tons of small details to soak in, it isn't actually much of a fan of giving the player any real influence over the course of events.  Despite the huge number of factions to join, quests to undertake and so on, there are almost never any long-term consequences for the decisions made.  The racist Nords in Windhelm will always be racist, and you can't call them out on it or make them see the light, even if you yourself aren't playing a Nord.  The feud between the two families in Whiterun can't be adequately resolved, despite it being such a strong theme of the location.  This pattern keeps up for almost the entire game: interesting locations and situations are painstakingly set up by the designers, but they never actually go anywhere or provide the player with any sort of pay-off.

Nowhere is this more obviously exemplified than in the College of Winterhold.  The College, located near the city of Winterhold, is rather notorious.  Nords hate magic, and due to a disaster many suspect the College of causing, Winterhold was almost entirely destroyed.  The people of Winterhold almost entirely hate the College as well as mages.  The player is able to eventually arise to the position of Archmage of the College, and with it come a whole new set of benefits, like some neat equipment and a lavish part of the College to use as a home base.  However, beyond this, very little about the gameplay or game world changes.  As Archmage, you are never called upon to oversee issues at the College.  You are never able to teach students, and indeed, it's possible to become Archmage with significantly less spell-casting ability than the other members there.  You never have to deal with anything remotely resembling what an Archmage might actually do.

Wait... so, uh, I can't send one of my underlings to do it?
Instead, you get to a handful of new quests to undertake, including, astonishingly, menial fetch quests and the ever-popular "go here and kill a few enemies" task.  As Archmage, I expect to see to administrative duties or academic ones which benefit the College and wider community, and yet everyone in the College still speaks as if my character is still a wet-behind-the-ears whelp.  What's more, given the strained nature of the relationship between the College and Winterhold, it'd make perfect sense to be able to sit down and have a nice, long talk with Winterhold's leader, and try to repair relations between the two, perhaps helping to donate gold in rebuilding the city, or using magic for good purposes.  Yet none of that is allowed either.  In fact, despite being Archmage, the people of Winterhold are entirely at ease with my presence.

Obviously, Skyrim is a massive game, and it's hard to provide meaningful choice and consequence for every single quest line.  I don't expect the game to devote the same resources to the main plot as it does to the side-stories taking place in the game world.  At the same time, though, this isn't an exception - throughout the entire game you will be given interesting scenarios positively begging for your intervention, with obvious end points for their resolutions, yet it is a rare day when anything you do is reflected in the game world in even a subtle way.

Even so, this lack of impact to decisions, or lack of decisions to make at all, dulls and even destroys much of the emotional impact of these scenarios, and reduce the player's agency in the game world.  It's often said that people need to see things in order to believe they're happening, whether that's the loading screen in a videogame ensuring the player that the game hasn't gone dead, or the progress bar during a program installation, and Skyrim suffers immensely from this, giving the player "implied consequence" rather than anything in the game world or gameplay itself.  It is a sad irony to think that a significantly more limited game like Divinity II can give players more interesting choice and consequence.

Conclusion

Building interesting scenarios for players to enjoy is always a challenge, regardless of the game or the genre.  RPGs typically have it even harder, as the amount of work required in building multiple outcomes to situations puts even more strain on a development team, and the higher standards for world design and writing inevitably invite more scrutiny.  Even so, while Skyrim provides one of the best fantasy worlds I've witnessed in years, and corrects many flaws in prior Bethesda games, the sheer size of Skyrim and Bethesda's attempt at a darker and more "realistic" story and world ultimately exhaust the game before it can ever really get started at exploring the people, places, and scenarios it provides.

There are morally ambiguous situations in Skyrim, and I'll be the first to state that many of them are interesting and compelling, but rarely are players ever given the tools to explore or solve them in convincing and satisfactory ways.  It's a common and integral point in fiction to never introduce concepts and problems that the setting itself can't sustain, but Skyrim does the opposite: it provides an extremely rich setting and then fails to do anything particularly interesting with it.  For all the improved atmosphere, art direction, and some of the most interesting backstory and lore I've witnessed this year, it is a bitter irony that Skyrim undermines so many of its own strengths.

Monday, November 14, 2011

User Interface Analysis: Skyrim

In my previous article, I took a pretty scathing and critical look at Skyrim's PC user interface, as well as some of the issues with the port in general, such as poor performance.  Bethesda released a day-one 1.1 patch just after I had written the article, which fixed a number of the interface problems (such as inconsistent keyboard and mouse controls), but it's clear that the shipping version of the game still had some major problems, and likely that quality assurance fell by the wayside in order to hit that majestic "11/11/11" shipping date.

Though user interface is something that one can write books on, and indeed has been the subject of a number of my previous articles, Skyrim's user interface is something which I feel deserves specific scrutiny beyond the PC compatibility and usability complaints I voiced.  Indeed, Skyrim has quite possibly one of the worst and most incompetently designed interfaces I have seen... well, to be frank, ever.  Skyrim, the game, is one of Bethesda's best works and a substantial improvement over previous ones, I do want to stress... but actually interacting with the game is an exercise in frustration, and the interface itself violates so many fundamental design tenets that it's downright upsetting.

Oblivion and Fallout 3, it's fair to say, did not have the best user interfaces.  Their layouts were a bit confusing and inconsistent, there were too many tabs, menus, nested menus, menus with multiple pages and sub-screens, etc.  Moreover, in Fallout 3, close to two-thirds of the screen space was taken up by the Pip-boy 3000, a fancy model with lots of shaders which had precisely no gameplay function whatsoever (but it sure did look neat, huh?).  One would think that after these two instances, Bethesda would go back to the drawing board and try to improve things for the better.

Initially, it looked that way.  Bethesda's bold new iPod-esque design, with plenty of clean, futuristic fonts and scrolling "cover flow" menus was clean and seemingly efficient, removing a lot of the excess baggage of previous menus and more effectively organizing information.  It's fair to say that this is one of the most radical redesigns of a user interface in a modern console game short of Fable III's interactive 3D Sanctuary.  However, like Fable III, Skyrim completely forgets that conventions exist for a reason... and demonstrates that Bethesda really have not learned very much about designing interfaces at all.

Poor Use of Space

The first, and most glaring fault, and a problem shared with their previous games no less, is an almost criminal misuse of space.  Though the heads-up-display is minimalistic and efficient actually getting into the menus demonstrates an almost complete ignorance of even the most basic design rules.

Upon opening up one of the game's menus (inventory or magic are the two most common), one is greeted with a single sidebar on the left or right side of the screen, containing a list of categories.  While there are ten distinct entries on the inventory list (depending on what types of items the player has), the default position for the list is not at the top of the screen, but at the center of the screen.  While this is immediately more readable, it quickly becomes apparent that not all entries can fit on-screen at once.  On a gamepad, this means that sometimes you'll need to do additional scrolling to be able to read some of the additional items in the menu.  On the PC, you'll need to actually scroll the list just to be able to click on the items that fall off-screen, even though there is more than enough real estate on screen to click each of them.

Despite all that extra space up top, the default list position makes no use of it whatsoever.
 Actually selecting one of these categories will reveal a second menu which lists all items within that sub-category, i.e. potions or weapons.  However, whereas a single column works for the smaller, ten-items-at-most list for inventory and magic categories, for the items underneath, it's a complete disaster.  While only a few items won't put any stress on the format, when you have potentially dozens or even hundreds of items, as in the case of various potions, ingredients, food items, and so on, this misuse of screen space and fixation on adhering to a specific aesthetic means that sometimes it can take ten seconds or more to even reach the item you're looking for.  Adding another column  would have mitigated the problem almost entirely, and placing the default list position at the top of the screen rather than the center would have further reduced additional scrolling.

Finally, there's the item or spell display itself.  Though it likely seemed a good idea at the time, over 50% of the screen space is taken over by a 3D model or particle effect of a given item, with attributes and a short description taking up close to 20% of the entire usable screen space.  Why this is, I cannot fathom.  Most of your time in the inventory will be taken up scrolling through items, not staring at 3D models.  Furthermore, a separate option to examine the models in detail already exists - so why do they take up so much room by default?  I imagine the goal was to show off those pretty models their artists no doubt worked very hard on, but to devote so much screen space to such a non-essential function is a major interface slip-up.


Text vs. Pictures

One immediately apparent characteristic of Skyrim's menus is that they almost entirely eschew pictures, instead replacing everything with text, sorted alphabetically in most cases.  This is a trend I've seen in a lot of modern games lately, and is often sold as "getting rid of the Tetris inventory" or the more general "streamlining."  Unfortunately, such a mode of thinking completely misses out on some of the many advantages that pictures and icons have over text.

While smart sorting options and using text aren't outright bad decisions, I want to stress, text, especially on a TV screen where real estate is more limited, takes up significantly more room than icons can, and have the immediate downside of being less easily identifiable.  Those lengthy lists which define Skyrim's menu systems could take up half the space if more traditional and RPG-like inventory icons were used instead - and it would have further eliminated the need for a large 3D model to take up the majority of screen space.

One of the most defining features of RPGs, especially in the West, has been a paper doll feature, or a graphical representation of the in-game character.  Traditionally, this was done (even in previous Elder Scrolls games) due to technical limitations, as highly-detailed and unique sprites were often beyond the graphical capabilities of many game engines.  Over time, this practice has generally waned, mostly because modern games are able to display a high-detail 3D representation of the player character anyway, either during gameplay or in cutscenes.

Though clearly not optimized for a gamepad, Icewind Dale and other Infinity Engine games accomplish far more with pictures than with text.
Though the paper doll was initially included in games as a compromise, a way to have a customizable character without needing to create high-detail animated sprites for every possible combination of races, sexes, equipment, clothing, and so on, it also ended up serving a very important purpose as far as user interface goes.  The paper doll, more than just a vanity, helped to instantly and immediately express exactly what items a player character had equipped - what suit of armor, what weapon, what magic amulet, and so on.  When coupled with an "equipped" inventory sorter of some variety, it meant that players could quickly and easily figure out what items they had equipped at any given time, literally at a glance.

Skyrim removes the paper doll function entirely in favor of the aforementioned 3D models, and the result is that it's actually harder to figure out what one's character is using at a given time.  Playing as a warrior, unless I have my weapon at the ready, I genuinely have no idea what I have equipped, potentially until it's too late and I meet the game over screen.  Playing as a mage, unless I have my spells at the ready, I have no idea what I can cast at a given moment, leading to much mashing of hotkeys - and furthermore, as many spells share similar visual effects, often I find myself casting the wrong spell for a situation because I can't even tell them apart until I've fired them off.

Comparing the interface in Skyrim to the interface in Icewind Dale, it seems that the old Infinity Engine was capable of producing a more immediately usable, quicker, and more attractive interface than all the modern technology and theft from Apple in the world could.  The pictures look good, it's easy to see what each item is, there are reams of more detailed information to be had at a single mouse click, quick-slots are easy to set up, and it's never a mystery what items I have equipped.  Even Arena did some things better than Skyrim, and that was over fifteen years ago.

The Worst Screen in the History of UIs

The above title is not hyperbole.  I think that Skyrim has genuinely managed to lay claim to the title of "worst interface element ever made."  It comes in the form of the skills menu, used primarily for leveling up.  It violates almost every single rule about designing user interfaces, and it does so for only one reason - to show off a pretty picture.

Among many other problems, the skills screen doesn't even give you an idea of how many skills there are to choose from.
 The gimmick with the skill screen is that it resembles a number of constellations in a night sky, with each constellation representing a specific skill.  I was under the impression that in previous Elder Scrolls lore, it was birthsigns that were the constellations, but I guess that idea was thrown out the window as birthsigns have been removed in Skyrim.  But I digress.  There are honestly so many issues with this screen that I am just going to list them one-by-one.
  1. It's impossible to see all the skills at once.  Want to know what your skill level in something is?  Prepare to do some additional left and right scrolling.  Depending on what skills you use, this could mean several seconds and close to a dozen discrete inputs to move the list along to where you want it.
  2. It wastes a lot of extra screen space.  By linking each of the headers to an image, instead of, say, displaying multiple rows or a vertical list with independent images, the numer of items on screen at once is further limited.
  3. It needlessly violates conventions both in games and in the real world.  From an early age, we are taught to read information left to right, and to list items top to bottom.  This convention may not be the ultimate in organization, but it works and most players are going to be used to it.  Instead, Skyrim presents a left-to-right list of items which is completely counter-intuitive to our existing understanding of how lists work.
  4. The default point is the center, not the left side.  Though it may seem more intuitive to place the currently-selected skill in the middle of the screen, in actuality it creates more work for the player, as the eyes have to travel both left and right to view other skills.
  5. The list scrolls both left or right, meaning there is no "starting" point to go from.  Usually in a game I want to know my information is organized in some sort of coherent way, but in Skyrim, the left and right scrolling ruins any spatial organization of information players might have.  Furthermore, anything that's off-screen might as well not exist at all, so if it's not immediately visible, you probably won't have a clue of exactly where it is in relation to the other items.
  6. On the PC, the controls are baffling and awkward.  Mouse clicks only move the list one position left or right.  Think you can click on one of those far-off items to select it?  Too bad.  I mean, really, what do you think that mouse even is, a cursor or something?
  7. When it comes time to inspect the perks in the skill trees themselves, or level up, only one perk's details are visible at one time.  This makes it impossible to view information at a glance, and furthermore means that it's harder to compare different perks to one another and weigh trade-offs.
  8. You have to go back from the perk menu to change to a different skill.  The way the controls are set up both on PC or gamepads, using the usual "back" button actually closes the entire skills screen, rather than going back to the main list.  Why the needless break from convention?  I certainly couldn't tell you.
  9. Navigating through different perks is a tedious and difficult process.  Rather than using a list, perks are represented by stars in each constellation, and must be "traveled" to using the analogue stick or mouse pointer.  If you're imprecise with your movement, be prepared to waste time as you travel to the wrong perk selection.  Furthermore, it takes around two seconds to move from one perk to the next, which itself can grow irritating if you want to find something at the opposite end of the perk tree.
  10. UI elements and camera perspective can actually block out perks that should be visible.  Instead of being able to see all the perks at once, the angle of the camera means that only a handful of them are even visible in the first place.  In some cases, such as the "Perks to increase" counter visible in the screenshot, the titles of perks that should be visible are actually blocked out entirely, requiring additional scrolling.
I honestly do not know who designed this portion of the interface, but it has so many elementary problems that I have trouble understanding how it even made it into the game - surely, somewhere, someone must have said "you know, this doesn't really work well"?  And yet it didn't - it's in the game, and players have to suffer through it.

As if scrolling everywhere wasn't bad enough, doing it in different directions presents its own share of issues.
I have a pretty good idea of what likely happened.  Somewhere, a designer came up with the idea... "it'd be cool if there were constellations, with all these stars on it representing skills."  Then, some artist whipped up a neat concept that looked really pretty, and everyone was on board.  However, in not sitting back and asking exactly how it would work from a user interface perspective, what the trade-offs were, and so on, the result was something not at all enjoyable to use, or intuitive.  Developers sometimes get married to an idea they really like, to the point where it can sometimes interfere with the rest of the game... in this case, Bethesda's designers were probably dead-set on this idea.  As a result, one of the game's more important interface elements was utterly ruined... all for the sake of a pretty picture.

Conclusion

I again want to stress that I have been enjoying my time with Skyrim.  The game is great, it's a lot of fun, and aside from my complaints with the interface and the PC version of the game, it really is a great experience compared to previous Bethesda titles.... and for what it's worth, there is one thing about the UI I do like - the mouse/stick gestures for selecting menus does work very well.  I also don't want to point any fingers at anyone in particular; I don't work for Bethesda, I don't know their company culture, and I don't know who makes exactly what decisions, or how much freedom and back-and-forth there is.

Even so, I have trouble understanding how such a, frankly, amateurish user interface ever made its way into a supposed triple-A game.  If Bethesda don't have a dedicated interface designer or engineer, then it's clear they need to get one as soon as possible.  If they're willing to sacrifice so much functionality and usability for the sake of aesthetic gimmickry, on the other hand... well, then I think maybe there are deeper problems at Bethesda that the company needs to work out, and in a way which doesn't leave their players saddled with the soiled fruits of their experimentation.

k-e-r-soc-problems-errors-help/306862-show-your-fps-ingame-step-step-instructions.html

Friday, November 11, 2011

Skyrim, or How Not to Make a PC Game

As with many, I have been looking forward to The Elder Scrolls V: Skyrim with both intrigue and trepidation.  Bethesda Softworks have long been a developer who both baffle and amaze in the most extreme of ways, and usually one has to overlook a lot of problems and outright bad design choices when it comes to their games in order to have fun (or install a thousand mods).  As a PC player, I generally expect Bethesda to deliver, if not the most polished and native PC games ever, at least a degree of quality that allows me to play their games without too many issues.

After spending a couple of hours with the game, there's no other way for me to say it: Skyrim is one of the worst PC ports I have ever had the displeasure of playing.  Though many issues will likely be fixed and improved as time goes on, and undoubtedly mods by more enterprising gamers will help additional problems, the release state of Skyrim is, frankly, inexcusable.  "Appalling" is not a word I use too lightly to describe a game, but it's the only one that fits here.

User Interface?  What's That?

After Fallout 3's interface left me with a bad taste in my mouth, I wasn't expecting Skyrim's interface to astound - indeed, preview footage lead me to believe that it was a classic case of Bethesda once again trading in functionality for looks.  Fair enough, that's nothing new.  I can struggle through the UI, even if it's a bit wonky, until someone comes out with a replacement.  Unfortunately, Skyrim's PC user interface is quite possibly the clunkiest and most difficult I have ever seen in a game made after the DOS era.  I have played games from 1985 that were infinitely more usable than Skyrim; that a modern game makes so many elementary mistakes suggests a certain kind of malevolence, not simple laziness.

The first sign things aren't as good as they could be comes before the game even begins.  The main menu is as sparse as can be, but is lacking features as basic as the ability to change settings before jumping into the game.  Why is this the case?  I can only speculate, but it probably would have required Bethesda to edit a couple of lines in an XML file somewhere, far too much effort to even bother considering.  Things take a nosedive soon after, as one realizes that not only are there standard menu items missing, but that the mouse doesn't even seem to work.  That's right, the first time you boot up Skyrim, you won't be afforded the luxury of using the mouse to start the game - you'll have to use the arrow keys like a schmuck.

Despite what it looks like, that "X to Delete" is not a button.  Also: click the load button and decide you want to go back?  Have fun figuring out how (hint: Alt+F4 is easier).
From there, things don't get much better.  After the game's highly scripted, lengthy and almost entirely non-interactive introductory sequence, you're called upon to create your character... only once again, something as simple as "mouse controls" was apparently too difficult to implement, as the dozens of sliders that allow you to customize your character to your heart's content can only be manipulated with the keyboard.  Bafflingly, turning between pages of customization options can be done with the mouse, but the settings underneath won't budge.

This inconsistency in where the game allows mouse controls extends to the main game even once you're past all the introductory stuff.  Many critical game functions, such as manipulating the world map, or assigning favorites and hotkeys, can only be done on the keyboard, while other shortcuts work just fine.  The game is also quite poor at communicating certain information in the interface itself (the first time I went to a shop, I accidentally bought a ton of equipment because it wasn't clear if I was buying or selling), but those are more general interface concerns I'll save for another time to pick on.

Other issues, like mouse acceleration being forced on and impossible to remove without an INI tweak, or the game defaulting to "Xbox 360 Controller - On" when started, even without a controller plugged in, or needless breaks from convention ("Tab" instead of "Esc" to exit menus) show just how little thought or care was put to the PC version's interface, and how little priority was given to this version in the game, or foresight given to what its players might want and expect.

"Customize" is a Four-Letter Word

Heavily tied to the horrendous UI design in Skyrim's PC port is the incredible lack of foresight demonstrated in the game's various customization options.  Some of these things, such as the inability to change certain game settings in-game, are unfortunate, but predictable considering previous Bethesda games.  There are, however, yet more problems which are clearly the result of extreme incompetence, ineptitude, or laziness, which extend to the most basic functions of the game.

Key remapping is something that you'd think would be a standard feature in a game, and that would work.  Bethesda, it seems, have decided that getting such a feature working isn't critical to releasing a triple-A product.  Though key remapping is in the game, whether or not it actually works properly is a crapshoot.  Many functions in the game can't be remapped, whether that's certain hotkeys or interface elements.  Fair enough - it's unfortunate, but not all games allow for player-customized hotkeys, so I can forgive Bethesda for that.

The "Favorites" menu is reserved for those who avoid the temptation to rebind their keys.  No spell for you!
Less forgivable is that many key bindings outright conflict with other interface elements and can almost entirely break the user interface of the game.  After setting up my standard "Bethesda game" control scheme that I've been using since Morrowind, I found almost immediately that those controls were incompatible with Skyrim.  The first sign something was wrong was when I went to "loot all" from a container using the hotkey of "R" and found that my camera perspective swapped to third-person.  Okay, whatever, rebind the third-person camera key, no big deal, right?

Unfortunately, things soon became worse.  After rebinding the hotkey for the Favorites menu to "F" and the auto-walk key to "Q", I found that Favorites menu didn't want to open anymore.  Curious, I pulled up my inventory and tried assigning some items to the Favorites - only it didn't work.  I spent the first hour of the game assuming the Favorites menu was completely non-functional, maybe because I hadn't unlocked the ability to use it yet or something.  Turns out, the "F" key is holy ground - because it controls the assigning of items to the Favorites menu, rebinding it to perform another function during regular gameplay, even a related function, makes it impossible to assign items and spells to it.  The Favorites menu, I should note, is a critical component of the game and outright necessary for any sort of convenience when using a large number of weapons, items and spells - that this bug wasn't caught shows a severe lack of play-testing on Bethesda's part.

Console = Cheater

Using the console in games to manipulate quest variables, add items to the inventory, to noclip through walls, and so forth has been a standard practice of players for many years, and even more so for Bethesda games - not because Bethesda players are damn dirty cheaters, but because the number of bugs in Bethesda games almost necessitates the occasional use of the console in order to avoid running into problems, whether that's getting stuck on the terrain, a critical item being lost or disappearing, or a quest not functioning properly.  Again, no hard feelings; it's an open-world game and sometimes things can go wrong.

Except it seems that Bethesda would rather not have its players making any changes using the console.  While it is fortunately enabled by default (unlike many other games), even so much as hitting the tilde key ("~") is enough for the game to permanently lock the player out of the Steam achievements the game so readily boasts as one of its key features.  That's right: get stuck in a Bethesda game due to a bug, and you have to choose between getting un-stuck, or losing out on your achievement.  Granted, restarting the game and avoiding the tilde key will allow you to continue on to earn achievements, but this introduces an additional level of tedium, as you'll need to restart the game every twenty minutes because you explored just a little too much in an Elder Scrolls game.

Should you "accidentally" fry a quest-critical NPC, you'd better not try using the console to bring him back to life.  Cheater.
This change was actually added in Fallout: New Vegas, because some players criticized Bethesda for their leniency regarding Fallout 3's achievements.  In Fallout 3, the achievements were actually unlocked using a set of console commands, just like manipulating any other standard game variable, so players could simply type in a few commands and get their Gamerscore maxed out.  Bethesda's response at the time was apathy, but apparently the outcry was great enough that they introduced the "console commands = cheater" measure in New Vegas.

However, what was a bad design choice then is still a bad design choice.  I can understand maybe they had to put such a restriction in place due to technical limitations in the New Vegas engine, but when Skyrim's new "Creation" engine has been so hyped up, one would expect it'd be possible to introduce some sort of counter-measure to prevent cheating, without forcing players to avoid the console entirely.  Apparently, expecting this was too much, and so now players who want to get around bugs in the game are being inconvenienced and punished for it, either by being denied their achievements, or by repeated game restarts.

There's another big problem with such a system.  I'm a gamer who's fairly sensitive to field-of-view; as a long-time PC player I'm comfortable with a standard 4:3 field-of-view of 90 degrees, and occasionally can tolerate lower FOVs depending on the game (I'm much more comfortable with low FOVs in third-person games than first-person ones).  In order to change the FOV to a usable setting in Skyrim, I have to use console commands... which in turn means that I have to choose between basic playability, and achievements plus motion sickness and disorientation.  Thanks, Bethesda.

Bomb the QA Department

One final damning point about Skyrim's PC port - it's horribly, pathetically optimized, and has major compatibility issues.  Right from the bat, I knew that something was wrong when my high-end system was getting framerates in the low 20s from time to time, and when performance did not improve upon lowering the graphics options.  Things became even more suspicious when I realized that there was no rhyme or reason for any of the framerate drops - whether I was outside in the overworld, the terrain stretching into the distance, or inside a tiny shop the size of a prison cell, the game's framerate fluctuates all over the place.  There's no question about this: Skyrim is badly optimized.

A second issue I immediately ran into was an intense audio distortion - crackling, skipping and popping most commonly heard in dialogue, but also in many of the game's environmental sounds.  No in-game audio options helped and there seemed to be no relevant settings in the game's INI files to help.  On a lark, I went to my Windows Sound Properties page and dropped my sound card's bit rate from 96,000 Hz to 44,100 Hz.  Instantly, the problem was gone - and instantly, I was frustrated at the fact that once again, it was clear Bethesda had foregone so much as basic compatibility testing.

Dragons?  Pfft.  This bread is the most challenging scene Skyrim has to offer your video card.
 While my problems from there on out were smaller, I've heard of a number of people have experienced severe graphical glitches, including corrupt textures, anti-aliasing incompatibilities, and strange flickering black marks across the screen while playing, all of which hint at a rushed and completely apathetic release with very little testing.  This goes beyond basic optimization and into the realm of "did not even make an effort."  If Bethesda think that this product is fit to sell, then Bethesda are clearly not fit to receive a single penny from me in the future.

All of the preview footage for Skyrim I'd seen was for the Xbox 360 version, with the PC version only featured in screenshots.  Many PC-focused web sites were reportedly denied their own review copies from Bethesda, as well, and in all those glowing 9/10 and 10/10 reviews, the PC version was either downplayed or not mentioned at all.  Now it's clear why - it's because it is at best a half-functional, poorly-performing wreck of a game.

Conclusion

"Mods will fix it" is a phrase commonly uttered when Bethesda games are mentioned, and it's certainly true that their often-buggy and occasionally-broken games have been substantially improved by their extremely dedicated fan community - I probably would not have got through Fallout: New Vegas or Oblivion if it wasn't for the countless hours fans spent overhauling the game's interface and fixing the bugs Bethesda refused to officially acknowledge.  Even so, Skyrim's PC release is a new low for Bethesda - it's not that the PC version was a low priority for them, it's that it was no priority at all.

As a PC gamer, it's hard to get my hopes worked up these days.  I've suffered through too many awful ports and broken games to expect every single title, especially one with strong console roots.  Even so, Bethesda have a strong PC history and their biggest fans have always been on PC, making their occasionally-great games even better.  But with Skyrim, it's clear Bethesda don't care too much about the fans that made them in the first place - not even enough to provide them a functional product.  Considering recent comments made by Bethesda's publishing side, and the poor quality of Rage on release, I'm not sure which part of the company is to blame.

I wish I could say that this was an isolated incident, but in truth, it reflects on the sheer apathy the games industry has towards the PC platform, even as many publishers come out claiming that they care about the market and that it's a priority for them.  If this is what things look like when developers and publishers supposedly put in an effort, I'm not even sure why I'm playing games anymore at all.

Wednesday, November 9, 2011

Opening Analysis: Morrowind

Earlier this week I took some time to lament what I feel is the lack of strong design in sandbox games.  In it, I used The Elder Scrolls III: Morrowind as an example of, effectively, a game that gets "sandbox design" right.  Truth be told, however, there's more I'd like to say about Morrowind, especially in the lead-up to Skyrim's release, as not only do I think it's one of the best-designed sandbox games ever made, but it also serves as a fantastic example of how to build a game introduction and tutorial sequence which all designers can take something away from.

Show, Don't Tell

Morrowind starts out, as is rather customary for the Elder Scrolls games, with the player as a prisoner, this time aboard a ship on the way to Vvardenfell.  The introduction, especially by modern standards, is surprisingly sparse: the player's view fades in as he/she comes to, watched over by a Dark Elf, who gives similarly sparse words on the present situation.  From there, an Imperial guard approaches and commands the player to follow on deck, where he/she must select a race before proceeding on.

Immediately, Morrowind is smart enough to adhere to one of the most important tenets of good narrative and good game design: namely, show, don't tell.  Most games would be content to begin with a lengthy backstory and cinematic opening setting the stage, a huge wall of text for the player to absorb as an orchestra swells in the background, or a completely non-interactive scene full of action, violence, destruction, death, and so on in some attempt to evoke an emotional response.

Morrowind's opening scene immediately establishes fundamental qualities of the game world and its characters.
 Morrowind does none of that.  Instead, it allows the player to slowly absorb the world and take in the situation.  The Dark Elf at the very beginning implies that the world the player is entering is irregular, different from the standard Roman fantasy theme of the previous games in the series, or regular high fantasy for that matter, while also hinting at the location itself (largely populated by Dark Elves).  The initial treatment by the Imperial guard and his strict, brief orders suggest the player is of a low caste or otherwise in a position of little power or authority, and the difference in appearance between the Imperial and the Dark Elf also implies a degree of race and class distinction which forms the undercurrents of Morrowind's game world.

Without directly saying anything, we are already able to learn a lot from the opening of the game... an opening which has gone on for less than two minutes, with minimal dialogue, no cinematic cutscenes, and no violence or ham-fisted attempt at drama.

Suggestion, Not Coercion

From the docks, the player is brought into a census office for the rest of the character creation process.  While perhaps not the most clever character creation process ever, it does its job well and gives the player an overview of all the stats, character classes, and so on.  Afterwards, the player is told to head on out, with scant tutorial messages going over the most fundamental of controls, like interacting with objects and equipping items.  During this sequence, the player can search a conspicuous barrel to find a magic ring, and will likely equip it before moving on.  After an exchange with another Imperial guard, who instructs the player to deliver a package to Caius Cosades in the nearby city of Balmora, with a suggestion given to take the nearby Stilt Strider, a massive insect-like creature used as transportation around Vvardenfell.
 
At this point, the game is completely open.  The player can go anywhere, talk to anyone, and do anything, completely ignoring the main storyline if he or she chooses.  The entire introductory sequence is over in about five minutes, with several of those minutes devoted to the necessary task of character creation.  While the game does push the player in a certain direction, and the suddenness of the situation certainly encourages the player to follow the recommended path, the game does not force the player to do this, even though at this point the game has given only the barest of tutorial functions.  Instead, Morrowind expertly crafts an opening scenario, using its first locations, the structure of its opening quests, and the tasks the player is given to do the rest of the teaching.

Fargoth is one of the first NPCs the player encounters, and interacting with him teaches more about the game mechanics than a dozen tutorial messages could.
This becomes abundantly clear in the opening town of Seyda Neen (whose name and details the player can learn from speaking to NPCs in the world).  Going back to the magic ring found earlier, one of the first characters the player will spot is Fargoth, a Wood Elf lamenting his missing ring (who is conspicuous in being quite short and rather foppish, ensuring many players will speak with him).  The player, had he or she found the ring earlier, is given the option to hand it back to its owner.  If so, Fargoth is overjoyed and thanks the player profusely.  Even this simple exchange teaches the player a number of things about the game and its world:
  1. Morrowind is full of different races.  Only a few minutes into the game and the player has already witnessed Imperials, Dark Elves, and Wood Elves, as well as potentially others in the area, such as Argonians and Khajiit.
  2. Exploration can and will be rewarded, and the player should make it his or her business to talk to everyone and examine everything in the world closely, because it might be important.
  3. Similarly, the player learns that prior actions can have consequences even if they are unforeseen.
  4. Dialogue topics can grow, expand, and lead to one another as the player learns of new things to talk about in conversation - if the player goes on to talk to other NPCs, he or she will notice that those topics persist to those NPCs as well, representing a growing knowledge of the game world.
  5. The player has choice in how to deal with various situations, and will receive different outcomes depending on this.  In this particular case, the player must have a choice between a magic ring and a less tangible reward, which might also tie into the player's own role-playing fantasy of his or her character.
  6. Doing good deeds will increase one's reputation on a character-by-character basis, while bad deeds will lower it.  This is plainly demonstrated in Fargoth's disposition level drastically increasing or decreasing depending on whether the player gives Fargoth the ring or denies it.
  7. Though not explicitly stated, the mere existence of a disposition mechanic, along with its change, with noticeable differences in Fargoth's dialogue and behavior toward the player, implies that disposition is an important part of the game and will have more significant effects later on.
All of this, the player has been taught with a single encounter with a completely insignificant and minor character, which may or may not even occur depending on prior decisions the player makes.  No tutorial pop-ups, no hints, no quest compasses or big arrows pointing over the heads of NPCs to indicate they're "important."  Morrowind presents a scenario and leaves the pace and way forward up to the player, trusts the player to make a series of decisions, and, should the player not follow the intended path, then it's no big deal, he or she will learn later down the road anyway.

This is the absolute best kind of tutorial - the kind that teaches the player something not just by incorporating it into gameplay, but in a way which feels natural to the player's own sense of discovery, rather than imposed upon him or her by a designer who assumes the player is too dumb to play the game the "right" way.

A World to Discover

Once the player has got his or her footing, one of the more significant choices presents itself: how to get to Balmora?  The player is given the option of taking the Stilt Strider for a modest fee (which, by the way, has its own minor side-quest uncovered further use of the persuasion system, and also establishes the importance of currency in the world), which is the easy and faster route... but it's also the least appealing, especially because in order to actually get to the Stilt Strider, the player will likely travel down a path, and find him or herself at a crossroads - in one direction, the Stilt Strider, and in the other, the open world waiting to be explored (with a nearby sign pointing the player in the direction of Balmora, no less).

View of Seyda Neen from the docks, including the Stilt Strider.  Reaching it requires the player travel just outside town, and in turn be tempted to explore the world.  Why ride when you can walk?
Even something as fundamental as the placement of the buildings, the direction of the road and the implied decision between the two routes all serves to reinforce that Morrowind is a game about choice: in this case, the doubtlessly more dangerous, but also more interesting and potentially profitable route through the swamp, or the quick, safe, but uneventful route via Stilt Strider.

Indeed, the player will shortly be rewarded upon wandering out into the wilderness just outside of town, with easy combat against Mudcrabs (the game's most basic enemy), as well as an interesting encounter wherein a man falls out of the sky and promptly dies - searching him reveals a few "Icarian Flight" scrolls (giving a taste of the powers magic will provide the player later on, but also likely ending in a swift death if used carelessly), and some fancy clothing to put on.  Almost immediately, the player sees the benefit in exploring versus taking the quick and easy route, but is also introduced to different types of danger in an entertaining and relatively consequence-free way.  From there, the player comes across the first optional dungeon just off the beaten path, and plenty more, all while still being able to follow the straightforward road (signposted for convenience) and reach the main objective in Balmora.

Upon arriving in Balmora, the player will likely be taken aback by the size of the city, which is much larger than Seyda Neen, without necessarily being intimidating or hard to understand.  The first buildings the player comes across upon arriving are likely shops, useful for offloading any unwanted equipment gained on the way over; if the player took the Stilt Strider, instead he or she will be treated to a nice view of the entire city from a high vantage point as something of a consolation prize.  Finding Caius Cosades in such a large place is difficult, so the player, likely in recalling previous interactions with NPCs, will find that asking for directions to his house yields useful information (reinforcing the value of talking to characters in the game world without making it mandatory).

Balmora is just large enough to get lost in, and requires the player take some initiative and employ the lessons learned in the first town to move the story forward.
 After meeting Caius, the initial quests will send the player off to buy better equipment (introducing the bartering mechanic if the player hasn't already figured it out), to sleep and rest (with a good chance the player will level up if he/she previously explored the outskirts of Seyda Neen), to the Fighters Guild (triggering the first quest-related dungeon-delving), the Mages Guild, and so on.  Importantly, in exploring the Mages Guild and Fighters Guild, the player is introduces to two fundamental character archetypes and philosophies, given a goal to strive for (mastery over martial or magical arts), and is allowed to join them and take the first steps in more substantial side-quests, which provide their own substantial rewards, a taste of combat, and require the player to do more exploration and investigation, as a bit of a test in learning the game's mechanics.

Suffice to say, the end result is effectively the same each time: the opening quests, interactions, and locations of Morrowind are all set up in a masterful way with the sole purpose of teaching the player about the game world, the game mechanics, and, provided the player has a bit of ingenuity, will also lead to additional rewards for above-average perseverance and exploration.  Everything you need to know about Morrowind is gradually doled out, not through endless tutorial messages, NPCs yelling orders at the player, or forcing the player down linear corridors, but simply by playing the game itself in the most natural and exploratory way possible... and it does so without ever actually forcing the player to do so.  When contrasted with Oblivion's opening sequence, an hour-long dungeon full of exposition, following NPCs from door to door, a tutorial pop-up every minute, and a contrived, almost arbitrary introduction of the main plot, it's hard to believe that the same developer is responsible for both games.

Conclusion

I really could keep going on with this.  Morrowind's early-game pacing is near-perfect and an amazing example of how to build not just a tutorial for any game, but also how to create a sandbox world with structure, cohesion, and which rewards the player in a way that doesn't feel like a dog being thrown a biscuit.  I often see deep analysis of the introductions and designs of some of the most famous and historically important games - Super Mario Bros., The Legend of Zelda, Mega Man, Castlevania - and rightly so, but rarely do I see this sort of thought given to more modern games, which is a real shame in my eyes, because as games become larger and more complex, the need for intelligent design and direction only grows.

Hopefully with Skyrim, Bethesda will be able to recapture some of the design magic that makes the opening of Morrowind so great, and more generally, that other developers will be able to recognize and take away aspects to incorporate into their own games in the future.  Building an open-world game is a massive undertaking, but as Morrowind demonstrates, putting it together and presenting it to the player doesn't have to be a task performed with an iron fist, or left a guessing game.

Monday, November 7, 2011

Sandboxes and the Rebirth of Grinding

With the release of The Elder Scrolls V: Skyrim just a short week away, many gamers are stoked for what they feel is the return of the king of sandbox games.  More so than being a competent RPG, Skyrim and other Elder Scrolls titles are concerned with providing players with huge, open-ended experiences where they can play as the characters they choose, and set their own goals for gameplay.  I, for one, am looking forward to seeing what the end product actually looks like once it hits shelves and the hype dies down.

More generally, though, sandboxes are all the rage these days for a variety of reasons.  The first is simple market trends - publishers are happy to adopt any characteristic they think might help enhance their games.  Second, a lot of players appreciate the feeling of control they have over their experience by being able to play in the way they choose.  Third, on a more esoteric, design-oriented level, sandbox games bring us closer to that nigh-unattainable ideal of a game which is able to provide us a truly interactive, responsive, and entertaining virtual world.

However, truth be told, I am not at all satisfied with the design of modern sandbox games, whether they're shooters, RPGs, or platform games.  While many of them are still enjoyable to play through and provide a large degree of freedom in what tasks the player is able to perform at any given time, I find that the modern sandbox design falls victim to significant problems - chief among these concerns, and the one I will focus on in this article, that they almost exclusively revolve around the act of grinding.  In addition to making for less interesting and compelling games which rarely lives up to that ideal, on a more basic level, it also makes me question whether such experiences even qualify as games to begin with.

Grinding for XP and Profit

Traditionally, when we think of grinding, we think of a game like Final Fantasy, or Dragon Quest.  In such games, a lot of time is taken up by fighting the same repetitive monsters and enemies over and over, as random encounters on a world map and inside dungeons.  Combat-heavy RPGs are nothing new, of course, and there's nothing wrong with this at its heart - attrition is one of the defining aspects of many RPGs, and being able to plan intelligently around the long haul rather than just the short term is a key distinction between players of different skill levels.

Castlevania II is infamous for revolving around grinding for currency, making up the majority of gameplay.
 Where heavy combat begins to turn into grinding is when a game presents the player with an obstacle so difficult that it can only be defeated or surpassed through a drastic leap in the player character's ability.  This can be manifest in a few ways.  Experience level is the most common one, especially in RPGs - the game will throw a tough enemy at the player he or she can't possibly defeat until growing to a greater level, usually only possible through fighting lesser enemies over and over again.  Another similar method is to bar the player from the next point until he or she has achieved certain arbitrary gameplay tasks (usually missions or side-quests), usually seen in several JRPGs and Western titles like Grand Theft Auto.  The third form of grinding most common is to require the player amass some quantity of money in order to purchase a necessary item, buy off an NPC, and so on - Baldur's Gate II does this early on, as does Castlevania II: Simon's Quest.

Usually, grinding is regarded - rightly so - as being an artificial barrier to advancing the game, put there solely to pad things out and stop the player from completing it "too quickly."  Sometimes, an alternate take sees it as a way to make sure the player is prepared for more advanced challenges, though I contend that this is just rephrasing the same problem in a more positive light.  Either way, however, grinding, especially of a mandatory nature, is rarely looked upon positively.  Some players enjoy the ability to grow in power and strength, but it's a fine line to walk between allowing upward mobility for the player, and forcing potentially hours of boring, completely insubstantive gameplay for the "privilege" of advancing the plot.

The Grindbox

As I mentioned above, usually sandbox games are well-regarded as a way of giving the player freedom to play the game in a way beyond what the designer explicitly set out, and ultimately to help reach that singularity point of a game whose mechanics and narrative are 100% reactive to the player's decisions.  The truth of the matter, of course, is that most games fall perilously short of this lofty goal, and most games settle for providing a linear storyline with optional content on the sideThe fact is that even the biggest games have only a fraction of the budget or time in development to ever capitalize on those sorts of promises, so they have to settle for something simpler.

The approach some developers take to get around this is to effectively create a large world with a ton of content in it, without linear restrictions on how the player can access this content. Though there are paths of progression, they are obscured by the fact that the player will often have many going at once, and that they frequently interact with one another.  For example, in Morrowind, the player may be concerned with leveling up, with getting new equipment, with advancing his or her standing with one of the game's guilds, and with moving the main plot forward all at once.  Progression in the guilds is linked in part to level progress, which in turn is made possible by better equipment, and in making progress in the guild, aspects of the main storyline can be uncovered and explored.  This inter-relatedness of different progression elements, including both at a narrative and gameplay level, is extremely rare to see, but when it's pulled off, it works extremely well; as far as approaching the ideal, this is about as close as one can get.

Far Cry 2 is one of the few games that suffers from a "50 km2, and nothing to do" problem.
The problem is what happens when a game doesn't manage to nail this complex interplay between elements, or ignores it entirely.  Most sandbox games are ambitious if nothing else - Far Cry 2 offers 50 square kilometers of land to explore both on foot and by vehicle, and Just Cause 2 has one of the largest game worlds to date - but usually they break down in actually giving the player anything interesting to do.  Far Cry 2 makes almost no use of its open world - at best it's a way to provide multiple tactical approaches, at worst it's a collection quest for diamonds and full of tedious hikes from one map marker to the next.  Just Cause 2 masks many of the same faults by draping them in a gown of explosions and 80s action movie cliches.  Contrast with a game like Fallout 2, where the interconnected world is governed by (and governs) the player's own unique progression, and the outcomes of the stories and quests of that world consequently change as well.

When such a design begins to break down, or isn't expanded beyond basic incentivization (i.e. collect the trinkets), the game begins to resemble less a wide-open sandbox, and instead more of a checklist, a set of arbitrary goals to accomplish before the game is "finished."  Just Cause 2, for example, houses thousands of collectable items and destructible buildings and objects for the player to hunt down.  This task may offer hundreds of hours of gameplay, but none of that gameplay ties into the story, and has no real influence on the player's own growth and progression in the game - it makes up the bulk of the content and yet as far as the game is concerned, it is effectively meaningless.  Indeed, at a certain point, the sandbox begins to resemble less a genuine player-driven game-narrative experience, and more a time-waster, an entire box full of filler content to grind through.

"Role-Playing" and Player Goals

It's possible to argue that player-driven goals should be able to trump designer-driven goals, and that if a player enjoys playing the game in a certain way, that's something a designer should play to.  I really don't buy this argument, however.  At a basic level, it's a designer's job to craft interesting interactive experiences for players, which includes interesting gameplay scenarios with meaningful stories, characters, and mechanics which drive the experience along - saying "we'll just put a bunch of stuff in the game and let the player figure out what's fun" feels like a cop-out to me, a way to wash hands of responsibility without actually designing a game at all.

Now, I'm definitely a fan of facilitating player customization and freedom in gameplay.  Some players practically live for this, choosing to role-play as a character and adhere to certain rules even if there are no actual mechanics supporting it, such as the aforementioned "eat three meals a day" example.  Certainly, if your game has room for it, give people the chance to do that.  Similarly, if a player wants to go out and hunt down all the collectables, destroy all the convoys, whatever, then the player should be able to do it.

Where I begin to draw the line is when the game itself begins to be designed around these notions of completionism and "role-playing," where the end result is a collection quest, or built around achievement farming, or on creating a consequence-free "virtual world" for players.  Frankly, I do not think that a set of scattered doodads to seek out and the ability to put on any single piece of clothing you like makes a particularly good game, and though it might approach the "sandbox ideal," it is a very sketchy resemblance at best.  A sandbox, after all, is supposed to be about providing the player with a living, breathing, reactive world, where the rules of reality are there for the player to manipulate.

Morrowind offers up an open-ended world, but unlike most sandboxes, seamlessly integrates the main quest with side-quests and other aspects of player progression and exploration.
Morrowind, as I mentioned above, provides enough goals, interconnected mechanics and definite modes of interaction with the world that even at its most unfocused, it still resembles an actual game, and rarely if ever forces the player into performing a repetitive task - there's always something else to see and do, and chances are the player will be able to make progress in one part of the game by making progress in a separate one.  I can certainly choose to play the game in ways not intended by the developers, and even use mods to create a more structured experience around that, but at the end of the day, there is still an actual game there to play.  Contrast with others, including Oblivion, and the examples I've mentioned above, which tend to fall more into the "grind" category, and the difference becomes quite clear.  Whereas one is an open-ended game which provides a huge number of options, and the extra stuff is merely there "on the side", the others resemble a series of arbitrary tasks the player is set loose to complete, and are devoid of any in-game meaning - goals without real context, influence, or finality.

Without any sort of reactivity by the game (narrative or mechanical, preferably both), or to any sort of interrelated progression mechanic, I have a lot of trouble calling the final product a videogame at all.  A game is, by most definitions (and mine), a system of rules which facilitate structure and challenge, and usually includes a completion state and failure state, determined by completion or failure of a goal or series of goals; yet in most sandbox games, any "sandbox" elements are wholly separate from this definition.  Instead, one ends up with a strange amalgam of a traditional game (missions, a story to follow, an end point, failure states), and a set of arbitrary tasks to perform merely for the sake of performing them, almost always entirely divorced from the actual game itself.  When they do link up, it tends to be in awkward and similarly arbitrary ways, entirely in service of the "real" game - Just Cause 2 unlocking new missions, for instance.

Conclusion

Grinding is and always has been a cheap, if completely unfulfilling way of making games appear bigger than they are, and of providing a false sense of forward momentum that wouldn't exist if not for better game balancing and pacing to begin with.  Now, it feels like those worst trends in gaming have become the central focus of so many other titles, whether that's because of an inability to properly rationalize trend-hopping, a lack of resources to actually flesh out a sandbox game in a substantive way, or simple lack of design forethought.  There are exceptions, where a sandbox truly does make for good game mechanics (I have yet to mention Minecraft), but these are few and far between.

I realize that this is a bit of a controversial stance, especially when we consider the many different permutations that videogames can take on and the unexplored potential of the medium.  For my part, I want to stress that I still can have a lot of fun with a sandbox if that sandbox is done well - I had plenty of fun with Just Cause 2 despite the fact that I will never, ever finish it (if it even can be "finished").  At the same time, in examining these games, it's hard to overlook the fact that many of them are, effectively, empty... and to be blunt, I am not satisfied with such experiences.