Happy solstice! It's time to celebrate the return of the light, the changing of the seasons, and the reintroduction of the Twinkie, which was nearly rendered extinct by the bankruptcy of Hostess Brands. And while we're letting in the light, let's shine a little on bad game design practices...
I’ll start with one of my own. I prefer to play solo in MMORPGs because I’ve already had more than enough experience with letting a team down in high school gym class. However, the solo player is at a disadvantage in many situations, and has to learn how to optimize his chances. I play by two sets of principles: those of Sun Tzu from The Art of War (which can be summed up in a nutshell as “be strong where the enemy is weak” and “don’t go where the enemy is stronger than you”), and those of single-ship naval combat (it’s all about range, speed, and firepower). To put these principles into practice, I play powerful archers who can run fast, and I try to find positions of maximum advantage to myself before engaging the enemy. As combat tactics go, this all makes perfect sense.
Imagine my disappointment, then, when the first time I climbed up to the top of a cliff to shoot down at an enemy in The Lord of the Rings Online, the game stopped and put up a popup box telling me that I was being naughty and trying to exploit it. The AI couldn’t work out any way for the enemy to get to me, so I was cheating. Of course it wasn’t able to get to me! That’s why I took the trouble to climb the cliff. Sun Tzu would have been proud of me, and I felt as if I had been punished for using my brain.
I realize the point of this rule is to prevent exploits, but I really think there’s a better way to handle it than shaking your finger at the player. What does any real-world combatant do when it’s outmaneuvered in this way? It retreats out of range, and that’s what should have happened. If the LOTRO enemy did that, I wouldn’t get the kill, so I wouldn’t get the experience points, but at the same time I would drive it off, rewarding me—in a small way, and only temporarily—for doing something smart. Once I climb down from the cliff, it could come after me again. The problem is solved within the game world, and as a natural part of it, rather than in an artificial, fantasy-killing way.
Part of the point about role-playing games, and massively-multiplayer ones in particular, is that they let players play in their own preferred style. The games are full of technology trees and personal traits and customizable weapons and so on that the player can spend endless time tinkering with and refining. As I said above, I'm all about long-range firepower. If I were a ship I'd be a guided missile cruiser.
Unfortunately The Lord of the Rings Online (which, don't get me wrong, I really like most of the time), sometimes decides that it's time for me to play a different way. As part of the epic quest chain I am occasionally obliged to assume the persona of some heroic figure from the game world's past and experience an adventure that he had. If I don't do this—and succeed—I don't get to go on with the quest chain.
Unfortunately, a lot of the time this heroic figure is a big dumb galoot with a sword. I have very little experience with this style of play and usually end up being clobbered. Repeatedly. What's more, there's no opportunity to learn—I just get dumped into an instance with a new user interface and a lot of options that I know nothing about. This is roughly equivalent to being yanked out of a soccer game and made to play American football instead. If I had wanted to play a hack-n-slash type, that's the kind of character I would have created in the first place.
It's OK to change environments, enemies, and other kinds of challenges—though as I've said in earlier No Twinkie columns, you shouldn't make mandatory wildly atypical levels, nor bosses that are so different that nothing the player has learned is of any use. This is a related error. Bad game designer! No Twinkie!
This one comes from Derek Manning of ORCAS. He writes, “[Something that] has really bugged me about Angry Birds, especially in Star Wars, is the absence of any indication of how to reach the next star at the end of a level. There have been multiple times where I’ve gotten done with a level after thinking I’ve done really well only to find out I’ve only gotten two stars, and I’m not told what the threshold is to reach the third star.”
This one doesn’t apply to all genres, since in some of them we don’t want to ruin the player’s immersion by dumping numbers on her. But Angry Birds is not L.A. Noire. It’s a heavily score-based game that offers its replayability by challenging you to do better. Most games that use a lot of numbers, such as CRPGs, are quite clear about what you have to do to get to the next level—indeed, in a lot of the casual free-to-play games it’s the whole point. “Be clear about short-term goals” is one of the bedrock principles of video game design, and making the player try the same thing endlessly without any idea of what he’s aiming for is a Twinkie Denial Condition.
This is actually two gripes in a single message. Elvira Björkman of Futuregames writes to say that Jak II (from the Jak and Daxter series) taught her how to use a hover-board in a tutorial, then didn’t let her use it in gameplay until hours later—by which time she had forgotten the controls, with no good way to learn them again. Games really do need to provide on-demand input device help, ideally in the pause menu. A lot of games permit players to reassign the controls (and ideally, all should). The control-reassignment screen, which should be available from the pause menu as well as from the shell screens, is an obvious place to let the player see what they are as well.
Also, the point at which you learn a skill should be shortly before you need that skill—not several hours earlier.
Kitsune Magyar writes, “Guild Wars 2 has excellent little circles showing you where not to be during boss fights and large encounters, and then drowns them in a billion visual effects to the point where they could have just skipped them completely. Some enemies in GW2 also have special animations to show the player they’re going to open up a can of wipe-your-party, which are then drowned in visual effects too. I remember me and my band of merry friends trying to take down an optional boss (thankfully) only to be completely murdered in a one-hit-KO. (Don’t have one-hit KOs either, please.) It seemed to happen at random. Upon reading the wiki-article for this encounter it turns out he did a special animation before wiping the team that was extremely subtle and similar to any other attack animations he had, and just to make sure it was extra hilarious the spell effects of five people attacking him made this even harder.”
It’s another bedrock rule of game design that the player needs to know when he, or his party, is in trouble. If the game screen is so filled with fireworks that he can’t see the essential clues warning him about this critical fact, you’ve overdone your visual effects.
This one is actually more positive than negative, with several suggestions about how to do it right.
I’ve already come down firmly on the side of saving the game—to those players who want the thrill of knowing that losing means starting over from the beginning, I say, “either play arcade games or never hit save.” But there are better and worse ways to implement the game save screen. François-Xavier Quencez sent me a screen shot from The Elder Scrolls V: Skyrim, demanding, “Why does one of the best games ever created have such a basic and ugly system of game save? I happened to create another character and then overwrite my first character by mistake. Of course, in order to avoid this and to see easily where my old saves are, I should rename my file, but in this game I can’t! And it’s even worse if a friend wants to play on my computer. The best way I found to preserve my files is to find the folder of the saved games in Windows Explorer and create a backup.” I agree: If the player has to go out to the file system and hunt around to find the save files, you’ve screwed up.
He also sent a screen shot from Go Go Nippon, showing a much better system. The color scheme is a bit garish, but the functionality is great. He wrote, “As you can see, each file has its own buttons and we can store five files per folder. The information is correctly classified and players can organize their saves as desired. Even quick save and quick load have separate files.”
This may all look a little to computer-y and not enough like the fantasy world that you’re creating, but let’s face it: Saving the game is outside the game world. You’re never going to preserve full immersion when the player is intentionally keeping a snapshot of the world for future use. (Single-button quick-save helps, though.)
On a similar topic, Harley C writes, “My kids and I have been playing through Disney Universe. This is a fun game for all of us, but it doesn't allow for multiple sets of save games to exist on the same machine. If I want to play through a part by myself, this progress will be reflected in the family's save game. There are other games for the Wii that do the same thing. They put all save games in a single place, instead of allowing players to tie the saves to their avatars.” We never considered this in the early days, but now that consoles routinely allow players to keep separate profiles, games ought to make use of it. Windows PC games should do the same thing: store game saves in a directory associated with the logged-in user, not in the application’s directory.
Finally, Stéphane Bessette writes to point out that the “Exit game? Any unsaved progress will be lost” warning message is annoying if you’ve just saved a second ago. Even the BIOS in Windows PCs offers you a choice: Exit Without Saving (with warning) and Save And Exit (which works silently). If the BIOS can do it, we can do it.
Build a save system that offers the players flexibility. Let different players keep their saves separate from one another, and offer them delete and rename functions so they can mange them properly. Compared to all the effort that goes into a large video game, this is trivial.
The constant buying and selling in CRPGs harms their heroic quest feel, so the least you can do is make this more efficient so that the player doesn't have to waste a lot of time on it. It’s not fun to have to click the sell button 100 times to get rid of 100 Vegetable Peelers of Unusual Bluntness looted from hapless kobolds. Stéphane Bessette suggested a set of qualities that a good loot-selling interface should have:
Players need a way to be able to buy stuff back that they’ve sold by accident, at least within the context of the current transaction. The opportunity doesn’t need to last forever, just long enough to be able to correct clicking on the wrong thing.
They need a way to lock items so they can’t sell them unintentionally.
It’s nice to be able to get rid of low-level junk without having to schlep it all the way back to the surface. Torchlight offers this.
Let players buy one item, a specific quantity of items, or a stack of items.
It would be good to be able to sell quantities of identical items that don’t stack.
Having complained about The Lord of the Rings Online above, I have to say that it does most of this stuff well, except #3 and #5. It even lets you automatically fill stacks so that you buy just as much as you need.
Someone calling themselves Tvwxyz wrote, “I play lots of Xbox games and many of them don’t even have an Exit option! With such games, you have to click the System button to return to the dashboard (This also happens with PlayStation games.) Come on—just give me a Quit Game button on the main menu. And when I click it, just let me leave without banner ads or videos. I am willing to accept this behavior with trial/demo games but not full games that I've already paid for.”
I don’t play enough different kinds of Xbox games to verify this for myself, but making you watch an advertising video in order to leave a game that you’ve paid for is crass. There really ought to be only two kinds of games that you can’t get out of: arcade games, and games in a machine with no other operating system to speak of (dedicated, single-game handhelds or cartridge consoles). It shouldn’t be necessary to press the Big Red Switch (or equivalent) just to leave a game. A game is a computer program; just as it needs a way to save the user’s work, it needs a way to exit quickly and cleanly.
And speaking of exiting quickly and cleanly, that’s the end of this year’s Bad Game Designer, No Twinkie! column. See you next year! Be sure to send your own gripes to firstname.lastname@example.org.