Archive for category design
Loose threads and general vagueness are often poor crutches in storytelling. These aspects tend to be weird for the sake of being weird, or serve as token springboards for potential sequels, or — worse yet — are indicative of the creator(s)’s lack of a narrative plan, e.g., Lost.
Mystery is inherently alluring, though, and it can also have a fulfilling payoff. The Souls games are a good example of that.
Each title begins with a seemingly disconnected CG intro, and proceeds to thrust the player into a crumbling world with barely an explanation. There are no lengthy expositions, conquests retold over animated world maps, extensive flashback sequences, etc. Instead, whatever pieces of narrative the player puts together are entirely optional and widely scattered about.
A tib-bit mentioned in passing by an NPC foreshadows a gruesome battlefield encountered later in the game. Flavour text accompanying an item hints at a long-standing dynasty and its wealth. Parts of defaced statues allude to an outcast regal heir.
There’s not much of a plot to the player-controlled protagonist, but there’s an incredible sense of depth and history to the setting itself. It’s all very cohesive and consistent, and delivered with understated elegance.
That’s something incredibly rare for a brand new series, but the Souls games actually have something of a 20+ year development history.
From Software’s other games such as Eternal Ring, Shadow Tower, Evergrace, Otogi, and King’s Field contain bits of gameplay and ambiance present in the Souls titles: stamina-draining melee attacks, stat-boosting equipment, sporadically dispersed NPCs, non-linear exploration, item durability and crafting, fog-of-war/dynamic lighting, loading screen and item flavour text, highly destructible environments, “soul”-harvesting progression, etc.
All of these previous games experimented with and revised what’s so confidently delivered in Demon’s Souls and Dark Souls, but the series itself also follows in the footsteps of another older title: Wizardry.
Wizardry’s arrival and subsequent popularity in Japan is fairly well documented, and King’s Field, From Software’s inaugural release, is said to have been closely inspired by the Western CRPG. The interesting part is that Wizardry’s success seems to have come in part due to a shoddy localization. The only clear example of this I can find is a Wikipedia entry that mentions Blade Cusinart — a silly nod to Cuisinart food processors — evoking an aura of alien mythology.
I assume the results were similar with subsequent Wizardry titles, which contained even more pop culture references, but it’s hard to find any concrete evidence of how these were interpreted in Japan. Perhaps someone else could shine a light on the subject?
Regardless, it’s still fascinating to think about how a simple misconception could be taken to an extreme. Many of From Software’s titles found a niche audience and followed their own paths instead of borrowing the homogeneous conventions of their peers; what else could we have seen if a misunderstood production memo or marketing bullet-point was left to evolve in a bubble?
In the end the significance of Wizardy’s Japanese localization might be a bit overstated, but its heritage is certainly evident in the Souls games. They’re positively brimming with relics steeped in a strange, foreign history, and greatly contribute to the series’ unique style.
It’s not difficult to find an article these days detailing the troubles facing the Japanese games industry. Things just aren’t as rosy as they used to be, and there’s plenty of finger-pointing as a result: budgets aren’t big enough, the cultural differences are too vast, software design methodologies aren’t properly utilized, the corporate world tends to stifle innovation, there’s a lack of outsourcing, the desire just isn’t there, etc.
While all these claims might be accurate to some extent, they’re high-level issues that no one can fix single-handedly. Instead of moping about them, I thought it might be a bit more constructive to offer some small, pragmatic advise. In a previous post I tried to do this with a certain localization issue, and now I’ll take a look an interface quirk common in many Japanese games: too many confirmation prompts.
As an example, continuing from a saved game in a typical modern title is fairly painless. Quite often a “Continue” option is the default selection on the title screen menu, and clicking it automatically loads the latest save file.
On the other hand, these are the steps required to resume my game of Resident Evil 5, one of the marquee current-gen titles developed in Japan:
- Entering the title screen menu immediately brings up a pop-up asking me to “Wait a moment…” followed by a message stating that there’s no storage device selected. This is accompanied by a “Yes/No” prompt asking me if I’d like to select one.
- Clicking “Yes” brings up the OS browser with the available options: hard-drive/memory card/the cloud. This requires me to scroll to my desired option and click it.
- Once the storage device is selected, a “Storage Device Configured” message appears along with an “OK” prompt that needs to manually dismissed.
- Following the previous prompt, a “Loading content…” message shows up and then a “Load successful.” message replaces it. This is accompanied by yet another “OK” prompt.
- When the title screen menu finally appears, the “PLAY GAME” option is selected by default. Clicking it takes me to the play game menu.
- On the play game menu, the “CONTINUE” option is selected by default. Clicking it takes me to an overview of the last save game.
- The save game overview displays a date stamp, the selected character, and some other miscellaneous info. It is accompanied by an “OK/Back” prompt.
- Clicking “OK” takes me to a network overview screen with various game options such as co-op settings and hit reactions. The default option is “START GAME”, and the screen is accompanied by an “OK/Back” prompt.
- Clicking “OK” takes me to a loading screen that’s quickly replaced by the inventory screen. Here the default option is “Organize” and I need to scroll down and click “Ready” to proceed.
- Clicking “Ready” brings up a confusingly labeled “Exit” confirmation with a “Yes/No” prompt. “Yes” is the default option, and clicking it finally loads my save game.
A large part of Apple’s success is elegantly accommodating for the most common use case. This basically means that an interface caters to the functionality that’s used most often, while the elegance comes from avoiding extraneous options, prompts, and technically-minded messages (and presenting an aesthetically appealing UI, of course).
Looking at Resident Evil 5 through this lens, the above steps could be truncated and/or altered to provide a more streamlined way of loading the latest save game.
- The “Select a storage device?” screen shouldn’t be there. Instead, the game should automatically select a default storage device, or better yet, select all the available storage devices. If none are available, a warning message could be displayed on the title screen without requiring a separate modal popup.
- The OS device-selection pop-up should only appear if the user chooses to manually change the current storage device.
- The “Storage device configured.” message shouldn’t appear. There’s no point in flooding the user with text if everything went OK. These messages should only pop up if there are errors.
- Same as above; there’s no need to display a “Load successful.” message. The transition into the save state should make it obvious that the data was correctly retrieved.
- If a saved game was found, the default options should be “Continue.” This option should immediately load the latest save game from the selected storage device. Optionally, the game could check all the available storage device and automatically load the latest save file in order to avoid any extra management on the player’s part.
- The secondary play game menu isn’t necessary if the “Continue.” option loads the latest save game.
- The save game overview should be removed as it provides non-vital information when trying to load the latest save game. Instead, this data should be presented in the load-game interface where the player browses through multiple save files. Optionally, it could also be shown on the loading screen itself.
- The network settings screen should be removed as well since it provides non-vital options that are configured at the beginning of the campaign. There’s no pressing need to change these every time the game is loaded, and this functionality could still be provided via an in-game menu.
- The inventory screen is also superfluous to loading a save game — the save data should already contain the proper equipment information. Presumably the screen is there so that the player can change their loadout following a game-over, but in that case the inventory-customization screen should only appear following the actual death. Alternatively it could also be accessible in-game from the save-checkpoint.
- The “Ready” confirmation is horribly labeled as it’s an ambiguous descriptor. Is the player exiting the inventory screen, or the actual save game loading process (it’s the first one, but it always makes me stop and think)? The prompt itself is also unnecessary, especially after the nine preceding ones.
Confirmations prompts in particular tend to be quite prevalent in Japanese titles. Of course these can be useful when it’s easy to hit the wrong button and the consequences of doing so are quite drastic, e.g., clicking the “close” button instead of the “maximize” button in a word processor after writing a lengthy, unsaved document. However, it’s rarely difficult to select the proper save-file in a game, and loading the incorrect one tends to waste only a short amount of time.
Despite this, Japanese developers seem paralysed with fear of the user accidentally selecting the wrong option. This only applies to UI-related interfaces, though; there’s no prompts for avoiding an accidental weapon-reload or putting a car into the wrong gear.
The convention also seems to be that “No” should be the default selection. I have no idea why this is the case, except to prevent the user from accidentally skipping through an important choice while blazing through a bunch pop-ups.
If that’s the assumption, then it speaks very poorly of the application flow as a whole. Perhaps the user wouldn’t be so quick to skip through these confirmations if there weren’t so many of them? And perhaps removing non-vital popups and prompts would provide a faster and sleeker way to get to the fun part of the game: the actual gameplay.
Have you ever played a game that you really liked, but certain parts of it disappointed you (for the record, I totally dig Tom Francis’ proposed ending to BioShock)?
Did the lack of knowledge pertaining to the developer’s budget/timeline/goals/etc., stop you from thinking “Why didn’t they do it *this* way?”
If you’re passionate about a particular title, then probably not. And why should it? As the end-user, you ultimately care about your own experience, and a game’s faults might seem all the more painful if seemingly obvious and feasible changes could have eliminated them.
For me, that game is Tokyo Jungle, and here’s what I think would have made it better:
Let’s start with the easy, somewhat less subjective field of UI. I don’t think anyone reading this enjoys manually scrolling through the thousands of words that make up a typical EULA (and sometimes studios don’t even want to write their own). The fact that Tokyo Jungle pops up a EULA every time you start the damn game is infuriating. It shouldn’t be there at all, really, especially since its only online component is a global leaderboard.
The leaderboard is not all that great either. It takes a very long time to load, and it’s retrieved whenever you finish playing Survival mode. Why not do it in a separate thread and let the user move on? Or at least only force this path if the player has gotten a new high score? What makes the delay even more frustrating is that it needs to be endured in order to register all the unlockables of the playthrough. Simply quitting a game does not record any of the collected items, story mode pieces, etc., which should be saved instantly.
Finally, the world map is quite useful, but also somewhat confusing. Its most zoomed-in level is quite small and doesn’t clearly indicate accessible areas. The location-labels are a bit misleading as well since they contain a bar that fills up and an icon inside the right edge of the bar. At first I thought the fill indicated my dominance of the area (how many spots I marked with my animal), while the number of icons represented the amount of food within its borders.
Turns out it’s actually the fill that reflects the quantity of available food, and the icon is just a label for the fill. To make this indicator more intuitive, the icon should be outside the bar on its left side, or alternatively a “food” caption should be displayed within the fill.
Aside from the herbivores’ double-jump and inability to consume other animals, there’s not a lot of mechanical variety between the various types of fauna. Sure, there are statistical differences, but the gameplay is exactly the same. Expect to see crocodiles scaling buildings by jumping from one extruding air conditioner to another. Creating custom gameplay for each animal would’ve been a sizeable undertaking, though, so I’ll give Tokyo Jungle a grudging pass here.
What’s less excusable is the stealth mechanic. For something that’s presented as a large part of the game — especially for those peaceful herbivores — there’s no clear way of telling what is an animal’s zone of awareness. This is exasperated by the fact that many animals spot you while they’re off-screen, especially in lower-left and lower-right corners of the view window due to the perspective of the camera.
The minimap helps to spot these potential threats, but not while it rains, and it’s more of a band-aid solution anyway. A circular outline for each animal’s field of vision would’ve helped, or at least some arrows on the edges of the screen indicating potential dangers. A further aid would be displaying the exact threat-level of each animal, and possibly a countdown timer showing how much longer before it reverts to a neutral state.
Toxicity can also be problematic to detect. Hiding inside of buildings or underneath bridges doesn’t seem to help when it’s raining, and contaminated food is hard to detect due to the very subtle purple visual that can blend in with the background. Simple icon indicators similar to the alert exclamations could have easily removed this ambiguity.
Surival vs. Story
Despite the annoyances mentioned above, Tokyo Jungle’s biggest failing is in how it handles its Survival and Story modes.
Tokyo Jungle was originally a retail game, and it’s painfully obvious that it was modified to fit a price tag. Story mode — the main campaign — consists of 14 short missions, and each one needs to be individually unlocked by grinding it out in Survival mode.
I suppose this approach greatly extends the overall playtime, but it’s quite frustrating to progress through the narrative one small step at a time after jumping through some hoops in a completely separate game mode. This is doubly perplexing as unlocking the story missions often involves a certain knowledge of the game’s mechanics, but those same mechanics are then explained in the unlocked missions. The whole arrangement reeks of a production change that was implemented late into the game’s development.
The story missions could use a few more checkpoints as well, but they’re quite fun as they contain lots of silly and amusing sequences that slowly unravel the game’s mystery: what happened to all the humans? It’s a neat premise, and it shouldn’t be so heavily gated (especially if it was a questionable way to justify the price since the game was released as an inexpensive downloadable title outside of Japan).
Instead, Story mode should be featured first and foremost, and the animals played/fought during its missions should then get unlocked in Survival mode.
Survival mode itself is an even bigger mess.
Its main goal is to live for 100 years and complete various side missions to get as high a score as possible. In order to provide variety and ensure that players get different scores, Survival mode employs randomization and high-threat events/encounters common to roguelikes. The problem is, all these gameplay systems conflict with each other.
Hunger is greatly boosted in comparison to Story mode (it takes 90-120 seconds to die of starvation) and the missions are on a strict time limit. This means you are constantly on the run if you hope to get a high score, which also doubles as the currency for unlocking new animals. Completing the side missions awards statistical bonuses and unlocks new costumes as well, providing further incentives to rush through the game.
This approach completely invalidates the stealth mechanic, makes exploration of the cool urban environment impractical, and prevents the player from messing around with fun, emergent events such as battle royales of bears fighting chickens fighting giraffes. The random toxic rains and food shortages add further frustration as they can make some of the side missions virtually impossible to complete.
A better approach would’ve been to tone down the unreasonable hunger meter and remove any other time pressures. Next, the randomization could be more prevalent, starting off each playthrough in a different area with a different mission set. New objectives could come in as old ones are completed, and the resulting pace would let players get comfortable with the game and experiment with its most fun components.
If this led to seemingly infinite playtimes, the randomization could be skewed to provide a gradually increasing challenge. Better yet, the statistically-boosted animals of other players could enter the gameworld as AI-controlled bosses to help crown the real king of the hill. Finally, new animals and costumes not present in Story mode could still be used as prizes for playing through Survival mode.
Agree? Disagree? Have any other examples of a game where certain design choices seemed downright baffling? If so, feel free to leave a comment!
- What Kratos Taught Me About Combat Encounters – Some very practical tips on combat from one of God of War’s designers.
- Hotel Dusk: Invisible Causality – I never had a name for it, but the term “Invisible Causality” seems fairly appropriate when describing one of the more infuriating aspects common to adventure games.
- Designing Better Levels Through Human Survival Instincts – Narrow, Intimate, and Prospect spaces, and how such concepts relate to level design.
Not too long ago I praised The Witcher for a plethora of things it did really well. The sequel’s not bad either, but its minimap is absolutely horrible. The main problem is that it rotates with the camera, and the lack of compass directions also exasperates the issue.
Rotating minimaps are great for following a linear path, which is why GPS devices use this design. The user hardly ever needs to worry about whether they’re driving South or South-East, but they need to accurately follow the generated route. Consequently, it’s a lot easier if the path is always facing the same direction as the car, i.e., if the arrow on the screen is pointing right, they need to make a right hand turn.
However, if the map doesn’t rotate, then driving South with an arrow pointing right actually means making a left-hand turn. To avoid this confusion and unnecessary work with mentally rotating the map, the view of GPS devices is synched to match that of the car.
FPS titles also tend to benefit from rotating minimaps. Their levels are often small or just linear, and it’s very helpful for the player to be synced with the minimap view. The reason for this is that split-second decisions often need to be made based on the immediate surroundings.
For example, if the player is following a team-mate turning right but there’s an enemy hiding just around the left corner, it’s beneficial to instantly know which direction to face in order to counter the ambush. Since FPS games also inherently don’t possess a floating camera, it’s that much more advantageous to be aware of what’s lurking beyond the player’s view as there’s no other way to peek around the scenery.
Static minimaps, on the other hand, are much more suitable for games with large areas that need to be traversed multiple times.
In these titles, it’s important to familiarize oneself with the layout of the land in order to travel through it efficiently. Goals are often described with compass directions in mind, and landmarks are used to aid in the building of a mental map for the overall area.
If the minimap constantly swings around, not only does it keep changing the direction north is pointing, but it also forces the player to digest a radically different topography each time they glance at the minimap. A static view is superior to this as it facilitates the parsing and memorization of an area’s layout. This in turn allows the player plot their own paths and comfortably maneauver through the game’s environments.
Of course some players are only used to one approach or the other, in which case why not simply include both options?