How to make an adventure game *really good*

[Gurok]

I'd like to see if we can compile some rules for *really good* adventure game design. i.e. What not to do. Reply with rules of your own.
Thread title inspired by the fantastic YouTube series "How to draw really good" by Oliver, Age 24.

I'll start.

(Before you read further, these aren't directed at anyone in particular. I think we've all considered or perhaps done these at some point. I'm just trying to shine a different light on them and perhaps bring a little levity. You are also free to take one of these and argue the opposite. I don't mind.)

Good games come in parts. Better games come in more parts.
You've got a great story idea, but you're petrified of writing an ending. What do you do? Release the introduction you spent a very long time on and call it the first part of a series. No, your vision is grander than that. The *prologue* of the first part of the first *act* of a series. There, that's better. And until you've written it, the ending will remain perfect, as envisioned, in your imagination. Remember, they're not plot holes if you never resolve the plot! Motivation's also running low and this release should help stoke the embers. That's how it works, right? People will be queueing for miles to get their hands on the next part and if they don't, it's because they don't understand your art. Nobody liked Robert Jordan's first book, but he kept at it. Persistence.

Who am I? Not the Jackie Chan film.
You don't have a story, but you have some cool environments and a few puzzles. Can you still make a good game out of that? Damn right you can. Okay, who is the main character? What's your identity? Let's see... uh... what about... you don't remember? You don't remember who you are. And how did you get there? You don't remember that either. Finally, what's the character's supposed to be doing? Let's just say the character doesn't remember that either. Whew. That's the bulk of the writing out of the way then. You're all set to tell that story about being alone on an abandoned spaceship with only a jumble of VHS-quality surveillance recordings too short to provide any substantial plot. No? Okay, a mysterious island with almost no human inhabitants that serves a paper-thin facade for some surprisingly setting-neutral puzzles. Bonus points if you make it first-person pre-rendered 3D.

When it comes to pixels, think big (and small).
We all know how great 320x200 games are. Part of what makes them so good, it could be argued, is their limited resolution. It serves to reason then that *more* limitations should result in a better game. 160x200 games were pretty much twice as good as the 320x200 ones. It gets even better when you halve the resolution again or invent your own resolutions that have no historical significance. Naturally, this improvement in quality reaches a peak at the point of singularity. All games are best played on a single pixel, flashing to inform you of what's happening. Low resolutions also make a REALLY good cover for your limited art skills. I mean, nobody can judge your art that harshly when you tell them that the resolution dictates a cat should be about the same shape as a kettle. To take the pressure off, consider a variable scale for your art. If you can't draw a cat, that means the cat should be drawn with larger pixels. Simple things like furniture and boxes should be drawn with smaller pixels. This is still "retro". Nobody cares about that authenticity thing anyway.

[Stupot]

I can't handle unhandled events anymore.

When it comes to unhandled interactions, games seem to fall into one of three extremes a) hardly any responses to perfectly good interactions, b) a stock of one or two random responses ('no you idiot, that won't work') or c) absolutely "hilarious" unique responses to absolutely every possible interaction.

a) is forgivable. Sometimes we just don't need a response to everything we click on and that's okay.
b) pisses me off. Don't tell me "that's a stupid idea" for daring to pick up a fire extinguisher despite the fact that there's a fire in the next room (that was an made up example for exaggeration, but it's not far off). In fact don't say tell the player he or she is stupid ever, for trying something unless it really is stupid.
c) is commendable, but actually gets quite boring to read and usually ends up with a load of terrible jokes that just fall flat. Not every interaction needs to be comic gold.

To avoid c) you need to make some stock responses. But it's lazy to just assign every unhandled event one if the stock responses without thinking about it first. Every possible interaction should be considered. There should be different subsets of stock responses for different types of items. Containers, doors, keys, big things, things which are too high, things which are impossible to do, things which might be possible but can't be done at this time, things which are inaccessible for whatever reason. Write four or five stock random responses for EACH of those categories. And consider every possible action to judge what TYPE of response it might trigger. And if none of them work, write a unique one.

[Edit] just realised this was in the rumpus room and I might have taken this post a bit too seriously, but I've been wanting to get that off my chest for a while, so sod it

[Haggis]

Wow, it's like you pulled them directly from my brain!

[CaptainD]

* Have a long, unskippable animated sequence every time something that regularly needs to be done in the game is done.

* Make moving from one location to another such a slow and tedious affair that you start to grow grey hairs between locations.

* Make long, unskippable dialogue sequences that you have to play the whole way through every time even though it accomplishes nothing useful and says nothing new.

* Ensure that the only way to complete a puzzle in room 10 is to have found an item which is a pixel hunt and for which no clues are given anywhere back in Room 1. 

* Make tricky puzzles for which the only clue is a passing reference much earlier in the game (in a dialogue that cannot be repeated).

[Danvzare]

In fact don't say tell the player he or she is stupid ever, for trying something unless it really is stupid.

I can't agree with that enough. That is the exact reason why I still haven't completed Randal's Monday. He calls you stupid every time you do something wrong, even if it makes sense to do that!
It's really off-putting.

Now for my additions.

A good story never ends, just keep it going forever and ever and ever. No one cares about endings or closure, just keep slapping higher numbers at the end.

Don't give hints, you should never mention what an item could be used for, or what it should be used for. As a matter of fact, you should only mention potential uses for an item that aren't in the game, to help serve as red herrings.

On the topic of red herrings, you should have loads of useless inventory items. It doesn't matter what these items are, even if they should be incredibly useful and should help complete the puzzles. Just give a default response of "I can't do that" whenever the player tries to use any of these red herrings. Putting an axe in your game which you can never use, is always a great idea.

And the best advice of them all. Randomly kill the player every chance you get. As a matter of fact, you should frequently put the player in a position where they are presented with two or more choices, and where only one choice won't kill them. So they have to use trial and error and the save load system to progress further.

[Slasher]

Hi Guys,

sometimes you can't be right for being wrong but i want to add my two pence worth.

Regarding interactions:
if there are repeats (objects/hotspots) then you should use stock phrases as Stupot pointed out so you don't keep repeating the same old same old, just make the text different, even if meaning the same thing. I agree that one standard phrase is bad practice.

Example stock if repeats else where:

'This place is dirty.'
'About time they cleaned up around here.'
'It's so messy in here.'

Also you don't need interactions for everything, only keep descriptions on what you can interact with.

Cutscenes:

Having to sit through 10 minutes of un-skippable dialog or sequences is a belly up for me especially when replaying. Having the player press Esc is the best way to end any cutscene.. They must have wanted to because they pressed Esc.
If the cutscene is vital and lasts less then say 10 seconds then that would be ok.

Needed object(s) for puzzles should not be more than 3 rooms back.

Having to click everytime to skip text is a no no for me and drives me nuts... just have it stay on the screen a bit longer.

Making references later on to previously mentioned .. this would be ok if the player had actually bothered to read the dialogs and the reference is later repeated so as to go into your memory

You get into a situation of doing nothing with no signs as to what to do...

Good idea
'I need light in here.' response.. hence the need to light the torch.

Trying to find  1 X 1 Pixel.. Aghhhhhhhhhhhhh

Ridiculous puzzles: keep them clear and solvable...even if you need to think about it more...

Don't try to be pretentious or smart.

etc etc

[Blondbraid]

Hello!
I agree with just about everything here so far and I'll give this a try:

Remember the classic 2D adventure games with beautiful graphics that still holds up really well today?
No, cheap 3D models that looks like creepy plastic puppets are far superior! They were cutting edge technology in the 1990:s! And don't bother with using custom models for important characters, standard assets will be just fine. And textures? Plain colors will do fine, preferably dull and grayish pastels.

In-game menus should be as busy and crowded as humanly possible, the more animations and different sound effects, the better!

Make all items you only use once in a puzzle stay in the inventory afterwards. Sure, the item will never be used again and it makes no sense for the player to keep it, but its a great red herring in future puzzles!

Speaking of puzzles, they don't actually need to be part of the game world in any way that makes sense. Players looking for an intriguing murder-mystery or epic fantasy adventure will love to randomly stop and do puzzles lifted from a children's activity-book! And let's not forget that puzzle where you have one 1-liter bucket, one 5-liter bucket and one 3-liter bucket and have to fill a bucket with 4-liters of liquid, I bet nobody has don exactly that puzzle in a video-game before!

Comedy relief is a must, context, story and basic common sense be damned! And the very best way to do comedy is to have a funny sidekick whit no personality whatsoever apart from being funny. And the best way to make them funny is to have them break everything all the time, and being a liability in general. And if you really, really want to make them stand out and be memorable, have them speak in a very squeaky voice and repeat themselves a lot. Or make them fat.

[Snarky]

The default AGS font is perfect in every way and for every purpose, and should under no circumstances be replaced.

Adventure games are all about leisurely, deliberate game play, and the best way to achieve this is to make sure that the character walks reeeaaalllyyy slooooowlyyy.

Even if a game has voice acting, no one cares about what the characters are saying, so make sure the sound mix is set so that the music drowns it out.

If you've written one dialog tree for a character, that's enough. Writing more dialog to update it in response to events in the game is a waste of time. So if there's a neighbor who starts out the game looking for a lost cat, it makes perfect sense to still have all the conversation options about that missing cat (and nothing else) even after you've found it, returned it, and discovered that the world is about to end.

After having spent months or even years toiling away at your game, why would you want to wait a few days just to beta test it, proofread the text, polish the UI, or anything like that? It's not like those are really easy improvements that can make a big difference to how your game is perceived.

[/sarcasm]

When it comes to unhandled interactions, games seem to fall into one of three extremes a) hardly any responses to perfectly good interactions ...

a) is forgivable. Sometimes we just don't need a response to everything we click on and that's okay.

I disagree with this. To click on something and have absolutely nothing happen is IMO the sign of a broken game. There should always be some default response, if only to provide feedback to the player that the click was detected. No need to make it too flip and clever; something simple like "Better not" "That won't work" or "I think there's a better solution" is fine. It's better not to vary it, because it makes it simpler for players to just click past it, except perhaps for actions that are so reasonable that a special excuse is needed for why it's not supported.

[Adeel]

How to make an adventure game really good?

- It's simple. Don't make one.

[Stupot]

For maximum indie Twitter cred, your next game MUST feature a strong, independent female protagonist. Anything else is terrorism.

@Snarky - I think I basically agree with most of what you say actually. I think by 'forgivable' I just meant 'slightly more forgivable than b) and c) but still rubbish'

[CaesarCub]

It's always a good idea to put an unskippable mini game that demands reflexes and quick thinking. Not only should it be difficult, but if possible, tie the game to the processor speed directly, so the difficulty increments hand in hand with technology.

[Mandle]

It's always a good idea to put an unskippable mini game that demands reflexes and quick thinking. Not only should it be difficult, but if possible, tie the game to the processor speed directly, so the difficulty increments hand in hand with technology.

Hahahahaha!!! That's my fave so far!!!

It's not a minigame, but there is a moment in one of the King's Quest games where you have to get rid of a time bomb before it explodes. You're supposed to have like 30 seconds or so. But the bomb timer runs off the processor speed, so on modern computers it just explodes as soon as you pick it up and you die. Classic!

[Creamy]

1) People love to click. A lot. Let them do things step by step to make sure they get their fix.

Spoiler

The light went out. You need to replace the bulb. Forget about using a new bulb just yet. You must remove the old one first. Not with your bare hands, it's still hot. Use rag with old bulb instead.

[close]

2) If you game takes place in the modern era, let the player hack anything - from cellphones to doors. Your game will gain in credibility.

Spoiler

More specifically, anything can be hacked with a technological gadget owned by the local nerd. You just have to convince him to lend it to you.
Don't worry, it doesn't involve anything too technical â€" it's a child's play to use.

[close]

3) Never joke with social conventions. Even in a life and death situation, no hero shall ever run, force a door, break a window or tear a crucial evidence from a little girl's hands

Spoiler

Let's trade it instead.

[close]

4) Common objects should come in one hidden copy only.

[Mouth for war]

"don't call the player stupid" haha oh Jimmy the troublemaker how you always did that. I actually had him giving an even harder boot to sensitive players in the sequel about that whole issue ...which i've restarted 3 times and probably won't ever happen...  I did put in options though for the player to choose from "Real Jimmy(I want it all)" or "Nice Jimmy (I stopped breastfeeding when i was 15)"

[Cmdr]

People love to click. A lot. Let them do things step by step to make sure they get their fix.

This reminds me of a great design choice I just saw in a fanadventure:

If you have a can (or bottle) with a liquid inside you need, make sure the player has to OPEN it first before he can use it.
If the player otherwise tries to just use the CLOSED can on the target object better give no response to the action to punish the player for his stupidness to forget to open it first.

[AnasAbdin]

I respect and appreciate all the comments and opinions here. But I always say it's the little differences between us (people) that make life an awesome experience. If you don't have a specific demography you are aiming the game for, then follow your heart, be different, be yourself.

Warning: Spoilers
For example, Stair Quest

Spoiler

Despite being a short game, Stair Quest uses one of the most frustrating puzzle techniques used in an AGI adventure game as the whole game's idea, and then the game forces you to go backwards to the beginning. Yet, many people including myself enjoyed the game and played it regardless how many times we died.

[close]

I do follow a lot of the suggested points mentioned in your previous comments because I do have preferences in adventure games like most of you do. And I'd hate to put something in a game I'm working on that I don't like in other games.

These are some of the things I learned to avoid in my style of adventure games:

• Long dialogs (skippable/unskippable)
• Blocking events that take too long (with or without good animations) -Anastronaut: The Moon Hopper: Pushing the big rock scene is 30 unskippable seconds
• Pixel hunting -Dexter Morning Routine: Killing the fly in the beginning of the game ^*
• Large inventory items that have to show in the character's sprite -Anastronaut: The Moon Hopper: I had to draw the walk cycles with the character carrying a paint bucket
• Impossible puzzles

Things I love in adventure games:

• Captivating intro scenes
• Combining/taking apart inventory items
• Mini games
• Subsystems: Computer interface, mobile device...etc
• Rewards for complicated puzzles -The old Sierra points sound!
• Comedy

* In my defense, Dexter was not meant to be a playable adventure game, it was made to be a video parody

[Mandle]

It's always fun to hide a vital clue (such as a safe combination) in a cutscene that takes place without the player-character even present, and then have the player need to use that clue later on to continue progress...

[tzachs]

My game features all of the above (plus more exciting stuff, like long repeatable mazes!) which pretty much makes my game the best and most epic game ever created!

160x200 games were pretty much twice as good as 320x200

That's true, but you inspired me to take it one step further: my next game will be in a 1x1 resolution. It will feature 200 different rooms, each with stunning hand drawn graphics and challenging puzzles (and no pixel hunts if you use a x16 scaling filter!).

[Jack]

Family games are too hardcore. Let's take it one step further and make a game for adults who want to be children:

- Solve every problem by providing in-depth catharsis, counselling and closure to all the world's inhabitants
- Achieve a 100% rating on your politically correct meter and unlock the special ending where your meagre flesh provides nourishment for starving polar bears
- Forget gameplay, you will be rated solely on how effectively you advanced the mainstream political agenda

[CaptainD]

I would like to add this fun-filled concept - A complicated copy-protection system that is actually more of a challenge to solve than any of the puzzles in the game itself.

[Danvzare]

I would like to add this fun-filled concept - A complicated copy-protection system that is actually more of a challenge to solve than any of the puzzles in the game itself.

And which relies on time-sensitive technology, such as paper whose ink will eventually disappear through age, or a prism that needs to be placed against a CRT monitor, or the requiment of a having a floppy disk drive connected. Preferably all three.

[Stupot]

You should continue to describe your game in all publicly viewable arenas as "retro" despite it's not really being very retro at all. This will help Adventure games to remain niche and ensure that no progress is made in the genre.

[CaptainD]

You should continue to describe your game in all publicly viewable arenas as "retro" despite it's not really being very retro at all. This will help Adventure games to remain niche and ensure that no progress is made in the genre.

You should also claim that with your game you are single-handedly "resurrecting the dead genre of adventure games". 

I would like to add this fun-filled concept - A complicated copy-protection system that is actually more of a challenge to solve than any of the puzzles in the game itself.

And which relies on time-sensitive technology, such as paper whose ink will eventually disappear through age, or a prism that needs to be placed against a CRT monitor, or the requiment of a having a floppy disk drive connected. Preferably all three.

Not forgetting of course to make a system that is absolutely impossible for anyone colour-blind or even slightly visually impaired to be able to solve.  (Actually add that to in-game puzzles as well as the copy protection!)


In addition, create an interface so ugly, complicated and confusing that even a polydextrous octopus with an IQ >200 and perfect vision would struggle with.

[Blondbraid]

This thread really reminds me of The worst muse on Twitter!
So, so much wonderful terrible advice, I couldn't resist posting a link.
Some of my favorite nuggets of wisdom:

If you make that kid the Chosen Hero, you can totally bypass any other character development!

Video game technology basically plateaued at the Super Nintendo, right?

You're right: A sanitized Medieval France where everyone's white and uses modern idioms is a very original setting. Go for it.

You won't have to do as much research if you replace the city's immigrant population with elves.

It's not really an exposition dump if a character is reading it off a futuristic device.

You don't want readers to solve the crime ahead of your detective--better withhold all the really critical clues.

It's doesn't really count as exposition if a character is just reminiscing out loud.

[Mandle]

Make sure the cursor changes from the active item back to the default pointer every time the player attempts an interaction, and make it so the player must then reopen the inventory window to reselect the same item again.

Bonus points if the game also insults the player every time a wrong interaction is tried (I know this one is mentioned above, but it's extra-super-great when used on an already annoyed enthralled player)

[Crimson Wizard]

Make sure an important event in the game, like the appearance of vital character or item, only takes place when player walks over precise but visually non-distinctive spot of a large room. Even better if he/she must walk in a very particular direction.

Implement score points in your game and make 1 score point impossible to get because of a bug. Or just for fun. You like fun, don't you? Your players will have fun too, searching for the solution to get maximal score.

[CaesarCub]

Implement score points in your game and make 1 score point impossible to get because of a bug. Or just for fun. You like fun, don't you? Your players will have fun too, searching for the solution to get maximal score.

Ohh I have been guilty of this one when I made public an unfinished game, had to apologize too many times.

Also:

Have the final puzzle of your game depend on a specific piece of technology. For example, having to eject the CD to solve the game, making it impossible to finish for all the future generations that bought it using GoG.
Also as an added bonus, make the game in really low poly 3D and force the players to walk around huge empty maps. (This might help get the reference if the previous line didn't).-

[Snarky]

Puzzle difficulty should be completely independent of story importance. Ideally you want your players to spend most of their time trying to figure out how to scrape gum off a desk or find the phone number of the video store, not stopping the Satanist conspiracy to drain the Pacific.

Also, if it's not a standard inventory or dialog puzzle, you don't want it in your game. No gameplay hybridization, original twists on the formula or unusual elements are advisable. Remember: A creative gameplay mechanic is why Insult Swordfighting is so reviled. Ideally, all puzzles should follow the exact same template: Pick up inventory item, combine with another inventory item, use on hotspot/give to character.

Adventure game players by definition have nothing better to do with their time, so take the opportunity to really set the stage up front. Will they be able to properly appreciate the story you've concocted without a complete history of the world in which it is set? I should think not! Can you really care about this Gabriel Knight fellow when you don't know what his childhood was like?

And for the sake of realism, make sure to have them play through a few "regular days" in the main character's life before anything interesting happens. A few hours in, the impact will be so much greater!

[Mandle]

And for the sake of realism, make sure to have them play through a few "regular days" in the main character's life before anything interesting happens. A few hours in, the impact will be so much greater!

I feel there can be some cases where this could be handled well:

For example, in the movie "Shaun Of The Dead", where the normal days are set up for social parody when compared to the later days after everything goes to hell...

But yeah, just for the sake of it, with no proper motive behind it...Not a fun idea!

[CaptainD]

It's not really an exposition dump if a character is reading it off a futuristic device.

It's doesn't really count as exposition if a character is just reminiscing out loud.

Additionally, don't forget to make these non-skippable.  And repeat in various random places throughout the game just in case the player fell asleep the first time.

[Blondbraid]

The key to make any creative work *really good* is to paste together a series of vague ideas of what the Kidz today think is hip,
and businessmen age 50 or older is an excellent source of information on what kids today like.

And if you are ever in doubt about how to make your work seem more contemporary and relevant,
make the protagonist wear a backwards cap and add rave music. Lots of rave music.

[Danvzare]

Another way to make your creative work *really good* is by banking on the nostalgia of the adults.
And to do that, all you have to do is add a few references to the old NES. Because everyone over twenty five years old, has nostalgia for that console. Absolutely everyone!
Just say "Power Glove", and you know it'll sell.

As a matter of fact, graphics peaked on the NES, so why not make your game with the same graphical limitations as that brilliant console, and get the bonus of calling it 8-bit.
And if for whatever reason you want to make your games more graphical than an NES game, perhaps making it look like something akin to a SNES game but with way more colours. Then call it 8-bit anyway. Besides no one actually knows what bits mean.

[Creamy]

And if you are ever in doubt about how to make your work seem more contemporary and relevant,
make the protagonist wear a backwards cap and add rave music. Lots of rave music.

[imgzoom]http://www.slightlywarped.com/crapfactory/curiosities/2013/february/images/bad-cat.jpg[/imgzoom]

[Blondbraid]

Some tips on how to make classic characters *really good* :

The protagonist always has an incredibly boring life before the adventure starts, and what better way to tell this than to slowly crawl through his everyday routine as slowly and tedious as possible until the audience is even more bored than the hero?
And he should whine a lot about being meant for greater things, but only until the adventure starts. Then he should whine a lot about how awful it is to be a hero and how he just wants to be a normal boy.
You don't actually have to show the protagonist doing good things, just have everyone tell him how awesome he is. And there is no such thing as a Gary Stu.

The strong female character is really good at fighting, so the male hero will look even more awesome when he has to rescue her towards the end of the story.
She should scream things like "I hate corsets" or "I don't wanna wear a dress!". Real strong female characters only like to wear practical combat gear, such as chain-mail bikinis or stiletto heels.
You don't actually need to show her being capable of fighting, just have her swing a sword at nothing in particular and say "I don't need a man. I can take care of myself." in a pointlessly rude tone.

The wise wizard/ Scientist should always look like Gandalf or Albert Einstein. They are a great way to deliver exposition to the hero, and the best way to do it is for them to hold a very, very long unskippable non-interactive monologue about it, and to make it extra good, have them bumble and forget their words and repeat themselves a few times. Also, it doesn't really matter weather the setting is sci-fi or medieval fantasy, all lab-equipment looks the same, funny looking bottles with tubes in them that are full of what looks like assorted soft drinks. A good idea is to have them fizz and bubble on a repeated loop while the old man holds his very, very long unskippable non-interactive monologue to the hero.

A villain has no need for a motivation other than Evil. If you want to be really fancy, have them be mad with power and want to take over the world/kingdom.
And there is no need whatsoever for the villain to have a well-defined or consistent plan, just have them do whatever is necessary to move the plot along.
There is no need to show them doing or saying anything remotely evil, wearing dark clothes and living in a Gothic castle is enough for the audience to want them dead.
And don't let the hero defeat them by the knowledge and powers the hero has acquired through the adventure when you can just use a Deus Ex Machina instead!