Curious blog post on adventure gaming seven deadly "sins"

[Datadog]

Okay.

[BBP]

Problem solving is inherently trapped to whatever the programmers can think of. Games like LSL5 and 6 make a variety in it: there's plenty of puzzles that can be solved in two or three or more ways, in LSL5 it might even be argued it makes the game too easy. Dev team thinks of anything - but anything they didn't think of (entering the women's changing room while Gary's at the bar) can't be programmed into the game. The same goes for any game that exists, save programming bugs (I'm thinking of the Prince of Persia sword things: bypass most of level 1 and get a sword anyway, and in lev 12 you have the ledge grab bug that can let you get the sword before somebody else can).

Maybe we should stop the "Yeti pie" setting as example, since that's such a particular unintuitive puzzle it stands out. KQ5 would be an example of the opposite since it does have a bit of variety in problem solving: the shoe and stick can be interchanged, and you can buy most things with the gold items you get in the city.

[adeyke]

In the context of adventure games, you're right. Those are programmed, in a very, very general way, as "If the player uses item A on hotspot B, do C". Every interaction has to be thought of by the designer in order to work. This isn't necessarily a bad thing, but it does mean that they need to spend a lot of time thinking of these interactions. If the player is presented with the something that looks like a solution, the designer should anticipate this and either actually make that also work, or give a good in-game reason why it wouldn't work. This is one of the challenges of designing an adventure game.

Note, though, that the "If the player uses item A on hotspot B, do X" programming isn't used in all genres. In many games, it's instead "Item A does X", "Object B does Y when X happens to it". For example, with Portal, they didn't have to specifically program a different interaction for each wall the player clicks on. Instead, the player is just given a portal gun that can create portals, and some surfaces have the property of allowing portals to be placed on them. Most of the gameplay then comes from the game's physics engine. What this means is that it's very possible for levels to be solved in a way that's different from what was intended. Instead of needing to program in every potential solution, there's the opposite problem: they had to be on the lookout for "wrong" solutions and try to redesign the levels to make them impossible (or leave them in as special challenges).

That sort of thing is also common in roleplaying or strategy games. Those have a variety of skills/units/items/weapons/cards/etc. that all just have some programmed functionality. The designer might have planned for a particular strategy to be especially effective, when it actually turns out that some other strategy is much better.

I'm not at all saying that "If the player uses item A on hotspot B, do X" system is bad. It works great for adventure games, provided care is taken to consider enough possibilities. I just don't agree that the resulting gameplay is especially out-of-the-box or creative.

I don't think throwing a pie at a yeti is actually that bad a puzzle. Temporarily blinding a charging beast on precarious cliff does make some sense. KQ5 had plenty of issues, and that's not high on the list. The bigger problem with the pie is that's it's edible, but that eating it results in a dead end.

[BBP]

It's the Lemmings factor. Actually finding alternative routes is a lot more fun: I'm amazed seeing those videos of alternative solutions where the player manages to, say, save 100% in Cascade, you know the one with the high release rate and quick long drop and only 10 umbrellas?

I suppose the audience for adventure gaming has changed a lot: checking TVTropes shows that a lot of Sierra puzzles are now considered to be on the "fake difficulty" pile as being "real bad".
I'd love to make an adventure game where you have the option between playing it hardcore Sierra style with timed puzzles and death at every corner, or soft-core can't lose but with a different ending. LSL7 allowed for that I believe, you could make a text-file that said "get hard" and get a more difficult game out of it.

The biggest problem in KQ5 is the voice acting, which I believed ranged from awful to extremely awful. The trial and error (prevalent in many puzzles in KQ5, even yeti could qualify), it may not be my favourite type of puzzle, but I'll do it. What makes it annoying is that it means you'll have to listen to all the voices again, which in combination with no text option and no clicking through dialogue is what makes it hard to bear.

[adeyke]

AGDI's remake of KQ1 was originally entirely true to the original (by which I mean the SCI remake), but a later version optionally removed dead ends (and some potential deaths) from the game. You'd be asked at the start if you want to play with dead ends enabled or not. That feature only happened because they wanted to simultaneously improve the game and provide a 1:1 remake. I can't see anyone doing that sort of thing for a new design.

I do agree that the voice acting was terrible. And there are some really bad trial-and-error puzzles. The desert is the prime example of this. There's no way to get through the desert legitimately. The only option is to try to get through repeatedly, dying and restoring each time, until you have a map. The end result is that Graham wanders through it seemingly at random and just happens to land on an oasis each time. Also, in addition to trial-and-error puzzles, there are dead ends (like the aforementioned edible pie). So even if you do start just systematically trying every possibility, that still might not give you a solution.

[BBP]

The desert is still a pretty doable maze. Unlike the one in the basement of Mordack's castle that is so confusing... Mapping out mazes is not something I really enjoy doing, but just one manageable maze in a game gives me no reason to nag. It's when the game itself is a maze or the mazes keep on coming (I'm looking at you, Frankenstein - Through The Eyes Of The Monster!).
I suppose that all that needing stuff for a later point when you're not able to get to it... a lot of games have that, and it's annoying... In making the Gold Rush! script I noticed there is a prompt for when you forgot to pick something significant up, which I suppose would be the gold coin. Should try it out.
Solution would be easy: just tell a person they still need an item as they try to progress. Or don't allow them to continue without said item.

[adeyke]

I never really had a problem with Mordack's maze. It can be solved by just always hugging the wall. I never needed a map for that maze, unlike with the desert.

There are a number of ways to avoid dead ends from missing items, some more subtle than others. In addition to just stopping the game from progressing if the player is missing the item, the game could just provide an alternative item if the player advances without it, it could have a puzzle that requires the item prior to the point of no return, or you could automatically pick up the item while picking up a different item. All of these are better solutions than just having the player miss the item and end up in an unwinnable state.

It occurs to me, also, that any such missing item dead end would also be an example of what Ron Gilbert calls a "backwards puzzle". So an even better solution might be to place the item at a point in the game after the player sees the puzzle that requires it.

[Tawmis]

I don't think adventure games really do offer that much in outside-the-box thinking. Everything you're able to do is something the game designer already thought of. You're never creating original solutions; instead, you're just finding the solution the designer planned for you. No matter how smart, creative, and resourceful you are, you'll never find a way to thwart the yeti other than throwing the pie, and you'll never find a way to reach Meathook's island other than via rubber chicken. This is the weakness of a system where each individual interaction has to be scripted. There are games in other genres where the player is just provided with tools that reliably work in certain ways, and these can then let the player find solutions the designers didn't think of.

Well, one thing. Even though a puzzle is designed by the creator - it still requires "out of the box" thinking. Out of the box does not mean, completely, unique, that NO ONE has ever thought of. It means, unusual thinking. So if someone played an adventure game, and one of the items they picked up was peanuts. Eventually they come to a hunter's trophy room with various animal heads - one of them being an elephant. Why on Earth would you give the peanuts to the elephant head? That doesn't make sense. However, perhaps in this game, doing so allows the elephant to reach a lever that you could not - and thus opens a door, that was, previously inaccessible. That is thinking outside the box.

[adeyke]

If it's been established that the world of that game is one in which mounted animal heads are somehow alive and hungry (or, at least, a world in which that type of thing is to be expected), that's a pretty straightforward puzzle.

If that hasn't been established, however, and the player has ever reason to think the head is dead, that's just a terrible puzzle.

[DeadPoolX]

Adeyke has summed up my issues with Adventure games and the insane so-called logic that's used in their puzzles. In particular, the part about how puzzles have to fit the established setting.

Here's an example:

REAL WORLD: Protagonist has a peanut butter sandwich in his inventory (let's say he's holding it and didn't just stick it in his coat pocket), but has come upon a difficult puzzle involving a VCR.

The protagonist decides that the best course of action is to stick the peanut butter sandwich into his VCR.

The result is... he's just destroyed his VCR.

ADVENTURE GAME WORLD: Protagonist has a peanut butter sandwich in his inventory (let's say he's holding it and didn't just stick it in his coat pocket), but has come upon a difficult puzzle involving a VCR.

The protagonist decides that the best course of action is to stick the peanut butter sandwich into his VCR.

The result is... the VCR transforms into a singing dildo, which is then used as a rocket-propelled grenade later on.

No, that's not thinking "outside-the-box." That's the sort of thinking that needs psychiatric assistance.

[adeyke]

The most important part of a puzzle, I think, is the "aha!" moment. After finding a puzzle, they should explore and think about it and eventually get that moment of realization. They should just know that they've found the solution, based on their understanding of the game, the world, and how things work. If someone gets an "aha!" that turns out not to actually work, that's a failure. But conversely, if the player is surprised that something they tried out works, or if the direct consequence of some action isn't at all what they expected, that's also a failure.

This doesn't mean that adventure games need to be realistic, just that they need to be understandable.

[Tawmis]

If it's been established that the world of that game is one in which mounted animal heads are somehow alive and hungry (or, at least, a world in which that type of thing is to be expected), that's a pretty straightforward puzzle. If that hasn't been established, however, and the player has ever reason to think the head is dead, that's just a terrible puzzle.

Is it, though? In an adventure game? I wouldn't say it's a terrible puzzle. To me, the player would think - "Well there's got to be some reason I was able to put the peanuts in my inventory."

And then, later in the game, coming upon a mounted elephant head - just might think, "I remember how they feed peanuts to elephants at the circus." (Or something along those lines; where real life logic is a part of the problem solving). So why not try to give the peanuts to the mounted elephant head and just see what happens - as crazy as that sounds.

Adeyke has summed up my issues with Adventure games and the insane so-called logic that's used in their puzzles. In particular, the part about how puzzles have to fit the established setting.
Here's an example:
REAL WORLD: Protagonist has a peanut butter sandwich in his inventory (let's say he's holding it and didn't just stick it in his coat pocket), but has come upon a difficult puzzle involving a VCR.
ADVENTURE GAME WORLD: Protagonist has a peanut butter sandwich in his inventory (let's say he's holding it and didn't just stick it in his coat pocket), but has come upon a difficult puzzle involving a VCR.
The result is... the VCR transforms into a singing dildo, which is then used as a rocket-propelled grenade later on.
No, that's not thinking "outside-the-box." That's the sort of thinking that needs psychiatric assistance.

LOL! I can see if it's to that extreme. But most puzzles (that I am thinking off, at the moment) have not been to that extreme. It's been loosely based on real life logic to some degree. (Well except for some from LSL, because they're pretty out there...)

[adeyke]

I'd say that it is a terrible puzzle, yes.

If the player has no reason to suspect that the elephant head is alive, they wouldn't be able to conclude "If I click the peanuts on this head, it'll eat them". It would, at best, be a guess. That means there was no "aha!". Trying crazy things just to see what happens is fine in the thinking/exploration phase, but it shouldn't actually be the solution to the puzzle.

It's also a case of a backwards puzzle. The way you describe it, the player had no actual reason to pick up the peanuts beyond the fact that they can, and they're instead reasoning that the ability to pick them up must mean that they're used for a puzzle. That sort of reasoning isn't at all about the game world and is instead only about the game itself. It would make much more sense if they first saw the head, then found out that it's somehow alive and hungry, and then intentionally picked up the peanuts for the purpose of feeding the head.

It's unfortunately true that a lot of adventure games do have puzzles like that one. Who could forget "throw bridle on snake to get magic sugar cube that protects against poisonous thorns"? It shouldn't really be held as acceptable, though.

I suppose another way to look at it is in terms of motivation and narrative. The character should have a reason for each of their actions, and if you write a novelization of the game, every action in it should make some kind of sense. Just saying that the character is a kleptomaniac who'll try anything isn't good enough .

[Tawmis]

I'd say that it is a terrible puzzle, yes.

Then don't play my text adventure game "Final Soul."

If the player has no reason to suspect that the elephant head is alive, they wouldn't be able to conclude "If I click the peanuts on this head, it'll eat them". It would, at best, be a guess. That means there was no "aha!". Trying crazy things just to see what happens is fine in the thinking/exploration phase, but it shouldn't actually be the solution to the puzzle.

It's also a case of a backwards puzzle. The way you describe it, the player had no actual reason to pick up the peanuts beyond the fact that they can, and they're instead reasoning that the ability to pick them up must mean that they're used for a puzzle. That sort of reasoning isn't at all about the game world and is instead only about the game itself. It would make much more sense if they first saw the head, then found out that it's somehow alive and hungry, and then intentionally picked up the peanuts for the purpose of feeding the head.

It's unfortunately true that a lot of adventure games do have puzzles like that one.

I guess, from playing Sierra games way, way, way, way back when - there was two things I remembered.

"Pick up anything that isn't nailed down."
"Save Early, Save Often."

So when I began playing Sierra games, once I quickly discovered the strange logic to some of the puzzles - the rest came naturally to me. (Some, naturally WAY more difficult than others). So I never thought it was a big deal. I still don't.

The gods forbid that someone make a game where you have to be a little creative and a little crazy to solve something - rather than just resorting to google to find the answer to riddles or something - or have a gun and shoot something. (Mind you, I love me some Gears of War, when I am in the mood for that!) But complaining that a puzzle is a little too far fetched - meh - I'm got no complaints.

But to each their own. Some people love Heavy Metal (me!), some people think it's just a lot of freaking noise (just about every parent out there!)

Who could forget "throw bridle on snake to get magic sugar cube that protects against poisonous thorns"? It shouldn't really be held as acceptable, though.

Granted, some were pretty far fetched. I think even crazier was the Rumplestilskin (sp) that was the reverse ALPHABET - not just his name. THAT was pretty crazy.

I suppose another way to look at it is in terms of motivation and narrative. The character should have a reason for each of their actions, and if you write a novelization of the game, every action in it should make some kind of sense. Just saying that the character is a kleptomaniac who'll try anything isn't good enough .

Yeah, but if you're looking to make it a novelization - that's totally different than a game. It'd be like writing a novel about a basketball game...

He passed the ball. Dribbled it down the court. Shoots. Misses. Bob gets the rebound. Shoots. Misses.

Not going to be all that exciting and make much sense, because it's a totally different medium.