Thursday, January 29, 2009

Braid: Understanding Difficulty

This Blog has moved to

This particular Page has moved to

If you are a recurring visitor, please update your bookmarks & rss feeds. We apologize for the inconvience.


Blogger Midwest Gamer Podcast said...

Although I loved the game. I have to agree that aspects of could very well be considered too hard. The parts of the game I enjoyed the most were towards the beginning when I was able to run through the level, consider my options and almost on the fly make a decision to solve a problem. When that worked it was an incredible feeling of satisfaction.

29 January, 2009 20:09  
Blogger Graham said...

Very interesting post. I agree; too many people are discussing the words, and not enough discussing the game. Thank you. :)

That said, I think that a lot of the specific problems you mention are actually more general problems. One you mention is how poorly 'used' keys are communicated.

Another one that really got to me (and several other people I know as well) is that the glowing green effect on irreversible items is not pronounced enough. The 'Don't Shoot The Puppy' puzzle you mention suffers severely from this. Everyone I've seen play this puzzle, including myself, takes a long time to even notice that one of the bars is glowing! This is even more fundamental than poor puzzle design. The high degree of accuracy needed in a lot of the platforming fits in this category as well.

However, there is something that I've had a lot of trouble reconciling in my own thoughts about the difficulty in Braid: of the 5 puzzles you listed, I only had issues with 1 of them. In fact, I thought that 'Movement, Amplified' was one of the easiest puzzles in the game. That said, there were other puzzles that had me stumped for days. In all my discussions with my friends about Braid, there is inevitably, "Oh man that puzzle was so hard!" "Really? I figured it out right away!" or vice versa.

I'm not sure how this could be resolved. This kind of thing often rears it's head in games, where some players pick up on a clue and others don't, but in Braid it seems to have much less pattern than in other games I've noticed this for. As you say, it does have to retain some challenge. If everyone says something different, how do you determine where the correct challenge to keep is?

Back to my original point, I think this can be dampened by tweaking elements outside of the puzzle design (many of which you mentioned): more obvious puzzle elements, better delineation of where a puzzle starts and ends, a simpler way of resetting a puzzle, more 'atomic' puzzles to teach skills and clear connections between puzzles using the same skill. All these things are enablers, that would allow players to focus on solving puzzles rather than fighting the game, bringing the designer closer to understanding which puzzles are hard vs. which are unclear.

In closing: I disagree that the game was too hard. In fact, I would have liked, overall, for there to be a bit more challenge! But I strongly agree that many puzzles were difficult for the wrong reasons, and to me, this is what people are actually complaining about.

29 January, 2009 21:14  
Blogger Graham said...

Something else that just struck me:

The game allows you to move through the different levels reasonably freely, but then requires absolute perfection to reach the last level.

You said that you needed to use a walkthrough for one puzzle. If the game let you finish with one unsolved puzzle, then you could have beat it without using a walkthrough. Then what if you could have two unsolved puzzles, and still seen the ending? How many more people would have enjoyed 'beating the game'? Three puzzles? Ten puzzles?

Again, speaking from my own experience, I found the monologue only mildly interesting; I played the game for the puzzles. But many people played for the ending. So players like me got what we wanted -- reward for completing the puzzles. But where was the reward for players who's goal was to puzzle through the story? Since completing the puzzles was not the motivation they were looking for, the game had these large, arbitrary-feeling obstacles in it. Thus: "Too hard!"

29 January, 2009 21:56  
Blogger Krystian Majewski said...

Well, I also finished the game fairly quickly and I needed help only on that one puzzle (which I couldn't have solved otherwise). You are right: From the puzzles I've mentioned only maybe 2 are ones where you could really get stuck. However, the others can establish a general impression of a very hard, unforgiving game... even if it isn't.

But I strongly agree that many puzzles were difficult for the wrong reasons, and to me, this is what people are actually complaining about.

I'm glad we've reached a common ground here.

The game [...] requires absolute perfection to reach the last level.

I found that there was plenty of wiggle room. It seems like getting one of the stars even depends on reaching the castle before the princess. I didn't bother to collect those, though. They are just insane.
You are somewhat right: there is some precision involved in the last level but you can achieve it fairly easy with the rewind power.

As for the narrative - see that's the thing. If we play the "what if Braid was a different game"-game briefly: if Braid was just a standard puzzle braintwister (like Timebot for example) without the narrative, I probably wouldn't complain so much. Those games are mental measures of strength and you somewhat expect them to be unforgiving and hardcore. But then again, I probably wouldn't bother to finish it, just like I didn't bother to finish Timebot.

But Braid DOES offer something substantial for finishing it and it is different from simply a mental challenge. Therfore, I expect the puzzles to be more carefully designed with a less frustration-tolerant player in mind. And this is ESPECIALLY true if the game has been polished for such a long time.

As for finishing the game without solving all puzzles - I didn't suggest that. I agree that it wouldn't make sense. I do think that some puzzles would need some tweaking, though. I mentioned how in my article.

01 February, 2009 21:33  
Anonymous Anonymous said...

Great post. I noticed most of these design issues myself in playing it, but I certainly did not write them up with the alacrity and clarity you have brought to the subject.

02 February, 2009 00:06  
Anonymous Anonymous said...

As someone who completed the game without hints, I wanted to comment on the issues you bring up. The majority of the levels you take issue with do provide hints to them in previous puzzles or provide simple workarounds that take a small amount of time.

3-6: "irreversible" is the theme of the entire level, the player is given a warning and a hint at the same time. Resetting the puzzles is as simple as leaving through a door and re-entering. Though this does have some "you can get stuck" moments, I think the designer tried to minimize this problem by making it easy to reset the level.

4:6 Movement Amplified. In 4:1, Jumpman, there is a puzzle piece locked by two doors, similar to this one. That puzzle warns you about the consequences of opening the right-hand door, with less punishing results.

5:4 Crossing the Gap. The author states that there is no time where an enemy will fall on the player's head to demonstrate what happens. The first part of this level has an enemy falling from a great height toward the player, and if the player starts running from the door, they are very likely to be hit on the head. This ability of the enemy is also likely to be discovered by players as they work through the game.

4:7 Fickle Companion. The level records the key's progress through the level at each instant in time(x position), until it is touched, carried, or dropped, at which point it overwrites the recording at that point. Only an enemy who can traverse the x axis while moving forward in time can send the key "back into time" so you can reach it when you go back there. This one was probably the most confusing puzzle for me, but there is some logic there, even if this behavior only shows up in one level.

I agree that some changes / indicators could have been more pronounced, but exploration of time in a gameworld, a dimension explored only superficially by other games, is its greatest strength.

03 March, 2009 16:59  
Blogger Krystian Majewski said...

@22Samurai: Thanks for your in-depth comment. I tried to answer it but my comment got eaten (grr) so here it goes again:

3-6: "irreversible": The title is too vague to make any difference as a warning. Calling a level in MegaMan "You might die" isn't going to make the game any easier.
Restarting that level is more difficult than others. Just recently, my collage Yu-Chung experienced a situation where he couldn't get out of the underground area because enemies collected at the exit. And while you can always use the menu, it feels somewhat awkward.

4:6 Movement Amplified: You are right. There is the same puzzle in "Jumpman". I mentioned that in my review. I chose this one as an example because it was clearer. The "Jumpman" puzzle doesn't warn you on anything. It's just the same puzzle with the same mistakes. If you failed to understand the mechanics back then, you will struggle again in "Movement Amplified".

5:4 Crossing the Gap: I agree that I haven't considered that there is indeed an enemy falling from above in this level. However, claiming that it is there to explain the mechanic is quite the stretch. It is not set up to explicitly hit the player, there is no effort made to make the player pay attention to how the enemy bounced off the player's avatar. The player will most likely try to avoid collision anyway. I don't believe the ability is likely to be discovered. At least I see no effort from the game designer to make sure it would.

4:7 Fickle Companion: I do understand the reasoning behind it. However, it is very obscure. Especially the fact that there are different rules for enemies and the player make this puzzle very difficult to comprehend. The problem is that if it would be consistent, it would look even more erratic. Best solution would be to leave it out.

16 April, 2009 12:52  
Blogger Unknown said...

This comment has been removed by the author.

18 May, 2009 01:00  
Blogger Unknown said...

In World 4-2 and 4-6, the door puzzle creates a divide by zero scenario. You're going backwards in time and simultaneously creating a forward in time action when you try to unlock the right door with the key. Maybe one way to fix this is to have the character face right the whole time while also allowing the freedom to move left. So the player ends up with his back to the right locked door with the key in hand and figures, "Oh, I'll never be able to unlock the door moving this way." This keeps the interesting paradox and doesn't punish the player.

There might still be problems with this quick fix such as, "Why am I stuck facing to the right?" or the paradox not being realized again.

I know this reply is ancient in this day and age of the internet. Been playing Braid recently and wanted to contribute. Loved Legend of Kay and Excit! Can't wait for your future projects!

Edit: Corrected the directions of the doors I was trying to talk about. >_<

18 May, 2009 01:44  
Blogger T said...

I love this post very much. I love the explanation of the puzzles, and I hope Blow has/will read this.

Your idea on the "message" is also the best I've ever seen. By totally ignoring any idea of the Princess being a person, or the atom bomb, or trying to make sense of the disjointed vignettes, I think you've probably arrived at the intended meaning, even though all the other meanings people have arrived at are equally as interesting.

However, you totally lost me at your discussion about whether the message is appropriate: "But finally, because I do have problems with the idea of the pursuit of truth..."

Before that, I was thinking to myself, "Yeah! It's a message about science, REAL science!" Not bubbling beakers and tesla coils and hating religions; real science as in the search for truth and the beauty of understanding."

You explain that the game is about finding the Absolute Truth, but that's wrong because a single truth doesn't exist, but I disagree (with the first part- we can't really say whether the second part is correct or not). Braid doesn't even care whether a single truth exists, it focuses solely on the search. Indeed, you never find the Princess, you only get closer and closer.

"I think we abandoned the idea of a single fundamental truth long time ago. Even a scientific mind should realize that there are many stones and many castles. All we can do is to construct models. ... There is no princess, only castles."

Stephen Hawking once said, "My goal is simple: It complete understanding of the Universe and it's reason for existing."

It's completely reasonable to think that there is some fundamental truth, otherwise what are these models approximating? We may never know it... I don't think any scientist reasonably EXPECTS us to ever know it. So thinking of The Princess as a metaphor for Absolute Truth really doesn't contradict anything at all.

Even though you say, "I think the princess is yet in another castle. The one that isn't in the game," you could also say the Princess is not in ANY castle. Tim/The Player will never find her, not in this game nor any other other. Nor will anyone else. She is unattainable.

The message is, that doesn't mean we shouldn't try to find her.

24 May, 2009 19:25  
Blogger Krystian Majewski said...

@Mark: That's actually a very simple yet excellent solution! Thanks so much for bringing it up, I certainly haven't been thinking of this but it work perfectly!
(And to be honest, my involvement in Legend of Kay was rather minor but I will make sure your kind words get to the right people ;-)

@Trevor: Finally someone commented on that part. I was eager to discuss that.

You bring up an excellent point. It's true that you never find the princess. At least not the Truth-Princess. So in that way, the game doesn't claim that the absolute truth is real. I haven't considered that.

However, there are a other aspects in the game which do support that ideology. It is especially the way every puzzle seems to be designed with a singular, specific solution in mind. And this has been critisized by other writers as well... but then again, that was the whole point.

By the way I find it quite ironic that even though Blow claims he had designed the game with a single specific interpretation in mind, it was able to generate so many different ones - like the A-Bomb one.

The game didn't leave me with the impression that it was about the search. After all, there was a clear winning condition. If it was about the search, I think it would have to (to stay within the metaphor) focus more on castle building. Then it would need to have some Sim City'esque qualities where there wouldn't be any winning conditions, just exploration of rules and designs and general pointers.

The Hawking quote you mention illustrates quite well the fundamental misunderstanding many bright minds in the scientific community exhibit - they expect the universe to have a reason, they (maybe even secretly) expect a fundamental truth. One of the great insights of the humanities was the exploration and ultimate realization of the concept that absolute truth doesn't exist. That neither the universe nor humans have a purpose but are merely means to themselves. I think the most striking example is Immanuel Kant with his concept of Human Dignity/wiki/Human_dignity which is the basis for the German Constitution and even the Universal Declaration of Human Rights (as opposed to a religious one).

It's completely reasonable to think that there is some fundamental truth, otherwise what are these models approximating?I think that's the tricky part where many people loose track of the argument. The models are just means of interpreting what is happening for us humans to somehow comprehend in a certain context. In order to do so, they reduce complexity and leave out details which can be important in different models. In any cases, they are simplifications. The question if there is a truth they approximate towards is useless. The function of the models works towards the human end - they work because they deliver us temporary answers. If they would to represent the world faithfully, they would in fact become like the world - incomprehensible. But there is the whole branch of Ontology with many way smarter and more eloquent people who wrote their whole lives about it.

And I don't think we need the notion of a fundamental truth to drive science. Society didn't collapse into anarchy when we started secularizing after all.

24 May, 2009 23:00  
Blogger Unknown said...

Hey, i also found this interesting to read, however, i find your level-critics too harsch:
World 3, Puzzle 6: "Irreversible" - the name gives the major hint, i knew before entering that level, that i was going to have to use "exit current world" in the menu, which by the way is always possible and appropriate.
World 4 Puzzle 6 "Movement, Amplified" - yes, the right door is a pure trap - so what? Re-enter the level and get the key takes less than 30 seconds, nothing I would regard as punishment.
World 5 Puzzle 4 "Crossing the Gap" - yeah, that enemy-shadow-double-jump is not apparent, but: There is nothing there but these enemies, when you switch the brigde to left, the new cannon-spit-out-enemy comes right to the area, where it's needed after killing the old one.
World 4, Puzzle 7: "Frickle Companion" - yeah, that was annoying: "Hey, wtf is the key doing? Why does it keep falling down ladders?...Oh, remember." BUT the enemy, that is able to keep the key, has a green aura. That's the reason, it is able to hold on to it. I agree, keymovent after killing the holding guy and rewinding seems buggy.

You keep saying, that those advanced machanics like unrewindable key/rewindable door - pairs should have been shown earlier for a learning curve - i disagree again: You (me?) realize it anyway and reentering and getting to that point is always a matter of less than a minute.

09 June, 2009 02:56  
Blogger Krystian Majewski said...

You (me?) realize it anyway and reentering and getting to that point is always a matter of less than a minute.

When you die in a game, or when you break a level, it's the game's way of telling you that you did something wrong. But not knowing something is not the player's fault, it's the games fault. Of course, you can also learn by punishment but that's a very bad way of teaching people things. My girlfriend is a teacher, she'd gladly talk for hours about this.

As for your arguments:

3-6: "irreversible": The name is not a hint, it won't prevent you from doing something wrong. I explained that above.

World 4 Puzzle 6 "Movement, Amplified": As I explained, the problem is not only the random punishment but also the fact that the game doesn't explain to you WHY it doesn't work.

World 5 Puzzle 4 "Crossing the Gap": I beg to differ. There are a lot of different objects in that level - levers, ladders, platforms and other backrgound objects that may or may not be of relevance. Also, it only works in a very specific moment. I think there are too many variables to expect players figure that out on their own.

World 4, Puzzle 7: "Frickle Companion": As I said, the fact that the enemy is green explains why it's not affected by your position. It does not explain why it can hold on to the key. In fact, he shouldn't as not even the player can.

09 June, 2009 03:15  

Post a Comment

<< Home