knytt stories is an amazing introduction to level design. it attracts a lot of amateur designers, which is rad. but i also see lots of amateur designers make the same mistakes, and it’s frustrating.
there’s a zzt game called zzt syndromes that tries to identify common design errors in zzt worlds, created in the hope that making designers aware of common errors and suggesting simple solutions would lead to an improvement in the quality of levels. i figured the same thing couldn’t hurt knytt stories.
knytt syndromes is a collection of design problems i see in a lot of knytt stories: making the player guess where a platform is off-screen, confusing background and foreground tiles, being stingy with save points. if designers are conscious of design pitfalls, maybe they won’t fall in – and drag the player in with them.
11 comments
oh man i haven’t thought about zzt in a thousand years, i spent so much time making zzt worlds back in the day. did megazeux every really take off i wonder? time to google
I passed the “impossible for the player!” screen on my first try……
well of course you did i’ve seen your meatboy levels
actually that’s not entirely correct when are you going to upload a video of meaty jill off so i can finally see it (or else get that ortoslon guy to do it)
megazeux blossomed into a small weird demoscene kind of community. i’m probably not the best person to describe it though!
Edmund claims to have recently patched it so that everyone can access Super Meat World even without 20 bandages. (But at the moment the patch seems to have only been delivered to a minority of users.)
Although, the unspoken implication that after all this time you still don’t have 20 bandages makes me wonder “what on Earth is taking you so long?!”
Regarding checkpoints:
Super Meat Boy is a game where each level is designed around a lack of checkpoints, and it is actually something of a lesson regarding the placement of checkpoints in a continuous action game. Levels in Super Meat World have a listed ‘par time’, which is the time taken for the designer to finish the level (as required to upload it), plus 2.5 seconds. After playing hundreds of Super Meat World levels, I’ve come to conclude that the maximum permissible ‘par time’ for a level, regardless of its actual content, is 30 seconds. Anything more – not letting the player rest after making 30 seconds of progress – pushes my personal bounds of reasonability. There are some scant exceptions – such as The End’s levels, which are deliberately intended as pure endurance challenges with very basic obstacles – but the rule holds in the main. This ’30 second’ rule, I feel, should also serve as a design goal for checkpoint distances in action games.
(In the case of maze- or environment-focused Knytt Stories, ones which more closely follow a ‘Metroidvania’ structure, it stands to reason that checkpoints should really be even more plentiful, but the case for which is beyond the scope of this comment.)
Now this sort of thing is the stuff I would like to see taught more often.
Along with a discussion as to what types of game favour what intervals – because most commercial games have a lot larger gaps between saves than this.
I agree with the first “Run, don’t Walk” thing, but if I remember right, that’s actually something I really liked about your old level, “The Lighthouse”. You start off only walking, then start hurriedly dashing towards the lighthouse on the next screen. It was really neat because it felt to me like she was kind of trodding off to do a routine task, only to pick up her pace once she realizes the urgency of the situation.
the lighthouse was a conscious attempt to make walking speed actually be appropriate in knytt stories. that’s why there’s much more vertical than horizontal movement in that story.
another story that uses walking well is “it waits.”
oh, and vitenka: most larger commercial games do not often have one-hit kills. everything in knytt stories is the same amount of deadly. ZAP. you die, go back a few screens. since there’s no attrition, going back a few screens means simply re-traversing things you’ve already seen until you get back to the place that killed you in the first place. and if you die there again, you get to re-traverse those areas AGAIN. the more we can minimalize the RE-TRAVERSING, the better.
That’s a good point – though most big games still offer some one-hit-kills (falling off the bottom of the screen being unreliably deadly has to be a pet hate)
And re-traversal is indeed very very annoying.
But, for example, RPGs usually don’t offer saves except a tthe start of the level and right before the boss encounter. And although the death is usually by attrition, the game can often put you in a position where you have to either willingly backtrack to the outside to recover, or continue and risk failure. And that’s a tense decision and (done properly) can be part of the fun.
I think I agree with your point in general; but it’s certainly worth investigating more deeply.
In my humble opinion, the right amount of save points in a Knytt Stories level is:
– One savepoint for every given amount of screens (I’d say between 15 and 20, others might think this is too little, others too much)
– One savepoint before every “challenging” area
– One savepoint on the screen after every “challenging” area
Any less is being miser.
Now, if the level is really really short and the whole purpose is to challenge the player into making a full run through in one sitting, OK, I’ll understand if you simply skip the savepoints. But otherwise, not adding enough is being mean to the players and to the testers (because, dammit, having people test your levels is important).
post a comment