Psychochild's Blog

A developer's musings on game development and writing.

10 July, 2016

What makes a good metroidvania?
Filed under: — Psychochild @ 1:16 PM

As I said, I picked up a few metroidvanias from the recent Steam sale. As I love that genre of games, enough to list it as one of my favorites, I figured I’d take apart a few things I think make for a good metroidvania.

The classics

The name “metroidvania” comes from the games Metroid and Castlevania, specifically the games after Symphony of the Night. The games tend to focus on exploration of a large space with progression. That progression is often tied to getting access to more areas of the map to explore. I’ll pull references from these games as examples. Other games might have done different things better, but these games are well-known enough that examples shouldn’t require a ton of explanation.

A big map to explore

Symphony of the Night set the standard here. Not only did you have a huge map with a wide variety of locations, but you also had the “twist” of having the reversed castle. Even if you explored 100% of the game, you needed to be completionist enough to find a few special items to get the “real ending” and reveal the second castle so you could explore up to 200% of the game. (And even beyond that if you exploited a few glitches!) The map tickles the exploration desire some of us have.

The player is often expected to go through areas multiple times. As you get new abilities, you might need to backtrack to reach some location you couldn’t reach before. And, in the original Metroid, each of the two minibosses was down a different path from the start and defeating one required you to backtrack back through the main area to get to the other boss.

Of course, having a large map has a few design consequences. The first is that it shouldn’t be a hassle to go through the areas repeatedly. Getting a new ability should make it fun to go back through the old areas. Finding shortcuts and defeating previously untouchable enemies should feel enjoyable. Also, it’s nice if there’s a way to get around the map quickly. If you want to go through a certain area again with a new ability, it’s better if the game allows you to get to that area easily without having to travel all the way back.

Finally, I appreciate a good in-game map. Being able to see where you’ve been in the larger picture is nice. The original Metroid didn’t have a built-in minimap, but the whole map is pretty simple when you look at it in its entirety. Plus, the larger Super Metroid did have a map. A map helps when you’re looking for where to go next, or where to look for those last remaining items you really need to find to get that 100% if that’s what you want.

Obstacles and puzzles

Any game needs obstacles to overcome, and in metroidvanias these take specific forms. Obstacles can range from the simple jumps and leaps to enemies and massive bosses you have to defeat. Part of moving around the map is overcoming these obstacles and enemies. Something stands your way of simply walking through the entire game.

Common obstacles generally include places that are unreachable when you first encounter them. A tall ledge, a locked door, or a particularly tough enemy might be a sign that you need to come back later once you progress a bit more. (We’ll get to progression in a moment.) A new tool or technique might be exactly what you need to get past that previously inaccessible part.

Good obstacles are obvious and immediate. You shouldn’t have to fight through a tough area of monsters just to reach a locked door you can’t open yet; this frustrates the player by wasting their time. An obvious obstacle also should be hint at what you need in order to overcome the obstacle: a tall ledge needs something that allows you to jump higher, a tough enemy needs something to do more damage or something to exploit a weakness, etc. The player should have an “ah ha!” moment when they get a new ability, remembering that obstacle from before and being eager to go back and conquer it.

You can also have puzzles in the game. These can range from simple puzzles where you have to kill all enemies in an area to unlock something, to more complex puzzles where you have to visit a certain location at a certain time to get an item used as a key somewhere else. The way to unlock the hidden reversed castle in Symphony of the Night is a good example of a pretty complex puzzle. In the day of online game walkthroughs, someone will figure out your fiendishly clever puzzle eventually.

A good puzzle should fit the game. If your game is straight-forward, then the puzzle should be as well. Trying to mix things up by making a puzzle super-obtuse doesn’t work in a metroidvania any more than it works in other games. In fact, it can be even more frustrating if the player is stuck on the puzzle and can’t progress. In general, I think obstacles work better than true puzzles in metroidvanias.


Your character needs to progress in some way. In metroidvanias you frequently find new abilities that open new routes to take. In Metroid, this was accomplished very simply with a few item: the Morph ball (or Maru Mari) to get through tight passages, the freeze beam to freeze enemies in place to jump on, the high jump to jump higher, etc. Each item gave you some power, but also let you get to places that were otherwise inaccessible.

When you get a new power, the game should show you how to use it without an obnoxious tutorial. For example, when you get the morph ball in the original Metroid, you run into a short tunnel right there where you have to use the morph ball to advance. This taught the player the simple use of such a power, and how to use it to get around. A great games gives you the tools, shows you how to use them, then lets you play around with them to figure out more advanced techniques.

Symphony of the Night also added RPG progression to the formula, years before it was cool to do that with other genres. You could gain experience and levels, and also gain better equipment that had better stats. This allowed you to withstand more punishment in harder areas and do more damage against advanced enemies. This also added an incentive to kill enemies in the game, whereas in Metroid it was often more efficient to ignore the enemies if you could. In fact, killing enemies might be counter-productive if you needed to freeze them and use them as stepping stones.

Pacing is important with progression. When giving out abilities, the player shouldn’t get too many abilities at once and feel overwhelmed, just like any other game. On the other hand, going too long without an upgrade can make the game feel a bit stale. Having the right pacing is important to keep the player interested but not overwhelmed.

Once again, this progression should make the game more fun. In the original Metroid, the screw attack ability was nice because it let you move through areas faster as you could attack most enemies while jumping. (Of course, it did add potential frustration if you accidentally attacked that frozen enemy, destroying your stepping stone to the higher areas!) A new ability or higher level should ideally let the player bypass old, stale challenges in order to find the new, exciting ones.

Evaluating metroidvanias

So, let me evaluate the two metroidvanias I picked up during the Steam sale using these criteria.

As I said in my brief review of Castle in the Darkness, one thing that frustrated me was the “instadeath” spikes in the game, as I felt it weakened the progression part of the game. I couldn’t use in-game methods to avoid the spikes, I would just die if could not perform the jump or I had forgotten about a trap. In theory I could get better at platforming and memorize positions of traps as tackled the same place again, this felt less satisfying than figuring out some in-game way to conquer the obstacle. I also felt the lack of a map and the limited number of weapons particularly since some weapons were much better to fight certain enemies; it was a bit too old-school even for my tastes. As for the map and exploration, it felt perhaps a bit too linear for my tastes. There were very few things to go back and discover with new abilities; although upgrades did make it a lot more fun to go back and kick the ass of things that were slapping you around before.

Valdis Story: Abyssal City does better. As I said, the map was huge and lots of places to explore. There were plenty of things that were tantalizingly out of reach the first time you went through an area. It also had a good flow to it so that you rarely felt lost, although the guidance you got from the story did get more vague as you went along. This was fine, as it was pretty fun to hop around and explore. Obstacles were great, with different enemies to threaten the player. Eventually you’d outlevel the enemies and you could take care of them easily. You’d also get abilities that let you skip past some of the harder ones. The advancement was more of the RPG sort from Symphony of the Night, where you’d kill beasties to get levels in addition to finding gear and items you could trade in for gear. A lot of the upgrades come from collecting a bunch of items needed to upgrade. Overall, I liked it a lot more as a metroidvania. (Although I’ll admit that I did get distracted by playing Lords of Xulima….)

So, what do you think? Are there other major features of metroidvanias that you think are essential? How do you evaluate metroidvanias? Which ones do you think are the best?


  1. I really enjoyed “Ghost 1.0″, a new metroidvania offering from the creator of UnEpic. It isn’t especially long, but quite satisfying.

    Comment by Gothmog — 10 July, 2016 @ 6:24 PM

  2. Very nice article!

    We are developing our own take on Metroidvanias and you article was very inspiring. We’ll use some of your advices for our game.

    Keep up the good work!

    Comment by AXE978 — 10 July, 2016 @ 11:30 PM

  3. One thing I have wondered is if it’s possible to make a Metroidvania with non-linear progression.

    Usually, getting a new ability would let you get to an area you couldn’t, which had areas you couldn’t get to until you got another new ability…there was a definite intended progression. I wonder if it’s possible to design the levels where you can make progress even without a particular ability. As an example, a high ledge that you would normally need a high jump ability to get to, but there’s also the possibility of some clever wall-jumping to get you there.

    I feel like making the game complete-able in any order would remove that “ah-ha” moment you were talking about. I also like my Metroidvanias to have a more cohesive plot (similar to Metroid Fusion or Zero Mission) and making it non-linear would definitely make that trickier.

    What are some of the other challenges in doing this? Can you even make a cohesive game by the time you’re done?

    Comment by Silvanis — 13 July, 2016 @ 1:43 AM

  4. Silvanis wrote:
    One thing I have wondered is if it’s possible to make a Metroidvania with non-linear progression.

    Yeah, I was thinking about this a bit the other day. I think there are a few problems.

    1 Harder to design the abilities well. If each ability can essentially do the same thing as others, then you start getting very boring abilities. Part of the fun of a good metroidvania is getting an ability that feels really crazy. If a wall jump is simply design to be an alternative to a double jump, then the abilities will probably not feel that exciting.

    2. Harder to design the levels well. If you don’t know what exact abilities a player will have by the time the reach a certain location, it’s harder to design challenges appropriate for it. As I said in my post, it’s a waste of the players time to make them fight through an area only to realize they don’t have the right ability or key to progress further. A player might use an ability to get past Obstacle A, but then hit Obstacle B with no way to get past it.

    3. Harder to tell a linear story, as you said. Although there are techniques for getting around that.

    That said, I think you can do some potentially interesting things. In Valdis Story, there are a few places where there’s a timed platforming puzzle to get an item. You have barely enough time to do it when first encountered, but if you get an ability or two it would be come a lot easier. I enjoyed challenging myself on these puzzles when I first found them. So, I think you could do something like this, where you could do an optional thing that becomes easier as you get more powers.

    Comment by Psychochild — 13 July, 2016 @ 11:15 AM

  5. The reason why I’ve been mulling the idea over is that I want to make a Metroidvania with an antagonist that is also going through the same areas. He would go through them in a set order, so you could really hamper him by beating him to certain places. Of course, he could do the same to you. This would culminate in a showdown where your relative strengths are determined by who got to each area first.

    I could make it so the ability item is not the power MacGuffin in each area (giving you two objectives foe each area, assuming the MacGuffin hasn’t been taken), but the antagonist’s route would need to make sense and you would need to be able to choose a different route through the areas than his. Thus the non-linear aspects.

    So it still needs work on how to make that possible.

    Thanks for your thoughts.

    Comment by Silvanis — 14 July, 2016 @ 1:49 AM

  6. Silvanis wrote:
    …I want to make a Metroidvania with an antagonist that is also going through the same areas.

    Oh, that’s an interesting idea.

    What I could see is having branching paths. So, the idea is you have one direct path from start to finish, but then you have different branches that combine back into the main path. Some branches might be “if you didn’t get power X, you need to take the branch”, others might be “to get bonus item Y, take the branch”; the bonus item might be a “consolation prize” if you miss getting the ability that lets you stay on the main path. The antagonist might decide to take some branches to get the bonus items, too, depending on preferences and if he or she is ahead enough.

    The one downside is that this might not work for traditional Metroidvania fans. Part of the fun is exploring around and finding something seemingly hidden away. You wouldn’t get to do much of that if you’re always racing against the clock. Perhaps have the antagonist be a special mode? Let people play the game and explore around, then challenge a competitor in an advanced mode.

    Anyway, let me know if you do make this. It sounds interesting. :)

    Comment by Psychochild — 14 July, 2016 @ 8:22 AM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:

Recent Comments


Search the Blog


June 2020
« Aug    



Standard Disclaimer

I speak only for myself, not for any company.

My Book


Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on