Spelunky 2: Fuel to the fire

Fabian Fischer
6 min readOct 28, 2020

“In a game with randomized levels, people will keep playing just to see what comes up. But to make it more than a glorified slot machine requires putting together a collection of systems and rules that is worth understanding, behind a world that feels interconnected.”

(Derek Yu in “Spelunky”, Boss Fight Books #11)

Spelunky 2 is great, because it continues to apply many of the things that made its precursors great. Players face randomly generated worlds holding many spatially significant subsystems that overlap in their behaviors in highly emergent ways. The fact that the game is essentially a pretty traditional platformer plays a strangely subordinate role. After all, the core mechanics, even though they are really polished in what is by now the title’s third edition (counting the original freeware version), are not what makes it interesting.

A game of failed plans

Despite its action-packed gameplay requiring quite some controller skills, Spelunky is a game of decision-making. However, these decisions are not aimed at specific actions that directly influence the state of the game or even lead to victory on their own, as you would for example expect in a classic turn-based roguelike. (For the historians: Rogue just hit Steam.) Instead, numerous micro-decisions are made every few seconds, all of which serve to minimize the risk of the subsequent execution these plans.

Do I deliberately trigger the dart trap below me in advance in case I fall into its range later? Do I throw the urn against the wall in front of me, at the risk of a monster breaking out of it and getting in my way? Do I rush ahead and use my whip to eliminate all enemies in my way or do I take the kitten with me and lead it to the exit right away? Do I wait for one more cycle of enemy movement or do I act as soon as possible?

Not only are all these decisions based around their necessary physical execution, but they are also subject to constant time pressure. Spelunky HD had the ghost that entered the level after two and a half minutes and chased the player incessantly — killing them at a touch. This mechanism harkens back to the “Hunger Clock” of classic roguelikes and can also be found in the latest version. Although the ghost appears after 3 minutes now, there is also more to do in the levels which are full of all kinds of “pseudo quests”.

Image taken after a variety of plans have failed already…

On top of that though, Spelunky 2 adds new elements with pretty much the same purpose: moles dig through the ground and make sure it’s rarely safe to just “wait and see”. Horned lizards roll towards you at high speeds if they catch even the slightest glimpse of you. Lava drips down edges slowly but steadily, so you really don’t want to linger below them a second longer than absolutely necessary. Mosquitos hover around the level on line-shaped paths, always looking for a way to rush towards the player from a great distance.

“Combine simple behaviors to give the impression that the monsters are working together. This not only creates challenging situations, but it also makes the world feel more like a living, breathing ecosystem. Wherever possible, I tried to add monsters that attack you from new directions, so that when they were paired with existing monsters the attacks would feel coordinated.”

(Derek Yu in “Spelunky”, Boss Fight Books #11)

All of this limits the player’s time for reflection (in smaller, more dynamic arcs than the wall-traversing ghost). And it makes one thing very clear: Spelunky 2 does not want to be a game of careful and thorough exploration, but one of efficiency. It’s about getting the most out of each level as quickly as possible. Rescue animals to increase your HP. Collect gold and gems to stock up on bombs, ropes and other equipment in the next shop. If possible, sacrifice a handful of enemies at the Kali shrine and be rewarded with additional items. On the way, deal with various traps and monster arrangements. Think on your feet, act decisively.

Decision vs. Execution

In light of this design intention, the gameplay elements make perfect sense. However, they also intensify the conflict that has always characterized the game. The physical execution of planned actions is uncertain, can never be mastered 100%, and is ultimately a form of randomness. If a plan fails due to a mistake of execution, this says very little about the quality of that plan. The other way around, reactions and twitch skill can sometimes make a deficient plan work. This phenomenon continues in the player ‘s learning journey (after all the core motivation, since there isn’t any metagame progress): While your execution improves, your plans may not, and vice versa.

Tom Francis brings up the issues of a “failure of execution”.

This can lead to unclear feedback and eventually to frustration. Especially in combination with the extremely high level of difficulty, this might drive away many players who would otherwise potentially enjoy Spelunky’s emergence and depth, but are not able to get to it in the first place. Yu calls the level of challenge “just a backdrop” for play and experimentation. However, difficulty is of course not absolute, but highly relative. Difficulty levels or, even better, a form of single-player matchmaking would certainly have helped.

By the way, the shortcut system from Spelunky HD is back, but remains a kind of half-hearted non-solution. It drops players in unfair situations without any equipment and resources in the later worlds. For a few minutes — until your inevitable demise — you may be able to study enemy types and environmental elements you don’t know yet. It would be way more exciting and effective to provide players with a random set of equipment in these situations, thus elevating the shortcuts to actual alternative modes of play as shortened, varied variants of the main game.

Andrew Aversa talks about the variety that is required in the early parts of a permadeath game.

Without this option however, players are forced to return to the starting world over and over, which is not only particularly difficult, but (unlike worlds 2 and 3), does not offer an alternative version of itself. This aggravates the reset problem even more and makes the journey towards building up enough skill to “unlock” the immense depth of the game all the more difficult.

In the end though, Spelunky 2 is worth all the initial frustration. It adds new, even more consequential nuances to its proven formula. It works not just despite, but also because of the central gameplay conflict it revolves around. Decision and execution support each other. Players continuously make plans, only to see them undermined immediately. This generates a uniquely emergent form of intuitive, forward-driven decision-making that can only arise through the holistic interplay of all the game’s systems and structures.

Spelunky 2 is not perfect, but that may just be the reason why it is such a deeply fascinating exhibition piece of game design and a ludic journey well worth the effort.