On Flow - Dynamic vs. Chosen Difficulty

While I will be discussing the game FlOw, I want to discuss the larger concept of Flow in games more. This concept of Flow, as I understand it, is basically finding a perfect balance of difficulty that engages the player without being too difficult (frustrating) or too easy (boring). Many games struggle with this, as different gamers have different ability levels, and learn at different rates.

FlOw seems to be widely hailed as an addicting game that creates Flow well. It has dynamic difficulty adjustment, but even more than that creates an aesthetic quality that relaxes the player and immerses them in the experience. The evolution aspect was rather clever, because while evolving your creature could make it look cooler and stronger, which was usually pleasing to the player, it also made it a bigger, easier target for enemies to hit. Players who wanted an easier time of it may actually have been well advised to avoid eating much of anything and just race through the depths as fast as they could.

I would criticize a few things in FlOw. Despite striving for that perfect DDA that'll match the game to the player, it still occasionally causes frustration, especially in the lower depths. When you fail an encounter with an enemy, your creature gets sent back up to a previous level, but this didn't feel like going easier on the player so much as just creating a setback. If you ate all the creatures on the previous level, they don't come back, so there's really nothing to do there except pluck up the courage to dive back down again.

Other than that, I felt the minimalist, explain-nothing approach of the game may be something of a double-edged sword. It may help immerse the player early on, but it also meant the player had no direct control of the difficulty. The natural inclination of eating things and diving deeper would gradually increase the difficulty, and the player won't realize this until after they start getting roughed up by enemies.

In talking about DDA, I'm reminded of an old favorite game of mine called God Hand. The goal of this action/adventure game was simply to beat up every bad guy that crossed your path. Many gamers loved (or hated) God Hand for its difficulty, but this game not only offered a traditional choice of difficulty level to the player, it also made an attempt at DDA at a time when such a thing was uncommon in such combat-centric games. The dynamic difficulty changes and the player-chosen difficulty changes worked hand-in-hand, or at least tried to.

God Hand employed a difficulty level system, consisting of four levels: Level 1, 2, 3, and Level DIE (which for most people was appropriately named). For every blow the player dealt on enemies, they'd slowly increase a gauge on the left of the screen. That same gauge would decrease if the player took a hit. When the gauge filled, it would increase the level by one, or if it was emptied by the player taking a beating, the level would drop.

So what did this level do (besides awarding more gold per enemy defeated)? Quite a lot. When the level increased, enemies seemingly got smarter. They reacted and counter-attacked more frequently, moved more quickly, and even the animations of their attacks sped up, meaning the player had to act and react more quickly. At Level 1, enemies would square off against the player one at a time. At higher levels, they ganged up on the player all at once, and also attacked the player from behind (which you couldn't see coming because of the camera's fixed over-the-shoulder view).

Now, the system wasn't fully automatic. Aside from the choice of Easy, Normal or Hard when starting a new game, for Normal and Easy modes, there was a move the player could do where they'd grovel at the enemies' feet, forcing the difficulty down to Level 1. They could also taunt enemies to make them briefly faster and more aggressive.

My point: the system wasn't perfect, but we should really see more of this in gaming. The DDA is great, but giving the player some direct control over the difficulty allows them to steer gameplay in the direction they want in case the DDA fails to fully engage them. A truly perfect DDA would be nice, but it's usually unrealistic to hope for.