Sekiro, after 80 hours

*** This post contains game spoilers ***

I’ve dumped most of my free time into Sekiro over the past month. Here are some thoughts. Apologies for the sprawling wall of text – “I didn’t have time to write a short letter, so I wrote a long one instead.”

Combat evolution

Dark Souls, Bloodborne and Sekiro represent 3 different iterations of the same foundational design principles of From Software. Each iteration after Dark Souls comes with significant additions but also subtractions. The end result is each product is based on a “proven” formula but also with enough differentiation to not just be a “bigger, badder” sequel.

Taking combat as an example, the higher level principles of SoulSekiroBorne combat in my opinion are the following:

  • The trifecta with regards to input – output:
    • Actions have consequence. The player is strongly encouraged to be thoughtful about every input, the wrong action at the wrong time will be severely punished, while right choices are highly rewarded
    • “Hesitation is defeat.” This Sekiro NPC quote perfectly sums up the real-time, split-second decision making required
    • Responsive, precise controls, in return for the strict requirements on player input (note the difference between responsive controls vs fast actions – this short video explains it visually)

With regards to enemies:

  • Enemies need to be respected and studied. Even the most basic enemy can have lethal combos or overwhelm you in numbers. On the other hand, even the most fearsome looking enemy has clearly exploitable weaknesses that are waiting for you to learn
  • Appearances can be deceiving; unexpected variations in animation timing create depth. The large, slow enemy that has a surprisingly quick slash. The extremely slow attack that punishes you for taking action too early. The suspense of different wind-up animations, some which have long pauses. This is also why each enemy must be studied and practice is required to develop muscle memory
  • Your instinct / fear works against you. The safest place when you face a huge menacing boss is often up front and personal. Instead of strafing and waiting for openings to whittle down your foe, you might find it better to go as aggressive as you can

Zooming in on the specific combat mechanics – Dark Souls at its core is the following:

  • Resources: HP, stamina
  • Defense: dodge roll / back step, block and parry
  • Offense: light and heavy basic attacks, kick, leap attack, backstab; large variety of weapons

(On paper this is not a particularly complex or exciting ruleset – for example, there are no special abilities and flashy combos. The generally slow animation timings may even make it look boring; but it provides enough space such that when you add in weapons variety the sum is a very rich system.)

Bloodborne‘s iteration:

  • Resources: Dark Souls + orange health (window to regain lost health by doing damage)
  • Defense: Dark Souls – block
  • Offense: Dark Souls + weapon transformation; smaller weapons catalog but much more unique weapons

These changes cohesively supported Bloodborne‘s design goals of offering faster-paced, higher risk combat with more offensive combo variety (in conjunction, character movement and animation times were also sped up).

Sekiro‘s iteration is a more significant deviation from the base formula, and all of these changes can be viewed in relation to how they serve the overarching design goal of simulating a deadly sword duel with a constant flurry of swords clashing:

  • Resources: Dark Souls – stamina, + posture
    • Removing a generic resource stamina (consumed both in offense and defense) with a defensive resource posture strongly encourages aggression
    • Posture is also a win condition in and of itself, and its slow recovery mechanic again adds urgency and aggression
  • Defense: Dark Souls + jump + mikiri counter; parry evolved into deflect
    • Deflection is a pivotal rhythm game that consumes most of the player’s attention, as how good you are at deflecting directly impacts how fast you are expending your own posture
    • The addition of jump (and double jump) as a regular button is significant, and necessary for thematic reasons as well
    • Deflect, jump and mikiri counter form a rock / paper / scissors mini-game in relation to enemies’ normal attack, sweeping attack and thrust attack
    • Dodge becomes extremely situational, as generally speaking a perfect deflect is strictly advantageous to dodging (both avoids damage, but you are “doing damage” to the enemy’s posture through perfect deflect)
  • Offense: single katana weapon, special attacks (combat arts), variety of prosthetic tools; grappling hook, ninjutsu skills (spells cast during backstabs), and some very specific situational attacks (lightning reversal)
    • While it looks like a lot due to the special systems added, they actually occupy very little design space (think of them as heavily situational or resource-gated supporting functions that unlock different gameplay patterns). On offense Sekiro has a tight design budget – there’s only one weapon and one basic attack button. In my view the bulk of the design budget – the amount of complexity that a player can handle in the combat system – went into the intricate defensive choices and posture / deflection as the critical interaction, while offense is largely “spam basic attacks at the right time”

It’s worth noting that Sekiro is a much more restrictive game in combat – players largely cannot opt-out of the posture / deflection system. Enemies, especially later ones, do not offer many vulnerabilities if you don’t engage in deflection / counter-attack, and there are several mini-boss fights that are primarily skill-checks on deflection. Additionally, the “no-charm” de-buff unlocked after the first play-through further emphasizes proper deflections, as blocking now costs chip damage – I suspect this was “the way the game is meant to be played” but it was found to be too difficult through play-testing, thus it was converted into an opt-in difficulty feature.

Tight scope management

As a game producer it’s hard not to think about scope when I play games. As far as I can tell, Sekiro exercised really tight scope management in a very thematically resonant way.

I’m of course, first of all thinking about the level design. Ashina Castle is an intricately designed location that serves as the hub of the game both geographically and thematically. The general play-through has the player reaching Ashina Castle, going to a connected location (e.g. Senpou Temple), and then coming back to Ashina Castle which has gone through a transformation due to plot triggers. In a full play-through the player would play the Ashina Castle “level” 3 full times, with different enemies each time. This is in contrast to most AAA games where an elaborate level is only experienced once and never revisited (much like an expensive set-piece in a Hollywood blockbuster).

It’s not limited to Ashina Castle – many locations, including boss-fight levels, are used twice, in what certainly feels like a nod to the game’s byline (Shadows Die Twice) and theme. Also, the game’s gating of the water-diving ability also encourages re-exploration of earlier areas.

Similarly, the game is very frugal in its roster of enemies. A quick and not scientific comparison of the list of enemies in Bloodborne vs Sekiro shows a 68:45 count (this doesn’t include bosses), which is a 34% decrease. There are some enemies that you fight throughout the whole game (and consistently give you a headache with their tight moveset), such as the Lone Shadow Ninja.

Some of this is thematic: in a game about sword-dueling in Sengoku-era Japan (even with some fantastical elements), there are tighter constraints around what makes a good enemy1. Some of this is gameplay driven – a single Lone Shadow Ninja can easily take you a good minute to defeat, and generally if you face more than a couple of enemies at once you will be overwhelmed, thus there’s a much higher emphasis on quality vs quantity.

Additionally, multiplayer was entirely removed. While Soulsborne multiplayer was optional (I’m not a PVP player in those games, since I never “git gud” enough), they are an important, integral part of the PVE design (covenants, co-op play etc.). For Sekiro, multiplayer doesn’t fit with the story (you are playing Sekiro, not a player-created character) nor the combat scope (all players use the same katana). Those are not unsolvable problems – but the single-player only decision must have been made very early in planning. The trade-offs of this decision are well discussed here, with the below benefits mentioned:

  • Bosses can be designed strictly for single player. It also means players must beat the game on their own (no summoning help)
  • Level design don’t have to take into account multiple players and the emergent interactions (especially in a PVP context). This is a substantial reduction in development scope
  • Players can pause the game at any time – beyond a general benefit to all players, this actually unlocks some interesting gameplay for a small segment of players, by enabling them to hotswap abilities in the middle of combat

A few blemishes

My biggest complaint with the game is around the camera system. At a glance, Sekiro shares the same camera system with other From Software games. This is a proven system that generally works well in supporting a controller-based 3rd-person action game.

A quick video comparison of Sekiro and Nioh (another samurai / shinobi themed game with some Souls-like elements) in two somewhat similar boss fights:

Sekiro’s camera is closer, and more fixed in position behind the player character; Nioh’s camera is more pulled back and more fixed to the world (the player character can move in a direction for a short distance before the camera starts to track). This can be summarized as an action camera vs a tactical camera.

Sekiro’s camera choice fits strongly with its gameplay – the sense of danger and feeling of intensity are superior. It does however have severe limitations, most notably in tight spaces where players often are fighting the camera:

This fight is the most flagrant case, where the tight arena and the extremely nimble enemy combine to make the camera behave erratically. (In this video the player does a great job of avoiding tight corners that causes the camera to break down and lose target lock, which is an even greater offense.)

The camera’s limitations also has a cost on boss design. In a couple of boss fights, the bosses have aerial movement that cause them to go out of camera and breaks target lock – these seem intentionally designed and force the player to frantically react to reposition the camera. I don’t find these to be rewarding skill-tests and cause more frustration than satisfaction. More generally, when you are in close proximity with the boss (especially those with a larger character model), it can be hard to read the enemy animation because it’s blocked or out of camera.

Aside from the camera, the game slightly suffers from bad early game difficulty tuning. It’s not that the early game enemies are hard; playing Normal Game+ it’s obvious how vulnerable / slow they are. It’s that in the first play-through the player character is severely deprived of HP and heals initially, which make mistakes in early sections overly punishing.

My other gripe is some optional mini-bosses (Headless and Shichimen Warriors) which require a consumable item (Divine Confetti) to effectively fight (they are almost immune to physical damage). It’s an unnecessary annoyance especially as this item is very rare until late in the first play-through.

Closing thoughts

Overall Sekiro is an excellent game. It’s a coin-toss between Sekiro and Bloodborne for which is my favorite From Software game. It’s the work of a studio that has been honing their craft over a decade in a sub-genre they created. And putting my game producer hat on, that’s a serious competitive advantage – not just in terms of know-how of the design space and sensibly managing scope, but also likely mature tools, workflows and pipelines, which collectively make developers’ lives a bit easier and the quality better.

Zooming out, that is the state of the industry we work in – in every genre you could probably name a studio that has been successfully tackling it for over 10 years. So for the upstarts and new entrants you really have to think about not only what you bring to the table (to players), but also how you are bridging the development learning curve versus the incumbents.

  1. Sekiro isn’t afraid to throw in a good joke here – one of the toughest regular enemies is a dual-wielding white monkey. It’s harder than some mini-bosses

1 thought on “Sekiro, after 80 hours”

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.