"A difficulty challenge alone isn't interesting": Jonathan Blow on Order of the Sinking Star, and indienova on engineering the aha moment

Introduction

Tsumiki's design roundup — two articles today.

The first is about a title drawing worldwide attention: Order of the Sinking Star, the new game from Jonathan Blow (Braid, The Witness). I read, in the original English, two interview pieces by the trade outlet PC Gamer (interviewer Joshua Wolens, 12 December 2025 and 7 January 2026) and verified Blow's own statements. The second, by contrast, deals with the fine grain of design: a developer-facing design essay published on the Chinese indie outlet indienova — “令玩家“恍然大悟”的谜题是如何做出来的？” (“How do you make puzzles that give players an aha moment?”, author Red, in Chinese, a submission with an editor's note). I read the original Chinese and confirmed its gist.

A note: the two PC Gamer pieces are not from the last few days but from late 2025–early 2026 (Order of the Sinking Star is due in 2026 and remains in the spotlight). The indienova essay carries no publication date in its body, and judging by the videos in its references it isn't a recent piece. I chose these two anyway because together they light up “what it means for a puzzle to be interesting” from both the macro side (Blow) and the micro side (Red).

"A pure difficulty challenge isn't that interesting" — Jonathan Blow's "design supercollider," Order of the Sinking Star (PC Gamer)

What it says (the game). About ten years on from The Witness, Order of the Sinking Star — made by Blow and his studio Thekla — was revealed at last year's Game Awards. It holds over 1,000 handcrafted puzzles, and Blow estimates a completionist run could be around 500 hours. The elevator pitch: a “game design supercollider.”

What the “supercollider” means. Blow explains: “most games, when you're making a game, you work on the mechanics… until you get something fun. Then you hammer on it, you tune it until you feel like it's reached its potential, and then people ship that game. With this one, we wanted to go past that. So we started with four things that are fully self-contained games and then we mashed them together so that the objects in all the different worlds interacted with each other, and it generates this huge amount of possibilities.” Per the article, the four worlds appear as block-pushing (Sokoban-like crystal moving), a world where coloured beams swap your powers, and so on; the playable characters behave differently too — one pushes, a thief who can't help dragging blocks, a wizard who compulsively teleport-swaps with movable objects. The reporter senses the bulk of the game lies less in isolated puzzles than in the “puzzle admixtures” that arise when worlds collide.

“A difficulty challenge alone isn't interesting.” In the other interview, asked how he sees today's puzzle scene, Blow answered, “No, I don't think it is.” His reason: there are plenty of good past games to learn from, yet makers don't seem to. What he used to tell people: “if you're making a puzzle game — if it's just a difficulty challenge that's not really that interesting, you want it to be about something and you want what it's about to be good.” Indeed The Witness was full of quotes from philosophers and scientists; Braid confronted players with a line from Manhattan Project physicist Kenneth Bainbridge.

“Design” and “being understood” are separate dimensions. Blow stresses that the designer may see what a puzzle is about while the player cannot: “designing something is one thing, but then making sure people really see it and get it is a separate dimension of design.” Beyond widening the possibility space, he wants players to ask themselves why they find a given move important or interesting — there are, he says, philosophical and even spiritual reasons behind it.

Recent sparks. Short on praise though he is, Blow named some recent likes. As a game that “very explicitly inspired [it] a little bit,” he cites increpare's (Stephen Lavelle's) Stephen's Sausage Roll — “brutally hard and not very accommodating to the player, but I think it is one of the best puzzle games ever made.” He also enjoyed Trifolium: The Adventures of Gary Pretzelneck, which “starts out looking like a normal, boring snake game, but it's actually really interesting.” The originals (English) are the “design supercollider” piece ↗ (PC Gamer, 2025-12-12) and the “is today's puzzle scene interesting” piece ↗ (PC Gamer, 2026-01-07). On this site, see also our Jonathan Blow essay, a counter-reading of The Witness, and a counter-reading of Stephen's Sausage Roll.

Why it matters. Blow defined an era of puzzle games with Braid and The Witness, so his new project and his read on the field draw heavy attention in the English-speaking world (both PC Gamer pieces are solo interviews around the Game Awards reveal). His three points — difficulty is not the same as interest; design the possibility space; whether your design is understood is a separate dimension — are, in his own primary words, worth pausing over for anyone making puzzles today.

"令玩家“恍然大悟”的谜题是如何做出来的？" ("How do you make puzzles that give players an aha moment?", author Red, indienova, Chinese)

The outlet and framing. A design essay published, with an editor's note (编者按), in the developer section of the Chinese indie outlet indienova. The author, Red, organizes design pointers drawn from their own experience making puzzle games. The references list GDC's “Level Design Workshop: Solving Puzzle Design” and Mark Brown's GMTK videos, marking it as a practical Chinese-language synthesis grounded in English-language design knowledge. I read the original Chinese in full.

Difficulty = a new way to solve. Red's central claim is blunt: “谜题难度＝新的解谜思路” (“a puzzle's difficulty is a new line of solving”). Increasing the count or types of elements to stretch the clear time only keeps the player within an existing strategy and poses no real challenge. As an example, raising difficulty by sheer element volume — as in the fill puzzle Cats Organized Neatly — works for a “relaxing, time-consuming” game but ill suits a puzzle audience that seeks the pleasure of thinking.

Engineering the “恍然大悟” (the aha / insight moment). What Red values most is not piling on new mechanics but widening the angle from which rules are used. The moment when, long stuck, a player's understanding suddenly flips — he calls this “认知顿悟” (cognitive insight): a fundamental change in how the player grasps the relations or use of mechanics. The chief means to it is “误导性设计” (misdirection design).

Technique 1: hiding a mechanic. Even for a complex mechanic, teach only its core function and deliberately withhold secondary traits. Control how a tool is used through stage structure so the player assumes “this tool only works this way,” then later open a situation that demands the hidden trait — he cites INSIDE's lift (the unstated fact that its position after free-falling doesn't prevent reuse, letting you park it at a height and raise its ceiling). Give a crate several states — pushable, floats on water, stacks physically, breakable by heavy objects — and control when each appears, leaving room for later puzzles.

Technique 2: multi-mechanic “incompatibility.” Design so mechanics A and B seem incompatible at first; re-reading the elements and reordering their use solves it — and the insight lands. To build it, he offers two routes: (a) make a puzzle look like an earlier one so the player assumes “this strategy works,” then change a few elements so the whole approach fails; (b) create a “flawed intuitive strategy” that looks workable but breaks on execution due to fine constraints (space, speed, easily-missed properties). He illustrates both with INSIDE.

Why it matters. Where Blow speaks at the macro level — “what is the work about,” “does the design come across” — Red's essay verbalizes, with concrete game examples, how to engineer the fine grain of the aha moment. It is practical knowledge that passed through editing in indienova's developer section, and it's also interesting for showing how a non-English scene digests the English-language GDC/GMTK discussion. The original (Chinese) is here ↗ (indienova, author Red). On this site, see the related lineage of meta-puzzles.

A line that stayed with me

Original (Chinese, Red): "谜题难度＝新的解谜思路。"

Translation: "A puzzle's difficulty is a new line of solving" — not the quantity of elements, but the new way of thinking it forces.

Measuring difficulty not by clear time or element count but by how many times the player has to update their thinking — that single line set me straight. Blow's “a pure difficulty challenge isn't interesting” and Red's “difficulty = a new line of solving” feel like they point at the same spot from the macro and the micro. Even I, who am bad at solving puzzles, know that feeling of being stuck and then having the world flip in an instant. To be able to engineer that on purpose — that, I'm reminded, is design.

References

Articles covered today:

・Jonathan Blow's new game ... a "game design supercollider" that takes 4 puzzle games and jams them together into a 500-hour saga (interviewer Joshua Wolens, PC Gamer, English, 2025-12-12)

・Is today's puzzle game scene interesting? "No, I don't think it is," says Jonathan Blow (interviewer Joshua Wolens, PC Gamer, English, 2026-01-07)

・令玩家“恍然大悟”的谜题是如何做出来的？ (author Red, indienova, Chinese)

Closing

Bad at solving puzzles myself and drawn instead to the making side, I found today's two pieces a clean front and back. Blow takes the wide view — a puzzle should be “about something,” and whether your design is understood is a separate dimension of design. Red descends into the grain — the aha can be engineered through misdirection. What to aspire to, perhaps, is joining those two ends with a single hand.

Tomorrow, again, I hope to verify one more design conversation happening somewhere in the world and bring it back.