PUEC 2024 Results Are Out
2024年PUEC出分，准备看邮箱查成绩了！
After what felt like a long wait, PUEC 2024 scores are finally out. Check your inboxes. A few notes on this year's competition before I get to the more pressing matter of having spent much of August playing Black Myth: Wukong.
The Problem Set
This year's problems were genuinely interesting — high in open-endedness, with the final section having no single correct answer. Judges awarded full marks for any physically reasonable response. In practice, this means students who engaged seriously with the final questions and gave a coherent argument should have collected those points even without hitting the "official" answer.
AI in the Competition
PUEC is now dealing with the same AI question every science competition faces. Roughly the first 40% of this year's problem set is solvable with good AI assistance, including the graphing sections (Python, MATLAB, Mathematica all work). The judging panel did not conduct formal AI checks this cycle, and the observation was that students who genuinely didn't understand the problem couldn't get useful output from AI either — the tool amplifies competence, it doesn't replace it.
The top-scoring papers were self-evidently done by the students themselves. The panel discussed this and decided not to make AI policy a central issue for 2024, but expect clearer guidelines in 2025.
Banded Awards
A meaningful change from previous years: the competition stopped splitting hairs at the margin. Rather than ranking students to 0.5-point precision and issuing distinct 1st, 2nd, and 3rd place awards within tiers, results are now reported by band. If you fall within a band, you receive that band's award — no finer distinction. The rationale is sound: encouraging participation and discouraging the demoralising experience of missing a higher award by a rounding difference.
If you are interested in competing in 2025 and need teammates, templates, or preparation support, get in touch early.
A Word on August
The plan for August was to finish the BPHO Section 2 manuscript. By the 20th I had about 220 pages done and was tracking toward a final 300+. Then Black Myth: Wukong launched, and priorities shifted accordingly.
The game genuinely earned the time. It does something from scratch that China's game industry had not done before at this scale — a full AAA production with real craft in every scene. The pricing (268 RMB, roughly half the Japanese equivalent) matches the intent. The Shanxi-inflected world of the second chapter hit personally; the region I grew up near is full of headless statues in cave temples, and seeing them rendered this way landed differently than I expected.
The critiques are fair where they land — some combat patterns borrow too directly from FromSoftware and Monster Hunter without reinterpreting them — but the overall experience is one of the most enjoyable things I played this year. BPHO Section 2 resumes in September. First week, definitely.
Post-game, I returned to admissions consultations for the start of school. The consistent observation: families who have done real research and know the landscape are much more productive to work with. Those who are starting from zero tend to stall at the trust-building stage before the actual work can begin. Something to write about properly, eventually.