I need to talk about how this author writes dialogue because the voice work here is genuinely impressive, and Ye Liuxi in particular is a gift to any writer who's ever struggled to make a character feel distinct.
Look at how each character sounds completely different when they speak. Hui Ba? All aggression and bravado. 'Who's playing tricks!' 'I've shed blood and lost men, asking me to give this up to you, I'm not happy.' He sounds like every粗人 boss in every Chinese thriller ever made—and that's intentional. He's supposed to be the loud one who talks big and dies first. His dialogue tells you exactly who he is: someone who measures his worth in blood spilled and threats made.
Huoya is the opposite. Hesitant, stammering, constantly deferring. 'Should we go back? What if something else happens...' He's the voice of the everyman, the guy who only signed up for this because someone else promised him money. His dialogue is full of questions and half-finished thoughts because he never actually thinks anything through.
But Ye Liuxi—oh man. Her lines are razor-sharp and completely devoid of wasted words. 'You're overthinking. I'm just here to watch the excitement.' That's it. She shows up, says three sentences, and already dominates the scene. No explanation, no justification, just pure, unnerving confidence. And then she drops the death preference line and just... goes back to examining a coffin like nothing happened.
Chang Dong's internal monologue provides this great counterpoint though. His narration is thoughtful, analytical, occasionally self-deprecating. He's the reader surrogate, the one trying to make sense of things logically while everyone around him is either running or dead. When he says 'Aren't you bothered by death at all?' it's not just a question for Ye Liuxi—it's a question he's asking himself, and the reader along with him.
The author also does something really clever with the dialogue rhythms. Long, flowing descriptions from Chang Dong. Sharp, clipped exchanges from Ye Liuxi. Frantic, overlapping panic from the background characters. It creates this almost cinematic quality where you can HEAR each character's voice without needing tags.