Slay the Spire

The comment sample skews heavily technical (HackerNews/Lemmy threads reacting to a developer post on RNG design), revealing an unusually engineering-literate player base. Users dissect seeded vs unseeded runs, criticize .NET's Random implementation as a flawed LCG, and debate PCG or hash-based (sha256) alternatives; they confirm the devs' stated design goal that two players on the same seed should get a similar, 'fair' experience by isolating map/card-reward/shop RNG from combat RNG streams. One user flags a regression: the follow-up passes seeds through a 32-bit hash, so only ~4 billion seeds exist (vs 64-bit in the first game), capping 'high roll' seeded runs. Others confirm Spire 1's desktop and mobile seeds diverged because each platform's standard-library PRNG differed. Difficulty is the other dominant theme: one player logged 40 hours / 40 attempts with only 4 wins (first win took ~25 tries on Standard) and hadn't touched Ascension; veterans counter that non-ascension runs are 'pretty close to guaranteeing' a winnable run, and prefer higher Ascension because low difficulty erodes decision relevance. Mechanical originality is praised relative to Dominion (Innate, Retain, Ethereal keywords, Watcher stances, Defect orbs), deck-dilution tradeoffs are debated (enemies shuffle curse-style cards in on higher difficulties), and there is visible community anxiety over the Unity engine after its retroactive pricing change ('they should honestly just move their engine').

No aggregate ratings were found for this product during the last harvest.