Indeed. Thank you!
When I determine my loot it uses damage + calamityDamage + luck + gem-luck + other modifiers to determine value dropped and then fills it with value of ores and amounts of them, plus luck can cause overfills and lucky hits on top of that. It's kinda overkill for a game like this. But basically DP is way more accurate, but slower, which in my case wasn't even remotely necessary. Let me just paste this google AI overview because it will explain it way better than I could haha. (Doesn't allow me to paste picture or long text here so posting in reply)
A DP (dynamic programming) table stores exact, optimal solutions to subproblems to solve a larger problem, guaranteeing an optimal solution but potentially requiring large amounts of time and memory. A heuristic is a shortcut or rule-of-thumb approach that finds a good, but not necessarily optimal, solution quickly. DP can be combined with heuristics to make the problem more manageable, such as using a heuristic to prune states in the DP table, or by creating a heuristic based on the structure of the DP solution.
DP table
- Purpose: To store and reuse the exact solutions of overlapping subproblems.
- Method: A step-by-step, exhaustive procedure that guarantees the optimal solution for the problem. It builds the solution from the bottom up.
- Pros: Guarantees the best possible solution.
- Cons: Can be very time-consuming and computationally expensive, especially for large problems. The number of states can grow impractically large.
- Use case: When finding the absolute best solution is critical and computational resources are not a constraint.
Heuristic flow
- Purpose: To find a good solution quickly, especially for complex problems where an exact solution is infeasible.
- Method: Uses a shortcut or "rule-of-thumb" strategy to approximate a solution. It doesn't guarantee optimality.
- Pros: Much faster and requires fewer computational resources than DP.
- Cons: The solution may be suboptimal or inaccurate.
- Use case: When a fast answer is needed and a "good enough" solution is acceptable, such as in real-time systems or large-scale optimization problems.