Version: 2021.2
public static Random.State state ;

描述

Gets or sets the full internal state of the random number generator.

This property can be used to save and restore a previously saved state of the random number generator. Note that state is serializable, so that determinism can be preserved across sessions. Determinism is an important trait in many scenarios, such as multiplayer games, reproducible simulations, and unit testing.

Generator state can be (re)-initialized in two ways:
Call InitState with a simple integer "seed". This is a one-way operation and cannot be retrieved. Setting state using a State previously retrieved from this same property. This type cannot be constructed by the user. See the following example for an explanation of how these work.

using UnityEngine;

public class ExampleClass : MonoBehaviour { void Start() { const int initialSeed = 1234;

Random.InitState(initialSeed); // cannot be retrieved

PrintRandom("Step 1"); PrintRandom("Step 2");

Random.State stateBeforeStep3 = Random.state; // can be serialized

PrintRandom("Step 3"); PrintRandom("Step 4");

Random.state = stateBeforeStep3;

PrintRandom("Step 5"); PrintRandom("Step 6");

Random.InitState(initialSeed);

PrintRandom("Step 7"); PrintRandom("Step 8"); }

static void PrintRandom(string label) => Debug.Log($"{label} - RandomValue {Random.Range(0, 100)}"); }

/* Output:

Step 1 - RandomValue 38 Step 2 - RandomValue 76 Step 3 - RandomValue 69 Step 4 - RandomValue 11 Step 5 - RandomValue 69 Step 6 - RandomValue 11 Step 7 - RandomValue 38 Step 8 - RandomValue 76 */

The values from step 5 and 6 will be equal to those from step 3 and 4 because the internal state of the generator was restored to what we saved in stateBeforeStep3. Also, the values from step 7 and 8 will be equal to those from step 1 and 2 because we are resetting the generator state with initialSeed via InitState, which leaves the generator in the exact same state as right before step 1.