再加上以上几点。在64位操作系统中使用ArrayList占用的内存是32位操作系统的2倍。同时，泛型列表list <T>占用的内存比ArrayList少得多。

例如，如果我们在32位中使用一个19MB的数组列表，那么在64位中它将占用39MB。但是如果你有一个32位的8MB的通用列表list <int>，它在64位只需要8.1MB，这与ArrayList相比相差了481%。

来源:数组列表与通用列表的基本类型和64位

简单的答案是，

数组列表是非泛型的

它是一个对象类型，因此您可以在其中存储任何数据类型。 可以在数组列表中存储任意值(值类型或引用类型)，例如string、int、employee和object。(注意,) 装箱和开箱将会发生。 不类型安全。 它更老。

列表是通用的

它是类型的类型，因此您可以在运行时指定T。 根据声明，您只能存储类型为T的值(字符串或int或employee或object)。(注意或) 装箱和开箱不会发生。 类型安全。 它更新。

例子:

ArrayList arrayList = new ArrayList();
List<int> list = new List<int>();

arrayList.Add(1);
arrayList.Add("String");
arrayList.Add(new object());


list.Add(1);
list.Add("String");                 // Compile-time Error
list.Add(new object());             // Compile-time Error

请阅读微软官方文档:https://blogs.msdn.microsoft.com/kcwalina/2005/09/23/system-collections-vs-system-collection-generic-and-system-collections-objectmodel/

注意:在理解它们的区别之前，您应该了解泛型:https://learn.microsoft.com/en-us/dotnet/csharp/programming-guide/generics/

我认为，ArrayList和List<T>之间的区别是:

List<T>, where T is value-type is faster than ArrayList. This is because List<T> avoids boxing/unboxing (where T is value-type). Many sources say - usually ArrayList used just for backward compatibility. (is not a real difference, but i think it is important note). Reflection is easier with nongeneric ArrayList then List<T> ArrayList has IsSynchronized property. So, It is easy to create and use syncronised ArrayList. I didin't found IsSynchronized property for List<T>. Also Keep in mind this type of synchronization is relatively inefficient, msdn): var arraylist = new ArrayList(); var arrayListSyncronized = ArrayList.Synchronized(arraylist Console.WriteLine($"syncronized {arraylist.IsSynchronized}"); Console.WriteLine($"syncronized {arrayListSyncronized.IsSynchronized}"); var list = new List<object>(); var listSyncronized = ArrayList.Synchronized(list); Console.WriteLine($"syncronized {list.IsSynchronized}");//error, no such prop Console.WriteLine($"syncronized {list.IsSynchronized}");//error, no such prop ArrayList has ArrayList.SyncRoot property which can be used for syncronisation (msdn). List<T> hasn't SyncRoot property, so in the following construction you need to use some object if you use List<T>: ArrayList myCollection = new ArrayList(); lock(myCollection.SyncRoot) // ofcourse you can use another object for this goal { foreach (object item in myCollection) { // ... } }

性能已经在几个答案中被提到作为一个区分因素，但是要解决“数组列表有多慢?”和“为什么整体上更慢?”，请看下面的内容。

当使用值类型作为元素时，ArrayList的性能会急剧下降。考虑简单地添加元素的情况。由于装箱正在进行——因为ArrayList的Add只接受对象参数——垃圾收集器被触发执行比List<T>更多的工作。

时差是多少?至少比List<T>慢几倍。看看在数组列表中添加10mil int值和List<T>的代码会发生什么:

在“均值”列(黄色突出显示)中，这是5倍的运行时间差异。还要注意为每个gc执行的垃圾收集数量的差异，用红色突出显示(每1000次运行没有gc)。

使用分析器快速查看发生了什么，这表明大部分时间都花在了gc上，而不是实际添加元素。下面的棕色条表示阻塞垃圾收集器活动:

我在这里https://mihai-albert.com/2019/12/15/boxing-performance-in-c-analysis-and-benchmark/写了一篇关于上述ArrayList场景的详细分析。

Jeffrey Richter的《CLR via c#》中也有类似的发现。第12章(泛型):

[…] When I compile and run a release build (with optimizations turned on) of this program on my computer, I get the following output. 00:00:01.6246959 (GCs= 6) List<Int32> 00:00:10.8555008 (GCs=390) ArrayList of Int32 00:00:02.5427847 (GCs= 4) List<String> 00:00:02.7944831 (GCs= 7) ArrayList of String The output here shows that using the generic List algorithm with the Int32 type is much faster than using the non-generic ArrayList algorithm with Int32. In fact, the difference is phenomenal: 1.6 seconds versus almost 11 seconds. That’s ~7 times faster! In addition, using a value type (Int32) with ArrayList causes a lot of boxing operations to occur, which results in 390 garbage collections. Meanwhile, the List algorithm required 6 garbage collections.

ArrayList是不同类型数据的集合，而List<>是其自身依赖的相似类型数据的集合。

To me its all about knowing your data. If I am continuing to expand my code on the basis of efficiency, I would have to choose the List option as a way of deciphering of my data w/o the unnecessary step of always wondering about types, especially 'Custom Types'. If the machine understands the difference and can determine on it's on what type of data I'm actually dealing with then why should I get in the way and waste time going thru the gyrations of 'IF THEN ELSE' determinations? My philosophy is to let the machine work for me instead of me working on the machine? Knowing the unique differences of different object code commands goes a long way in making your code as efficient.

汤姆•约翰逊 (一项…一个出口)