为什么内存使用2G的苹果手机比内存使用4G的安卓机更流畅?
这是在国外一家网站看到的,分析得不错,原文如下:
Why does the iPhone require less RAM than Android devices?
There are people who always fight when it comes to comparing the Android devices vs iPhone. Iphones are way more expensive than Android devices even with the same specifications. However, the iPhone works a lot smoother and faster than an Android device with the same amount of RAM. That means an iPhone with 2GB RAM works much faster than an Android device with 2GB RAM with same features.
If you compare the performance of iPhone 7 (2GB RAM) with Google Pixel (4GB RAM) or Samsung s7 (4GB RAM), You’ll see no difference in performance. iPhone 6 was the best performing phone in its time with just 1 GB of RAM. Now, the question is why does the iPhone require less RAM than Android devices? Why does the Android device need more RAM?
Why does the iPhone require less RAM than Android devices?

It involves many factors. Basically the Garbage collection, App Management, User Interface, and Hardware.
1. Garbage collection
In the case of Android, since it involves garbage collection it requires more memory. The Android apps use Java and as a result, Android does garbage collection. The problem with garbage collection is that memory usage grows until it’s collected, so there might be more memory allocated than necessary. That’s bad for devices with restricted memory and no option to swap.
When the garbage collector runs, it scans the heap to find memory that’s no longer being used, and that’s an expensive process, that slows down the device until is has completed.
The iPhone do not use Garbage Collection. It uses Automatic Reference Counting, which is an innovative way of managing Objective-C objects on iOS. It does away with explicit retain, release and autorelease messages, and it behaves the same on both platforms. Unlike garbage collection, ARC does not handle reference cycles automatically. This means that as long as there are “strong” references to an object, it will not be deallocated. Strong cross-references can accordingly create deadlocks and memory leaks. It is up to the developer to break cycles by using weak references.
Here is the full Difference between ARC vs Garbage Collection
2. User Interface
An Android user often complains about phone lag but the iOS UI runs so smooth. The main reason is that the iOS UI rendering happens in a separate thread with real-time priority but on Android, this happens in the main thread with normal priority. This means that other apps in Android can take over the processor resources and hurt basic UI interactions, translating into a noticeable lag.
3. App management
Android lets developers run processes in the background more freely than iOS. iOS kills any process when it thinks it doesn’t need to be running. Apps that play music in the background or similar can stay alive.
Android gives a lot of freedom to app developers that cause memory leaks or resource hogging background processes which means you need to be careful what you install on your Android device. Android has made some attemps to control this, such as killing apps that use too much CPU, but still, these problems can be noticed by the user.
The iOS puts a lot of limitations on the developers and what they can do causes them not to do much harm.
4. Hardware
Apple’s iOS is optimized to run on the hardware, which is also designed and tested by Apple, and manufactured by Apple’s manufacturing partners. Android doesn’t have that level of control because the software OS developers are separate from the hardware makers
为什么内存使用2G的苹果手机比内存使用4G的安卓机更流畅?的更多相关文章
- 程序最多能new多少内存(2G内存里要放程序的5大区,HeapAlloc比new要快多了,而且超过2G的时候会告诉你)
根据<Windows核心编程>得知:X86操作系统提供每个程序最多只有4G的虚拟内存,其中2G虚拟内存提供给系统用(具体用来干什么还待考察),还有2G的内存留给用户使用.那这2G内存能拿来 ...
- 升级win10,提示(RAM)内存不足2G的解决的方法,亲測可行
前两天升级win10,检測我内存不足2G,可我的电脑是8G的内存如图 百度,google了非常多方法,有些是两根内存,去掉一个就好了,但是我的就一根8G的,其它就没什么好的方法了,改biosCPU选项 ...
- Windows系统CPU内存网络性能统计第一篇 内存
最近翻出以前做过的Windows系统性能统计程序,这个程序可以统计系统中的CPU使用情况,内存使用情况以及网络流量.现在将其整理一下(共有三篇),希望对大家有所帮助. 目录如下: 1.<Wind ...
- 【嵌入式开发】裸机引导操作系统和ARM 内存操作 ( DRAM SRAM 类型 简介 | Logical Bank | 内存地址空间介绍 | 内存芯片连接方式 | 内存初始化 | 汇编代码示例 )
[嵌入式开发]ARM 内存操作 ( DRAM SRAM 类型 简介 | Logical Bank | 内存地址空间介绍 | 内存芯片连接方式 | 内存初始化 | 汇编代码示例 ) 一. 内存 ...
- linux内存(二)高端内存
来此网址 https://ilinuxkernel.com/?p=1013 Linux内核地址映射模型x86 CPU采用了段页式地址映射模型.进程代码中的地址为逻辑地址,经过段页式地址映射后,才真正访 ...
- 死磕内存篇 --- JAVA进程和linux内存间的大小关系
运行个JAVA 用sleep去hold住 package org.hjb.test; public class TestOnly { public static void main(String[] ...
- 【腾讯优测干货分享】如何降低App的待机内存(四)——进阶:内存原理
本文来自于腾讯优测公众号(wxutest),未经作者同意,请勿转载,原文地址:http://mp.weixin.qq.com/s/3FTPFvZRqyAQnU047kmWJQ 1.4进阶:内存原理 在 ...
- Linux内核笔记--内存管理之用户态进程内存分配
内核版本:linux-2.6.11 Linux在加载一个可执行程序的时候做了种种复杂的工作,内存分配是其中非常重要的一环,作为一个linux程序员必然会想要知道这个过程到底是怎么样的,内核源码会告诉你 ...
- JVM内存模型、指令重排、内存屏障概念解析
在高并发模型中,无是面对物理机SMP系统模型,还是面对像JVM的虚拟机多线程并发内存模型,指令重排(编译器.运行时)和内存屏障都是非常重要的概念,因此,搞清楚这些概念和原理很重要.否则,你很难搞清楚哪 ...
随机推荐
- (转)Spring boot——logback.xml 配置详解(三)<appender>
文章转载自:http://aub.iteye.com/blog/1101260,在此对作者的辛苦表示感谢! 1 appender <appender>是<configuration& ...
- jar包的一生
经常会头疼于一个jar包是如何制作的,包括maven的打包方式,springboot的打jar包的原理,jar包稍稍有错误就会完全无法运行.在网上折腾了很久终于有些思路和步骤,在这里做个笔记 本文大纲 ...
- 【SignalR学习系列】1. SignalR理论介绍
什么是SignalR? ASP.NET SignalR 是一个让 ASP.NET开发者可以简单地给自己的程序添加即时通讯功能的开发库.即时通讯功能可以直接从服务器端给在线的客户端发送数据,而不用等待客 ...
- Git版本控制系统之基本使用
最早是通过接触著名的开源社区Github了解到Git的,但一直没有系统学习过.这次下定决心从头到尾系统的学一学,也将学习过程记录于此,供大家批驳.本篇文章先从以下几个方面简单了解一下Git: Git的 ...
- Linux的nginx环境的vue 部署
1.使用WebStrom编程好的vue 进入到Terminal运行npm run build 2.使用WinSCP进入到nginx 目录配置文件下面,找到nginx.conf文件: ...
- ssl Diffie-Hellman弱密码问题
title: ssl Diffie-Hellman弱密码问题 date: 2017-08-16 16:41:55 categories: 网络安全 tags: ssl Diffie-Hellman - ...
- android 适配器 ArrayAdapter,SimpleAdapter的学习
今天认真看了下android适配器,学习了下它的使用方法. 一,ArrayAdapter ArrayAdapter 比较简单,只可以存放一行文本信息.下面是简单的实现 private ListView ...
- oracle常用视图介绍
oracle常用视图介绍---------------------2013/10/04(last updated) dba_开头: dba_users 数据库用户信息 dba_segm ...
- C/C++ 知识点---C语言关键字(32个)
C/C++ 知识点 1.C语言关键字(32个) <1>.基本数据类型 [5] void :声明函数无返回值或无参数,声明空类型指针 char :声明字符型变量 int :声明整形变量 fl ...
- 远景面试算法题——FolderSize
描述 文件被存储在磁盘上的时候,通常为cluster方式.每个cluster具有固定的大小,一个文件所消耗的空间量始终是cluster大小的整数倍.因此,如果cluster的大小为100字节,165字 ...