Tesla-> Fermi (550Ti) -> Kepler(680) -> Maxwell (750Ti) -> Volta(was Pascal)
Pascal GPU
Pascal (from French mathematician Blaise Pascal) is Maxwell successor. In this news, we learnt that Volta was the post-Maxwell architecture, but it seems that Pascal is the new official name. One of the main feature of the Pascal architecture is the 3D memory or stacked DRAM that should provide terabyte bandwidth.
Update (2014.03.26): according to techreport, Volta is the successor of Pascal:
Turns out Volta remains on the roadmap, but it comes after Pascal and will evidently include more extensive changes to Nvidia’s core GPU architecture.
Maxwell GPU architecture (post-Kepler) is not available yet, but NVIDIA is already talking, at GTC 2013, about post-Maxwell architecture called Volta that features 3D stacked DRAM in order to provide terabyte bandwidth.
According to the guys at videocardz, NVIDIA Maxwell GM107 features 128 CUDA cores per SMX (streaming multiprocessor) against 192 for a Kepler-based GPU. The GeForce GTX 750 Ti has 5 SMX (640 cores) while the GTX 750 packs 512 cores in 4 SMX. The L2 cache has also received some changes: in Kepler the L2 cache has 256KB while on Maxwell, the cache has 2MB!
In the power consumption side, the GM107 has a TDP of 60W. And to finish, the GM107 will offer at least the same level (if not better) of performance than the GF100, the GPU behind the old good GTX 480.
Update (2013.03.20)
NVIDIA has just launched the first graphics cards based on Maxwell, NVIDIA’s new GPU architecture. TheGeForce GTX 750 Ti and GeForce GTX 750 are both powered by a GM107, a cut down version of an upcoming flagship (GM110 ?) GPU. The Maxwell architecture is focused on power efficiency:
Kepler SMXs (streaming multiprocessors) are now replaced by Maxwell SMMs. A SMM packs 128 CUDA cores(SMX = 192 cores). The L2 cache size has been increased from 256KB in Kepler to 2048KB in Maxwell.
The new GTX 750 and GTX 750 Ti targets the old GeForce GTX 650 Ti and the recently launched AMD Radeon R7 265.
GeForce GTX 750 Ti specifications
- GPU: GM107 @ 1020MHz (base) / 1085MHz (boost), 28nm
- Cores: 640 (5 SMMs)
- Texture units: 40
- ROPs: 16
- Memory: 2048MB @ 5.4GHz GDDR5, 128-bit
- TDP: 60W
- Price: USD $149
GeForce GTX 750 specifications
- GPU: GM107 @ 1020MHz (base) / 1085MHz (boost), 28nm
- Cores: 512 (4 SMMs)
- Texture units: 32
- ROPs: 16
- Memory: 1024MB @ 5.0GHz GDDR5, 128-bit
- TDP: 55W
- Price: USD $119
Related Articles and Reviews
- 5 Things You Should Know About the New Maxwell GPU Architecture
- NVIDIA GeForce GTX 750 Ti whitepaper
- Nvidia lance les GeForce GTX 750 Ti / 750 Maxwell
- NVIDIA GeForce GTX 750 Ti Review – Maxwell Architecture debuts at $150
- The NVIDIA GeForce GTX 750 Ti and GTX 750 Review: Maxwell Makes Its Move
- NVIDIA GeForce GTX 750 Ti 2 GB
NVIDIA GeForce GTX 750 Ti global performance
Tesla-> Fermi (550Ti) -> Kepler(680) -> Maxwell (750Ti) -> Volta(was Pascal)的更多相关文章
- CUDA ---- GPU架构(Fermi、Kepler)
GPU架构 SM(Streaming Multiprocessors)是GPU架构中非常重要的部分,GPU硬件的并行性就是由SM决定的. 以Fermi架构为例,其包含以下主要组成部分: CUDA co ...
- NVIDIA GPU架构与原理分析(一)——GPU简介与主流Fermi、Kepler架构GPU概述
1 GPU简介 图形处理单元GPU英文全称Graphic Processing Unit,GPU是相对于CPU的一个概念,NVIDIA公司在1999年发布GeForce256图形处理芯片时首先提出GP ...
- AI芯片
课程作业,正好自己也在学深度学习,正好有所帮助,做了深度学习的AI芯片调研,时间比较短,写的比较仓促,大家随便看看 近年来,深度学习技术,如卷积神经网络(CNN).递归神经网络(RNN)等,成为计算机 ...
- CUDA程序设计(一)
为什么需要GPU 几年前我启动并主导了一个项目,当时还在谷歌,这个项目叫谷歌大脑.该项目利用谷歌的计算基础设施来构建神经网络. 规模大概比之前的神经网络扩大了一百倍,我们的方法是用约一千台电脑.这确实 ...
- CSDN日报20170312——《成功人士的七个习惯》
[程序人生]成功人士的七个习惯 作者:liumiaocn 所谓温故而知新.重读往往会带来新的收获,即使没有新的收获,可以带来新的思考也是非常不错.在个人成长的历程中.习惯所扮演的角色不言而喻,Step ...
- 显卡,显卡驱动,nvcc, cuda driver,cudatoolkit,cudnn到底是什么?
在使用深度学习框架的过程中一定会经常碰到这些东西,虽然anaconda有时会帮助我们自动地解决这些设置,但是有些特殊的库却还是需要我们手动配置环境,但是我对标题上的这些名词其实并不十分清楚,所以老是被 ...
- CUDA Programming Guide 学习笔记
CUDA学习笔记 GPU架构 GPU围绕流式多处理器(SM)的可扩展阵列搭建,每个GPU有多个SM,每个SM支持数百个线程并发执行.目前Nvidia推出了6种GPU架构(按时间顺序,详见下图):Fer ...
- NVIDIA FFmpeg 转码技术分析
NVIDIA FFmpeg 转码技术分析 所有从 Kepler 一代开始的 NVIDIA GPUs 都支持完全加速的硬件视频编码,而从费米一代开始的所有 GPUs 都支持完全加速的硬件视频解码.截至 ...
- Life of a triangle - NVIDIA's logical pipeline
Home GameWorks Blog Life of a triangle - NVIDIA's logical pipeline Life of a triangle - NVIDIA's l ...
随机推荐
- C++里的静态成员函数不能用const的原因
From http://blog.csdn.net/freeboy1015/article/details/7634950 static在c++中的第五种含义:用static修饰不访问非静态数 ...
- Material Design入门(二)
本文主要包括以下内容 侧滑菜单DrawerLayout实现 CardView实现 DrawerLayout介绍 drawerLayout是Support Library包中实现了侧滑菜单效果的控件,可 ...
- Linux网络编程入门 (转载)
(一)Linux网络编程--网络知识介绍 Linux网络编程--网络知识介绍客户端和服务端 网络程序和普通的程序有一个最大的区别是网络程序是由两个部分组成的--客户端和服务器端. 客户 ...
- ubuntu用户添加adduser, useradd
ubuntu和windows一样,可以任意创建或者删除新的用户,windows下比较简单,ubuntu下需要使用命令,不过操作起来不是很繁琐,所以我尽量写的详细一些. 如何创建ubuntu新用户? ...
- kindEditort图片自动上传
参考:http://www.cnblogs.com/jaxu/p/3824583.html (赞一个)
- Android Design
在android的讲解 Android Design
- Android权限设置android.permission
android.permission.ACCESS_NETWORK_STATE: 允许程序访问有关GSM网络信息(Allows applications to access information a ...
- 网站构建学习笔记(0)——基本概念了解及资源学习(copy自w3school)
一.学习方面 1.WWW - 万维网 什么是 WWW? WWW 指万维网(World Wide Web) 万维网常被称为Web Web 是由遍布全球的计算机所组成的网络 所有 Web 中的计算机都可以 ...
- SQL Server:查询当前服务器有多少连接请求
有时DBA需要检查当前服务器有多少连接请求,以及连接请求的登录名,客户端版本,主机名,程序名等等之类的信息,我们就可以对服务器的连接状况有所了解,防止不明用户使用. SQL Server本身提供了这么 ...
- android 自定义相机画面倒立解决方案
有部分手机的影像是倒立的,如何解决这个问题呢? 请看下面 public static void setCameraDisplayOrientation(Activity activity, int c ...