Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
branch: master
Fetching contributors…

Cannot retrieve contributors at this time

66 lines (54 sloc) 3.044 kB
<html>
<head>
<title>CIS 565: GPU Programming and Architecture. Fall 2012.</title>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<style type="text/css">
body {
font-family:"Helvetica";
}
</style>
</head>
<body>
<table border="0" summary="" width="800">
<tr align="center">
<td>
<h1>CIS 565: GPU Programming and Architecture</h1>
<p align="center">
Fall 2012
</p>
<p align="center">
University of Pennsylvania<br />
<a href="http://cg.cis.upenn.edu/index.html">Computer Graphics @ Penn</a><br/><br/>
<img src="images/banner.png" width="640" height="120" alt="Student Projects" /><br />
</p>
</td>
<td valign="top">
<img src ="images/shield.gif" width="63" height="70" alt="University of Pennsylvania" />
</td>
</tr>
<tr align="center">
<td>
<br />
<a href="index.html">Course Overview</a> &bullet;
FAQ &bullet;
<a href="schedule.html">Schedule</a> &bullet;
<a href="studentwork.html">Student Work</a> &bullet;
<a href="http://cis565-spring-2012.github.com/">Previous semester</a>
</td>
</tr>
</table>
<br />
<table border="0" summary="" width="800">
<tr><td>
<h2>Can I attend class if I am not registered?</h2>
Anyone interested in GPUs and real-time rendering can attend lectures. However, we don't have the bandwidth to review work for those not registered for the course. The slides and audio for lectures are also available on the <a href="schedule.html">schedule</a> page for the benefit of the entire graphics community.
<h2>Can I take this course if I don't have a graphics background?</h2>
This semester the focus is on real-time rendering. For a student to do well, they most likely will have taken <a href="http://www.cis.upenn.edu/~badler/courses/cis560.html">CIS 460/560</a> and done very well.
<h2>Is this a computer architecture course?</h2>
Not really. We study GPU architecture and see its influence on languages like CUDA and APIs like OpenGL, but this is not an architecture course for GPUs in the same manner that <a href="https://www.cis.upenn.edu/~cis501/">CIS 501</a> is an architecture course for (mostly) CPUs. We are concerned with writing high-performance code that runs on GPUs using high-level languages and APIs, not designing the GPU itself.
<h2>What kind of GPU do I need to do the projects?</h2>
Both the Moore and SIG labs have systems with suffient GPUs. For a personal system, anything capable of OpenGL 3 is suffient (NVIDIA GeForce 8 series or AMD Radeon HD 2000 series). OpenGL 4 is even better (NVIDIA GeForce 400 series or AMD Radeon HD 5000 series). Make sure your graphics drivers are up to date. Some projects use CUDA, which requires an NVIDIA GPU; however, an AMD GPU is fine for most projects, and their drivers are very good now.
</td></tr>
</table>
</body>
</html>
Jump to Line
Something went wrong with that request. Please try again.