Ernest Cockrell, Jr. Centennial Chair in Engineering Professor of ECE University of Texas at Austin Austin, TX 78712-1084 Phone: 512-471-4085 Email: patt@ece.utexas.edu
Ramblings after all these years
After more than 30 years of teaching, while at the same time having some success at research and consulting in the high-tech
microprocessor area, I have acquired some opinions on education. If you let me, I would be happy to share some of them. This
talk does exactly what the title says -- rambles over my views on education. I will no doubt get into the problems with distance
learning, JAVA vs.
other religions, equal opportunity, and my personal set of rules for being a
good teacher. How much time we spend on each, and what else we get into will
depend on the audience.
Are there any questions?
Perhaps my favorite lecture, this talk is completely free format. It is my response to the question: "How do you know you are
talking about something the audience wants to know about?" Topics can be whatever the audience wants them to be. In some sense,
this talk is the "Ramblings..." talk described above with no compass to get us back on track since there is no track. Some topics
you might consider are those listed above, but that is a very incomplete set. If I think the question is inappropriate, I may
not answer it. If I don't know the answer (since there are far more things I don't know than I do know), I will simply say,
"I don't know." This talk can last as short or as long as those in charge want it to. There have been times we have quit after
45 minutes. Other times, they have thrown us out in order to close up the building.
The Microprocessor: Its Characteristics Ten Years from now
The number of transistors on a single chip has grown from 2300 on the original Intel 4004 to almost 300 million today.
In just a few years, the solid-state circuits people are projecting more than one billion transistors, and a clock frequency
in excess of 10 GHz. With so much capability possible to put on a single chip, what will we find there? This talk explores the
options, always from the point of view of addressing fundamentals; i.e., elementary, but hopefully not superficial.
Faster and Faster Microprocessors: Are the challenges getting too hard for us to continue to deliver
For the past twenty years, the individual microprocessor's performance has continued to skyrocket, partly due to the physics
which has put more and more transistors on each chip operating at higher and higher frequencies and, partly due to the
microarchitecture, which harnesses those transistors in useful ways. Many are saying the end is in sight. We are running out
of gas. This talk explores that issue: why they are saying it and what, if anything, can be done about it. Some of the recent
breakthroughs in microarchitecture will be discussed, and where we might expect others. Examples from new microprocessors will
illustrate some of the points. As always, the approach is intended to be elementary without being superficial.
The Importance of the Freshman Course for Majors
Now that I have published a textbook on the subject (Introduction to Computing Systems: From Bits and Gates to C and Beyond,
with Sanjay J. Patel, McGraw-Hill, second edition, 2004) all my credibility is at risk. Nonetheless, I push onward. I believe
the correct way to introduce computer science and computer engineering majors to computing is through what I call the *motivated
bottom-up* approach, which is contrary to the conventional wisdom of "start with programming in a high level language." I introduced the model at Michigan in 1995, and at Texas in 2000. More than two dozen colleges and universities have already
adopted it. This talk explores the course, its influence on the rest of the curriculum, why we feel it makes sense, and our
experience with it.