Welcome, so in this course we're going to
be learning about computer architecture.
And this course is an adaptation of a
course which I teach at Princeton
University called ELE, or for electrical
engineering 475.
And I'm David Wentzlaff, I'm a professor
here at Princeton.
And, my background is mostly looking at
how to build minicore and multicore
microprocessors.
And in the past, I've actually built two
of the worlds fastest minicore
microprocessors, in industry and before
that I've worked in academia building
minicore microprocessors.
So, about fifteen years of processor
design experience.
In today's lecture.
We're going to be talking about an
introduction or some opening to, what is
computer architecture, why do you want to
learn computer architecture, how it's
different than previous courses that you
may have taken, so something like a
computer organization class or a logic
design course.
And then we'll talk about some content
today, which is looking at instruction set
architectures and how a instruction set
architecture or big A architecture, is
different than implementation or micro
architecture and why it's a good idea to
split out these two ideas.
So let's take a step back, and look at the
course administration of this class and
how this class is going to be organized.
So as I said I am David Wentzlaff.
This class is roughly gonna be the
equivalent of two 80 minute lectures a
week.
So this is the same format that's used at
Princeton University for this course.
And it's, we are going to try to segment
in shorter segments to sort of give you
bite sized nuggets with questions and
answers intermixed to, to check your
understanding as the, as the video goes
on.
Two books that I wanted to talk about.
This is the Computer Architecture, A
Quantitative Approach by John Hennessy and
David Patterson.
This is a very, very good book.
If you are going to be doing computer
architecture, I highly recommend it, or,
and it's a heavily, heavily, heavily
suggested book for this course.
I did wanna point out that there's a lot
of different versions of this book
floating around.
We're gonna be using the fifth edition,
and you should go get the fifth edition.
The fifth edition is very different than
the fourth edition, and it's updated as of
2012, so it's very fresh.
And then a auxiliary text, which is useful
for a portion of this class, I'll mention
it as the Schind-Lapaste book.
These are the two authors.
Modern Processor Design and Fundamentals
of Superscalar Processors.
The reason that we're gonna use this book,
or the reason that we're going to
recommend some readings out of this book
is that it has a lot deeper coverage of
superscalar processors than the computer
architecture book.
So this is a great book to begin with but
it doesn't cover how to build superscalar
processors in great depth.
This book goes on to do that, and that's
something we're going to be talking a lot
about in this course.
A lot of the contents of computer
organization, or, or traditional computer
organization class, I'm gonna repeat in
the first three lectures of this course,
so the first 3-1/2 lectures of this
course.
And the reason for this is because I teach
everything from first principles, and want
to get everyone on the same page.
But it's, we're gonna go very, very fast
through that material.
So if you've not had a computer
organization class, it may be possible to
take this class, but I highly recommend
taking a computer organization class
before this class because this is the
second class in sort of computer
architecture series, where you have
computer organization and computer
architecture.
And we really do rely on the prereqs, but,
if you are watching the first three
lectures, and you can say I know all of
this.
Yes, that is correct.
You should know all of this.
If you don't know all this, then you
should probably go back and retake
computer organization or take a computer
organization class.
But don't drop the class if you take the
first three lectures and say, oh I know
all this and just, just leave at that
point.
Because we're just gonna breeze through
that content very fast as building from
first principles.