About ten years ago I was at a dinner party where I met an Intel engineer who was working on the design of one of the Pentium series of microprocessors. He told me many fascinating things about the design process, and stated that there was no one person who came even close to understanding the chip in its entirety. Instead, Intel has fashioned a very clever social process where that is not necessary, and the engineers working on the chip only understand it collectively.
I’m wondering if any of my readers know of any illuminating references on the social aspects of the design process for modern microprocessors? I’d be interesting in anything from a few sentences through to an entire book. My immediate need is for a single number – how many engineers are involved in designing a new chip – but I’m also interested to learn more. I’d be fascinated if anyone knows of any connection between those social process and the famous Pentium floating point bug. Thanks in advance for any assistance!
Update: John Dupuis points me to the book The Pentium Chronicles:The People, Passion, and Politics Behind Intel’s Landmark Chips, which looks very promising. Other suggestions would still be very much appreciated, though!
Update 2: More reference goodness from John Dupuis. This time it’s a paper by Bell and Kastelic entitled “Inside Intel-coping with complex projects”. Many fascinating facts at first glance – the Itanium team involved up to 650 engineers and mask designers, and 4500 person-years of work – but I’ve got to run, and have no time to digest right now.