Design007 Magazine


Issue link:

Contents of this Issue


Page 14 of 99

DECEMBER 2020 I DESIGN007 MAGAZINE 15 Bob said, "The worst thing that happened to Silicon Valley was the Harvard MBA." We said, "What do you mean?" He explained, "They teach these people that if you have an MBA, you can manage anything." Of course, that's not true. If you don't understand the technology, you're not going to make the right choices. The people who are put in charge of these companies have MBAs, and they don't know anything about technology; they just write schedules, and you have to live with them, whether they're realistic or not. Then, it doesn't work on the first spin. There's all this agony in, so they do it again. No one learns from that. If you look at those companies, they just keep doing it that way until a competitor eliminates them. Matties: What do you think is the average number of respins? I know it's a broad ques- tion, but if you look at the industry, what's the typical respin in the industry? Ritchey: I can only speak about the companies my partner John Zazio and I work with, and that's zero. We have failed if we have to respin a board. But I will tell you about a company that I have been working with. I can't say the company's name, but they spin the board 14 times. Matties: How are they still in business? Ritchey: I asked the same question. About half of them weren't working, and one person who wasn't working was the chassis designer. It didn't fit in the box. But I almost fell out of my chair when I heard the answer: They don't have a competitor. They wouldn't be here if they had a competitor. Matties: They're allowed to not learn from their mistakes. Ritchey: When I was at Maxtor, I tried to explain to them that if we take a little longer and get this thing right, we can get ahead of the com- petition. I told them, "The first competitor who does that will wind up defining the field," and sure enough, it happened. Matties: In your classes, do you ever ask the attendees how many respins they do? Ritchey: I never thought to ask that question, but I can give you some yardsticks. The first step you have to take for respin is you must fix the artwork. Then, you have to go through a fab cycle and then an assembly cycle. It is rare that you get a turnaround of a week at a fab shop, and the same is true for assembly. Of course, how long a CAD department takes depends on how big the error is. If you're lucky, it will be three weeks. Now, you have to multiply that by the cost of all the people on the project because they're all sitting on their hands until that prototype shows up, including sales. Matties: That's the lost opportunity. Ritchey: They're sitting on their hands waiting for this thing, and that's money the company is going to spend. Dan Feinberg: What do you think the average number of respins is? Ritchey: I don't know how to answer that. But as soon as you can accurately describe what the real cost is, then people can put their num- ber of respins in context. Feinberg: That's the break-even time metric; the MBAs understood that. Happy Holden: Hewlett and Packard both under- stood engineering, but they also understood first to market and profitability and total sys- tem costs, not just how much you spend on wages. Ritchey: HP was really good at this for a very long time. I have a scholarship program in our university called a Hornet Leadership Program. The reason is that I've seen dozens of companies crash because the engineers

Articles in this issue

Archives of this issue

view archives of Design007 Magazine - Design007-Dec2020