Issue link: https://iconnect007.uberflip.com/i/1014812
16 DESIGN007 MAGAZINE I AUGUST 2018 to facilitate the packaging industry?" We were looking at CerDIPs, ceramic dual in-line pack- ages. I forget his last name, but Tom from Corn- ing and I went out for dinner and talked about our volumes, and where they were going. He came back about six months later, and we talked again. He said, "The problem is that at Corning, after looking at the market, we figured out that we could make enough glass for the entire worldwide CerDIP market in one of our labo- ratory batch ovens. We could make a 100-year supply in one batch." The problem with glass as a substrate is, how does it capture any value? It's such a low-cost material and the process- ing is where all the value-add is going to come in. It has a lot of potential, because it's very stable. Not really that hard to work with, but just getting a good, solid supply chain in place is going to be extremely difficult. Matties: Chuck, what practical advice would you give on packaging in general? Bauer: If you ask an expert if something's pos- sible and he says yes, he's probably right. If you ask an expert if something's impossible, and he says yes, he's probably wrong. I guess my advice would be that the most important two things that one needs to consider in looking at the future direction is, first, "is the approach or technology that you're pursuing scalable?" Scalable can mean many things. One thing obviously is if it is scalable to a manufactur- ing environment? More importantly, is your packaging technology scalable to the future of the circuit board and the chip? One of the big advantages that I saw in the GE embedded chip build up technology was that, because it relied on the same imaging metallization techniques that the chip guys were using, it would scale as the chips went to finer and finer geometries, and the package would can do that as well. So scalability is one thing. The other thing is that if you're looking at disruptive technology, you need to be able to see at least a 75% cost reduction within five years, or nobody's going to talk to you because everybody knows that they can reduce the cost by 50% every two or three years just by con- tinuing to beat on their suppliers, and become more efficient, more automated. Costs always are going down on the assembly side of things. Happy saw that and had to live with it when he was at Foxconn. So I would say to focus on scalability and on the cost impact of your technology. Matties: Is there anything else you'd like to add? Bauer: I would probably say that too many times, when we're talking about technologies, whether we're talking about wafer-level pack- aging or 3D packaging or whatever, we get too focused on the technology. We have to recog- nize that there are a lot of other risks besides the technical risk. There's a market risk, there's a financial risk, and you need to make sure that the market risk is minimized as well. You can't disrupt the technology in a way that destroys your own market. I guess those would be the three points I would make. Matties: Thanks, Chuck. We appreciate your time. Bauer: All right. Thank you. DESIGN007 If you ask an expert if something's possible and he says yes, he's probably right. If you ask an expert if something's impossible, and he says yes, he's probably wrong."