Issue link: https://iconnect007.uberflip.com/i/1401769
34 PCB007 MAGAZINE I AUGUST 2021 X things in cost in dollars. By having an artifi- cial currency as the relative cost index, it was most useful in comparing choice A with choice B with choice C. It wasn't useful for pricing a board, but it was useful in design decisions like, "What route on this important milestone here should I travel down?" Max Clark: Right. If I use a via that's too small and I've spent three days routing this board, then you come back and tell me, "It's going to cost you more because you did." I'm going to say, "at's too bad, because I'm not going to go back and reroute this whole board." Holden: Like you mentioned, you have some goals to begin with. e goal of this board is to be fabricated by a tier two, not a tier one. We have some excellent tier twos that are always good on quality, but their prices are great. So I want to stay away from the tier ones that are prima donnas; they can build anything, but I want this. You don't have to put a particular name when you encode the data into the cloud database that drives advice and recommenda- tions. Joe Clark: We're always pushing the envelope, especially in the mil-aero space. But I can't tell you how many customers we deal with across the spectrum worldwide that don't even do blind and buried vias. And, in the PCB world, what's really changed? What's the biggest change in the last 20 or 30 years? Almeida: e last big change in PCB was prob- ably surface mount. Max Clark: Now you have the embedded com- ponents. Almeida: But those are the outliers. I mean, if you look at the main market, it's two-layer, four-layer, six-layer designs. It's exactly what you said earlier, Happy, the commodity PCB. at's what's being built most of the time. Matties: I think we're going to start seeing some rapid change, though, with M-SAP and these others, especially with inflation. If we start seeing the inflation hit hard, the design- ers are going to have to get smarter to reduce cost. I think we're going to see some innova- tion, right? Max Clark: It's going to force it. I do see some changes occurring even though we're still in our infancy with the progress that we're want- ing to make. One of our competitors started this idea of shiing the manufacturing. ey had something in place as well. Everybody is going in this direction. I think it will continue to shi in that direction where more and more of the analysis is done less based upon what the designer put in for the design tool to design to, but more about what the manufacturer is ca- pable of and looking for that intersection. e designer will never design a board just because a fabricator can make it. ey're originally designing the board because the technology needs it. ey build it because the technology needs it. Now they look to see if their manufac- turers can manufacture it. Today, they're doing it mainly aer the fact, when it's too late. What they need to do is shi that over so that some- body comes and says, "We need these criteria." And they start working on it, but if you want to make it at this board shop they can't do it; I can tell you that right now, so start looking for somebody else. Who else should we look for? Joe Clark: Yes, but I really like that idea. It nev- er occurred to me. I really like the idea of the encrypted fabricated data. I can bring that up- stream and it's neutral, meaning anybody can read it; it's just encrypted. If the fabricators are nervous about releasing their family jew- els, if it can be encrypted in a way that we can read it upstream regardless of whether it's your tool, my tool, anybody's tool, that's the infor- mation exchange that supports exactly what Happy said. You could do your analysis up front as you're designing the board. You might