Thomas Edman: Yes, thanks Will. So, we’ve been focused and really for — particularly since the acquisition of Telephonics, as you move up in terms of complexity. So, if you start at the printed circuit board foundational level, that’s not where the major constraints are. It really is concentrated in that integrated electronics area that Dan highlighted. That’s about 50% of our defense business. So, that’s where we’re focused on. Assembly is relatively complex but when you move to a full system build, that’s when it really gets complex. We have been focused on incremental quarter-by-quarter improvements. We made some critical decisions soon after the acquisition in terms of larger orders with our supply base to get their attention on critical bottleneck areas.
I’m glad to say that those parts are now arriving. So, think about that lead-time. So, we’re looking at essentially a year that it took to get a number of those bottlenecked parts and start to see them showing up on our docks. But they are showing up now. That’s great. So, the decisions that we made then starting to pay off now. We’re going to continue to work on that incremental quarterly improvement, but we do still see supply chain constraints on that side of the business and better than it was, partly because of decisions we made partly because the climate has improved a bit, our suppliers have been able to ramp a bit, labor availability has improved, allowing them to do that. But we still have those constraints, Will. And I think they’ll carry out, as I’ve said, through the end of this year into early next year, we’ll continue to see those constraints, but making progress.
That’s the key.
William Stein: I do have a follow-up, if I can. I’d like to ask about the new facility that you’re building in Syracuse. Is that intended to address incremental demand in the future? Or will this be more a matter of shifting and consolidating where you build into a new, more efficient factory?
Thomas Edman: So, two-part answer, Will. Number one, absolutely, this is new program requirements. These are future requirements for obvious reasons, I can’t go into great detail. But what I can tell you is, as in our commercial world where customers are driving towards miniaturization. So, trying to get more components onto a smaller space, smaller footprint. That’s critical, carrying faster and faster signal speeds, critical. Then material changes to handle the thermal management required to carry those speeds and is also critical. So, all of those facets are feeding into program demand. This is future program demand and while I say that future ramped demand. So, the second part of that question is we are building some ultra HDI boards today in our existing footprint.
It is rather inefficient and the yields are not optimized. So, relatively small volumes today, thank goodness, because as you know, some of this demand could overwhelm a given small facility. So, the facility that we are going to build in Syracuse will be purpose-built to build ultra HDI printed circuit boards in an efficient process that deals with the bottleneck areas that we struggle with today. So that’s how we’re looking at this developing. That will free up a bit of the capacity in some of our smaller facilities today that are struggling to build these boards, successfully building them, but not as efficiently as we will be able to in the future.
William Stein: Thank you.
Thomas Edman: Thank you, Will.
Operator: [Operator Instructions] Our next question comes from Chan Park with Stifel. Your line is open.
Chan Park: Good morning and thanks for taking the question. I just had one on your data center business. You saw some nice upside in the quarter. So, could you talk a little bit about how you differentiate versus some of the Asian peers or competitors you guys compete against? And if I’m doing the math correctly, your Q4 guide implies a sequential decline. So, could you talk about the sustainability of some of these AI ramps into 2024 and the visibility you’re getting from some of your customers?
