And we do our back of the envelope numbers. Sometimes those numbers are bigger than what the customers are saying. And candidly, that has happened in the past where we built when a customer said we’re going to buy X, we were building 2X and they came back and said, I need 3X. So and we were able to actually quickly ramp to them. So I know you’re looking for better color. I hope that gives you some, but it’s not that clear to you I realize.
Jed Dorsheimer: No, no, listen, it’s helpful. I know I’m going to be asked these questions. So I just wanted to get it from the horse’s mouth. I guess just before I jump back in queue your high voltage product. Congratulations. That takes you into opens up a few doors for you.
Gayn Erickson: It does.
Jed Dorsheimer: I was wondering if you might just elaborate on that. I know you mentioned it for the silicon. You mentioned silicon photonics, but does that also change the calculus for trench in silicon carbide? And then how does that play for memory too?
Gayn Erickson: Yeah. Okay. Well, for memory, it doesn’t apply for trench or gallium nitride. There are different burn-in processes that can be used to accelerate different failures and depending on what the customer failures may be seen, this gives them all the tools for them to be able to apply it. And so candidly we’re just there for them. A lot of people are trying to get away from high voltage. I mean, it’s more expensive than the gate bias, even from us, even though we’re a fraction of the other guys. But it’s still cheaper to do the gate bias only. And so people are playing around with recipes and mixes of how much high voltage do they need, how much gate bias do they need and different things. And they’re all trying to optimize.
And of course, we can’t share anyone else’s secrets. So we’re just giving them the tools and being supportive. The nice thing is, you can buy a system from us that can do all of the above. In fact, you can buy a blade that can do all of the above and one blade. You can optimize it depending on test times. Let’s say your test time for high voltage is X, but your test time for gate is 10X. You can optimize the systems and you can reconfigure them in the field on. So it gives us a lot of, it gives customers a lot of flexibility and we kind of arm them with those tools and are supportive and we don’t have to say no to anything that’s key. And then things like gallium nitride where the primary, if not maybe only the only real failure mode has to do with the high temp reverse bias, which is a high voltage gate breakdown.
And so in that case they need the high voltage for it. And so that’s a critical piece to it. And the fact that we can do it at much higher voltages than anyone else without arching across a whole wafer GaN devices also tend to be smaller, so they have a much higher quantity per wafer. Even the competitors of ours that talk about having the ability to test a wafer, do it with lower quantity than we do. So we have an advantage there as well. And it seems to be coming out from our — the customer inquiries. So it’s nice to have these options. You mentioned memory. Memory, if you mentioned to a memory guy, 2000 volts, it might, they’ll get a eke sound out of ours. And so they’re like don’t get that near my device. So high voltage is really more to do with these compound semiconductor devices for power management than is for memory.
