Judy Warner (00:01) Well, Steve, Heidi and Ben, so good to have you again. Our annual DesignCon, now we're in 2025, can you believe it? And so excited to share all of you with our audience and talk about all the exciting things that you're doing at DesignCon and beyond. So welcome to all of you. Great to see you. Steven M Sandler (00:21) Thanks, good to see you too. Heidi (00:22) Thank you. Benjamin Dannan (00:23) Yeah, thanks for having us. Judy Warner (00:25) Well, know, Matt Burns of Samtech, one of his colleagues calls DesignCon the Super Bowl of electronics. So, you know, I do this podcast series every year, which is our pregame tailgate party. And so when I think of that silly theme, I think of you guys being the MVPs of not only my podcast series, but in many ways the show. So I feel very lucky to have you. to jump off that theme. If you were MVPs of DesignCon, what are your positions, Steve? Heidi (00:59) . Steven M Sandler (01:01) I don't know much about football, I think it'd have to be the quarterback. Because if I'm not mistaken, he's the one who gets to catch the ball and run a lot. Heidi (01:04) you Judy Warner (01:09) Yeah. And so you're the quarter, you're the, power, power integrity, power management quarterback. Okay. Heidi. Steven M Sandler (01:12) Thanks. Yeah. Heidi (01:19) I guess I'd have to be like a wide receiver or something trying to catch all those balls that Steve's throwing at me. See if I can run them down the field for a touchdown because Steve is just amazing with his insights and what he sees and he keeps trying to push us into the future and we keep going, what? Why are we going here? No, hey, wait a second. Let me catch up. But every year it's... Judy Warner (01:42) How about you? Yeah. Heidi (01:44) It's like, how are we going to do that? And then he manages to come up with some crazy solutions and then all of a sudden everybody's jumping on board. So it's a pleasure to be working with quarterback Steve. Judy Warner (01:59) Yeah, I'm always trying to keep up with him too. I bet you are too, Ben. Where do you think you'd be on this field? Benjamin Dannan (02:05) I'd be a receiver. Same as Heidi, right? The three of us together would make great team. Honestly, Steve really sets the bar, which is amazing. We have so much fun. We're just lucky to be able do what we're doing, be honest. Heidi (02:07) You Steven M Sandler (02:18) Thanks guys. Judy Warner (02:21) Ben your audio is a little soft. I'm gonna pause right here You were sounding okay, so I think that'll solve it Okay MVPs I know for a few years now you've been co-presenting and Steve y'all blew our mind Actually Steve all of you right with but Steve in your booth you were presenting the refrigerated Steven M Sandler (02:28) No. Thank you. Thank you. Judy Warner (02:47) probes for 2000 amps, which just sounded like science fiction. So like, where do you go from there? What is your presentation about? And why should engineers want to attend? And what are they going to learn? Heidi, let's start with you, and then we'll go around the horn. Heidi (02:51) you So like you just mentioned, Steve, last year it was amazing. He was able to pull off the 2,000 amp with water cooled in his booth at DesignCon showing that 2,000 amp step load. The thing that we're going to do this year is actually take a deeper dive and do a tutorial on how did we actually accomplish that and how can engineers do this themselves. In some respects, it's just back to the basics. Steven M Sandler (03:08) Thank Heidi (03:33) know simple things like power delivery is not DC, it's AC. So just understanding impedance and you know simple IR drop things that get you a long ways along the challenges of how to deal with the thermal. So that's what we're going to be doing is a deep dive for the tutorial on 2000 amp power delivery, how to simulate and model and measure and validate. Steven M Sandler (03:54) Thank Heidi (03:57) The other one is the Chiphead Theater. So another place or presentation where we're really trying to show people how to do the measurements. I think that's one thing I've learned with Steve. It's not just simulation, it's not just measurement. It's simulation and measurement together that's enabling this AI, know, high performance computing, know, the 2000 amps. Steven M Sandler (04:20) And so, I have been on a 10 year, you know, 10 year, Heidi (04:25) to maybe 9,000, 10,000 amps coming in our future, that's going to be a combination of simulation and measurement together. And that's what you're going to see in our presentations, our tutorials, our Chippet Theaters, and in the panel sessions this year. Steven M Sandler (04:25) 10 year, 10 year, 10 10 10 year, 10 10 year, 10 year, year, 10 year, 10 year, year, Judy Warner (04:41) Steve, you and I had a talk not too long ago and you were talking to me about the velocity of innovation and how since last year, you know, doing sort of this almost science fiction realization of technology, but you said, no, it's gone there and beyond. And I remember last year you had like every semiconductor company talking to you. So why don't you set the, you know, why did you guys decide to make this presentation? Steven M Sandler (04:54) you Heidi (04:57) Thank Steven M Sandler (04:57) Yeah. Judy Warner (05:10) And tell us a little bit about the technology drivers as you're seeing them sort of from the front lines. Steven M Sandler (05:17) Sure. So, you know, at the beginning, PicoTest is a company that provides tools to bleeding edge companies that are trying to figure out how to measure things that they don't know how to measure. And so I think by definition, we're always at the bleeding edge, trying to figure out what our customers' problems are and what they're going to do about it. And, you know, I envisioned the paper last year, we were already helping our customers on these 2000 amp power rails and everybody else is saying, what? 2000 amps? What are you nuts? Heidi (05:46) you Steven M Sandler (05:46) In fact, when we decided to do the paper, Yidan Ben-Isra from Broadcom, he joined our paper. And the very first call we had to talk about the paper, he said, I don't understand. Nobody's doing 2000 amps. You we built some of the craziest stuff. We're not at 2000 amps. I think it was the third call we had when he said, I can't believe it. My next board is 2000 amps. And I said, I told you, you can go kicking and screaming, but you're going. It's not like you're getting a choice. Heidi (06:05) you Steven M Sandler (06:14) But one of the things that's really interesting about now, and I talk about this a lot, is that this is only the third time in my 50 year career where we're standing in a place that nobody knows the answers. Nobody knows what to do. We're all kind of dumbfounded. And I'll never forget this. It was around 1979 and I received six sample MOSFETs from Alex Lideau, the very first MOSFET. And know, he invented it. um. I looked at these things, we blew them out, 15 minutes, they were all gone. I said, know, MOSFETs, nobody's going to like them, that's it, you know, it's done for. By 1985, I don't think I ever saw a bipolar junction transistor again, everything changed to MOSFETs. But we were experts at bipolar junction transistors, we knew everything about them, we knew how to keep them from saturating, we know how to make them fast, we know how to make them efficient, we knew everything. And so we had all of this stored up knowledge. Heidi (06:47) you you Steven M Sandler (07:13) that nobody needed. And I kind of feel like we're there again now. And everybody's trying to figure out, yes, we're at 2000 amps and yes, it's growing really fast. I we have customers, they're already at 6000. 2000 now is kind of mainstream. Judy Warner (07:29) That's crazy. Heidi (07:30) Yeah. Steven M Sandler (07:31) But nobody really knows what to do. How do we make the measurements of these crazy currents? Our measurement techniques don't work. What are the VRMs gonna be? How are we gonna make them smaller? How are we gonna get them more efficient? And so everybody's looking to everybody else to figure out what the answers are. And so I think in our team, we're not saying we have all the answers. We're saying we do have all the questions. We do. I think we're very well positioned to understand the challenges. Heidi (07:55) Thank Steven M Sandler (08:01) And we're able to clear away a lot of the smoke and the dust so we can kind of see what pieces are stable and what pieces aren't stable. And how do we build a infrastructure workflow that allows us to build on what we need to get to where we want to go. And that's kind of where we are. That was the goal for the paper. It was thrilling to win the best paper award. And so this year we thought if there was that much interest that we could win a best paper award for the paper, maybe what we need is more time. With more time, we can do demonstrations, we can do a deeper dive, we can talk to the actual detailed processes of how we're figuring out where to go from where we are and how to get there. And so that's kind of the plan for our paper. Judy Warner (08:47) What I love about what you just said is it's as important to know what not to pay attention to because there's so much noise. And if you guys get better fidelity about the things you should be focused on and the parts that are still in development, I imagine for the engineers going to that session, that would be a huge way to get clarification and to sort of turn down the noise and... Steven M Sandler (08:57) Yeah. Heidi (08:58) you Judy Warner (09:17) and not listen to everybody that's talking incessantly all the time. So Ben, you took a huge step this last year. You've been working with Heidi and Steve and others for a long time, but you took the big step this year by launching out on your own after a amazing career at Northrop Grumman and others. Steven M Sandler (09:21) Yeah. Heidi (09:23) you Judy Warner (09:44) decided to start Signal Edge Solutions. So I noticed a lot of that focuses on power integrity, but in relation to what you're seeing with your consulting business and that focus, what are the things you're seeing in the field and how you integrated those learnings into this presentation and paper? Benjamin Dannan (10:05) Yeah, thanks, Shruti. Yeah, so good question. So for folks who don't know who Signal Institution is, we're a services product business. So we are partnered with PicoTest. We use Keysight ADS. We sell ready-to-work equipment, other test equipment from other vendors. goal is being an ecosystem provider for SIPI and EMC. That being said, there are a lot of challenges that I was solving over and over again in Northrop and even in previous lives. And I realize as this has become a passion for me, Steven M Sandler (10:08) Thank Thank I'm not going tell you. previous slides, but I know that this is a real challenge for when I'm in this whole area. And I need to focus more on my secrets, all of the campaigns that I'm doing, and to things that And so, you know, I'm going to be able to say, I'm going to be say I have a good relationship with all of you. I'm be able say that I've a lot of things that I've done in the and I hope you're going be excited to be part of this, and you in the next video. Benjamin Dannan (10:34) whether it's power integrity or signal integrity, that if these problems are manifesting in all these designs and programs at this company, what about everybody else? And so really that was the motivation, right, to address that gap in the industry as a whole. And I know Heidi and Steve can attest to this. And in fact, since I've gone out, we've grown extensively, right? We have a whole team of just engineers doing simulation and measurements and modeling. Steven M Sandler (10:59) to the United States and the Benjamin Dannan (11:01) And so we're here basically to address those gaps and these gaps are everything from, how do I do a two-portent genes measurement? Wait, which probe do I need to use to provide an education, right? My goal is to compliment what Steve and Heidi have really started and really help just kind of push that even further and add more synergy, right? To, okay, so I'm doing this measurement application, which probe do I want to use? how do I simulate this in ADS? The challenges now are really being driven as we've talked about by the data center. look at, let's use the Verso as an example. The Verso has been around the latest version, or I should say the first version came out in 2019. You look at like a demo board from the Verso, like the BCK 190, there's between 20 and 30 power domains coming out of that package on that demo board. Let's take that and scale it a bit further now. Let's add chiplets into this mix. Steven M Sandler (11:30) Thank you. I'm Benjamin Dannan (11:57) Right. You go and you do chiplets, right? It's not one die. You have multiple die and you're tying these dies out. Right. I mean, lot of these chiplet applications, if you're looking at the UCIE standard, we're talking, um, like one customer I'm working with, Steven M Sandler (12:11) going Benjamin Dannan (12:11) it's 256 IO at 16 gigabits per second. The UCIE standard supports going up to 64. So you're talking at 16, you already have four terabits a second. Right. Uh, and then going up to 32 next and 64. So eight and then Steven M Sandler (12:12) I have a CCA that's presenting. So I was talking to my CCA about this actually. And he said, so you are in a project in Mississippi, right? I said, no, I'm not project in I'm in project in Mississippi. Benjamin Dannan (12:25) suddenly up to 16 terabits. That's insane. Well, what people aren't talking about, the challenges are not so much on the signal integrity side. We had that down, right? Modeling that package, those challenges, we'd lay out constraints, good rules, routing things very cleanly and precisely. Steven M Sandler (12:25) I'm a project in Mississippi. I'm project project project in project I'm in project in in I'm I'm in project project I'm in in project in project project in project in in project Heidi (12:28) you Benjamin Dannan (12:42) The SI looked fine. We might have had to add just a few more return BIs to mitigate some crosstalk and some other things, but in a nutshell, those challenges, we have good recipes on how to understand that. Steven M Sandler (12:43) that is very similar to the one that we saw earlier. So, you know, it's not a comparison to the original side of things. But in the past, many of the challenges have been up? And so, you can kind of understand that, but what is the end of the day? So, the whole idea is that the first thing that's being done is to make for the architect to write about it. And the first thing that's being done is to the team come up with something that's a little child-like to do. And how are we going to have that happen? And so, that's the way it's going. So, here are five of the things do. Benjamin Dannan (12:53) and how to implement that from a rules perspective. From the power integrity, there's nothing. There's no standards. The gaps are, you have gaps in vendor models. There's no standardization for models. PDN crosstalk is a real challenge, right? So you have core power domains, sensitive IO power domains, PLL power domains, RF power domains, depending on the application you're talking about, and you're trying to model those together and put these in the system and make sure it's gonna work without doing a bunch Steven M Sandler (13:20) That's a great point. That's a great point because that's what we hear from our customers. What scares the hell out of them right now is PDN crosstalk because the currents are going up and the packages are getting smaller and that means everything is closer to the signals that they're trying to protect. And at the end of the day, you know, I had a boss 40 or 50 years ago and he said, you know, keep in mind power supplies don't make money. Nobody's going to love you for your power supplies. They're not sexy. Signals make money. Heidi (13:46) you Steven M Sandler (13:47) Your job is to make sure the signals look great. And so when I talk to customers today, that's their big fear in their next generation chips is crosstalk. How are they going to manage PDA and crosstalk? And you know, the developing boards now for chips they won't have for a few years. And so how do they do that? How do they try to predict that? That's when people like Heidi and ADS come in and the challenges that they face and the tools that they're giving us to help with that. Heidi (14:01) Mm-hmm. And Steve, couldn't, you know, last year I was so surprised. We were building a step loader. I thought we were doing a step loader to test the PDN. And I remember you saying, no, we had more customers coming up to the Pico test booth at DesignCon last year asking for the step loader so they could look at crosstalk. And, you know, in response to that, we're also, you know, setting up simulations that can mimic that same sort of crosstalk analysis that, you know, your step loader's doing in measurement. Steven M Sandler (14:17) It's a great lecture. Yeah. Great. Benjamin Dannan (14:45) Yeah, these challenges are becoming really hard. I had a design where I worked on for EW application and a fence contractor. Steven Heide saw some of this work and this is where ADS was able to solve the problem in minutes. This team had been working on this problem for a while, over almost two years. And with ADS, building good models and building all the pieces together, we were able to show this crosstalk from a power domain to a RF rail in just a couple minutes. Steven M Sandler (14:55) You should have heard that. And this Benjamin Dannan (15:14) And it just took having the right models and showing with bomb changes that they can implement real time on the hardware on the line. They want to get 10 DB improvement. The other tools struggle with that because the ended modeling is another challenge, right? Having those models is one thing, but being able to put them together and the simulators are other. And so that's where we use what we call the Sandler state space average model, right? Which was invented by Steve. And Steve likes to say that Heidi and I like to dial it up to 11. Steven M Sandler (15:16) to the following features. Yeah. Yeah. Judy Warner (15:29) Yeah. Benjamin Dannan (15:40) And so that's a nutshell of what we've done, right? We've taken that model and we're just building on it and doing more than we ever could or imagined in these simulators, specifically ADI. Judy Warner (15:52) don't know this for sure. Steven M Sandler (15:54) The interesting thing about that is that I published that model for the first time in a book in 1996 and it finally became popular I think a couple of years ago and Heidi and Ben decided to run with it. So it took 30 years for people to notice that model. Heidi (15:58) you Mm-hmm. Judy Warner (16:11) By the way, for our listeners, I'm gonna stick that link again, and Steve and I have talked about it before, but again, Ben and Heidi have really examined that in a close way, but I wanna put that link in so you guys remember to send me that link. What I think is interesting, and I need your insights because you're out there on the front lines all the time, It appears to me sort of like the story you told in the beginning, Steve, that This technology, someone said it's at this time AI data centers HPC that we're running on a bridge that's not yet built, you know, at the speed of light, right? And so the whole ecosystem hasn't grown up. And like you said, there's things we can do things, things that are coming, but all the standards, you know, the whole ecosystem has to grow up around it to, to make these things fully scalable and usable. Steven M Sandler (16:56) Yeah, that was good. Judy Warner (17:14) to us and it's important that we, like you said, here's the parts that are here, here's the parts that aren't here and let engineers know that in some way I feel like a trains head of engineers that they don't fully grasp because of the siloed approach to engineering. And it kind of feels like it's forcing everyone to kind of have a systems based approach. It's forcing them now. We've talked about it and whined about it for a long time. Is that true? Steven M Sandler (17:21) Yeah. Heidi (17:30) Mm-hmm. Steven M Sandler (17:40) Well, it is true. And it's really easy for us to get lost in our own world. We tend to be very focused on what we do and what the current problems are. so a year ago, a PDN was 50 or 100 micro-ohms. And we got really good. had a design 50 or 100 micro-ohms. We knew how to measure it. We got down to 10 micro ohms and I published a few times under the chair. It's nearly impossible to measure it. And I do have a couple of presentations at DesignCon and elsewhere this year on how we're going to measure 10 micro ohms. But don't forget that there are still people like Isfan Novak running around and Isfan contacted me about six months ago and he was working on his DesignCon paper and he said, can you help me measure below one micro ohm? And... Heidi (18:34) you Steven M Sandler (18:38) I mean, I was, I was set on this 10 micro oh my God, how are we going to do this? And it's easy to lose sight of the fact that there is no end. Um, and no matter how crazy what we think we're doing, there are people doing so many crazier things. And, um, I am going to actually present his paper for him since he's not going to be a design con, but we did successfully measure a hundred nano ohm noise floor. Um, and it just. Heidi (18:43) Thank Steven M Sandler (19:06) I think it's a reminder that it doesn't stop. And it's always been like that. You know, I used to have people in my lectures when I was talking about ADS and they would say, you know, I like spice ADS is so hard to use. You know, there's so much to know about it. And I, you know, I would say to my lecture rooms, I said, you know, let's try to get some perspective here. ADS is a very powerful program. And yes, I'm asking you to learn how to use it. Heidi (19:20) Thank Steven M Sandler (19:33) Don't forget somebody else got asked to write it. You get the easy part. You only got to use it, right? And so I think it's really easy to get lost in this maze of what we all do to realize that we're just really one piece of a very large puzzle. And no matter how fast we think we're running and no matter how far ahead we think we are, there's always people right on our tail and we just can't ever stop. Judy Warner (19:39) Right. Mm-hmm. Steven M Sandler (20:03) I think that's one of the reasons that I love working with people like Ben and Heidi. You my hair is pretty gray, I'm pretty old. They have a lot more energy than I'll ever have. And so I think it's really great to have, you know, these people that are propelling this forward. And with the realization that it's not ever going to stop, there really isn't an end. It's like you said, we're running across a bridge that hasn't even been built yet. We don't even know where the bridge is gonna go, what kind of bridge is gonna be. But people look to us to have answers, that's what they want. And I don't even know that they need necessarily hard-firm answers. They wanna know that they're on the right track. They wanna know that they're not idiots and that the problems they're having are the same problems everybody else is having. And if people can say, wow, Steve's struggling with that too. Judy Warner (20:49) Exactly. Steven M Sandler (21:00) You know, there's a certain permission that comes with that, a permission to feel uncomfortable with where we're at. And I think that's really a powerful thing. I started myself in my personal life. You know, I'm a big pizza fan, right? And for my 60th birthday, my wife rented out my favorite pizza restaurant with the pizza oven and the pizza yolo and the whole bit. And I was talking to him about the challenges that I have making pizza. And he looked at me he said, Heidi (21:15) you Steven M Sandler (21:29) I have all the same challenges. just a little bit better at dealing with them. And it's true. And so I kind of look at us that way now and say, all right, so it's not that the challenges that we're facing are different than everybody else's. Yeah, we're a little bit out front. We just learned how to deal with them a little bit better. And I think we're a little bit calmer about it. And so we have this opportunity to share that calmness. and to let people know that they're not crazy with the discomfort they have. We understand it, and you should feel uncomfortable with where you're at, but it doesn't mean you can stop because you can't. Heidi (22:06) . Benjamin Dannan (22:08) And that's why we have all these questions for design to help people address some of those questions. Judy Warner (22:08) Well, Heidi? Exactly. And along with your analogy, and I was just going to go to you Ben and Heidi, is like, yeah, so what are you seeing? I like the analogy of say a master pizzeria, right? And that you have the same challenges, but at least you can say, here's some things you can do. You can use this kind of water, this other kind of flour. Steven M Sandler (22:17) you We Exactly. Judy Warner (22:40) Steve, I know that you leaven your pizza dough longer than anyone I know. you're delivering that message, which I think is highly valuable by sort of locking arms and sort of, you know, the ecosystem, the podcast, that's the spirit. Like put everybody together so we can see all the pieces, right? And in hopes of facilitating engineering success and sending them to the right experts and places. Heidi (22:44) Thank Steven M Sandler (22:58) Thanks. Thank Judy Warner (23:07) Ben and Heidi, since you talk to engineers directly and see the pressing challenges, are there one or two things that you're seeing that is particularly current right now as we go into design con and ways that perhaps things they can plug into, whether it's your talk or a paper? I know you have the cover of SIJ right now. Are there places we can send folks Steven M Sandler (23:27) So, thank very Heidi (23:34) . Judy Warner (23:37) So, what are those problems and challenges and where can we send folks to onboard more of this wisdom besides your talk? Steven M Sandler (23:38) So I that's a good question. Benjamin Dannan (23:48) Do want to go first, baby? Heidi (23:50) sure, no problem. I mean, like, I mean, I have to put in a plug for, you know, it's been, I think, 10 years now with Steve Sandler and we've teamed up. But he did the whole series, How to Design for Power Integrity on YouTube. And that is, you know, it takes, what, maybe an hour to watch all of them. And that's a great prep work just to understand power integrity and some of the basic challenges. Steven M Sandler (23:54) Thank Heidi (24:18) Other thing I think that I find very exciting and it's really what's enabling us to the simulations and even measurements and working together, but understanding the importance of looking at impedance, designing in the impedance or the frequency domain and looking at how things are responding there and being able to translate over to the time domain. Steven M Sandler (24:21) I Heidi (24:45) I I would highly recommend every engineer out there, whether you're signal integrity or power integrity, you need to be an expert at what that transformation looks like and how that works, jumping between the frequency domain and the time domain. Steve has some very exciting examples of using that with measurements and we do that in simulation all the time. I have the harmonic balance simulator, let's just simulate in the frequency domain and jump instantaneously to steady state in the time domain. There's things like that. Steven M Sandler (24:56) to Heidi (25:15) And I'm of course more on the simulation side, Ben's leveraging simulation and measurement, and I'll turn it over to Ben with his response. Benjamin Dannan (25:25) Yeah, so there's a lot of aspects that kind of overlap with what Heidi said. I'll start with really at the source of the power. And let's start with these voltage regulator modules. One of the challenges is everything's going towards these higher current density packages. You don't have access to the control loop anymore. How do you assess stability now if you can't inject a signal? Those are challenges that think people are really starting to realize. Steven M Sandler (25:26) So, that's a little bit of nice picture of the whole thing, the whole whole thing. Also, it's nice to hear the other people's stories about it. So, I think it's great advice. Both are good decisions. I think it's really important to help you advance the current decisions you're making. I it's really important to hear your sense of what's going to be the challenge that going to face. Benjamin Dannan (25:52) come to grips with and try to wrap their hands around. And there are solutions out there for that. NISM is one of the solutions, which is another mathematical approach that Steve came up with that's really become ubiquitous across multiple platforms such as VNAs and simulation tools today. There's a good example where a big semiconductor company who makes voltage regulator modules sent me one to say, Ben, can you measure this? I said, yeah, sure. Steven M Sandler (26:16) So, I'm going to call it the same way that I just did. I'm I'm going to this as a country, but I'm going do it as a country. And you're going think of some of the records that you've got, but you think, I'm going say, I'm do it I'm going do it as a country. I'm it as a country. Benjamin Dannan (26:20) I measured it and it was unstable and it was a 70 amp module and they said, well, how unstable is it? said, well, you need five millivariates of capacitance to fix it. Now this was a P-Mix so it had a bunch of internal registers that you could tune and dial. didn't have time to adjust it. So they gave me a new bill with some adjusted registers and you still needed two millivariates of capacitance to fix it. And so I had a call with a big EV company and they were looking at this exact same regulator. And I said, you should probably see these slides, right? Steven M Sandler (26:35) Thank you. That's how we move forward. Benjamin Dannan (26:50) This is this they hadn't even done that. And so the takeaway here is really as a designer, as an engineer, you really should be measuring what you are integrating into your system is or getting a good model. I can tell you from a design conversation I last year with one of my mentees, Will McCaffrey from Northrop. We showed what all of these vendor eval boards look like from a VRM perspective. We showed multiple. Nobody's innocent. They're all having stability issues. And these are supposed to be the reference designs. Steven M Sandler (26:59) Thank you. Yeah. Benjamin Dannan (27:19) And that's the problem, right? That's one of many problems. Judy Warner (27:19) Yeah. Heidi (27:20) Thank Benjamin Dannan (27:23) Okay, so now let's move forward in the system, right? The challenges with modeling the board effects, getting good capacitor models. There's no standard for that. There are tools out there that can help you. There's measurement solutions out there that can help you. The challenge today as an engineer, there are a lot of gaps in that knowledge. And in our ChipEd session, Steve, Heidi and I will talk about that. And then if we go forward into the package, Steven M Sandler (27:25) Okay. So Benjamin Dannan (27:47) In our tutorial session, we're going to talk about how do you model with the package? How do you model from a large signal perspective? And we'll even go a step further and show you large signal phenomena. With these edge rates now becoming as fast as they are below a nanosecond in terms of speed, we're seeing large signal phenomena in the real world and showing how to model in simulation for the first time ever. And then there's another piece of that. Well, we've talked about ground loops in the frequency domain when you're doing, say, a two-port impedance measurement. Steven M Sandler (28:16) This Benjamin Dannan (28:16) What about ground loops in the time domain? And so this is another, there's another fun product that Pico Test came up with. And so I'll show how to use that product in one of my talks with Rodeo Chorts. And so this is where these things become very challenging. And I like to say it's never been a better time being engineer than now. Cause we're solving and doing more things. And Steve said, it just doesn't stop. I just say that's fun. Steven M Sandler (28:17) is what I practiced this way in school. This is what I was working I that's the problem with it. It's a false field. It's not quite what I thought it was going towards. so this is where it fails because of the energy of the transits. I think so. And so we're going to have to make sure that we're just doing what's on our hands. Basically, I think it seems that it's flexible. But it's not. I don't know. There are mornings I wake up and say, man, I wish I was a Heidi (28:43) you Steven M Sandler (28:46) young engineer right now. And there's mornings I wake up and say, thank God I'm not a young engineer right now. So I think it depends a lot on the day to week. But you know, it is an exciting time and there's a lot of new momentum. don't know if you know Masashi Naga, but Masashi and I also go back a long way. He worked at Texas Instrument more than a decade ago and I was one of his mentors and now he's a... Heidi (28:51) you Judy Warner (28:52) You Steven M Sandler (29:12) you know, senior guy at Corvo dealing with high power systems, not for HPC, but mostly for high power pulse radar systems and pulse amplifiers. And he grabbed things like NISM and SAPIA and he's kind of pushing those forward and getting them into their own tool, QSPICE and pushing the instrument manufacturers to get them into their tools. And so I think that, you know, as a group, we're doing our part and And it is rubbing off on other people. There are people like Masashi that are getting the message and saying, you know what, I can help you run with this and I'll do my part. And I won't rule out Masashi being in our paper next year. You know, I don't know. He's got a lot of energy too. And you know, the more the merrier. think that our job is to share our knowledge with people that know how to move the ball. And I think that's the best thing that we can do. Heidi (29:53) . Benjamin Dannan (30:08) Yeah, Heidi, Judy, I could have one more plug, it would be on the vertical power versus horizontal power. That's probably one of the most interesting polarizing topics in the industry right now. To take a little thunder away from Steve, know, he'll say he's on the fence, right? I'm right there with him. I don't know what's the solution. What I believe it's going to be a combination. Steve has seen a lot of those. I'm seeing some of those with my customers and there's no one size fits all solution. Steven M Sandler (30:36) So, we'll Benjamin Dannan (30:39) We're seeing power supplies or modules for, you know, these, these, these integrated voltage regulated modules that are now over a hundred amp current density in a square centimeter for the first time. And that's increasingly. Yeah. And so this is where we're going. So you have packages that's PicoDisc is designing step loads for that are a hundred millimeter. And those are 2000 Watts. And then you have the same wattage in a package for like Steven M Sandler (30:42) I think it's good idea to a grant IT as part of process and it's actually a good idea. Judy Warner (30:52) Crazy. Steven M Sandler (30:58) I'm going to speak as I come to the end of this year and try to bring together all 30,000 employees that are in the state of Florida to see the Americans that are playing. So, thank you, SDSU. Benjamin Dannan (31:08) I don't know, what's the 50 millimeters, like 50 millimeters over in that ballpark. So you have a module that's a square centimeter. Where do you put it if you're putting it directly below? And then where do you put the cap? So these are those topics that we're gonna talk about in this panel. It's gonna be really fun. We have a great group. We have a gentleman, Guli Khan from Google. Steve's on the panel. He's gonna add his two cents and his perspective. We have a pride host letter from HPE. Steven M Sandler (31:11) Yes, I love it. So. Benjamin Dannan (31:35) So he's doing other stuff with Enterprise and Data Center perspective. And then we have Professor Tin Long from the University of Cambridge. And so I'll be moderating and kind of hopefully doing the metaphorical cat herding and really just facilitating a great discussion between some awesome panelists. Steven M Sandler (31:53) That's one of the exciting things about this year is look at the panels that we have at this conference. mean, you know, we're attracting people like we've never attracted before. And this is the academic world. It's the bleeding edge, you know, Google lights and HPEs. This is really high horsepower people that are getting involved. and so I think that's, that's really awesome to see. And I think that, you know, helps to propel this industry. move yet faster. So I mean, we're complaining it's moving too fast and yet at the same time, we're helping it to move faster. The one thing that I hope I do share this year is that, you know, my dad and I, never really got along, but I always remembered one thing he taught me when he taught me to drive. He said, don't look at where your car is. Look at where you need your car to be and the car will go there. Heidi (32:43) Okay. Steven M Sandler (32:47) And so when we talk about, you know, what's the right answer? Is it vertical power? Do we need smaller modules? We're looking at our feet and that's a really bad place to look because maybe it'll solve today's problem, but it's not going to be a long-term solution. It's going to be a very short-term solution. Everything that we decide now for how we deal with thousands of amp solutions has to be scalable. And when we think about the, to use Heidi's word, velocity of development, You know, we think about how fast things are moving for it to be useful in a long range. That that's saying a lot, but we need to be assessing the scalability of solutions that we're coming up with. Otherwise we're to be right back here next year saying that lot. We really need to have solutions that are going to carry over and scale. Heidi (33:32) Yeah. Judy Warner (33:38) Well, we're, we're go ahead, Heidi. Heidi (33:38) Now, over. just real quick Steven M Sandler (33:39) You're a good boy. You're Heidi (33:42) on Steven M Sandler (33:42) a good boy. I think I'm going to go sleep. Heidi (33:42) kind of on that same topic of what engineers need to understand. I think one of the light bulbs that went on this year or this past year for me was just looking at, we always say, power delivery is DC. Well, if you take an inductor, even like a PicoHenry inductor, really small inductance, a tiny inductance on your printed circuit board, and you take that impedance Steven M Sandler (33:55) you Heidi (34:06) and extrapolate it down on your frequency versus impedance and that inductor slope and you bring it down to one micro ohm, you're suddenly getting into kilohertz low frequency behavior that is transmission line theory. So this idea that we can just think of it as DC is no longer. When we start delivering thousands of amps, Steven M Sandler (34:22) unfortunately, I'm not sure what you're hearing. Heidi (34:31) We're down in the kilohertz worrying about transmission line theory and reflections and impedance. And so that's where the sensitivity to really small parasitics becomes, you need the simulators. It's no longer intuitive. And that's really where the simulation starts to really enable engineers to gain intuition fast and look at the end-to-end and try to gain insight and debug problems going forward. Steven M Sandler (34:57) here, and think we're going start to some of the things going forward. Judy Warner (35:01) To piggyback on Steve's analogy, Steven M Sandler (35:02) So I'm going start with the slides. I'm going start the slides. Judy Warner (35:06) it's like a racetrack. You can have a pebble on a racetrack, go in 100 miles an hour and you'll go right over it, right? The faster you go, those little things that weren't a problem, everything becomes a problem, right? Those tires, whatever it is, everything becomes so critical and so sensitive in those places. Heidi (35:09) you Steven M Sandler (35:14) Thank Heidi (35:16) Mm-hmm. Steven M Sandler (35:19) I I actually have that Heidi (35:19) problem. Yes. Steven M Sandler (35:28) slogan in one of my slides and it says everything matters. Heidi (35:31) Hahaha. Judy Warner (35:32) There it is. You probably taught me that, Steve. So I know we're about a time now, but I'd like to go around the horn. And cause I know we've just scratched the surface. We could talk about another hour about all the things you're presenting at DesignCon. And I promise to our listeners, I will get all these links of these things that we're talking about, NISM, Space-based model, DesignCon papers, all of that. I promise you're going to want to go check into our show notes, but. Let's go around the horn. Steve will start with you. Everything you're presenting at DesignCon and Steve, have, PicoTis has a booth, correct? Steven M Sandler (36:09) Yes, PicoTest does have a booth again this year. We're going to have exciting stuff there. We're even going to have our latest 2000 amp board and it's going to be amazing to see how much it shrunk since last year. But it's about a third the size it was last year. And I think that's a good representation of how industry is moving. I have seven sessions this year at DesignCon and I'm not sure exactly how that happened, but I am taking over for Isfanovac that unfortunately won't be able to make it. I'll miss him terribly, but I'm pleased that I was asked to give his presentation. We have a chip head paper that's going to provide some really good tips on how it is that we do what we do and measurement and design tips. We have our tutorial that I think we're most excited about. We have a couple of very controversial panels, both related to AI. One is what... What is the stimulus behind AI and what are the problems with it and where is it going? And the other one on vertical power delivery and does that actually make any sense? And that's going to be very interesting because we have people at both ends of the spectrum and all the way through the middle. And I'm going to try to behave myself, but you know, I do tend to cause trouble. Ben mentioned that Professor Tang Long is there and we are going to have an exciting announcement at DesignCon, but Tang Long and I are now partners as well as friends. Heidi (37:20) you Steven M Sandler (37:33) And so we're going to share some really interesting news about a new business venture as well. And we'll also be working with people like Nasir Jinnagawa at this year's event. And it's really nice to see these new faces. So there's a lot to see. Judy Warner (37:49) Okay, well, I'm not surprised. You always act like you're surprised. How'd I get here, Steve? But I'm like, yeah, it's another year. So, Heidi, what are you involved in at this year's DesignCon? Steven M Sandler (37:55) Yeah, I know. Heidi (38:03) Well, Steve keeps me busy, so definitely we're teaming up together on the tutorial session on the Tuesday, and then the Chip Head Theater on Wednesday. But also, stop by the Keysight booth if you want to learn about end-to-end power integrity simulation. also, you'll see at the Keysight booth, we're making a heavy investment into the multi-die chiplet sort of packaging world. That's been one of our strengths with Power Integrity. Our simulators can handle those high-density packages, and we have a lot of customers very happy with being able to do the multi-die packages simulations with our PI Pro simulation tool, and then being able to bring that into the ADS environment and do end-to-end simulations. So please stop by the Keysight booth, and of course, also the Pico test booth to see Steven M Sandler (38:55) you Heidi (39:00) some of those actual two thousand amp design running running line Steven M Sandler (39:00) Yeah. Judy Warner (39:06) Heidi, I talked to Roberto Piacentini, who's the head of marketing for your EDA division yesterday. And he also mentioned all the sessions which I'll post in the Keysight Heidi (39:10) Mm-hmm. Correct, yes. Okay, for Keysight engineers. Judy Warner (39:20) Education Forum, right? So there's All Day, Chiplet, Phi, and all kinds of stuff. So that'll be shared in Roberto's version, but I'll make sure and share that from your team. Ben, I know there's no grass growing under your feet, buddy. What are you up to at DesignCon? Steven M Sandler (39:21) So, think that's a start out. So, I'm stop here and just you see. Heidi (39:31) Thank you. Benjamin Dannan (39:36) Yeah, there's a lot of big nomination for me this year. So just honored to be in the running for engineer of the year. That's quite an honor. I'm just honestly just following even Heidi to prior award winners if people have forgotten, there's I'm just standing on the shoulders of lots of great people who wouldn't be here without working with these two and many others and other mentorships. So grateful for that. And I just try to get, get back and pay it forward. So I'm doing four sessions. Steven M Sandler (39:39) So, I just wanted to help you guys see that. Judy Warner (39:42) Woohoo! Heidi (39:47) you Steven M Sandler (40:01) That is a place that you can come and see for yourself. Benjamin Dannan (40:06) Three with Steve and two of them are with also Heidi. And then one more, it's a sponsored session with Rody where we talk about large signal measurements and phenomena, showing ground loops in both the frequency and time domain. So lots of exciting things. I'll be floating around between Keysight, PicoTest, and the Rody and Sports booth. Just really in my element enjoying design con like I always do. Yeah. Steven M Sandler (40:06) So, much in the immediate, to a feeling of, Judy Warner (40:30) Indeed. Well, guys, I can't wait to see you and I can't wait to this year. The ecosystem isn't having a booth, so I'm a little footloose and fancy free. So for the first time in a few years, I get to come to your sessions and come to the booths and not be so locked down to a physical space. So I'm looking forward to seeing all of you and seeing all of your presentations unfold. Ben, a big congratulations to you for being voted. Steven M Sandler (40:36) Thank Judy Warner (40:59) nominated for engineer of the year. And like you said, that's why I call you guys the MVPs. You've all either been nominated or won. Steve won, was it two years ago, Steve, that you won? Two years ago, Steve won. Heidi won some years, 2017. And now Ben's coming up. So, again, Steven M Sandler (40:59) I'm going to be little more correct. So, the fact that I'm going to lawyer, I'm going to be a lawyer for a very long And I'm to be a lawyer for a time. And two years ago, I think. Heidi (41:11) Mm-hmm. 2017. eight years ago already. Yep, Ben's going to make it. Judy Warner (41:25) for our listeners. These are people you want to tap into whether it's on LinkedIn, know, download papers. If you're able to come to DesignCon, I'll put all those links for you. If you're not, you should plug into all these resources, Keysight, PicoTest, Signal Edge Solution, and I'll put all their websites and information below. Again, you guys, I look so forward to seeing you as I always do at DesignCon. Thank you so much for giving us just a tiny taste of all the Steven M Sandler (41:49) Thanks Judy. See you Judy Warner (41:52) the craziness and the goodness really. And thank you so much for all you do for the industry. I really appreciate all of you very much. Heidi (42:00) Well, thank you for the opportunity here, Judy. Steven M Sandler (42:01) soon. Benjamin Dannan (42:02) Thank you, Judy. Judy Warner (42:04) For our listeners, you will regret it if you don't go check out the show notes. And sorry, they're gonna be long. There's gonna be a lot of links. But I wanna load you up. Also, if you use my promo code ecco25, you'll get 15 % off DesignCon conference and a free expo pass if that's useful to you. Or refer25 is another one with the same discount. So I'll put that below for you. Steven M Sandler (42:05) Thank you. Thank Thank Judy Warner (42:32) I hope to see you at DesignCon, but if not, make sure you tap into all the resources for these three DesignCon MVPs. We will see you next week. Until then, remember, always stay connected to the ecosystem. Okay. Heidi (42:46) Thank you. Steven M Sandler (42:48) Awesome.