How ARM Became The World’s Default Chip Architecture (with ARM CEO Rene Haas)
01:14:09 Share

Hello, acquired listeners today we have with us. Rene has the C E. O of ARM holdings. ARM is the company that develops the instruction set architecture and many of the designs underpinning CPU all over your life today, from our phones to our cars.

And David, I actually did an episode way back in twenty fifty on the history of the company, which had a facilitate art out of cambridge university. The company was publicly traded, then taken private in twenty sixteen by soft bank, then last year in public again, and is now value that around one hundred and fifty billion dollars. Rena has quite the career himself in sebi conductors. He's been at ARM for the last eleven years, and before that was a VP at NVIDIA reporting to Jackson. Welcome to acquired.

Thank you very much.

Pleasure to be.

Pleasure is all well, I thought a fun way to start this off, ince. There's a lot of people listening to this that are going to sea ARM holdings and say, I know exactly what that is. I know about the rated gic shift that they have going on.

I've been following every earnings call since they went public. And there's people that are going to say, ARM, what is that? So maybe till level set, everyone on how important arma is in the world. What are the types of devices where the ARM instruction set architecture and ARM designs are used?

So ARM those CPU. And what is the CPU? The CPU is the digital brain of every modern electronic device that is, your television, said your thermostat, your car.

We are having a chat, maybe about talking about a day in the life inside ARM. And I can walk through some of all the ARM devices inside my home. But simplest way to think about IT is we do CPU. And that CPU is the digital brain of every modern electronic device.

And so what is your relationship then with, you know, in my head, apple makes the CPU. In my phone it's the eighteen or in my mac thi. And four, what do you mean?

ARM does CPU, yeah. So drilling down one level aper, we do the design, the I sa, which is the instruction set architecture. And we license that as either an introductions set architecture to a partner, they can develop their own CPU based on ARM, that's what apple does, or we design to build our own CPU and license CPU.

The companies like samsung media attack tesla cocom, amazon eta. So we delivered IT in two different ways. But those CPU that you mention inside your iphone and inside your macbook, those are all are in based.

And so where we were going with this, I mean, today, if you were to imagine your house, my house, bench house, any house, how many devices have an armed chip in them?

And is that a different question then? How many armed ships are floating around .

my house is a hard question to answer in terms of just how many armed chips are in my house. So how many ARM chips get delivered in terms of a typical application space because IT really varies. But again, let's let's go back to first principles, right?

ARM designs the CPU and that is the digital brain of every device, which means IT runs all the the complex software that either runs the dashboard or IT runs the Operating system. IT runs an application. So I was thinking about the question and i'm going to drop a bunch of brand names here, but let's just kind of walk through.

I pulled my auto into the garage that aud has ARM m processors. Those armed processors are what you see running the display. And at digital dashboard, they're also helping with some of the driver assist and they're probably in the power locks, power windows at seta.

I have A A nest doorbell camera that's ARM and that's ARM that basically runs to camera interfaces with the door belt seta walking by the L G. Refrigerator or wolf stove. I can sure you both of those have armed side too.

They're probably running to display. They're probably running the temperatures and the stove. They're definitely running to display. They're running everything in turn to the oven, turn on the television set, which is a samsung that samsung digital TV. It's actually running an Operating system.

So when you run all those apps and everything you see kind of shows up on there, that's a version of android. That's all ARM. Let's say I want to go downstairs and do some gaming.

My PS five has ARM inside, most likely running some of the display controllers and and running some of the stuff with with the game controller. And if I want to flip through on my pixel phone, that is ARM inside running android, and i've got my ipad next to me, that's all ARM. So you can imagine just about everything that you interacted with that does something that either runs an application, recognizes your face, gives you some display information.

armed IT, all of that. I think it's probably true there, hundreds of armed chips or devices with maybe armed ships. How would you describe? There are hundreds of instances of ARM around my house.

Probably hundreds. Yeah probably hundred. As I mean, if you think about the more your homeless connected, all those connected things have ARM inside. It's hard to avoid IT because you almost have to go back to these old mechanical type of controls on on machines that actually don't have something to digital because if it's digital, I can pretty much assure you that it's all it's pretty wild.

So one stat that I pulled LED just from your last quarter financial presentation is that in F Y, this is estimated twenty twenty four. There is almost twenty nine billion ARM chips shipped. And that is for every human on earth. There is four ARM based chips shipped in the last twelve months.

It's a crazy number, right? When you think about the laptop market, which is a big market, right? Everyone wants to ship. And the laptops, big market at sea, at two hundred million units, plus or minus a year, which is a fraction of that twenty nine billion, right?

Very small fraction.

very small fraction, right? So you look at as well, how is that? How's that possible? Because laptop computers, they seem to be a pretty ubiquitous, but just walked to that example I just gave you in terms of those eight or nine examples inside the house.

And then you start to see, well, how do you avoid IT? ARM is an aircraft you go on airport and you check in for your flight and you look up at those displays that are the gate information and the flight information that's all ARM powered that's running that step and in the background. So it's everywhere.

And at this point, counter intuitively, it's also in all the cloud architecture that are running the web services that all of these devices are communicating with. And that is, as we'll get til later in the episode, kind of a narrative violation from the way that the world thought about ARM a decade ago versus what is true today.

That's right. I mean, the identity the company we grew up, as you mentioned, in the opening thirty plus years ago out of cambridge and and the company's original product that we were designed into was the apple newton. And for those who may or may not remember, that was a pda before anything, had a right to be a pda before, before there was the internet, before you advice recognition, before you had fingerprint recognition.

But the ship that was designed, that was based upon ARM inside, had two important characteristics that had to be running off a battery. So as a result, IT had to be defined to be low power. And secondly, performance and cost was really important.

And back in the day, they used to build ships in two different ways. We had plastic packages, which was prety rare, and ceramic packages, which were much Better in terms of heat dissipation, but were costly and not that great in terms of thermos. So one of the directives in terms of the origin design was let's get IT in the plastic package. So as a result from the very early days, the early ARM processor, the ARM one that was defined was basically to run off of a battery yeah.

which at the time didn't feel as critical to the world. The sense all computers were basically plugged in all the time or most of the computing that people did was at computers that we're prompting all the time.

And now obviously, that's very different, absolutely. If I think back in time to the first time, that one could take one of those large satellite phones and walk around with them twenty minutes while having to plug them in IT just seem like magic.

You know, back in the day, if you could get thirty to forty minutes of Better life off of anything that was doing something sophisticated, IT was considered to be just a complete game changer because mobility was simply not something that was very, very ubiquitous back in the early days. If I think of stories around this, one of the jobs I had my career was a field application engineer, and i'm going to date myself here. But we used to call in to the offices for messages.

And in fact, we would be driving from account to account. We find ourselves, get to a payphone. Once we got to that payphone, we could then dial the office. The office would list a whole bunch of messages.

Detail of those messages were something like call me back or i'm not exactly sure what you west or i'm busy when we suddenly had a phone in our cars that would allow us to do all these things remotely with that, oh my gosh. This is the ultimate productivity gain relative to what seemed like western union looking mag in terms of making these phone calls back and forth. But true story. That's exactly how my first field applications shop. That's how we used to correspond .

with the home office. So the arc that we're going to keep calling back to over the course this episode, as you mentioned, the original ARM processor was designed with extreme low heat requirements in mind, kind of low power requirements to in order to not quickly during the battery in a very inefficient world with way fewer advances in on versus what we have today.

And so you sort of think, well, this crappy processor architecture that is extremely limited. Its capabilities will never be the dominant architecture used in all of the most sophisticated in advanced computing applications in the world. Yet IT is.

So over the next hour, we're going going to wander through, how did we get here? But in acquired fashion, I want to go way back to the beginning and introduce this idea of the reduced instruction set computer. And I wanted to sort to turn IT over to you.

You are wildly over qualified to do this as the CEO of ARM, but maybe play computer science professor, computer science history professor with us for bit. What was the development of the risk versus the sister? The complex instruction set computer, like the .

concepts of a risk. And I think they were originally kind of conceived by professors at university of california, berkely, David Peterson. And the whole notion around risk versus c was. These original processors that were invented and we're going to go way back in time to processor architectures is such as the x ty six or the sixty eight thousand from motorola redeemed as sic processors, which stand for you kind of complex instruction set computer, which basically meant that they had lots and lots and lots of instructions that they had to Carry forward because the software that was written for them prior relied on them. So they were Carrying a lot of baggage to do these very, very complicated instructions, which burned a lot of power. Because the simplest way to think about a sesc implementation is an instruction by its definition because its complex means that has to run multiple Operations from a clock time point to execute the instruction, which means those transistors are running more than they probably should and you're burning up .

a bunch of power by example, in any given clock ycl. I need to allow for the possibility of doing something complicated, like in this instruction or in this Operation, i'm going to go fetch something from memory and loaded into a register so that I can be added and I can return the answer all within the same clock cycle. So he sort of to have, like extra band with everywhere to accommodate doing complicated things. In one simple assembly language, you line, yeah.

that's a good way to describe, but or another way to think of a complex struction set a complex instruction is go three steps, forward two steps to your left directly two steps, write three steps. Now if you can find an Operation that benefits from that specialized activity, that's pretty good, but not a lot of programs can.

But once the program has been written and relies on that instruction, then by definition, the architecture has to Carry that forward. So you've got all this kind of heavyweight stuff that's involved. So the concepts were risk.

We're really around simple movements. Move one step forward once step backward, one step to your left, one stepped to your right. I'm oversimplifying, of course, but these are things like ads of tracked settat.

And the idea they are being that if you have a simple er set of instructions that can be combined in such a way to be much more efficient than this is the concept of risk versus sk, which guys I am thinking now. Probably back to the to one thousand nine hundred and eighty were MIPS was invented and and things of that nature that were the original kind of risk processors. And this was all around reducing instruction set complexity.

But interesting enough, that was back in the day when a lot of programs were written, main frames or many computers with previous architectures. And IT was really interesting. If I look back in that time, you had a lot of energy being spent on developing new processor techniques when actually you didn't have nearly the mountain of software that you have today.

But yes, if you go back in time, risk was seen as a much more efficient way to do computing. And one of the benefits you had to that was just not only lower power systems, but also going back in time, one more, more expensive things was actually the memory associated to run all these programs. If you could fit the program and a smaller memory footprint, which again, and with a risk machine, you can do that, there was some benefit to that. So yeah, was was way, way, way back. I would say probably seventy eighty time .

for him is so interesting. And you can totally see why this was conceived of first, or at least in the early days, believed to be Better. It's this really incredibly powerful system where you can any give in, you know, instruction can actually do a lot of cool stuff behind the scenes.

When you sort of jokes opposit to risk, which in early days had very few instructions, a simple Operation would be load, hey, go grab this thing from memory. Put her in a register up. I can't do anything else that is all we've allowed for in this instruction. Just load that IT PS add that's also IT, especially if these .

larger programs were compiled and then made use of those big instructions from an assembly standpoint because the other thing was happening was you were having a change from everything being done and what was called low level programing assembly language to higher level programme models such as fortran, pascal and then c and c plus plus.

And when you're programing at the higher level languages, you have these compellers and then that what does the compiler do? The compiler takes that high level language and tries to put IT into lower level language, which are what these instructions are. So the compilers end up making use of these heavy instructions. And as a result, you just got heavy and more inferior code. And again, one of things that people are trying to do back in the day was get to smaller memory footprints.

Everything you're describing of the old sister world sounds like you said fits perfectly with the main frame. And the mini computer are a big iron, big architecture. Nobodies worried about power requirements.

Complexity is fine. IBM designs. The whole thing. You would think that heavily, that the shift to the PC era would have created the right opening for risk.

But actually this continued to the P. C. air. So like what happened was risk developed just lake slater too late. Did ARM not exist yet?

So continued on this, this history lesson here for for a moment. One of the most amazing things that took place with the IBM computer, I B M P C, back in the day was IBM, which was the world's leader in computing. If you go back in time, right, if you think about IBM three, sixty and the IBM mainframes and the IBM minies, IBM was a one stop shop, right? IBM to the software, IBM to the service, IBM to the hardware.

IBM was was everything. So now, in one thousand nine hundred eighty one, IBM decides i'm going to enter the PC market. And they were behind you. Go back to the late seventies, early one thousand nine hundred and eighty. Apple hasn't invented the first in and quote, home computer based on on the motorola architecture.

And back in the day, you had lots of, I wouldn't all them toy computers, but things like T R S A D and comedy and any all had these smaller, weird little processors into IT. So the irony, the whole IBM story was IBM. The behemoth of computing decides that we're going to now enter in to building computers for the home.

And what does IBM decide to do? IBM decides to not in house the processor, nor do they decide to in house the Operating system. They decide that they're going to make this platform and quote, open and they need an Operating system, doss, so something they can run off the disk. And they started talking to a company that was actually not microsoft .

of computer products, products.

the classic cpm d. They were talking to gary kill dolin in his company about doing that. But they chose microsoft and they were also looking at motor olla, which was kind of considered that the kingpin at the time to do the processor, but for various different reasons, they to stat on intel.

And eighty eighty six, the IBM PC is born. And the irony of IT is that there's nothing about IT that's very IBM like because that uses external memory, IT uses external hard drives, IT uses an intil processor and IT uses an Operating system for microsoft. So a little crazy if you kind of look back in time that you would look at and say, why would IBM actually? And this is what took off with the birth of all the the clones, because you could build a clone on that system.

Because if you bought processor from hotel and you bought a hard drive him corner or max door or c gate, and you bought a monitor from one of the third parties in the taiwan and you've got a license to doss, you're in business and off you go. So to your question though, in terms okay, so why why didn't somebody do something on risk there? In lies, the magic of software compatibility and software legacy. Because all these early programs, IT was stumped, like lost one, two, three, are now written to run on the sixty six processor and optimized on that.

So what happened was, over the one thousand and eighties, as the IBM PC compatible Marks started to take off, you get all this software that was written for that platform and the dirty little secret about CPU architectures and and there has been a lots of them over the days, whether it's, again, back to Spark or MIPS or art or ten sila, ten silica used to work twenty nine thousand sixty eight thousand deck alpha a CPU is only as good as the software that's written on IT and how long that software survives. So the IBM PC and its clans, ultimately building by companies, by by compact and by deal, and by gateway and all these other companies that have been a long gone research, if you remember those guys, this is what created the birth of not only the IBM PC platform, obviously, but the intl lexi six architecture. And that's why, as a default sister and quote, because that's what exact six was, is the defector IT really wasn't a oh risk was Better because IT probably he was, but kind of didn't matter.

Once IBM select in eighty, ty six and doss was optimize for that. And then subsequently windows, you know off IT went. Now the one company was quite interesting that has probably made the most pips in this area was was apple, right? Because apple was originally sixty eight thousand .

based on relative yep.

And then they created a consortium .

for the power PC.

power PC right with with IBM .

and right with IBM with IBM.

Yeah exactly which was kind of a risk sc hybrid.

Think that's very ninety's apple to have something that is like neither risk nor sis, but entirely reinvented in proprietary .

yeah and that's what power was. That was a big switching cost. But I think I think is interesting about that is IT was a large switching cost because the amount of software worked there was there, but not nearly the amount of software that exists today. I was having a discussion on on a podcast that I did with jensen. Yeah.

you guys just launched your own podcast.

right? We did. We did and jensen made to comment on the podcast that software never dies.

That continues to be a very true theme relative to just the amount of heavy lifting required to switch an architecture. But yeah, long with the answer to your story, you to go back in time. Why did this kind of make IT? He was the IBM PC. And once that took off, that's been a very sticking platform.

It's so funny and cause risk was there and arguably would have been Better for pcs of like, hey, new paradigm, a new software going to get written. But IT was that decision to go with sixty six, that los kid for the P. C. era.

And in C, P, U. And I would argue for any any program able architecture to get to something that drives a major switching cost, you need A, A fairly large paradigm shift. In terms of benefits on power or benefits on on cost, you know people will talk about you need Peter nex advantage to make the switch.

I'm not sure it's tenets, but it's not fifteen percent. It's got to be something that's quite material that's going to change in terms of left and or IT has to drive us a level of innovation that could not be done when you're starting out, which kind of goes all the way back to the newton. There was no way that next eighty six could have been an option.

You simply could not build the product. You want to have to start in a space where something is very, very new and you need some very, very unique computing paradise and door. You've got to drive some different level of innovation.

IT took quantify the if it's not text, what is IT? I bet if you just go look at the geek bench scores from whatever apple's latest and greatest intel based macbook prose war before switching to the m one, that's probably the exact quantification of how much Better to something need to be to stay in an existing paradise and switch from one horse to another.

That's about right.

Uh, we want to think our long time friend of the show, vantine, the leading trust management platform, venta, of course, automates your security reviews and compliance efforts. So frameworks like soc two, I saw twenty seven o one gdpr and hypo complaints and modeling ing vent to takes care of these otherwise incredibly time and resort training efforts for your organza and makes them fast and simple.

Yeah, fanta is the perfect example of the quote that we talk about all the time here and acquired jeff basis, this idea that the company should only focus on what actually makes your beer taste Better. I E spend your time and resources only on what's actually gonna, move the needle for your product and your customers and outsource everything else that does IT. Every company needs compliance and trust with their vendors and customers. IT plays a major role in enabling revenue because customers and partners demand IT. But yet IT add zero flavor to your actual product that IT takes .

care of all that for you. No where spread sheet, no fragment to tools, no media reviews to couple together your security and compliance requirements, IT is one single software pain of glass that connects to all of your services via is and eliminate countless hours of work for your organization. There are now A I capabilities to make this even more powerful. And they even integrate with over three hundred external tools, plus they let customers build private and integrations with their internal systems.

And perhaps most importantly, your security reviews are now real time instead of static, so you can monitor and share with your customers and nars to give them added confidence.

So whether you are started up or a large enterprise and your company is ready to automate complaints and streamline security reviews like vantage seven thousand customers around the globe and go back to making your beer taste Better, head on over to vent a outcomes lush required and just tell them that then in David sent you and thanks to friend of the show, Christina anta CEO, all acquired listeners get a thousand dollars of free credit.

Vantage com slash acquired. okay. So we've perfectly set the table for sister in the PC era, pretty locked in, not go and anywhere ARM is founded, it's using a risk based approach. What is ARM doing for its first couple decades in existence? What markets does IT serve?

So go back to kind of the the invention of ARM. And one of the unique things that armed drove also back in the day that I think couldn't be done today. But perfect time, perfect place, perfect strategy.

All of this is also luck and timing. All of those processors that I just described to you, sixty six, sixty eight thousand, A M D, twenty and nine thousand, the list and list goes on. They were all vertically integrated and believe you're not a lot of people used to spend a lot of time designing their own microprocessors.

And ARM had an idea that, that's a lot of work. That's a lot of effort. There is not a lot of differentiation that one microprocessor can have versus another microprocessor.

So why not we come up with a business model that rather than building our own and trying to enter the market against that, what is very, very crowded. I'm going to license IT and i'm going to make IT available to companies rather than developing their own. Just run on ARM and i'm going to license i'm not going to charge and no pun intended, and ARM in a leg for IT.

I'm going to have a business model that's going to require an upright licensing free, which is modest. And i'll take a royalty when you ship in production. And the idea back then was on a shared success model, which I think, again, back the founders is seeing back to people like Robin saxby e and and to the round.

There was really a rather brilliant idea, because the notion was painting in up front license fee, which is kind of a proxy for R N D. In other words, you're not going to spend the money on the engineers anymore to to do the development. The licensing fee will be sort of a proxy for R N D.

So it's not it's not an exhibit free. And more importantly, it's not money you wouldn't be spending anyway. So by licensing the technology, you're not going to need to hire the engineers to develop the products ever to done that for you.

And then on the back end, if you ship a whole bunch of products which is good for you, then pay me a percentage of bit because it's good for me too. So it's a share success model. You look back, say, well, brilliant, of course.

Why would anybody do that? But back when arms started in the early one thousand and ninety, one of things that was really not there yet was all of the tools and methodologies and flow is needed in the ecosystem to make IT work sopsy and high level design language. Pretty new cats doing back in design where you could just take someone else's design and integrated IT into an overall flow.

Pretty new. I said a software tools that was involved in all pretty new. So ARM was really driving a lot of innovation.

And because we were so new again, going back to the super power of a CPU was really the software. We had no software. There were no application ecosystem that ran on ARM.

They weren't Operating systems that ran on arms. So IT was very difficult in the early days to get some stickiness from a software standard point. So our very first design win that kind of made the company, and it's again, it's a classic story of of accidental empires were right time, right place.

But now mobile phones are taking off in the midd one thousand nine hundred ninety. And text instruments is one of the largest suppliers of baseball ships for two g phones and gsm phones. And what they needed inside the phone was a small microprocessor that could help the base band machine run. So the idea of the processor was not to run any kind of applications because back in one thousand nine ninety, there were no applications that rent on a gsm phone.

The application was the phone.

The application was the phone, right? So the customer was ti, but the big customer was nokia. There was the first nokia gsm phones that used the ti chip that had an armed CPU inside.

And T I chose ARM because they looked around everything they had, and they didn't really have anything that was as illegant as ARM. And they thought, why would I designed my own seat? You, because the value back then with two eyes product was in the radio, wasn't really in the processor. Every company, if you look back in the chip world, has a design that was the market maker for them that was IT for us.

IT really reminds me of the t sm c story and journey. Just do a couple years later, right of like starting with like okay, we're going to take a layer of the stack here, you know them at the most lowest level layer of production and you guys one layer up from that. And we're going to make IT available to all these people who want chipps spilling.

We're not going after the P C. market. We're not going. And after you know anything big that's gonna be what IT is today.

We will start with this small stuff and applications like these T I C, P U in a component not leading edge in the fab terms. great. We'll take. That is just amazing over the next twenty thirty years how far it's come.

And it's the same echo of the windows story, which is it's fine to not make that much money early on. But once everyone standardized on you, you have a lot of power in a market.

Well, exactly right. Once we found our way into the T I set chip set that went in the nokia phone, now we have traction. And now other folks who are trying to build baseball chips for gsm phones, ARM becomes the defect standard.

Not so much, quite Frankly, because we ran any Operating system where we ran into the apps because there we're done. IT was just simply, hey, IT works pretty well. It's got the right power.

It's got the right performance and off you go, which is a lot of waiting of designs ultimately sort to take off. So then you get into what was the lift, if you will, underneath the wings of the architecture. Well, fast forward, these gsm phones got a little bit smarter, and they began to run an Operating system called simba.

So we actually began to have some level of stickiness in terms of there was a software community and development ecosystem that, that started to learn and run on ARM. But I would say if I was to look back and say, well, what was the design that took ARM completely into the next level? IT? IT was the iphone. If you look back at the iphone, because ARM now had some three create, if you will, in terms of low power, and we had three grade in terms that we could run small Operating systems and small applications. We were chosen as the engine inside the first ipod.

Oh, I didn't realize that.

yes. So if you go back to early two thousands, and when the first ipod .

came out early, little toa hard drives that that's right. Other use case.

except for that's right. So if you remember that ipod, right, that ipod had a kind of a crew display.

And I would have I had an Operating.

had a little Operating system, IT had a some wheels. You had A U. I had had all the things of a tiny little computer.

The ipod was based on ARM. So fast forward. Now, this is early two thousands. As the two thousands are moving forward in apple starts the funds around without we're going to build a phone and we're going to build an ipad and revision in the history.

There's all kinds of stories which when they were going to build for forest, but it's probably less important in. They had a decision to make in terms of what was the process are going to be inside the iphone industry. The legend is that they did talk to intel about using intel, an intel processor of choice spec. Then was something called the adam.

which was their low power or attempted low power device. Respectfully.

IT was not really so low power, and IT was not really so low cost. IT was kind of a very, very stripped down eighty six. And they were building all the history and kind of going back in time here, you guys remember product called a netbook.

Oh, of course.

the P. C. Industry was lined up. That net books were the future. And that was just .

flat out wrong until .

the iphone .

IT was right, until the iphone and an adam was the chip inside the that books. Intel was coming from a very lofty place of selling you a very, very high performance and very, very good cry. Seven cori fives the adam. This is the classic c innovators.

Dilma, right? Innovate from the bottom over to the top. Intel was having to come all the way down from I seven, I five, I three pentium sellon down to a little idiot, adam, which was designed for the netbook and was probably OK for a stripped down, low power net.

Book a laptop, excuse me, but for a phone that needs to run IT even more lower, not great, right? But intel has got all of the street read inside of apple at the time, because they made that transition by now away from power to eighty six. So all the the laptops inside of apple are all all running on sixty six.

which is on its own, a miracle. Like they changed the compiler to make IT all that applications written, targeting a power platform. The power architecture could suddenly, now with some changes compiled to intel. Oh my god, that is a compiler miracle.

Massive amount of work, years and years of work by apple. So you can imagine the debates inside of apple in two thousand and six.

two thousand and seven, the stated goal for the Operating system, whether phone, tablet, whatever. Suppose we initially what was basically running as ten or a version of IT on a mobile device? I know I ten ran on intel at that point. Yeah.

you guys are are bringing back all kinds of stuff that I completely thought I had forgotten in my memory is a whole different exercise on how neuroscience work because you going to other stuff, but you had Operating systems like leopard and no leopard and all, all these things that were pretty powerful, hefty Operating systems, right? They're running all on on eighty six.

So intel and apple made the shift now in the mid two thousands away from power into intel. You have all this investment has been made on these mac Operating systems, as I mentioned, all of these tigers and levers that are all optimists, zed to intel. You have a big franchise inside of apple that is all based on intel and the mack Operating system. And then you ve got this little fuzzy, the ipod that runs on ARM with a kind of a crew display.

which is basically an .

ebel ded system, which is basically a bedded system. So you can imagine that an easy choice would be, we're going to build this on adam, and we're going to have the Operating system of macos and this new thing look the same, because software will be easier, will strip IT down and will just basically take our laptop and our test top Operating system strip IT down to the phone and run IT on email.

Or we can build up from this ipod, use ARM and build something called IOS, which is the Operating system for the phone. And it's going to be different than the macro s but you know what? This market is very different. It's going to require different level efficiency, different ever power.

If we clean sheet and do IT right this way or the bias was at the time from the ipod team, well, as this is the right way to do IT, we'll end up with Better product at the end of the day. So that was the debate inside. And ultimately, the ipod team .

won sort of right. They could put the baby IT wasn't our own processor, but IT was a version of macos s corneal that had a new compiler written to target ARM instead. Oh yeah.

yeah, yeah for sure. But they didn't start from scratch. But yes, they started cutting things down and the simplified and build.

But yes, that was the key design win for us, the ones that happened then very quickly, you had followers from the android ecosystem, the samsung of the world, and you go back in. Time comes like HTC with andy rubin. And android started to take off.

Now ARM was seen as as the defect to standard. And you had a lot of work that was already now being done around linux and such. So we had the one two punch of having the iphone and ultimately android ecosystem designing around ARM. This is two thousand seven, two thousand eight time for him.

And at this point in time, just so listeners can kind of anchor on what ingredient to the stood as ARM provide, they were standardizing apple and these android vendors on ARM as the instruction set architecture who was actually making the processor in the first iphone or in these other android phones.

If one thousand nine hundred eighty one is two thousand seven, ARM is intel. Except the benefit that ARM has is that instead of into being l in other words, into builds the x six and owns the architecture, ARM is licensing the architecture to companies like samsung. To your question, if you go all the way back in time, I believe or not that first iphone ship, I think was built by samsung for apple.

And then ultimately, I think apple went to tmc, the chip vendors back in the day or companies like samsung, qualcomm believe that are not in video. The tegor stuff was all armed based. So IT was crowded.

And why not you have this, the smart phone market that's now starting to take off and chip vendors now have an opportunity to build chips for these phones based on ARM. And again, if I do my IBM PC parallel, IT would have been as if intel would have a license x eighty six made, did to one guy, AMD, because they were forced to. And this is kind of interesting if you just do the parallels, because IBM was so worried about multiple sourcing, because the actually six was such critical part is that they exercise.

And I I think I have this right to work on a second source for x ty six. So what you had with ARM was multiple sources, so you can see why the business model suddenly became very powerful. Because now to your point, what did we provide in the school kind of that whatever the most basic ingredient isn't? I don't like to do personally, so I don't know what best ingredient is.

But let's assume its water, that without water you have nothing. We supply the water. There was no way anybody could do anything to enter the smart phone market unless he went through ARM.

And there's an element portability. It's beautiful. If you're apple and you wanted design the next version of your phone, you're thinking, well, the there are a bunch of ARM base processors out there. So as long as we pick ARM, we have this whole different sea vendors, including eventually ourselves after we acquire P A M I, that we can sort of pick as our chip vender.

That's right. So they can either pick companies that build ARM chips or if they're brave enough and talented enough and smart enough, ARM will give you the rights to build an ARM compatible chip yourself. So rather than licensing IT from or buying the chip from samsung who use one of our designs, you can just go build your own, which is what apple did.

So we're in these early twenty ten period. This is play a good place to explain the dull ARM business models. So at least at that point in history, how does ARM make money?

So back to the simple concept of licensing and royalties. Our business model way back in the day, and it's still pretty much holes, is that we have favor licensing and royalties. As you can imagine, when you're starting out in a lot of companies aren't actually shopping any volume, the vast majority of your revenues come from licensing.

And the proxy for that in the chip world is design wins. So you get a lot of design wins. You get people committed to the architecture, but they don't actually ship and a viam.

So you don't really get to a mix of of royalties until you envision so and took a long time. But licensing was bigger than realty for many years. And you could look at that glass, you could look at that glass, have full and say, wow, the future is going to be bright. If you ever get there.

glass of everything would be this stuff doesn't work. Yeah.

i'm betting on the front end. And are these things ever gone to see the light day? Now one thing that we we changed, I would say changed.

But another version of the business model is the license. You can either license and core that we built, we call that an implementation. That is we basically do the blueprint and says the house looks like this and .

you let this means in house. You have your own chip designers. They're using canes since and opposite and they're floor planning and they are actually, you know they're doing what everyone sort of imagine and video doing over there.

That's right. There were set of customers that believe that either due to the link between hardware and software or the ability of their engineers to develop something that would be high performance of what we could build, we had these architectural licenses and IT allowed customers to build their own implementation.

Now one of the things that sometimes get confused about these licenses is that are they able to run software that's not armed, compliant? In other words, can they add some special instructions that nobody else has, which gives them a unique advantage and they're not allowed to do that. And the reason is very, very simple.

Once instructions looked different across a number of different architecture that a customer has software can't understand IT. And let me drill on that little bit further. If customer a has instruction that says accelerate and customer b has instructed accelerate two x and customer three has instructions as accelerate.

If i'm a software developer and i'm writing software for ARM, I really don't have my program taking advantage of the three x instruction because I don't know that everybody has IT. So I end up going to something we kind of call inside as a lowest common the nobody approach that the software developer would not make you to those instructions. So as well, the great things the company has done its early days and we've maintained its early since i've been been running at, we're never going to break the say.

We're not going to allow people to add custom instructions because once you do that. You break software compatibility, which is one of the superpowers of ARM. And if you think about know why did sixty six of sticky on the IBM PC? It's because into was the only game in town.

So of course they're onna run and that's why compact and deal and all these other clone guys were able to copy the PC because the software just around. And if they were not able to do IT in such a way that IBM did that, they can never be successful. So we offer these licenses.

There are architectural licences, but all they really do is allow people to build their own implementations. And I will say, just adding on to that, they know we're going back and forth tween the future, the past. We used to do a lot of them because customers used to believe that a, they could build a Better design and ARM and or b, there were something specifically the software they want to take advantage of.

Not many people do them anymore. They're really hard. And back to the ten to fifteen percent advantage, even five percent advantage. The R A, Y isn't all that high. And if you're going to have three or four engineers designing an armed CPU that you can buy from ARM anyway, why not take those three or four hundred engineers and put them on IP that you do as a customer that only you do?

Nobody he's building a CPU with three or four engineers. So it's is three or .

four hundred or four hundred. Yeah not four. I said three, four, that's a big way. Yeah three, four hundred at least it's a lot of work.

It's hard what I imagine for probably almost every customer out there now, the ecosystem and compatibility of software across all vendors to all applications out there is worth so much that they wouldn't even consider going out in altering the instruction separate because then they would lose compatible with the rest.

And I know we're hopping around you in terms of history dates, but that's one of the things that I think gets lost in terms of what's gone on with with CPU and software compatibility over the last fifteen twenty years because as we are talking about in one thousand nine and eighties, early one thousand nine hundred nineties, I mentioned a lot of microprocessors, right, sixty eight thousand power PC, twenty nine thousand deck alpha Spark.

There's a pretty large graveyard of CPU. And they're very good products, very, very good in terms of performance, very, very good in terms of their design. And they just entered the graveyard of CPU.

And you say, if so, why do they all die off? Well, once the fly wheel of software gets built on to a certain architecture is very, very difficult for a if you're developing a new piece of hardware to say, well, i'll choose one of the ones I just mentioned because there really isn't a software story around IT. So they all began to to weather away once the internet took off.

And particular, you've got into the dot come era and a little bit after IT, huge amounts of investment started to go into software companies in software as a service, subscriptions, SaaS models occurring, revenue, everything around the software industry, which was wonderful. Two things happened with that. Number one, IT drawed.

An increasingly innovation and investment into software and all levels of software, complexity of software, software stacks that run the cloud, that run in a network switch, that run in an automobile. And at the same time, semicon inductor investments, which is changing a little bit, now, began to wait and very little venture money started to go into startups. And some other startups are particular.

Well, that's the further ground where new innovation happens, whether it's around new computer architecture, including CPU. So you had very, very little innovation taking place with companies building CPU of startups. In fact, I was with one of the very last ones funded in late ninety ninety come to called ten silica.

And we were a bunch of x and options and x myc guys building configurable processors in the idea they are being that you could build a custom piece of processor with your own customer extensions at sea. And we started nineteen, nineteen ninety seven, I think, and I left in two thousand four company. He was ultimately bought by caes, I think in two thousand and twelve shipped a lot of cores, I think hundreds and how, maybe over a billion cars. But the point was, after ten silica, a another company called arc that was doing the same thing, I was very, very little innovation taking place or investment in some of your CPU startups.

The great irony of the name sake of silicon valley is that if you were a silicon startup, you could no longer raise venture capital dollars there.

Exactly, exactly. So what you have is, as all these architecture start to win away and the amazing amount of investment is now going into the software industry in general, and all of the investment going into stuff, going into the cloud, two architectures really ultimately remain x eighty six, which has been around for forty plus years. And ARM, we were talking earlier about the data center.

Why ARM in the data center? Well, the two things, first off, the choices aren't massive. Not like there's seventeen different choices as what we just talked about. And number two, one of the things are becoming extremely important in the data center is this power efficiency. Because when you're running these extremely large loads, whether there is general purpose compute and now the advent of running accelerate to compute with A I models, you need incredible efficiency in the processor space. So I think we've arrived at this place both as a combination of having A A really, really good low power architecture, be an incredible amount of software innovation has been done on ARM and see just opponent has kind of gone away because investment has has went that last .

one is just like a last man standing. You know, why is the winner? The winner? Well, there was going to be a winner as all the sort of competitors fell by the wayside. It's always totally logical that, you know, whoever becomes the winner, there was going to be one who was left standing or two. In this case.

I would argue it's not one of these industries where last man standing has occurred because the market is uninteresting. It's actually the reverse. The markets never been more interesting. But because of the massive amount of investment required from a software standpoint, optionality is limited because if you were to rock up today and say, I want to go build a system on ship based upon the motorola sixty eight thousand architecture, what software exists is going to run.

It's so what he really is, just like the fab industry, right of like the capital investment required and the software investment required is so massive that like you get to wear, we are now where you've got T S M C.

you got samsung g, the founder.

But at the leading edge, like you know, it's kind of that's left, right.

There are definite parallels. You know the feminine story is direct capex. And you'd look at and say if i'm going to build a two enemy or fab and beyond, i'm going to have have to have thirty or thirty five billion dollars of capex. Our industry is not that. But on the flip side, it's not unlike that when you think about the apex of all of the twenty million developers and plus that had developed on ARM, you're actually having .

to talk that incredible momentum there. I still am flared by this architecture that was originally built not to melt plasti C2Be sup er low. Power ended up becoming, i'm sure you have Better stats than I do, but a dominant architecture running in data centres, doing this heavy compute load AI training, inference.

Maybe A A, I could ask you with your most honest assessment on where is there still a place for X A D six architectures versus where is there a place like should the whole world be ARM? Is IT just actually Better? Or are there different use cases for each?

You know, i'm going to try hard to be unbias even though my job is the CEO of ARM. There's a lot of things that, that are in our favor. One of them is, quite Frankly, the fact that we have an open model where our products can be built at any fab by any chip company.

So if you are looking at X, Y, six, you're looking at two people who build IT now. One of them build A T M C M D, and the other one builds in house at in l though they build a hunch that the m they do these days, but it's just two people. And not only are you betting on those two people, but the IP around the chip that they build, whether it's around because whether it's around accelerate a computing, whether it's round network storage, you're banking on on that to to bring a lot to the party.

And then one might look at and say, well, why isn't intellect empty, just license sexy six and flat out the playing field. And maybe that playbook probably could have been run a while ago. Well, also.

when you've a high market business, mol is a very hard to switch to a low march business model.

bingo. ARM came from a very different place. So as a result, we have a huge advantage just just with our model. Now in the data center, we have another fairly significant advantage in that. If you look at customers like microsoft or google or at all who have customer efforts on ARM, all who have talked about getting sixty percent benefit in terms of performance on a like for like basis, that's not just the ARM isa. That's not just the fact that we are more efficient than the sixty six.

That's in that they can build a customer oc with a custom piece of memory, let's say, or customer storage or custom blade or customer interconnect or custom offload where from a tco, same point, their optionality is incredible. And as a result, their flexibility in terms of building something that is absolutely right for an asia, a state or A G C P S day or an A W S A state because they have the kind of volume and spend that can drive that. So again, one of the benefits we get with the hyper scale is because no one intended the scale is so large. And doing custom chips, they can get in our way on IT. You can do that with thirty six, right?

Go to until and they say, here's my product for you.

here's my product, right? And then you've got to to put the pieces together and see how IT all fit. So that in itself gives us a big advantage. And we have optionality with with people like amp er for example, who do stand products, but that kind of optionality of there's a standard market play and or a custom play or Grace.

For example, the CPU from IT from video, you could buy a Grace and or the way they ship IT today increasingly with Grace black well, where it's highly integrated. And again, why Grace blackwell verses intel plus blackwell or a AMD plus blackwell? Well, if you looked at the architecture and some of the things that they do with N V link, and how they couple the CPU to the GPU and how the interface between h bm memory and CPU memory, they can do that in in x eighty six world.

And then, by the way, in a Grace black world system, the other benefit you have is that Grace can run all major pieces of the Operating system. You can run A A I cluster, A I cloud, and the software stacks that are sort of native that run for an ARM general propose compute can run in your AI cluster. So that in itself gives me a huge, huge optionality.

So I know how we started this. I was I advocating what to do about actually six and share time about army day. But hard, just just hard.

Yeah, yeah, that makes total sense. okay. So we'll all the space bed were at the present. We have come forward today, and I want to talk to about a couple things.

One, how the business model has evolved and how you deal with your customers differently and you're sort of products that you sell to customers now in the way in which you work with customers. And the other of which is last quarter as a recording, you did nine hundred and thirty nine million in revenue. So right around the run rate of four billion dollars, the market cap is about one hundred and fifty billion dollars.

Investors think the future is very, very bright for this company. As we move into this world of AI and connected devices everywhere. Why are people so insane bullishness on ARM, what is the incredible future hold? And why is that valuation evaluation?

We been talking for forty minutes or so, but hope these last forty minutes have .

been helping sort of build.

yes, very much in that case, daddy. I think IT kind of goes back to the fundamental advantages, both from a technology standpoint and probably more importantly, as IT tends to me with with this world, the market forces that are, are in our favor. If you just start with the fact that more and more chips are shipped every year and more and more those chips are based on ARM.

And you look at the end markets, whether the examples I gave you in in my house, from my car to my camera to my stove to my, they are all ARM based and they are all in a growth mode. You look at and say, gosh is out. There's a ton of taye win associated with with this company.

And maybe people are a bit more excited and see I P, I don't know, is around the fact that A I has created this next level of compute need. Now one can argue incessantly around, well, gosh, know forty dollars for copilot and I really getting the arai on that. And what of the new term conomo models in?

You sound like mark.

Better off, I just think the mere economic models on on I is kind of the wrong way to think about IT. I look at IT much more in the parallels of the automobile, the industrial revolution, the smartphone revolution, the internet revolution.

And for a company like ARM, because AI requires a next level of compute capacity and capability, in other words, and it's not just you running strawberry training models in, the massive amount is required to to train all these next generation. Alams are even beyond large in which models video of related models, but it's actually then running those applications. The influence in your car, on your stove, in your head pat, on your wearable able influence is going to run across all those workspaces that all requires a lot of compute.

And one of the things that we've talked about when I was at in video, I was, what is the death for anybody who's either in the computing category or accelerating computing category? And that's when you get to end quote, good enough. I remember being in good enough for i've been in the semitic inie since I got at a school in thousand nine hundred eighty four.

Ren started T. I. And and there's definite periods of good enough. I think the late two thousands, early two thousand ten, felt like good enough, like a nett books were a good at definition of good enough, where at that time I didn't seem like you had the application space in area to drive the need for more compute.

So what did you end up? Building a little crimm hundred ninety nine dollar computer, because IT could do everything your big computer did. So we've definitely had periods and ever in our industry where good enough has existed and the need for compute innovation has slowed.

It's never stopped, but it's slowed with A I in the foreseeable future. You look at and say this appears to be almost unabated because when you think about the benefits A I could bring, whether it's on education, drug, research, investment is mind boggling. So arms, give me in the center of that, whether it's in the data center, whether it's in your automobile, whether it's on your smartphone, whether it's in your wearable, the AI compute path, this is going to run through ARM on some way, share, perform.

So it's kind like the basotho ment of you. I can't imagine a future where my customers ever say, you know, gosh, I wish this were a little more expensive. You can't imagine a future where, gosh, I wish GPT seven or just a little dummer. You know.

I actually like the fact that people look at and say i'm not really seen much benefit from this yet because that actually says, oh my gosh, what a fantastic opportunity to innovate and do more. And a big part of IT is the hardware that you're seeing today, particularly the edge based hardware. Those were designed couple years ago. These large language models weren't needed to run locally. So you have completely unopposed zed architecture everywhere to take advantage of the AI capability that we're going unharness.

So to me, I look at this and it's like White space in terms of the the compute opportunity, which back to the question that then asked in terms of the White people so bullish on the company, i'd like to think that's why IT is we play in a super large market and I conductors are a trillion dollar market. By the end of the decade, you said we're four billion dollars. We probably could take a bigger chunk a that one trillion dollar market at some point time because of the importance of the company.

And this is a good lead into this question I have for you. I've heard you a spouse, this idea.

I'm sure there's a way to rationalize these two things, but I almost feels herrera, you open the episode by saying we do CPU the whole industry over the last five or ten years, including David anne on our invidia episodes, had this obsession with GPU, with accelerated computing, with get those stupid cereal workloads off the CPU, get among of the GPU where you can do pure magic with IT. This enabled the whole eye revolution. You're the C.

P. U. Company, and i've heard you talk about, okay, now that we know some of the use cases that are happening on GPU, history has shown us that those can attend to migrate back to the CPU over time and CPU.

The definition of CPU kind of changes. How do you view the state of things right now with everyone being so excited about GPU and incredibly parallel P. S of the future? And C P S, E P S are fine, but there are none quantity.

I think accelerated ated computing and the advent of GPU is fantastic for ARM because what IT indicates is that there's lots of compute out there and more compute needs to run in such a way that you have not only based computer but accelerate a compute.

The reality, I think it's kind of oversimplified because it's almost the notion of oh, and i've ve met with investors who have had this questions to us and say, well, everything's moving to the GPU. Do you need a CPU anymore? It's almost like I saying, well, i've got this basic engine going to A V A.

I don't need tires and the steering will anymore, do I? It's nonsensical. Just think about the architecture of IT.

So what the advent of of all of these accelerator computing models that are doing gig in, it's primarily the data center. Let's just be very real about this, right? It's it's all happening in the data center.

It's a fantastic outcome for poor CPU. Why is that one? Number one, all these data centers need CPU, obviously.

And I just gave the example in Grace black. Well, why? That's a great positioning. Peace for for. But more importantly, all of that training converts in inference, if training is the teacher influenced, is a student, and there are far more students than teachers in the universe. And that's why the'd be far more inferences workloads than than training.

And that's going to run everywhere relative to the smallest devices, whether it's wearable ables, whether it's headset augmented reality, you're not gna run one hundred watt GPU on your head. I am sorry this is not going to happen, right? You're going you're going to have to get into very, very different form factors.

Now naturally, a CPU is going to be there. You can't have an accelerator out there without something this running the main in the system. That's a fantastic Operating for ARM because that means a couple of things for us.

We can solve that in a few ways. We can add more and more capability to our CPU, which we are today around extensions that help with A I acceleration. And this goes back to risk assist and things that we can add in terms of just extensions that I will help with AI, but also back to the customer ation. You could add small AI acceleration, which we do today with our ethos nps.

There are four tops, 8 tops eta that will do some level of offload。 And I think the model, the model will be for these edge devices to run in conjunction with cloud, where you're going to have some processing happening locally, some processing going to be happy cloud, you're going to need to have some level of security and authentication and at a station locally so that the models know that it's to you and it's not somebody else and the information get private to you. So game on.

I mean, all this, all this GPU accelerated compute is wonderful up for us because it's it's just going to drive incredible, incredible demand. And the idea that the only way you'll ever run a computer is through A A large GPU in the data center. It's just not the way the world works. In the last thing i'll say on this, my love jensen, he's got a brilliant job with the company. But remember.

he tried to buy ARM. I to say there's no Better date in video, tried to buy ARM. And when he tried to buy ARM.

ARM was a two billion dollar company and he was a twenty five billion dollar. He certainly didn't do IT because he wanted to be revenue of creative. He knew the importance of what ARM meant to the industry.

Was that really the. The valuations of both companies at the time of the well.

I was revenue rates, I think but in video was in video tried to buy us for forty billion dollars back in two thousand twenty. And I think their market cap was three fifty to four, wasn't anything close to IT. Is now if you looked at from the outside in back down two thousand and twenty, we had not yet gone public and we hadn't really started to turn around and in our core businesses yet.

So there are a lot of people at the time looking at the deal, moser bt ARM in twenty sixteen for thirty two billion dollars and basically sold at four years later for thirty two plus change, forty billion dollars. There are a lot of critics of the deal that said in video overpaid for this thing because it's not really a growth company. Thank overpaid no, I said that in video overpaid for ARM um yeah that had the acquisition .

got do in video would have been over petting for IT yeah i'm sorry.

The Price that they put down forty there was a lot of criticism that they had overpaid back in the day. Now you look back and now and .

IT seems laughed, yeah, laughs.

yeah, laughed. In terms of their market cap, robi ten x, their revenue is for x. And by the way, the last thing I would say about that, about that acquisition, first of a lot of people thought the video were paid. Secondly, a lot of people hated IT. And there was a lot, a lot of opposition that we got from regulators, customers, ecosystem partners, at which I think be lie kind of the importance of the company and in the kind of a round about way that said, gosh, this is a company being bought for this amount of money at this evaluation. And so if you against that, maybe the company is more important than than folks said originally gave us credit .

for this kind of seems like an area where regulation did exactly what it's supposed to. You were abroad horriston al provider that served a whole bunch of customers that was in a role to an industry and is kind of essential for the further advancement of human kind. I mean, truly in our sort of most important innovation area. And one of your customers wanted to own all of IT, which over time presumably means all the other customers wouldn't quite have the same access to IT.

IT was a fascinating case study because I learned a lot about emin and regulatory and one of things that had surprised our teams that were advocating on the deal was that generally, most of the blocking takes place. And then we this way, I was a vertical merger, right? So wasn't a horizon ona merger.

IT was a vertical merger. And typically in a vertical merger, people will object to emerge if four closes a market or the stifles competition in in a given market. But at the time, we were predominantly smartphone revenue and the video is not a smart phone company. And the folks looked at and said, well, because IT doesn't really violate a vertical integration mantra, and regulators tend to care more about the near term than the long term, this should be okay. But what they actually did in that case was Carried much more about the long term of the what may happen someday versus what we think will happen in the new term.

I'm curious, actually, if you no sense, you are additional dio for a long time. The ARM journey for in video also seems like an improbable one, right? Because in video started as you obviously a graphics card company for pcs, which ran on eighty six and then did this incredible shift into the data center. But at the time, as they were making that shift, data center was also in six environment. When did the company start really realizing, hey, this armed platform is gonna be a lot more than just not melting plastic in, you know, to g phones in video.

Been an amazing partner for ARM when I was working there. We made a very distinct pivot to try to accelerate our mobile business with tega and really accelerate everything we were doing with ARM. Invited about a can be many a portal player.

I remember those guys, they were actually doing the audio chip for ipod back in the day. And we in video now we are doing, we are actually doing the S O C rezone, if you, that's right. Oh yeah, yeah.

That was the the microsoft equivalent ipod we have been flooding with all kinds of stuff that were ARM based, whether with microsoft, with windows ce and zoo. But the real thing that had in video double down on IT was a when the smart phone thing took off, that was number one and then number two. And this was the business I was managing at the time.

This two thousand nine ish time frame is when microsoft made the commitment to do windows on ARM. And we felt at in video at the time that we were very well position to do very well in that market because of all the history that invidia had with the windows ecosystem, all the work that they had done with PC gaming direct. Yeah, yeah.

I was running the business for all laptops back then for in videos, right? I took over all the windows on ARM stuff. So I was doing that first hand myself.

which windows on ARM is like another miracle, if you can make that happen. All that translation layer, all those compilers, everything that you know been written for decades. Specifically for x ty six chips, theoretically you're going to be able to press one button, compile your code differently. And now that runs on our I mean, that is quite the promise.

Yeah, no, allow. The native stuff now has all been ported to ARM. And that really benefit from stuff on mobile, right?

If you think, well, all the apps, all the microsoft dabs that run on ipads today, you know whether it's office and what not. So we ve got a huge benefit of that. But going back to a question, they've in terms of of invidia, they stuck with ARM for quite some time.

We stuck with ARM with windows on ARM. And then after I laugh, ARM became the default platform for everything they're doing on automotive. So if you look at the invidia dry platform, everything in video does around robotics, that's all ARM based.

So everything that they do that uses and quote h accelerated computing, that whole software stack is all all runs on ARM, which is know why ARM is so. Ubiquity is not a motive. If you look at worked done by rena says, or worked done by NVIDIA, or worked done by qualcomm, a lot of those software stacks are now native and all run on onto ARM. So it's a why we're so strong in the automotive space. In the bacteria video question, they were very committed to ARM for a long, long time and combination of a tega b windows, then all the stuff on auto and then the data .

center probably really starting to come online.

The data center really arted take IT off. And I think they they know back to the customization, the way they architect ted Grace hopper and now Grace black, well gives them A A degree of innovation that they can't get any other way right.

On a closing topic, I want to ask about what seems to me to be a little bit of a strategic evolution. Can you tell us what you're doing with sub systems and how .

that came to be? Yeah, so sub systems are kind of a natural extension of an IP business model. So the core model of doing CPU, and I say we do CPU, they over simplified.

There's a lot of their products we do inside the company. We do GPU, we do pu for a AI. We do all of the complex interconnect that's required to build a severe ship, the cmn, with our coherent mesh networks.

And these are essentially the plumbing, fewer building A S oc that has one hundred twenty eight CPU. You need this mission network that helps connect the CPU together and then interfaces them in the memory. It's just a lot of a lot of plummet.

And this is analogy. I know your audience is pretty technical, but I used during our road show think of all those things as desperate legal blocks to very open istis ted customers. You can basically sell them or provided them these legal blocks and they will provide a beautiful copy of the statute liberty. Or you can basically say, look, connect everything exactly this way in this particular reform. And you will get the statue of liberty a hacking of a lot faster than if you built that yourself.

IT really is like lego.

That's where compute up systems are. We're basically take the hundred and twenty eight CPU. We take the complete mission network, other controllers and memory interfaces.

And not only do we stich and together, but we also verify that this is all gonna unctions work. And be correct that when you put IT into your design is just going to work. And that can save three months, six months and nine months of engineering time. You can get a product up to market a heck of a lot sooner.

We can take that a step deeper, which we do in terms that we may work with A A T sm c or a samsung or an intel and say, we're going to actually now say, if you build up this way with this type of characteristics, we will guarantee that you will get four point four gigg hurts of frequency output. We know that you can get this kind of performance. So we are taking IT much further than we have. It's it's almost, I would say, a virtual chip set, but it's not quite to the final building a chip, but it's it's pretty down close. And then you say, well, why would you do that?

Ah this is a lot of integration, is a lot of bundling from just the instructions architecture to uh design to now this kind of complete solution. Hey, connected all this way and know four point four gg hurts are yours. Yes.

i'll call IT packaging instead of bundling. But IT is a way of it's a way of providing a full solution that will simply allow customers to get to market hack of a lot faster. IT provides us a lot of benefit because we can do early prototyping from a software standpoint earlier.

But for customers, the big benefit is they get to market much faster than they would end back to the IP standpoint, connected up all the CPU taking the IP that we deliver. That's not really value add from an end customer, right? An end customer or its building a phone's wants to focus on the I sp in the camera. If you're a cloud customer, you may want to focus on the accelerator or something on analog IO. So for us, our position is if it's around the computer and essentially what's running the main software, the system and how that performs in a certain fab were probably in the best position to be able to define what the best performance output will look like.

All right? So you now have this essentially reference design. For how to make an amazing trip. Are we ever going to see ARM call up T, S. M, C and say, hey, go make a few million of these.

Nothing I can say about that today.

fair. great. Well, and this has been awesome. Thank you so much.

I was great. Thank you .

awesome listeners. We will see you next time.

See next time.