The Rise and Rise of ARM Holdings

British computing company from the nineteen seventies. If you haven't listened to the first part of this, I recommend you go back and listen to that one first because it's sequential.

But we left off in our last episode with the launch of the ARMS six family of microprocessors, confusingly the fourth generation of them, and the failure of the Apple Newton product, which partly prompted ARM CEO Robin Saxby to adopt a strategy to license our architecture technology to other companies back in in other words, to license the the intellectual property to other fabricators, so ARM would not be

a manufacturing company. It would design microprocessors, and it would design instruction sets for those microprocessors, and then license that information to other companies that could then make these chips themselves, and even make them and sell them to other customers.

That's kind of how this could work, and ARM would end up receiving an upfront licensing fee and some royalties as well, joining v L s I, which had previously been the sole fabricator of ARM chips, where the Japanese company Sharp and the British semiconductor company g E C plus E Semiconductors. Now this would just be the beginning of companies manufacturing our microprocessors. And as I mentioned, the

licensing deal had two parts them. First was the upfront licensing fee for the technology, so this was a flat fee, and if a company like Sharp wanted to manufacture ARMED chips, it would first have to pay that flat fee in order to do so. But on the back end was a royalty fee which might not kick in for a few years, in fact, in some cases up to five years after products were being sold. That's when the royalties

would start to kick in. So as companies would sell Armed chips or selling products that used ARM designed chips, then ARM would end up earning a small amount on those sales. Would prove to be an enormous year for ARM.

Not only did it see the launch of the ARMS six in the form of the first Apple Newton, which admittedly is perhaps not the most auspicious of beginnings, and not only did it transition into a sort of intellectual property company, it also landed a deal with Texas Instruments, a company famous for lots of electronic devices is including

popular scientific calculators. T I would license the ARM architecture and would become not only a valuable partner for ARM, but also a sort of almost like a sales rep. Like I was saying. You know, companies would make ARM processors and then some sometimes sell those to other companies. So when hand set manufacturer Nokia began looking for a processor solution for its upcoming mobile phones, Texas Instruments suggested ARM.

Nokia balked at the idea at first because the cost of the ARMS six chips was higher than the company wanted to pay for components for its handsets. It would mean either Nokia would have to eat that cost, thus cutting into profits, maybe even leading at selling at a loss, which isn't really acceptable, or they would have to price the handsets at a level higher than the market would maybe support so they could price themselves out. The phones

would be too expensive, no one would buy them. ARMS response was quick, which was helped by the fact that the team at ARM was still relatively small. The company could actually be really nimble and respond to things quickly, and they got to work designing a new instruction set, one that was just sixteen bits rather than thirty two. If you need to have a refresher on that, listen

to the last podcast. But it lowered the memory demands for the technology and thus reduced the cost of the components required, and it was called the ARM seven T d M I. The T d M I stands for a thumb debug fast Multiplier and enhanced i c e. The only bit I really want to talk about among those right now is thumb and that was a special instruction set, and it was this instruction set that allowed

for the sixteen bit approach. To go into more detail would frankly require a deeper understanding than I possess, So rather than fumble about with attempted explanations, I'll just leave it at this. Essentially, it was is able to take thirty two bits set of instructions and dumb it down

to sixteen bit. That is drastically oversimplifying what it did, but it's it serves its purpose for this episode, Texas Instruments licensed the technology from ARM and then sold the chips to Nokia, and then Nokia then used that chip to power the Nokia sixty one ten hand set, which was phenomenally successful in Europe. Nokia would introduce a version of this phone for the North American market with some

minor changes to the hand set. This one was called the six nine, and the candy bar style phone would become one of the early big successes in cellular phones. I remember having a Nokia. I don't think it was this model. I think it was actually a little later, but I remember playing Snake on it. You guys remember Snake Anyway. More than that, the success of the Nokia sixty one ten led to the big success for ARM

as the ARMS seven technology. She became the go to architecture for mobile phones, and to call it a success really understates things. More than ten billion with a B ARMS seven chips were produced since they were first introduced in nineteen and the Nokia phone itself wouldn't debut until now. Keep in mind those chips are being made by companies

all around the world. ARM is licensing the design to those companies now in the meantime, ARM was developing a new generation of this technology, ARM eight, which confusingly marks the fourth generation of ARM processors, would emerge in nineteen nine six. While it followed the same process as the ARMS seven ten line, it packed in twice the performance

with a more optimized design. It also revamped the instruction pipeline, which, hey, that opens up another opportunity to talk about how that works. The idea behind an instruction pipeline is to maximize the usefulness of a processor by delivering LLL instructions to the

processor simultaneously. So rather than have parts of the processor kind of going dormant because they're not needed for a particular process, the goal was to keep all the parts of the processor busy in an effort to reduce the amount of time that the processor takes to complete a certain task or program. Pipelines do this by delivering sequential steps for specific tasks that different processor units can handle. The ARMS seven architecture had a three stage pipeline divided

into the instructions for fetch, decode, and execute. These are basic instructions. The ARM eight pipeline introduced a five stage approach which added instructions for memory access and writing to memories, so essentially reading from memory and writing to it. Really, what this means is that the design optimizes how instructions arrive at the processor and how much more of the processor is in use at any given moment to complete tasks faster. As Night was coming to a close, ARM

was entering a new phase of its existence. It had up to that point been a privately held company. The income for that year measured up to around two point nine million pounds, which if we converted that to US dollars and adjusted for inflation, is somewhere around seven and a half million dollars. But to be accurate, I really should just give a range of values, because we're dealing with both, you know, translating from one form of currency

to another, plus adjusting for inflation. So really it could be anywhere between six point four and nine point one million dollars, which is pretty big range. The company itself was worth in excess of twenty million pounds, a princely sum. Indeed, it was time to take the company public. Now I've talked about this process in other episodes of Tech Stuff, but generally speaking, the process of taking up private company public.

First of all, It involves an awful lot of paperwork, as people have to determine the value of the company when all its assets and debts are taking into account, and that value in turn guides how many shares and how much money per share will be on offer when the company goes public and people can can buy a stake in the company. The idea here is that the number of shares and the price per share will reflect

the value of the company. You know, you can't just keep printing out shares at the same price, because it all has to kind of link back to how much is the company actually worth. It's actually a pretty complicated thing, and it's partly dependent upon our perception of a company's worth as opposed to a you know, hard, concrete, universal, quote unquote real number. Now, at this point, I think

it's good to reflect a moment on Acorn Computers. If you listen to the first episode, you know that ARM spun off from Acorn Computers, which was a company that used the risk based processors and the personal computers that manufactured the Archimedes computers. Well, things had taken a turn over at Acorn as ARM was making a move to

go public. Much of Acorn's management was focused on that process. Now, remember, ARM was originally a joint venture between Acorn, Apple Computers and v L s I, which was a fabrication company. Acorn management changed and after the company experienced a substantial loss in the new management decided on a massive restructuring that involved selling off several divisions to other companies and ultimately renaming Acorn Computers itself. The company transformed into a

new one called Element fourteen. Now, if you take a quick glance at the periodic table, which I imagine you keep handy just as I do, you will see that Element fourteen is Silicon Clever. Element fourteen wouldn't be around in that form for very long as an Element fourteen the company Silicon, As it turns out, it's still in good shape. But Broad Calm Corporation acquired Element fourteen in

two thousand. Broad Calm, in turn would later be acquired by Evago Technologies, though it still operates as Broad Calm Corporation. But this all reminds us that, as Quigon Jen once said famously, there's always a bigger fish. While Acorn Computers effectively vanished due to acquisitions, ARM continued to go strong The company listed on both the London Stock Exchange and the NASDAQ on April seventeenth, changing its name to ARM

Holdings in the process. The reason for listing on both stock exchanges was due to a mixture of practicality and appreciation. America was leading in the text space, so listing on the NAZDAC just made sense, but the ARM executives also wanted Acorn shareholders back in England to remain involved with the company. Interest in the company pushed the stock price higher, and before long, ARM went from being a twenty million pound company to a company that was valued at more

than a billion dollars. However, not all of that would end up being good news, at least in the short term. ARMS rise and valuation was in step with a general trended technology that proved to be unsustainable, and I'll get back to that in just a moment. Another thing that happened in was the release of the ARM nine group of processor cores. This move also saw a departure from the way ARM had been designing processors. Up through ARM eight.

The company had followed a von Neumann model sometimes called the Princeton model, and the basic concept of this model is that you have a system that has a control unit, a logic unit, and a memory unit. Input would prompt the system to perform operations on data, which results in output, which in turn gets sent to some sort of output device like a display or a printer. And it's named after John von Neumann, who described such a system in a paper called the First Draft of a Report on

the EDVAC. EDVAC was an early electronic computer that ran on binary data. With a von Neumann system, a processor cannot both fetch instructions and run a data operation at the same time because both of these processes share the same bus. Now, you can think of a bus as a data pathway, and it connects different components in a computer system or a circuit, and you can't have two things share the same pathway at the same time. Typically, I often think of it similar to like pipes in

a plumbing system. The ARM nine family of processors changed to a Harvard architecture, which has dedicated buses for stuff like memory and fetching instructions. And since this approach has paths specifically for each of those tasks, there's no bottleneck

if you need to do both of them simultaneously. Now, for certain implementations that doesn't really matter that much, so von Neumann approach is perfectly fine in some cases, but as you get into circuit complexity, being able to fetch instructions and read or write data to memory can really speed things up. ARM recognized that it was time to transitionto a more robust circuit design as lightweight computational devices

were putting heavier requirements on processors. The ARM nine chips also saw improvements in heat production, meaning they were producing less heat than their comparable ARMS seven predecessors. ARM ten would not be far behind. It launched in late nine, and as you would expect, the new micro architecture included some benefits over older designs, but frankly, there wasn't anything so incredible that I feel that I should really break

it out in this episode. We'll just keep in mind that the company was keeping up its process of research and development and then licensing the resulting designs to various fabricators around the world, because otherwise this episode gets way too dry away too fast. In the company made an acquisition. It purchased a software consulting company called Micrologic Solutions. Like ARM Holdings, this company was also based out of Cambridge, England.

The following year it made three more acquisitions. Euromps that was a company out of France that designed smart cards. Those are integrated circuit cards have an embedded circuit chip. They act as a kind of authorization device. The second was a Lot Software. It was a company that made debugging software, and the third was Infinite Designs, another English design company, And in two thousand one the company purchased a division out of Neural Micrologics, another debugging company in England.

But around this time the X sector had become the new gold Rush, with entrepreneurs launching new companies by the day, many of which were tied in some way to the growth of the Internet. In particular, in the late nineties, the general public was really just starting to understand that the Internet was a thing. Previously, the Internet had largely been the stuff of research labs, college computer labs, some

specific industries, the American military. But then we got the Worldwide Web in the early nineties, and slowly but surely, the average person began to catch on to what the Internet was. And one thing lots of people were sure about was that the Internet was the future of all commerce and business. You could argue convincingly that they were all right. But the problem was, back then there wasn't as much of an understanding around how this would come about.

It was just generally considered to be a foregone conclusion, and so numerous businesses popped up with the intent of cashing in on the tech craze in general and the Internet in particular. Investors went bonkers, and so did some of these companies, as many of them would go on to lavishly spend money on stuff like office spaces and perks without you know, figuring out how the company would ultimately make money. And this trend proved to be unsustainable

in the long run. How's that for a cliffhanger. I mean, you guys know what's coming, but still, you know, structurally it's a cliffhanger. Anyway, we'll be back to talk about it after this quick break. Eventually, due to a few different factors, reality would pull the rug out from under the tech sector and we got the great Dot Com crash.

Companies that have been valued in the millions of dollars lost all their value rapidly as investors lost confidence, and ventures that just showed no signs of having a business plan or means of making revenue. But it wasn't just the questionable companies that suffered the effects of the crash rippled through the tech sector, hitting more established companies with actual, solid business plans, and that included ARM. Now part of the issue was that ARM itself was seeing its value

inflated well beyond the company's own earnings. In before the dot com crash, the value of the company was more than three hundred times what it was earning. But even after the crash, ARM was still hitting targets, but its value was plummeting because the market was readjusting. The world went into a recession, and that recession affected ARM just as it was affecting other companies, and it was, according to some employees at the time, a pretty rough time

to be working for ARM. To be fair, it was a pretty tough time to be working just about everybody around this time. Robin Saxby, the CEO who had sort of guided ARM into the intellectual property phase of its existence, transitioned into the role of chairman of ARM, and David Warren Arthur East, better known as just Warren East, came in as the new CEO of the company. East had previously worked at Texas Instruments until n And joined ARM

and created a consulting business division within ARM. He then rose to the rank of vice president of Business Operations. Then later on he became the chief operating Officer or CEO, and then he became the CEO, and he would lead the company for more than a decade, which included guiding ARM not just out of the two thousand one recession, but then a subsequent recession in two thousand nine. These were enormous recessions. It wasn't like ARM did something wrong.

It was more that these were global events. And while he would move on from ARM in two thousand thirteen, I wouldn't worry so much about him because now he's the CEO of Rolls Royce Holdings. So one of the things East did in order to try and and re center the company after the dot com crash was to create five year road maps with a plan on how to guide the business beyond just the short term, you know, more of a long term look at production and and

research and development and business plans, which I think is phenomenal. Meanwhile, the business was beginning to change again because in the early days, a big part of The work around designing processors was getting the individual components to a smaller size so that you could fit more of them on a

single chip without making the chip any bigger. Manajorization was really the goal, while at the same time you had to keep an eye on stuff like heat management, because packing more components closer together usually means that a powered processor is going to generate more heat, and heat and electronics are not a great pair. And at a certain point, the microprocessors were reaching a stage where the need to make them more powerful was starting to diminish, at least

for the time. The microprocessors could be shrunk down to a smaller form factor because the components were smaller, but there wasn't a need to keep the next generation processors at the same size to cram more of them on there. In other words, you didn't need to keep the chip itself the same size. If your chip measured a centimeter square and you were able to shrink down the components, but you didn't need to add more components, you could

make that square smaller. Maybe it's you know, nine tenths of a centimeter instead point nine centimeters. Well, companies began to take the opportunity to build out software based approaches to chips and build out what are called system on a chip or s OC solutions, and that really made it useful to have these very very tiny microprocessors as

part of it. A system on a chip, by the way, is an integrated circuit that creates an entire system on a single chip, and that includes stuff like a central processing unit UH, internal memory, input and output ports, and more. The system on a chip approach cuts down energy consumption, it can add more capabilities to smaller devices that don't

have the physical space to fit multiple chips. So the system on a chip approach is what would ultimately allow for the evolution of something like smartphones and other small computational devices. Many companies interested in developing system on a chip systems for various devices from communication systems to AI platforms look to ARM processors to provide the low power,

high efficiency processing part of that system. So again, it wasn't that ARM was creating systems on a chip, but rather the designs of ARMS micro architecture was an important part of the system on a chip that was made by these other companies. ARM Holdings have been releasing new micro architectures pretty frequently leading up to the dot com crash, But after ARMED ten, which released in late it would be nearly four years before the next generation of processor

designs would emerge from ARM. ARM eleven would launch in April two thousand two. This family of processors didn't come out of the Cambridge office. This one actually originated in a French office at Sophia Antipolis, where ARM had expanded. At this stage, a lot of the optimization was centering around playing media files efficiently as MP three players. You know,

most notably the iPod were becoming popular. These devices needed small, powerful and efficient processors to avoid problems like overheating or delays. You know, no one wants a super hot electronic gadget burning a hole in their pocket or have the experience of navigating a menu in an iPod or similar device and then having to wait for the device to catch up. In two thousand five, ARM would make a really big change.

The company recognized that in order to continue to grow and to succeed, it needed to diversify beyond just making smaller and more powerful processors. The ARM eleven family would be the last group of ARM microprocessors using that particular simple numbering system. The company changed gears and introduced a new classification, several new classifications of processors under different lines called Cortex. So you had Cortex A, and Cortex A was a continuation of the chain that had left off

with ARM eleven. So you could think of the first generation of Cortex A almost like it's ARMED twelve, because that's it was a continuation of that process. But there was also Cortex ARE. These were processors that ARM had optimized for real time applications, so in other words, for applications that needed super fast responsiveness and really low latency

but not necessarily super powerful processing power. Then there was Cortex M still is actually there is Cortex M. That's a family of processors that use very little power and are inexpensive, and those are mostly meant for embedded technologies. And this was right around two thousand five, and to me,

that is incredible. ARM Holdings recognized a trend before a lot of other companies and people did, and that that was gonna be a real need for lightweight processors that could run on very little power on all sorts of

electronic devices. So ARM Holdings anticipated the Internet of Things era, and this was two years before we even got the first iPhone, So by focusing on these three major strategies, the company could continue to succeed in markets like mobile while expanding its offerings to new opportunities like bettable technologies. By two thousand and eight, the demands of technology required ARM to innovate even more. The iPhone launch was, as we all know, a phenomenal success, kind of understating it.

More smartphones would follow, with Google Android entering the fray and Microsoft doing its best as well. Microsoft wouldn't work out, but Google's certainly did. But smartphones are quite demanding devices. Sophisticated apps need a decent processor, and the small form factor of a smartphone means that battery life is a premium as well. To meet the increasing demands of smartphones, ARM Holdings created its first multi core processor, the Cortex

A nine MP core. It's a good idea to run over what multi core processors are and what they are good at doing. These are processors that have at least two separate processing units, and each of those units is capable of reading an exit uting program instructions on its own. The effect is the same as if you had a device that had multiple CPUs, you might sacrifice a bit

of processing speed per core. For example, let's say I design a single core processor and it can run at three point five giga hurts, but then my multi core processor version I have is limited to three point two giga hurts. Now, remember when we're talking about processing speed, we're talking about the number of pulses the CPU generates in order to carry out instructions. And a giga hurts is one billion pulses per second. So the single core on its own is faster than either of the two

multi cores. But the multi cores can work independently and thus solve certain problems faster than a single core CPU. A multiple core approach is great if you can divide the problems that you're working on into parallel tracks, and we call this parallel processing. And there's an analogy I love to use. Long time listeners of tech stuff know what's coming because I use it every time. But the

analogy involves math students. So let's take two different scenarios, and in each scenario, we have a class of five math students. One of those students is a true math genius. She can solve problems whip fast. Now, The other four math students are good math students. They're smart, they perform well, but they cannot solve math problems as quickly as our

genius can. So in our first scenario, these students all get the same math quiz, and the quiz has one long problem on it with several steps to the problem, and each step of the problem is dependent upon the answer or the solution from the previous step, so you can't skip around because your work completely depends upon the stuff that came earlier. In this scenario, our math genius

would finish first. She's just super efficient at answering each step and she can move on to the next one, while our other students are still working on the earlier parts of the problem. But now let's move to scenario too. Now. In this scenario, the teacher has decided to have a race just for fun, and the teacher hands out a math quiz, and the math quiz has four math problems on it. These problems are not related to one another, they are completely independent. Our genius has to answer all

four problems. However, the other four students each get assigned one of the four problems, So student one has Problem one. Student too has problem too, and so on. The genius has to see if she can solve all four of the problems on her quiz before the four other students

each solve their single problem. And in this scenario, we would expect the four students to win because even though they cannot solve an individual problem as quickly as the genius can, they are only working on a single problem each, rather than a collection of four problems. That's kind of how multi core processors work. If the device is handling

multiple processes that are independent of each other. A multi core processor approach can make things faster and more efficient, but not all computational problems fall into the category of parallel problems, just as a quick tangent. Parallel problems are where quantum computing could potentially make an enormous difference. You've heard me talk a lot about bits in these episodes, the units of binary information that are either a zero

or a one. Quantum computing systems use quantum bits or cubits, but not cub Bert's that's an arcade game character. Because of the properties of quantum mechanics, a cubit can essentially be both a zero and a one, and all value is in between technically all at the same time, and that means a quantum machine can attempt to solve a problem in as many ways as are allowed by the

number of cubits that the system has. So with that approach, we can make some pretty drastic changes to our world, including the complete elimination of encryption as we know it today. But that's getting off topic. I just couldn't resist it. So arm creates its first multi core processor, with smartphones foremost in mind due to their hefty processing needs. In two thousand eleven, the company would introduce another innovative approach

for devices like smartphones. This one was called Big Dot Little. And the funny thing is is that big is all in lower case and little is all in upper case. And that's cute. I mean, it reminds me of tests where you know, our friend finds a word that spells out one color like the it spells out the word blue, but the actual letters are all a different color, like green, and then they show it to you really quickly and say what color is that? And that's when you decide

you need new friends. The big Dot Little approach, however, was meant to provide processing power on kind of an as needed basis. So the idea was, if the device launches an application that requires a lot of processing power, the primary processor kicks into high gear, but when that task is over, it can then shift operations to a

lower power core, conserving battery power. This lower power core doesn't need as much electricity essentially to run, it's just running simple processes, possibly in the background, and that way it could conserve battery energy. Now, one of the reasons companies like arms have to do that is that while we see processors advance on a trajectory that more or less follows the vision of Moore's law, particularly as we start to fudge what Moore's law actually means, not all

technologies can follow that same trend. Battery technology, for example, lags behind. Batteries do get better over time, but ultimately they depend upon electrochemical processes, and while we can make better design batteries, you can't actually improve physics. You know. It's not like we could go to a skate park and say, let's make gravity more better here, and then we're able to jump higher and there's no risk of injury if we fall, because physics and chemistry don't really

care what our needs are. They just are, and we have to work within those limitations as best we can. So while we search out better means of storing energy to later releases, electricity processor designers have to keep working on ways to reduce the demands for electricity in order to make batteries last longer. I've got a bit more to say about ARM before we wrap things up, but first let's take another quick break. In the world of PCs,

Intel is dominant. In March, analysts estimated that Intel processors made up eighty one twenty five percent of the market share, and rival a m D would take the other eighteen point seven five At times a m D s market share can top. But Intel is clearly in the lead when it comes to processors in PCs. But that is in PC land, and Intel just wishes it had the market position that ARM has when it comes to mobile devices.

See in mobile ARM has more than five percent of the market share when it comes to devices running on ARM designed processors. Like the mobile devices out there are on an ARM designed processor at some point in their in their architecture, even if it's a small part of it. The other five percent are the only ones that are not. So keep in mind. Again, these processors are not manufactured by ARM. ARM remains a fabulous processor designer, so it

does not fabricate hardware. The company designs the architecture and then license it out to these companies, So companies like Qualcom and Samsung and hundreds of others do this. If you've heard of snap Dragon chips, those include ARM processor designed cores. Samsung's ex and Nose, which was formerly known as Hummingbird, also has ARM designed processing cores. Even Apple, with its A nine system on a chip, had ARMED design processing cores, So companies can also synthesize the ARM

design into their specific system on a chip architecture. So it's not like all of these are idea knuckle to one another. They are very different to one another, but

at their heart is this ARM designed technology. I should point out, however, that with more recent Apple chips like the A twelve Z bionic chip from Apple, UH that one uses an instruction set licensed by ARM, but Apple is actually responsible for the actual design of the processor itself, so in that case, the micro architecture isn't dependent upon a hardware design from ARM, just the instruction set that runs on the hardware is from arm Apple engineers designed

the actual chip in that case. Meanwhile, as the company ARM Holdings secured its near monopoly on processor designs in the mobile space, it also became more attractive to bigger fish. In two thousand and sixteen, one such big fish came

a chomping at ARM. This one was soft Bank, which in July anounced its plan to acquire arm for twenty three point four billion pounds, which at the time was equal to about thirty one point four billion dollars a princely some Indeed, from announcement to completion, the process took less than two months, and in the world of large acquisitions,

that's pretty darned fast. Soft Bank is a conglomerate based in Japan, and the company owns a steak in lots of other companies, particularly in the tech and energy sectors, as well as the financial sector. So, for example, among its other subsidiaries are Boston Dynamics, you know, the robot company, Sprint, various Japanese branches of companies such as Yahoo or Ali Baba, not that it owns the whole company, but rather it has a stake in the Japanese branch of those companies.

The price of that acquisition surprised some people because you know, while ARM was in a really secure spot, the annual venue for the company was just a fraction of that acquisition price. In ARM made about one point five billion with a B dollars in revenue, and that is a princely some, no doubt about it. But revenue is not profit. You still have to subtract all expenses from that number

before you can start talking about profit. However, the expansion of the Internet of Things and the proliferation of more devices and need of processors everything from thermostats to VR headsets, meant that acquiring the company was a good way to plan for the future. In two thousand nineteen, ARM announced it had signed a partnership agreement with DARPA, that is, the R and D branch of the U. S Department of Defense. Typically, DARPA doesn't do a lot of the

hands on work in developing technologies. The agency tends to create contract opportunities and other companies and research organizations do the actual groundwork. So DARPA will define a goal or a challenge and then offer up contracts to fund companies that want to try and meet that goal or challenge. But in this case, Pentagon research teams are getting access to ARMS, deep knowledge and expertise on low power, high

efficiency processors for research purposes. One more recent development includes a few newer processor lines joining the Cortex family. UH. In addition to Cortex, there's also ethos In and ethos you. These are intended to work in systems that relate to machine learning and artificial neural networks. UH. There's also the neo Verse, which launched in twenty eighteen, that aims at computer servers and data centers. And then there's secure Core.

These are processors that are meant to work with stuff like smart cards and embedded security systems. As it turns out, soft Bank is standing to make a tidy little profit by selling ARM holdings to another big fish, this time in Vidia, which is best known for its graphics cards and video, announced its intention to buy ARM from soft

Bank on September twenty. This deal is valued at forty billion dollars, so nearly nine billion more than what soft Bank paid back in I think, even if we adjust for inflation, soft Bank is making while it's making bank on this deal. So what is actually going on here? Well,

it's actually really astounding to me. So, according to in Vidia's CEO Jensen Huang, the plan is not to change the way ARM does things because that would endanger all those business partnerships that the company has formed over the course of its history. It considers the various companies it licenses,

it's i P two partners. So instead of messing with that, Wong says he wants in Video to kind of follow in ARMS footsteps, making in Vidio gpu tech anology and add on to ARMS i P. So companies that license with ARM could potentially licensed in Vidio gpu tech. So in other words, if I'm understanding this announcement correctly, and

it's it's entirely possible, I'm not. But it sounds like in Nvidia is opening up the chance for other companies, like fabrication companies to take Vidio designed graphics processing units and make their own versions of it, which actually sounds a bit crazy to me. I mean, in Vidio partners with fabricators in order to actually make their graphics cards.

They do that, you know, they design their graphics cards based on the capabilities of the fabricators they work with, but they actually are the ones behind that whole process, right, It's almost like they're their partnerships mean that they get to use the fabrication equipment of these other companies. It's more complicated than that, but you see where I'm getting at. Anyway, what I mean is that Video has more of a role in the production of the final product, whereas ARM

is all about selling or licensing intellectual property to other companies. Now, could this mean that in Video as a company is ultimately looking to transition out of being part of the whole fabrication process and to move into more of a design role. I don't know the answer to that. The only other alternative I can think of is that in Vidia is creating the opportunity for other companies PETE within Video, which also seems crazy to me. So I don't know

the answer to this. At the time of this recording, that deal is not done yet, it may not happen. Regulatory agencies could potentially end up blocking the move, though it's not necessarily likely because in Video and ARM do not directly compete with each other, so it's not like we're seeing the consolidation of a market here. We're not seeing, you know, two competitors in the same market space become

one single thing. I suspect. I'm going to have to do future episodes to follow up on whatever happens next, but for now, that is the Arms story. I hope you guys enjoyed this pair of episodes. UM, I'm starting to lose my voice, but I have to record one more episode after this, so I can't wait to hear what that sounds like. Right, guys, If you have any suggestions for topics I should tackle in future episodes, send me a message. You can get in touch with me

on Twitter. The handle is text stuff h s W and I'll talk to you again really soon. Text Stuff is an I Heart Radio production. For more podcasts from my Heart Radio, visit the I Heart Radio app, Apple Podcasts, or wherever you listen to your favorite shows.