The Nvidia Story So Far

go listen to that one first. Anyway, In Video's first product, the n V one, had not really sold super well in the market. It hadn't really captured a whole lot of market share, and the company had to cancel the in V two chip after it's big partners, Sega, announced it was going with a different chip designer for the

graphics for the upcoming dream Cast console. But the engineers at Nvidia rolled with the punches and they went back to the drawing board and designed a new graphics chip that would rely upon the humble triangular polygon rather than the quadrilateralals that the Envy one employed for texture mapping.

This would make their new card compatible with Windows based machines that were running the direct three D a p I, and the new card would include polygone texture mapping, which reduced the amount of time it took for the graphics card to render frames, but it had a negative impact on image quality. So what the heck is texture mapping? Does that even mean? It might sound like gobbledygook, Well it pretty much is what it sounds like. It involves

taking a two dimensional surface called a texture map. Think of it like wallpaper or wrapping paper. Think of a pattern on that, and then virtually wrapping a three dimensional object with that texture. So it's it's a virtual three dimensional object. So it's kind of like taking wrapping paper and wrapping it around a box. Let's say you've got a box, You've got a gift you're gonna give somebody,

You have some wrapping paper. You have to take this wrapping paper that's in a two dimensional format, that being a flat sheet of paper, and wrap it around a three dimensional object, that being a rectangular cubicle type box. And you have to fold the pattern around the contours of the box in order to wrap it, which is easier to do with simple geometric shapes, but as you introduce more complicated shapes, it gets harder to do very well. And it doesn't take very much to do to make

this really hard. I mean you can if you wanted to map a simple two D shape like a pattern of squares on a three D object like a sphere, you're gonna run into problems because if you took an actual sphere, like a physical sphere, and you had a wrapping paper that had boxes on it, like let's say it's just black and white boxes on it, a checked pattern almost, and you wrap that sphere with that wrapping paper, you you're gonna find that it's going to require you

to cut the paper in such a way that some point on the surface of the sphere the pattern would no longer be continuous. There'd be a break in the pattern. The same thing can happen with texture mapping, and it means that your graphic quality can suffer as a result. It can become noticeable. The political texture mapping strategy did require less processing power to cover three D objects with textures, however,

so you have a trade off. The quality of the graphics might not be quite as high, but you can render them much more quickly, which could improve things like the refresh rate. While in Video was working on this, another company came out with a product that had a big impact on the graphics card industry, and it also led to conditions that convinced me to get out of

PC gaming for quite some time. Not putting blame here, just that it created an environment that I got increasingly disenchanted with the company, and their product was not a bad one, by the way. The company was three d f X Interactive. It launched a year after Video had incorporated, so he launched in and they introduced a three D

accelerated graphics card called the Voodoo. And this card was different from other graphics cards that had come before it because up to that point, graphics cards would typically integrate two D and three D graphics processing on a single motherboard. But three d FX created a card solely dedicated to three D graphics. So you could not just have a Voodoo graphics card and a computer and have that be

your only graphics card and have that work. You actually had to make it work with in parallel with a two D graphics card or really in series, because what you would do is you would take a v g A cable from your video controller card, your your basic

graphics card in your computer. You would feed that cable into your Voodoo card, you know, plug it in on the back of where the expansion card is, and then you'd take another cable and you'd hook the Voodoo card to your monitor, so you effectively had two different graphics

cards doing work inside your computer. The Voodoo handling all the three D stuff, and it gave gamers a new option when building out computers to play games like Doom and Quake, which were progressively putting more emphasis on three D models, and they were pushing the boundaries of what PCs could do. At that time. Before the Voodoo, the most common way to keep up with gaming developments would be you'd go and get an upgrade to your CPU,

which get it really expensive. So I remember when my family went from having a two eight six IBM compatible to a four a D six, and then later on we got a Pentium. But these graphics accelerator cards meant you could just buy a graphics accelerated card for for much less money than it would cost to upgrade the whole CPU, and you could insert it into an expansion slot inside your computer, plug up a few cables, and provide the processing of you would need to have really

great graphics. But it also helped usher in an era in which new graphics cards would debut at an alarmingly frequent rate and make the previous cards obsolete very very shortly after they had debuted. Some were two D cards that also had three D accelerators on them, some of them were dedicated three D accelerators, like the Voodoo, and more games were coming out that would push the limits

on the fastest processors on the market. So it became this vicious cycle where you would have brand new cards coming out that were much more capable than the ones, so everyone wanted one because you thought, who faster means better, more features means better. Developers would start developing games that

would take advantage of those greater capabilities. So while it would seem in the very short term that you were getting a head start by getting one of these powerful cards, pretty soon developers would end up creating applications that would push those cards to their limits, and it just became this kind of seesaw approach. And that's when I decided I don't want to deal with this, and I stopped being a PC gamer for a while, and I switched

to consoles for for quite some time. I did eventually go back to PC gamer, but it was in this era of numerous graphics cards, and it weren't they weren't just graphics cards too, that you also had sound cards that were doing a very similar thing at the same time, and it was causing all sorts of issues, things like questions about compatibility and affordability. I just too frustrating for me.

But it was a boom time for PC gamers who were, you know, eager to pursue that hobby, and they weren't weren't put off by the idea that there was suddenly a ton of different products out there now. It did not help that by were looking at around seventy companies

that were marketing graphics cards of some sort. Some of them were working with technology that was compatible with the open g L standard, many more were developing cards that were specifically geared to support Microsoft's Direct three D Application Programming Interface, and more than a few had their own proprietary application programmer interface that that developers could use to create software that would look fantastic if it ran on

those cards. There was a ton of market confusion, not just for end consumers, but also for programmers who were making software in the first place. Well a Voodoo from three D f X. Actually it would officially debut in nine and the following year in Video would come out with their second commercial product. This was their follow up to the n V one. This was the n V three because the n V two was canceled. This one

was better known as the RIVA on. RIVA or r i v A stood for real Time Interactive Video and Animation Accelerator and this was a hundred twenty eight bit three D processor. That's why the one was in the name, and that means the processor and busses could operate on one eight bit integers or data units that were sixteen octets wide. This gave the RIVA a bandwidth of one point six gigabytes per second, which is the same thing

as twelve point eight gigabits per second. Incredible bandwidth. The processor was a big hit with O E M or original equipment manufacturer customers. So these were companies that either made computers or they may graphics cards and they would incorporate the REVA in them. It had four megabytes of

RAM and a two hundred six mega Hurts processor. Um. There was a faster model that came out shortly thereafter, called the Revo z X. That one had an additional four megabytes of RAM, bringing up to eight and had a clock speed of two hundred fifty mega hurts. This chip was sold to those original equipment manufacturers, and one of them was Diamond, which had also been one of the O E M s for the original n V

one chip. They introduced the Diamond of Viper three thirty which had a Reva one chip as its three D graphics accelerator, and I read some reviews of the revoight from that era. I actually found you can find reviews online the date back to when this chip came out, because the web was a thing in and the reviews were pretty positive. Critics were impressed with the three D performance.

The review is also said that, you know, obviously the quality of the performance didn't come just from in video, but also from the manufacturers that produced the other equipment, the the graphics cards themselves, like the motherboard for the the that the chip was sitting on, so it was more than justin video, but in video overall came out looking pretty good. From the Reva however, it did create a brand new problem for in Video but this one

was a welcome problem, not a bad one. It turned out the company was not able to keep up with demand because the Reva was a huge economic and critical success for the company, and so in Video would sign a manufacturing deal with T s MC to supplement manufacturing

that was already going on with their established partner. They had been manufacturing chips through a company called st Micro or Microsystems, and they decided to supplement that and use a second second partner to provide even more of these. By the end, in Video had gone from a struggling company on the virginal collapse to owning twenty fo of the market share in three D graphics processors. Not a bad turnaround, and as it turns out, they had higher

heights to climb. I'll say more in just a moment, but first let's take a quick break to thank our sponsor. At this stage, the market was starting to clear out a little bit. Some of the companies that had entered the graphics cards space we're starting to go out of business. But there were three companies that were essentially dominating the high performance graphics cards space. Those three companies were a

T I, three D f X and Nvidia. Some companies, in an effort to survive in this increase singly narrow market, began to focus on creating budget graphics cards that had fewer features, They had a slightly lower performance level, and they were generally of a lower cost, so that would put them in the range of a different market for gamers.

But in video was on pretty solid ground now. They wanted to follow up on the success of the Reva with their next card, which was the Riva TNT also known as the n V four T n T stood for twin Texel because this card actually added a parallel pixel pipeline. It's a lot of ps there to get the pop filter going, but this would double the rendering speed for the card, and it was a two D and three D graphics card. It had support for thirty

two bit true color graphics. They had expanded the RAM up to sixteen megabytes, and it was available both in PC I and A g P formats. Those would be the buses that you could use with personal computers, and there was two different formats for buses. The A g P was the newer one of the two and the superior one. As far as performance goes of the two,

but they in video. This time we're mostly focused on PC I simply because a g P had not rolled out to a very wide user base yet it was still new enough where it was not widely adopted as the news standard yet. The one downside to in Video's design was that the complexity of the chip they had made with the tn T meant they weren't able to run it at what they had originally planned with their

memory core speed. They originally wanted to have a memory core speed of a hundred twenty five Mega hurts, but they found out that because they had made the chips soaked or the card so complex, they had packed it with so much stuff that if they tried to run it at a Mega hurts, the card would over heat, so they were forced to throttle the speed down to nine d Mega hurts to avoid those overheating issues. The big competitor to the t n T was three D f x IS Voodoo Too card, which in benchmark tests

pretty much dominated the three D acceleration world. But the Voodoo Too still required a second two D graphics card to work. It was not a standalone all in one card, so if you were to add up costs. If you were building a brand new PC from the ground up,

you're buying all the different components. If you went with a Voodoo too, the costs actually would go up a bit because you had to have a secondary graphics card in order to run the Voodoo two, whereas in videos T and T wasn't all in one you could just buy one card and get all of that, So it all depended on what your needs were and what you already had at your disposal. Also, behind the scenes, three d FX was trying to head off competitors like in Vidio and A T I by marketing and selling their

graphics card boards themselves directly. So three d f X, in order to do this, purchased a company called STB Systems for a hundred forty one million dollars in the stock deal. But it turned out the cost of manufacturing ended up being higher with STB Systems foundry than the foundries that A T and nine and Video were using, So the the actual expenses of producing the boards was higher than competitors costs, and so that would eat into your profit margin. If you're still trying to price your

your cards competitively against A T I and Nvidia. So three d f x is misstep became in Video's opportunity because some of three d f x's former partners began to drift over to work within Video instead. But while in Video was doing well against his competitors, specifically through the t n T card, it was dealing with other issues that were not so great. Silicon Graphics Incorporated now s g I filed a lawsuit against Nvidia. The company alleged that in Vidia had infringed on a patent related

to texture mapping. That dispute would stretch from the spring of n all the way into nine but eventually the two companies would reach a settlement, and as part of that settlement, in Video would actually get s g i s professional graphics portfolio, and they hired on several of s g i s low level graphics team members, and SGI sort of transitioned out of the graphics business at that point and started to focus on other areas of business.

It never really worked out for that company. The company s g I was already reeling from other problems, and it would ultimately go into bankruptcy in two thousand nine. And Video released the t n T two and the t n T to Ultra the following year, which effectively rested the title of most powerful three D accelerator away from three D f X and the Voodoo three card.

At least in most bitchmark tests. Graphics cards are weird by the way you typically test these by running various tests that that check the different settings and different features of these graphics cards, and it's not always a one to one match. You don't always have features that measure up where there's an equivalent on a competing card. So depending on the test, you might have one card come out on top, and then you change tests and you run those same two cards through it, and the other

card comes out on top. So it's a little complicated to talk about, but this was a point where in Videous performance was look looked at as best in class,

at least at that moment. Uh. And also keep in mind, we're still talking about a world that had different A p I s out there that developers were using to build software that had three D graphics in it, and so sometimes you would find a program that would run really well on a computer that had a three D f X card, but not as well on a computer that had in video card, and then meanwhile you might get a different program made by a different developer that

the opposite is true. So it's very complicated. On October one, and Video would define a new era because it announced a new graphics card called the g Force two. It previously had the code name in V ten. This was the first chip to get the designation of GPU, or graphics processing unit. The very first GPU was an Nvidia product called the g Force two. So what was different about this card? What was it that merited the creation

of a new term. Well, it had a transformation and lighting engine which tackled floating point calculations to figure out how to render transforming virtual three dimensional objects and display them in a two dimensional representation of the image. Wait what all right? This is actually not as complicated as I just made it sound. In fact, a lot of you probably are way ahead of me. But just in case,

here's what I actually mean by that. Let's imagine for a second that you can pop into the virtual world of a computer, kind of like Entron. So you are in a virtual world, and in this virtual world you can see three dimensional objects. Right, there are three dimensional objects all around you. They truly have those three dimensions now, and in the computer world, that's all represented by math. They are just virtual objects. But let's pretend that we

are in that virtual world. We can see them, and we can see how those objects transform as they move through a space and environment. Maybe they change shape. Maybe it represents a character in a video game, and as the character moves, we see how this shape changes. We can see how light from fixed sources can play over the surfaces of these three dimensional images. We can see

where shadows form, where highlights are. Now, let's say that we we have these three dimensional images that exist in this virtual world, and now it's time for us to display those images to the real world. And our our method of doing that. Our medium is a computer display. But computer displays are not three dimensional. They are two dimensional. They have width and length, but they don't have depth.

And so you have to figure out how do I portray this virtual three dimensional image that has stuff like lighting effects going on with it in a way that makes sense within a two dimensional medium, so that you get the idea of what that object, how that object is shaped. Before the g fource two six, the CPU handled that sort of work, this transformation and lighting work.

But the CPU was also doing all the other work too, besides the other graphics processing, and that meant that if you added in too much detail in your game, everything would bog down and run super slowly or maybe even crash because the CPU would be so busy trying to calculate all the information required to display that kind of

stuff that it couldn't do anything else. But by moving that requirement over to the graphics card, the GPU could free the CPU up to do other calculations, and the bandwidth the GPU was super high, high enough to handle a massive amount of data, so as long as the processor was fast, it could go through this information very quickly.

In fact, that g Force two six had a throughput that was about five times greater than that of a five fifty Mega Hurts Pentium three processor, So for certain types of applications, a GPU is going to be much better than a CPU. Plus, the GPU had on board memory, it included a second set of parallel pixel pipelines that meant you had four total that was double the TNT design and game design developers now could create games with much more detail without fear of causing a total slow

down on a system. It was a literal game changer. The clock speed for the g Force two fifty six was technically slower than that of the Riva T and T two, so they actually took a step down in their clock speed. They didn't make it faster than the previous graphics card. However, the improvements they made in the architecture meant that the g Force two fifty six still performed at at faster speed than the t n T two,

even though the clock speed was slower. Now, up to this point, I've gone through the Nvidia cards more or less one by one because they were really important for the foundation and success of the company and its rise from startup to industry giant. But from this point forward, I'm going to be jumping around a bit. Otherwise, all I'm gonna be doing is giving you a long list of graphics cards and their specifications, and that is not

terribly interesting. It's like reading a catalog. However, around this time, things were still in that pretty chaotic stage in the gaming rig world. You had a small number of companies that were dominating the graphics cards scene at the very top, but there were lots of different budget companies out there. There were other companies that would occasionally create a high performing graphics card to try and compete with a t I three D f X and Nvidia, And there were

also the companies that were specializing in sound cards. There were the different application programming interfaces, there were the different drivers that were meant to try and make more compatibility between different hardware and software. It was still a huge, confusing mess for a lot of people, including myself. I remember the days of looking at a game and wondering if I could even play it because it might not be compatible with one or more of the components inside

my PC. It was a real mess. And not all the companies that were around at that point would serve five, but in Video would. Now, when we come back, i'll give you a little last look at what in Video has been up to, and we'll kind of rush through the rest of its of its history to kind of give you the highlights. But first let's take a quick break to thank our sponsor. In late nine and Video

announced a new line of products called Quadripe cards. These were meant not for gaming rigs, but for professional workstations, so in videos branching out. At this point, the GPU was showing that it had some interesting uses outside of rendering graphics for gamers, although at this point it was really still about high end graphically intensive workstations. But speaking of games, and video also landed a pretty sweet contract

with Microsoft. Microsoft was developing its own game console, which at that point was still a super secret project and it wouldn't come out until November two thousand one. Obviously, that was the original Xbox, and Microsoft went in Video to build the graphics chip that would be inside the system. So the initial payment for that chip, which was called the n V two A internally at Nvidia, was two hundred million dollars, healthy chunk of change, one might even say,

a princely some. By the end of the fiscal year, in Video had made a profit of forty one million dollars, which was ten times more than the previous year's profit. That's a heck of a lot of growth. It's really nice to be able to point to a number and say we made ten times as much money this year, not not ten times the revenue, ten times the profit.

The following year saw in Video sweep up a former competitor, or begin to anyway, three d f X, which had transformed the PC gaming industry in the mid nineties when they launched the Voodoo three D accelerator car had essentially

started to burn itself out. It had made some bad acquisitions, uh, some of the cards that had created just weren't performing well in the marketplace, and in Video made a move to acquire three D effects, mostly for the intellectual property and for some of their engineers, and it began the acquisition process starting at the tail end of two thousand.

But this acquisition process was incredibly painful and troublesome. There were a ton of different court cases that came out from creditors who said that, you know, three D f X had owed them a lot of money and therefore they deserved a larger slice of cash from the acquisition. So that process did not officially conclude until two thousand eight, So it's like an eight year acquisition process that's pretty painful.

At this point, A t I was still the dominant name in graphics cards in Video was in second place, but that would quickly switch where in Video was finally able to kind of overtake A t I. The follow up of the G Force two fifty six was the g Force two just seems like a prequel to me, but anyway, g Force two line of cards were the first from Nvidia to support multiple monitor configurations, which is

something we take for granted these days. Then after that came the g Force three that debuted in February two one and in video introduced some new technologies with these cards, implementing processors saving features, like one called light Speed Memory Architecture or l m A. Essentially, what lm A did two pixels is the same thing that MP three files due to sound. It's a it's sort of a compression strategy. So what I mean by that is MP three compression.

One of the strategy is MP three uses is it analyzes the sound within a file and it says, anything that appears to be outside the range of human hearing, we're just gonna get rid of that data. We don't need it because no one could hear it anyway. If you can't hear it, we don't need to save it. And that save space, right, We aren't saving all that extraneous information. L m A does the same thing, except

with pixels. It would indicate like, well, these pixels are being blocked on the screen by some other component, so there's no need to save this information because those pixels are not going to be displayed. Let's say it's pixels that represent a character and the character passes behind something that is uh blocking your view of the character. Well, there's no reason to keep the information of the pixels that relate to the character at that point because you

can't see it anyway. So uh. It was called a process called z occlusion culling, and it would help save save bandwidth because you don't need as much You're not sending as much data through, so you can, uh you don't have to worry about taking up so much bandwidth space. This marked a turning point for in video. This was the point wherein Video overtook a TI as the industry leader. Actually, there was a couple of shifts here. At this point,

in Vidia commanded of the graphics market. Intel was in second place with and a t I had fallen to third place. Now that would not remain steady over the next few years. Companies would swap places a couple of times, but it did show that in Vidia had now become a truly dominant player in this space. Then we get the g Force four that came out in February two.

There were several budget models that came out along with that series, and they were in a line called m X. They were actually focused or built largely upon the architecture from g Force two, not g Force four. So even though they were part of the g Force four series, they were based on an on an earlier and some would say outdated architecture, and they had limited features, but

they also had a lower price point. So here in Video was saying, I'm seeing all these other companies coming out with budget graphics cards aimed at a different target audience. What if we win after those people and we just start creating budget, lower feature graphics cards. And that's sort

of what this this was. Fortune would name in Video the fastest growing company in the United States in two thousand two, and in two thousand three, in Video acquired a company called media Q that specialized in graphics and MultiMate technology for wireless devices. Is where we start seeing

in Video getting into mobile technology as well. This is also the year when in Video co founder Curtis Prium retires, and it's also the year and Video would release the first cards in its g Force f X series, including one that became infamous. The g Force hundred, had a nickname in the industry, it was the dust Buster. Why well, the dust Buster or g Force fifty hundred had a dual slot cooling mechanism double and cooling takes up two slots,

high powered, very loud. So people said, oh, it sounds like that you've got a vacuum cleaner inside your computer, because it's it's so distracting, so loud. This was a a piece of technology that was built in part by some of the engineers that in video had pulled over from three Dfx, and that it got a lot of ribbing in the industry for the fact that it was it was so loud, but others in the f X

line would end up getting some acclaim. The company created a demo as well in this in two thousand three. The demo was called Dawn d A w N and it was all meant to show off the various features of what the the g Force Effex line of graphics cards was capable of doing. And the way it did this was it showed a female ferry, little wings, translucent wings,

translucent outfit. Um. The the naughty bits were covered by opaque costuming, but the rest of it was see through and the ferry was very curvacious a lot of people called out a video saying it was the most blatant use of sex appeal to try and sell a graphics card up to that date. UH, So they got some criticism for their use of this particular visual effect in

order to try and sell cards. In two thousand four, s l I technology would debut, and that would allow users to link together multiple GPUs to boost performance even further. That it's where you start seeing these crazy gaming rigs that have multiple GPUs in link together, plus typically some sort of crazy cooling system, because as you start adding more CPU and GPU power to a system, it's going to generate more heat, and heat and electronics don't go

well together. If they get hot enough, things tend to break down. So there are a lot of innovative cooling systems that came out around this time as well to try and keep those temperatures at at nominal levels so things would still operate well. In two thousand five, in Video would UH get the contract to develop the graphics processor for Sony's PlayStation three console. So they had already

developed the one for Xbox. Now they were developing the one from PlayStation three, and in two thousand six and Video introduced the seventy x two, which was the first of in videous cards to feature two GPU single boards packaged as a single product, so it was a dual GPU all in one package. On October two thousand six, a m D would acquire a t I, the competitor to in video. That acquisition was a huge deal. Is

like five point four billion dollars, an enormous deal. And now it was no longer in video versus a t I. Now it was in video versus a m D. And remember that H. Jensen Huang, the president and CEO of Nvidia, the co founder of Nvidia, had you had worked for a m D for a while. In June two seven, and Video introduced a line of math coprocessors called Tesla.

Now these initially used architecture that was originally built for the g Force and Quadrille products, and they are general purpose GPUs because now there was this emerging opportunity to use GPUs to tackle other problems besides rendering graphics. GPUs can power supercomputers, and they can perform calculations that have high floating operations per seconds or or flops. They can handle huge amounts of data and and and perform lots

of operations in a short amount of time. The big difference between the Tesla GPUs and the ones that we're powering gaming rigs is that the Tesla versions didn't have any output to displays because that's not what they were for. They weren't meant to create graphics. They were meant to crunch numbers. And uh, I think this really does mark the beginning of using GPUs to do some pretty phenomenal stuff.

And we see it and a lot of parallel processing applications where you have these problems that can be broken up into smaller parts. A lot of GPUs are really good at handling that kind of stuff. The g Force eight series would come out from Video and included chip sets that had high lead solder and inadequate cooling mechanisms. So there were a substantial number, not all of them, but a substantial number of cards that had g Force eight series chips in them that ended up suffering failures

because of overheating issues. This led to in Vidia writing off about a couple of hundred million dollars worth of product. Not to mention it led to lawsuits that alleged that the CEO Jensen Huang and the CFO Marvin Burkett were aware of these manufacturing problems, but they didn't say anything

about it. They attempted to go forward despite these issues, and according to one source, the full amount that the the company incurred as a cost relating to this problem ended up being around four hundred seventies six million dollars, and one of the UH sources I was reading said, it's unfortunate that this happened. It ended up being such a black mark against in Video's reputation that a lot of

people to this day hold it against the company. Um because it was it came across is so deceptive and um and see like it was exploiting the customer base, and Video would stumble again, at least according to a

lot of critics. With the G Force nine series, a lot of people said that it didn't really have very much very many innovative factors, but in Video would introduce new micro architecture code named FIRMI in with a launch of a news series, the G Force four series, which did not It didn't equal the performance of a m d S flagship cards, but it was more affordable and

it performed well enough. It just wasn't at the same level of metrics that a m d S flagships were performing at, and in Video had also throttled the clock speed of the cards on that on that series, largely in order to make sure that power consumption stayed at nominal levels. But it did mean that if you were a serious gamer and you didn't mind getting your your hands a little dirty, uh metaphorically speaking, you could over clock your GPU. You could make it run faster than

what it was supposed to run. Because in Video had been very conservative and had throttled, the actual top speed of the processing units was much higher than what in Video put it out to be, So if you wanted to, you could overclock it get some better performance out of it. And Video followed up for me with other micro architecture designs you had Kepler had, Maxwell, Pascal and Turing and just in general, the way micro architecture tends to work.

Intel does this as well, is that you get what Intel calls the TICK talk approach, where in TICK you create a new layout of your of your microprocessor that has smaller components. It's when you've shrunk things down so that you can fit more individual components on the same amount of chip space as the previous generation. So UH think of it as like cramming more employees into the same office space. And and so you can keep doing that, especially if you get smaller employees right they take up

less space individually. Then the tox side is where you end up fine tuning the layout of the office space to accommodate the fact that you've crammed more employees in there. So when I talk about these different micro architecture designs, it's a combination of things. It's creating dye sets that have smaller components so you can cram more of them onto that chip, and it's rearranging the orientation of those

components so that they operate at a greater efficiency. And so each of those changes in micro arch texture represent an attempt to produce more power more efficiently than the previous generation of chips. And that's generally how we see all microprocessors, whether they're CPUs or GPUs, how they how they continue and we're still seeing it today in video is still one of the top names in graphics cards and UH their CEO, Jensen Wong, is worth more than

six billion dollars. He is the sixty one highest paid CEO. Not bad for somebody who back in said yeah, why don't we give us a shot? So a pretty amazing story. And like I said, I really skipped over quite a

lot in this last section. You know, I didn't go card by card because frankly, to go through all those specifications and explain their relevance would mean having to do companion episode odes about the evolution of graphics in UH, in computers in general, UH, and programming and the whole thought process behind graphics cards as well, and it would

require a lot more, a lot several more episodes. The important thing to remember now is that in videos not just making these graphics cards for for gamers, although they still do that obviously, but that they have branched out into other areas. GPUs these days are being used in all sorts of fields, including artificial intelligence and deep learning,

machine learning, that kind of stuff. So we're seeing a lot more relevance of GPUs, also used in bitcoin mining, u in in UH cryptography, we're seeing tons of applications of GPUs far beyond making Laura Croft look amazing as she runs through the jungle. So prey the phenomenal story. I hope you enjoyed this episode or these episodes about Nvidia. Uh. In our next episodes, we're gonna cover totally different topics, so stick around. If you guys have suggestions for future topics.

Maybe there's a company I should cover, a person in tech, a technology in particular that you would like to hear more about. Send me an email the addresses tech stuff at how stuff works dot com or drop me a line on Facebook or Twitter to handle with both of those. Is tech stuff h s W. Go to T public dot com slash tech stuff that's T E E Public dot com slash tech stuff for all your tech stuff

merchandise needs. We've got some cool stuff over there, and every purchase helps out the show, so we appreciate it, and don't forget follow us on Instagram and I'll talk to you again really soon for more on this and bathands of other topics. Is that how stuff works dot com