WEBVTT - The End Of Moore's Law? (Not Really) 0:00:00.160 --> 0:00:07.280 Brought to you by Toyota. Let's go places. Welcome to 0:00:07.440 --> 0:00:15.360 Forward Thinking. Hey there, and welcome to Forward Thinking, the 0:00:15.520 --> 0:00:17.920 podcast that looks at the future and says out on 0:00:17.960 --> 0:00:23.480 the wily Misty Moore's I'm John Constrictland and I'm Joe McCormick, 0:00:23.520 --> 0:00:26.720 and our other host, Lauren vocal baumb is not with 0:00:26.800 --> 0:00:30.360 us this time, but hopefully she will be back next time. Jonathan, 0:00:30.440 --> 0:00:34.599 I I suspect you of equivocating on more maybe a 0:00:34.640 --> 0:00:37.440 little bit, teeny tiny bit. Okay, So today we are 0:00:37.479 --> 0:00:39.559 going to be talking about Moore's Law. It won't be 0:00:39.560 --> 0:00:41.720 the first time on this podcast, probably won't be the 0:00:41.800 --> 0:00:45.280 last time, because it contributes very much to how we 0:00:45.400 --> 0:00:49.640 grapple with the future of consumer electronics and computer technology. 0:00:49.640 --> 0:00:53.440 And we've used, I mean we being the humans in general, 0:00:53.520 --> 0:00:56.720 have used Moore's Law to kind of be shorthand for 0:00:57.080 --> 0:01:01.360 the progression of computational power for pretty much as long 0:01:01.400 --> 0:01:05.840 as it's been around. So we are specifically focusing on 0:01:05.920 --> 0:01:07.560 how it's gonna go through a little bit of a 0:01:07.560 --> 0:01:11.080 transformation in the not too distant future. To quote MST 0:01:11.200 --> 0:01:14.360 three K. Jonathan recently wrote a piece for How Stuff 0:01:14.360 --> 0:01:18.280 Works Now about a recent development in the world of 0:01:18.319 --> 0:01:21.360 people thinking about Moore's law. Yeah, so first, before we 0:01:21.400 --> 0:01:24.440 even get into that, we should probably chat a little 0:01:24.440 --> 0:01:27.080 bit about what Moore's law is. The observation was originally 0:01:27.120 --> 0:01:31.360 made back in nineteen sixty five. That's when Gordon Moore 0:01:31.880 --> 0:01:34.039 came up with this. Uh, well, it really was just 0:01:34.080 --> 0:01:36.600 an observation, right, It wasn't a law. He was he 0:01:37.080 --> 0:01:40.120 noticed something and he wrote a paper about it. But 0:01:40.200 --> 0:01:44.360 maybe we should make it a law. Well, we certainly have. No, 0:01:44.520 --> 0:01:46.840 you're right that it is not a physical law. It 0:01:46.959 --> 0:01:50.400 is more a yeah, an observation or a prediction. Right, 0:01:50.440 --> 0:01:53.480 It's just a prediction that has held more true than 0:01:53.560 --> 0:01:56.480 not throughout the years, and so people referred to it 0:01:56.560 --> 0:01:58.680 like a law, kind of like Murphy's law. Murphy's law 0:01:58.720 --> 0:02:00.720 is not a real law. Moore's law is not a 0:02:00.760 --> 0:02:04.400 real law. But when Gordon Moore made that observation back 0:02:04.440 --> 0:02:07.480 in five, he was the director of research and development 0:02:07.520 --> 0:02:10.360 for fair Child Semiconductor. He was also a co founder 0:02:10.400 --> 0:02:12.520 of fair Child. He would go on to be a 0:02:12.560 --> 0:02:15.600 co founder of Intel. So here's a guy who was 0:02:15.760 --> 0:02:21.600 really really smart about integrated circuits. Uh, integrated circuits were 0:02:21.760 --> 0:02:24.919 a relatively new thing because transistors had not been around 0:02:24.960 --> 0:02:28.120 that long before then you're talking about vacuum tubes. And 0:02:28.200 --> 0:02:31.400 so he was taking a look at this and seeing 0:02:31.440 --> 0:02:35.600 that there was an interesting situation going on. He wrote 0:02:35.600 --> 0:02:38.680 an article for Electronics magazine and it has one of 0:02:38.680 --> 0:02:42.840 my favorite titles of all time for technical uh articles, 0:02:42.880 --> 0:02:48.560 it's cramming more components onto integrated circuits, because there's nothing 0:02:48.600 --> 0:02:52.519 delicate about that cramming. Cramming is such a great word. 0:02:52.760 --> 0:02:55.000 You get this image of Gordon Moore with a bag 0:02:55.120 --> 0:02:57.959 full of components and he's reaching in and pulling out 0:02:58.000 --> 0:03:00.640 fistfuls and trying to cram them in to a machine. 0:03:00.800 --> 0:03:03.320 And maybe he has like a real large wooden mallet 0:03:03.360 --> 0:03:06.680 sitting by just in case. Yeah, but that's not quite 0:03:06.880 --> 0:03:11.560 the actual way that this is being done. But he 0:03:11.600 --> 0:03:14.240 was talking in fact, I think he was talking as 0:03:14.320 --> 0:03:17.440 much about economics as he was about technology. In fact, 0:03:17.440 --> 0:03:21.080 he was talking more about economics. Yeah, more m O 0:03:21.240 --> 0:03:24.079 r E, not m o O r E in this case. Yeah, 0:03:24.120 --> 0:03:29.480 he was talking about economics primarily because he wanted to see, uh, 0:03:29.760 --> 0:03:33.520 where the price point fell on an individual component, At 0:03:33.520 --> 0:03:36.920 what point in a in the density of a an 0:03:36.920 --> 0:03:41.400 integrated circuit. Do you have the best price per component 0:03:41.480 --> 0:03:44.880 so that your your circuit that you're done with is 0:03:45.000 --> 0:03:48.480 at its lowest cost, right, the lowest price to to 0:03:48.880 --> 0:03:52.680 make it. And there's a certain rule that he saw 0:03:52.800 --> 0:03:55.840 that you had to hit a certain volume in order 0:03:55.840 --> 0:03:57.360 for that price to come down. It's kind of like 0:03:57.400 --> 0:04:00.160 if you're buying in bulk. Yeah, you know, if you're 0:04:00.160 --> 0:04:02.520 buying in bulk, you know, you buy more the individual 0:04:02.600 --> 0:04:07.160 cost of each individual item is lower. But that's why 0:04:07.200 --> 0:04:09.560 it makes so much more sense to buy that barrel 0:04:09.640 --> 0:04:12.880 of cheese balls, small little can of cheese ball by 0:04:13.000 --> 0:04:17.760 by the hot tub sized mayonnaise jar and put it 0:04:17.800 --> 0:04:21.160 in the garage. Yeah, it's always a great idea. Uh. Well, 0:04:21.360 --> 0:04:24.279 he noticed that there was a certain volume at which 0:04:24.680 --> 0:04:28.720 the the components would be at their ideal cost. But 0:04:29.000 --> 0:04:31.000 it was a sweet spot, like if you went too 0:04:31.120 --> 0:04:34.840 much over that, then the finished integrated circuit would be 0:04:34.880 --> 0:04:38.280 too expensive. If it were too too much under that, 0:04:38.360 --> 0:04:41.000 it would also be that the individual price per component 0:04:41.040 --> 0:04:43.800 would be too high, thus making the finished circuit too expensive. 0:04:44.480 --> 0:04:46.200 But he also knows that not only was there a 0:04:46.200 --> 0:04:49.680 sweet spot, that sweet spot moved over time. There was 0:04:49.720 --> 0:04:55.000 an incentive for companies to develop more powerful processors. There 0:04:55.040 --> 0:04:58.080 was a need in the marketplace, which meant that there 0:04:58.120 --> 0:05:02.039 were economic factor is that pushed the development of more 0:05:02.080 --> 0:05:08.159 sophisticated um approaches to manufacturing transistors and and UH and 0:05:08.279 --> 0:05:12.200 integrated circuits. That meant that the individual component prices began 0:05:12.200 --> 0:05:15.240 to shift over time. So at the time that he 0:05:15.279 --> 0:05:17.960 made this observation, he said that the ideal number of 0:05:18.000 --> 0:05:22.640 components for an integrated circuit was fifty. Folks were in 0:05:22.680 --> 0:05:25.839 the billions now, but at the time it was fifty 0:05:26.000 --> 0:05:29.560 fifty components. That was where you hit the lowest price 0:05:29.720 --> 0:05:33.400 per component. But he said, this was twice as good 0:05:33.440 --> 0:05:36.400 as the year before, and then that and that was 0:05:36.440 --> 0:05:39.279 twice as good as the year before it. So he said, well, 0:05:39.320 --> 0:05:42.400 I think this is going to continue. I project that 0:05:42.480 --> 0:05:45.840 within five years the lowest cost per component on an 0:05:45.839 --> 0:05:48.640 integrade circuit would be realized with circuits having around a 0:05:48.720 --> 0:05:54.320 thousand components. So to the day he makes the observation, 0:05:54.360 --> 0:05:56.680 it's fifty five years from then, he's saying it's going 0:05:56.720 --> 0:05:58.479 to be a thousand that we're just gonna see this 0:05:58.520 --> 0:06:01.040 trend to continue. He said, there's no reason you can't 0:06:01.040 --> 0:06:04.120 project this out further, and he said that by the 0:06:04.160 --> 0:06:06.760 time you would get to nineteen seventy five, it would 0:06:06.760 --> 0:06:10.320 be sixty five thousand combinans. He was talking about it 0:06:11.040 --> 0:06:15.800 essentially doubling more or less doubling every single year. He says, 0:06:15.839 --> 0:06:19.400 eventually you would probably have to make some adjustments. You 0:06:19.480 --> 0:06:23.159 might hit a point where this this trend continues, but 0:06:23.240 --> 0:06:26.640 it slows down. And in fact, that is what we saw, right. 0:06:26.680 --> 0:06:29.720 We saw it go from uh, every twelve months to 0:06:29.800 --> 0:06:32.720 really more like between eighteen and twenty four, depending upon 0:06:32.800 --> 0:06:37.440 the times span you're looking at um. And that was 0:06:38.000 --> 0:06:40.520 incredible that he made this observation and that it was 0:06:40.560 --> 0:06:44.200 so prescient and so accurate. And that's why we often 0:06:44.200 --> 0:06:47.360 call it Moore's law, although we don't necessarily I think 0:06:47.440 --> 0:06:53.000 the wider interpretation isn't focused so heavily on the economic 0:06:53.120 --> 0:06:57.039 side of things. So as a point of illustration of 0:06:57.360 --> 0:07:01.120 exactly what this looks like in terms of the hardware 0:07:01.120 --> 0:07:03.479 in our devices. There was a great example I found 0:07:03.520 --> 0:07:06.880 in a piece that was in The Economist, and it 0:07:07.080 --> 0:07:11.400 highlighted this fact. When Intel released its first microprocessor in 0:07:11.480 --> 0:07:14.320 nineteen seventy one, this was called the four zero zero 0:07:14.360 --> 0:07:19.400 four and the four thousand four That chip was twelve 0:07:19.480 --> 0:07:24.280 square millimeters and it had two thousand, three hundred transistors 0:07:24.320 --> 0:07:29.480 and the gap between each transistor was ten thousand nanimeters, 0:07:29.480 --> 0:07:31.640 which is about the width of a red blood cell. 0:07:31.760 --> 0:07:33.960 They say, you know, a kid with a decent microscope 0:07:34.200 --> 0:07:40.800 could see the individual transistors. Intel's Skylake chips in six 0:07:41.880 --> 0:07:45.440 are a little bit different. These chips are the spacing 0:07:45.600 --> 0:07:51.400 of fourteen nanimeters between transistors, so that's ten thousand nanimeters 0:07:51.480 --> 0:07:55.840 to fourteen and they're not only invisible to the naked eye, 0:07:55.880 --> 0:07:59.800 they're they're even invisible to any normal microscope right light. 0:08:00.520 --> 0:08:03.320 The wavelengths of light are too big for us to 0:08:03.320 --> 0:08:04.840 be able to see this. You have to use like 0:08:04.880 --> 0:08:10.000 a scanning microscope, electron scanning microscope. So here, here's two 0:08:10.040 --> 0:08:12.960 factors that are important in this. Like one is that 0:08:13.040 --> 0:08:15.960 the individual components, we were able to shrink them down 0:08:16.520 --> 0:08:19.720 to increasingly smaller sizes, which is a weird way of 0:08:19.720 --> 0:08:22.600 putting it, but it gets a point across, and we 0:08:22.600 --> 0:08:25.040 were able to pack them together more densely. We were 0:08:25.080 --> 0:08:29.320 able to decrease the spaces between those components. Because we 0:08:29.440 --> 0:08:33.440 both both miniaturization and architecture. Right, we're getting really, really 0:08:33.440 --> 0:08:37.679 good at various types of lithography and other methods of 0:08:38.360 --> 0:08:42.520 developing and laying out these these circuits. Um and that 0:08:42.600 --> 0:08:47.400 was what allowed us to continue on to a ridiculous degree. 0:08:47.720 --> 0:08:50.120 When you think about it, I mean you're talking about 0:08:50.120 --> 0:08:53.800 from a time where it was fifty components to literally billions. 0:08:53.840 --> 0:08:57.760 Now it's it's hard to wrap your mind around it. 0:08:58.160 --> 0:09:01.840 And again this was really all about how More was saying, 0:09:01.840 --> 0:09:04.559 this only makes sense if it makes economic sense. Right, 0:09:05.760 --> 0:09:10.480 It will only work to the point where companies can 0:09:10.800 --> 0:09:14.080 do this and have it be a product that they 0:09:14.080 --> 0:09:16.679 can make a profit selling if it gets to a 0:09:16.679 --> 0:09:20.840 point where it's too difficult to do. Too difficult translates 0:09:20.840 --> 0:09:23.880 to too expensive. Right, the harder something is to do, 0:09:23.960 --> 0:09:25.840 the more expensive it is to do it. Well, there's 0:09:25.880 --> 0:09:27.720 all kinds of stuff you can do in the lab, 0:09:27.840 --> 0:09:31.160 and it just doesn't make sense from a consumer perspective. Yeah, 0:09:31.240 --> 0:09:33.160 and it if it's one of those things where you 0:09:33.200 --> 0:09:35.360 get to a point where it's just too much trouble, 0:09:35.400 --> 0:09:39.320 then what it's worth then Moore's law breaks down, because 0:09:39.360 --> 0:09:42.719 again it's not that it's technically impossible, it's that it's 0:09:42.800 --> 0:09:48.400 economically not productive. And if it's not economically productive, no 0:09:48.440 --> 0:09:51.680 one's going to no one's gonna lose money just trying 0:09:51.720 --> 0:09:57.000 to uh to get across an engineering hurdle that has 0:09:57.040 --> 0:09:59.600 been problematic. It's sort of like saying, you know, could 0:09:59.600 --> 0:10:02.760 you in near me a car that goes three hundred 0:10:02.800 --> 0:10:06.440 miles per hour? You know, I bet somehow if we 0:10:06.440 --> 0:10:09.480 were gonna be willing to pour all of our resources 0:10:09.480 --> 0:10:12.920 into that, we could do that. Just why, there's no 0:10:13.000 --> 0:10:15.520 reason to do that. It doesn't make any economic sense. 0:10:15.520 --> 0:10:18.920 There's no economic imperative for it. I can do that easily. 0:10:19.240 --> 0:10:21.559 You just have to find a height tall enough to 0:10:21.640 --> 0:10:25.360 drop the car off. And then I mean, you know, 0:10:25.520 --> 0:10:27.640 I'm just say it. Technically I can get it up 0:10:27.679 --> 0:10:30.520 to three hundred miles per hour. Yeah, has you're not 0:10:30.520 --> 0:10:33.360 gonna be able to do anything with it. Well. You also, 0:10:33.559 --> 0:10:36.400 if you were to make a car, a consumer car 0:10:36.480 --> 0:10:38.880 that goes three hundred miles per hour, you you'd introduce 0:10:39.000 --> 0:10:42.960 problems based sort of on physics and the limitations of drivers, 0:10:43.000 --> 0:10:45.440 Like I would say, you'd, how should I put this? 0:10:45.520 --> 0:10:49.400 You'd you'd increase the error rate of that car, and 0:10:49.440 --> 0:10:53.360 that in fact leads us into one of the fundamental 0:10:53.400 --> 0:10:56.720 issues that are that we face now in our production 0:10:57.360 --> 0:11:01.280 of these incredibly complicated microprocessors. Now, one thing I should 0:11:01.280 --> 0:11:06.560 say is that we haven't necessarily seen the number of 0:11:06.559 --> 0:11:11.960 components double every eighteen to twenty four months over the 0:11:11.960 --> 0:11:14.480 past few years. Right, That was the original definition of 0:11:14.480 --> 0:11:17.960 Moore's law, though in MOR's law does still seem to 0:11:18.000 --> 0:11:21.000 hold in a more abstracted sense, which is the way 0:11:21.040 --> 0:11:24.040 it trickles down to the consumer, is that you can 0:11:24.080 --> 0:11:27.800 expect every eighteen to twenty four months, you're computing devices 0:11:27.840 --> 0:11:31.320 will be twice as fast, twice twice as fast, or 0:11:31.440 --> 0:11:35.720 the processing power is twice as much some some variation 0:11:35.960 --> 0:11:38.240 of that. That tends to be how we define More's 0:11:38.320 --> 0:11:40.439 law now, is that every eighteen to twenty four months, 0:11:40.480 --> 0:11:44.080 like if you buy a computer today, in two years, 0:11:44.120 --> 0:11:46.400 the computer you buy that day will be twice as 0:11:46.400 --> 0:11:49.400 powerful as the one that you just bought, And that 0:11:49.520 --> 0:11:53.200 it just shows the rapid development of technology and the 0:11:53.280 --> 0:11:57.440 speed at which we can improve processing. That that tends 0:11:57.440 --> 0:11:59.320 to be how we focus on More's law. But even 0:11:59.360 --> 0:12:01.280 that is starting to get difficult. So even when we 0:12:01.320 --> 0:12:02.880 got to the point where we were no longer saying 0:12:02.920 --> 0:12:05.000 all right, We're not not so much worried about how 0:12:05.000 --> 0:12:09.079 many components are you adding, but how you how you 0:12:09.480 --> 0:12:12.360 best utilize those so that you get the most out 0:12:12.360 --> 0:12:15.880 of them. We still run into problems. So one of 0:12:15.920 --> 0:12:19.120 the things we've seen is that we've seen companies like 0:12:19.200 --> 0:12:22.920 Intel take a TikTok approach. That's what they actually call 0:12:23.000 --> 0:12:28.320 their strategy when they're creating new types of microprocessors and UH. 0:12:28.440 --> 0:12:33.480 The TikTok approach means that you have two different generations 0:12:33.480 --> 0:12:37.760 of processors UH that are kind of piggybacked onto each other, 0:12:37.800 --> 0:12:41.280 and in fact they chain up to the previous generations. 0:12:41.920 --> 0:12:46.840 So the TICK generation is not blue and nine vulnerable 0:12:46.880 --> 0:12:50.439 and has a sidekick named Arthur. The TICK generation is 0:12:50.720 --> 0:12:55.240 when you shrink down those components to a smaller size. 0:12:55.280 --> 0:12:58.520 We're talking on the nano scale now, right Like it 0:12:58.640 --> 0:13:02.520 used to be that forty now cometers was considered the 0:13:02.520 --> 0:13:06.200 the super small components. Now we're talking about getting down 0:13:06.240 --> 0:13:11.800 to like less than ten nanometers per component. That's insanely small. 0:13:13.080 --> 0:13:15.760 When you get down to that size, that becomes the 0:13:15.760 --> 0:13:17.920 TICK where you say, all right, we're gonna build it 0:13:17.960 --> 0:13:23.440 on the same architecture as the previous generation's microprocessors. But 0:13:23.600 --> 0:13:25.760 now everything is just smaller, so we can pack stuff 0:13:25.840 --> 0:13:30.000 in more more pieces in so we're using the same chassis, 0:13:30.200 --> 0:13:34.120 but now we've got way more components in there. The 0:13:34.160 --> 0:13:37.080 talk is when they figure out, hey, now we know 0:13:37.160 --> 0:13:39.800 how to design all those little components so that they 0:13:39.840 --> 0:13:42.880 work best they work, and because the architecture from one 0:13:42.880 --> 0:13:46.600 generation to the next may not be the most efficient, right, like, 0:13:47.040 --> 0:13:50.080 you may need to change that, especially for stuff like 0:13:50.120 --> 0:13:55.480 heat dispersal. Heat is a real problem with microprocessors. So 0:13:55.760 --> 0:13:57.840 the talk that would be when you figure out, all, right, 0:13:57.960 --> 0:14:00.400 here's the architecture that works best with this eyes and 0:14:00.400 --> 0:14:02.200 then you would go to the next tick, which means 0:14:02.280 --> 0:14:04.880 everything is even smaller, and then the next talk where 0:14:04.880 --> 0:14:08.600 everything has been laid out in the most ideal way possible. 0:14:09.480 --> 0:14:12.800 So uh so here's a question though, Yeah, is that 0:14:12.840 --> 0:14:16.200 TikTok of the clock counting up or counting down that's 0:14:16.240 --> 0:14:21.040 coming down? Buddy? So we're hitting some fundamental limitations just 0:14:21.160 --> 0:14:24.640 because of our good old buddy physics, right, Like, you know, 0:14:24.720 --> 0:14:26.760 it's not like the nano scale is as small as 0:14:26.800 --> 0:14:28.960 it gets. You can go down to the atomic scale, right. 0:14:29.440 --> 0:14:31.040 The only problem is that as you're getting down to 0:14:31.080 --> 0:14:34.080 the nanoscale and the atomic scale, something else starts to 0:14:34.160 --> 0:14:38.520 play a role in your designs, and that's quantum physics. 0:14:39.160 --> 0:14:41.760 You don't need to worry about that. On the classical side. Really, 0:14:41.840 --> 0:14:44.960 quantum physics at that point becomes negligible. You're not you're 0:14:44.960 --> 0:14:48.120 not so much worried about weird quantum effects to bring 0:14:48.120 --> 0:14:50.000 back the car. I mean, you don't have to go 0:14:50.200 --> 0:14:54.080 to relativity to understand the physics of designing a car, right, Yeah, 0:14:54.160 --> 0:14:56.520 we we don't have components on our cars that go 0:14:56.640 --> 0:14:59.000 down to such the nanoscale where you have to think, oh, 0:14:59.000 --> 0:15:02.120 wait a minute, quantum tunneling. But we do have to 0:15:02.120 --> 0:15:05.840 worry about it with microprocessors and quantum tunneling in particular. 0:15:05.960 --> 0:15:07.840 I mean, there are a lot of quantum effects that 0:15:07.880 --> 0:15:10.480 we could talk about, but quantum tunneling in particular is 0:15:10.520 --> 0:15:13.920 a problem. Uh. It leads to what some folks call 0:15:14.000 --> 0:15:18.520 electron leakage, which just sounds gross. But here's what's happening. 0:15:18.560 --> 0:15:22.120 So quantum tunneling is this phenomenon you get where you 0:15:22.200 --> 0:15:25.840 have a quantum particle. In this case, we're talking about electrons, 0:15:27.400 --> 0:15:32.280 and a quantum particle has uh kind of a range 0:15:32.920 --> 0:15:36.640 of of probabilities of where it can be at any 0:15:36.680 --> 0:15:40.360 given time. So we've talked a little bit in previous 0:15:40.360 --> 0:15:43.440 episodes about the Heisenberg on certainty principle. Right, we don't 0:15:43.440 --> 0:15:46.320 know where it is. We know where it could be, right, Right, 0:15:46.360 --> 0:15:51.160 We've got a general idea of the area where within 0:15:51.200 --> 0:15:53.760 that area somewhere, that's where the particle is. But it 0:15:53.800 --> 0:15:57.000 could be anywhere within that one area. So instead of 0:15:57.040 --> 0:16:02.240 thinking of an electron as a point, think of a 0:16:02.240 --> 0:16:06.920 a kind of a nebulous fog of war like circle, 0:16:07.480 --> 0:16:11.960 and the electron could be anywhere within that circle. All right. Now, 0:16:12.000 --> 0:16:16.480 with microprocessors, you have these things called gates. The gates 0:16:17.120 --> 0:16:19.880 allow either allow electrons to pass through or do not 0:16:19.960 --> 0:16:22.720 allow electrons to pass through. In terms of effects, I 0:16:22.720 --> 0:16:25.960 would say their logic gates. The gates in a in 0:16:26.000 --> 0:16:29.160 a process or sort of the the ability of your 0:16:29.240 --> 0:16:32.440 brain to tell the difference between yes and no. Right, 0:16:32.520 --> 0:16:37.000 And so if you think about computer processing ultimately, you're 0:16:37.000 --> 0:16:41.280 talking about very very very very complicated approaches to just 0:16:41.440 --> 0:16:44.560 having lots and lots of yes or no questions, Right, 0:16:44.600 --> 0:16:47.520 I mean logic gates. You build logic gates by having 0:16:47.560 --> 0:16:51.560 all these different channels, and by opening some channels and 0:16:51.600 --> 0:16:55.760 closing off others, that's where you create the the language 0:16:55.760 --> 0:16:59.480 that ultimately becomes the commands for the computer and you 0:17:00.000 --> 0:17:01.560 of getting to play your Call of Duty game or 0:17:01.560 --> 0:17:06.320 whatever it may be. So let's say you've got this 0:17:06.480 --> 0:17:09.440 Heisenberg uncertainty principle at play, and let's say you've shrunk 0:17:09.760 --> 0:17:13.800 down the components to such a degree that when you 0:17:13.800 --> 0:17:16.680 get close to one of these gates, it's so thin 0:17:18.480 --> 0:17:22.840 that part of the cloud of probabilities that that electron 0:17:22.880 --> 0:17:26.119 could inhabit overlaps the gate, so that part of it 0:17:26.160 --> 0:17:29.639 is on the other side of the gate. So if 0:17:29.680 --> 0:17:32.440 you can think of it like a flashlight, imagine that 0:17:32.480 --> 0:17:36.320 you're shining a flashlight on on an actual little like barrier, 0:17:36.640 --> 0:17:38.440 and part of the flashlight you can see like the 0:17:39.400 --> 0:17:41.720 circle of light. Part of it is on the is 0:17:41.760 --> 0:17:43.800 on one side of that barrier, and part of it's 0:17:43.800 --> 0:17:46.679 on the other side of the barrier. Think of that like, 0:17:46.760 --> 0:17:49.919 that's the potential of where the electron could be. Now, 0:17:50.359 --> 0:17:52.520 if there's a probability for an electron to be in 0:17:52.560 --> 0:17:56.359 a specific location, I mean sometimes the electron is in 0:17:56.400 --> 0:17:59.960 that specific location. So if the probability field of the 0:18:00.000 --> 0:18:03.840 electron actually overlaps the gate. That means sometimes the electrons 0:18:03.880 --> 0:18:05.840 on the other side of the gate whether you've opened 0:18:05.840 --> 0:18:08.760 the gate or not. And so sometimes the brain of 0:18:08.840 --> 0:18:12.400 your electronic device can't tell the difference between yes and no. Right, 0:18:12.520 --> 0:18:14.480 It may think that it's a yes, when in fact 0:18:14.520 --> 0:18:16.639 it was supposed to be a no. That's where you 0:18:16.680 --> 0:18:21.120 get electron leakage, where electrons are leaking through the system 0:18:21.200 --> 0:18:25.639 and inserting errors into your calculations. For computers, this is 0:18:25.640 --> 0:18:28.640 what we call a bad thing, Like you want your 0:18:28.720 --> 0:18:32.640 calculations to be reliable, and if they're not, then programs 0:18:32.680 --> 0:18:35.399 are gonna crash, files will get corrupted, you won't get 0:18:35.640 --> 0:18:39.080 you won't get good behavior out of your computer. And 0:18:39.119 --> 0:18:42.400 we've been hitting up against this. But I was trying 0:18:42.440 --> 0:18:44.480 to say, no, don't go into the room with the group. 0:18:44.920 --> 0:18:48.399 Yeah exactly, Like yeah, so I wanted I wanted to 0:18:48.440 --> 0:18:51.680 go east and open the mailbox. Um. Yeah. Know, when 0:18:51.720 --> 0:18:55.760 we get to that point, then that's that's an issue, 0:18:55.800 --> 0:18:57.600 and that's where we've been bumping up against Over the 0:18:57.640 --> 0:18:59.879 last few years now. We keep seeing engineers come up 0:18:59.880 --> 0:19:03.400 with new materials that help edge us away from that 0:19:04.160 --> 0:19:07.080 but it's something that has been an issue for the 0:19:07.160 --> 0:19:10.960 last several generations of microprocessors. Another issue with the design 0:19:10.960 --> 0:19:13.719 of microprocessors you already touched on, but I was gonna 0:19:13.880 --> 0:19:18.160 mention wasting energy's heat, Yeah, and not just wasting energy. Yeah, 0:19:18.440 --> 0:19:20.639 so there's wasting energy, but also just the problem of 0:19:20.680 --> 0:19:23.360 accumulating heat. Right, if if a chip gets too hot, 0:19:23.400 --> 0:19:26.239 it'll it'll lock up, it'll shut down exactly. So you 0:19:26.280 --> 0:19:30.760 continually shrink these components and increase transistor density, but this 0:19:30.840 --> 0:19:34.200 cuts into your ability to disperse all of the waste 0:19:34.320 --> 0:19:36.880 energy that they create as heat. Yeah, how do you 0:19:37.000 --> 0:19:41.160 end up cooling all those billions of components so that 0:19:41.359 --> 0:19:44.560 they can continue to operate without getting to that that 0:19:44.600 --> 0:19:47.679 critical threshold of heat. It's not a perfect analogy, but 0:19:47.720 --> 0:19:50.800 I would say it's sort of like trying to cram 0:19:51.280 --> 0:19:56.000 more computers into a server bank. Sure, and if you 0:19:56.080 --> 0:19:58.640 just keep cramming more and more in, well, that's great. 0:19:58.680 --> 0:20:01.440 You can fit more computer is into that tiny little room. 0:20:01.520 --> 0:20:04.600 But eventually the room's gonna get really hot, and if 0:20:04.600 --> 0:20:07.080 you don't have a good capacity to cooling that room down, 0:20:07.119 --> 0:20:10.919 then eventually those machines break down. They stop. Um, you know, 0:20:11.000 --> 0:20:14.880 you may have heard that in overclocking competitions, where people 0:20:14.880 --> 0:20:18.440 are trying to massively push the limits of what their 0:20:18.480 --> 0:20:22.359 processors processors can do, they might go to such extremes 0:20:22.440 --> 0:20:26.760 as cooling their systems with liquid nitrogen in order to 0:20:26.960 --> 0:20:29.920 get rid of that heat, because since they're running them 0:20:29.960 --> 0:20:33.440 so hard, they're generating even more heat than they normally would. 0:20:33.440 --> 0:20:36.920 And these are like top of the line processors. Oh. 0:20:36.440 --> 0:20:40.400 I I have a friend actually who is buying himself 0:20:40.440 --> 0:20:43.640 an insane computer, and he's going to get a liquid 0:20:43.720 --> 0:20:47.639 cooled computer. Liquid Cooled is pretty standard for things like 0:20:47.680 --> 0:20:50.320 a high end gaming rig these days because it's it's 0:20:50.320 --> 0:20:52.640 more efficient than air cooling. I did not know that. 0:20:52.680 --> 0:20:54.399 I mean I knew that you could do that, but 0:20:54.440 --> 0:20:57.600 I did not know that liquid cooled computers were things 0:20:57.600 --> 0:21:00.320 somebody just have in their house as opposed to in 0:21:00.359 --> 0:21:03.679 a labs. Right. Yeah, I would say that about maybe 0:21:03.880 --> 0:21:06.560 four or five years ago, it became kind of like 0:21:07.520 --> 0:21:11.160 the gold standard of how to cool your gaming rig. 0:21:11.520 --> 0:21:14.560 Before that, it was all you know, GPUs that had 0:21:14.600 --> 0:21:17.160 their own dedicated fans, so you get more and more 0:21:17.200 --> 0:21:21.560 fans inside your computer, which would make it louder and louder. H. 0:21:21.800 --> 0:21:24.680 But uh, it's not as unusual now as it used 0:21:24.680 --> 0:21:26.439 to be. Used to be. Like that was, if you 0:21:26.440 --> 0:21:30.119 had ridiculous amounts of money and you wanted to really 0:21:30.520 --> 0:21:35.320 increase your swagger around the PC gaming circles, you would 0:21:35.400 --> 0:21:38.680 invest in a water cooled system. It's it's less, it's 0:21:38.760 --> 0:21:43.520 less extravagant. Now it's still pretty extravagant, but it's not ridiculous. 0:21:43.600 --> 0:21:46.280 It's not like the Rolls Royce like it used to be. 0:21:46.760 --> 0:21:49.359 But yeah, we've seen engineers work really really hard to 0:21:49.400 --> 0:21:54.840 get around these problems. But that can only go so long, right, 0:21:54.920 --> 0:21:57.920 you do start to bump up against this issue, and 0:21:57.920 --> 0:22:00.879 and by by saying engineers work really really hard to 0:22:00.880 --> 0:22:03.400 get around these problems, that's where we're starting to see 0:22:04.520 --> 0:22:08.080 where the the issue really is. Like we said at 0:22:08.080 --> 0:22:13.280 the very beginning, Moore said, as long as it's economically advantageous, 0:22:13.320 --> 0:22:15.959 as long as it makes sense from a money standpoint, 0:22:16.240 --> 0:22:20.640 this will continue. So those engineers working really hard, somebody's 0:22:20.640 --> 0:22:22.879 got to pay their salaries, exactly, And if they have 0:22:22.920 --> 0:22:25.280 to work really hard for a really long time, that 0:22:26.000 --> 0:22:29.560 technology gets really expensive. So this kind of leads us 0:22:29.560 --> 0:22:33.280 to a recent report called the International Technology Roadmap for 0:22:33.359 --> 0:22:37.560 Semiconductors or i t r S, which was an annual 0:22:37.640 --> 0:22:43.320 report or ITTERS. It was an annual report until until 0:22:43.320 --> 0:22:45.720 this most recent one, which was technically the two thousand 0:22:45.800 --> 0:22:48.040 fifteen i t r S, but it came out this year. 0:22:48.080 --> 0:22:50.119 I'm sorry, I didn't mean to step on your emphasis. 0:22:50.119 --> 0:22:54.200 It was you're saying. It was like it's done. This 0:22:54.320 --> 0:22:56.320 was the last one that came out. That the most 0:22:56.320 --> 0:22:58.840 recent one is the final one. Bye bye, shed it 0:22:58.880 --> 0:23:03.480 here for ITITTERS. Yep. So they found that by twenty 0:23:03.600 --> 0:23:07.560 twenty one, so not far at all, we will no 0:23:07.800 --> 0:23:11.520 longer be shrinking transistors. We will have reached a limit. 0:23:12.160 --> 0:23:14.920 That's five years, folks, five years, five years. We will 0:23:14.960 --> 0:23:19.159 no longer be making smaller components for these microprocessors. We 0:23:19.200 --> 0:23:23.320 will have gone as small as we can go. Now, 0:23:23.320 --> 0:23:25.840