WEBVTT - TechStuff Tidbits: What does CPU architecture actually mean? 0:00:04.440 --> 0:00:12.360 Welcome to tech Stuff, a production from iHeartRadio. Hey there, 0:00:12.400 --> 0:00:15.560 and welcome to tech Stuff. I'm your host Jonathan Strickland. 0:00:15.640 --> 0:00:18.840 I'm an executive producer with iHeartRadio. And how the tech 0:00:18.920 --> 0:00:21.439 are you? You know? Over the last few years, there's 0:00:21.480 --> 0:00:24.880 been a lot of conversation around microchips in general, and 0:00:25.200 --> 0:00:30.440 CPUs and GPUs in particular. The pandemic led to bottlenecks 0:00:30.520 --> 0:00:33.680 in the supply chain. Manufacturing facilities had to shut down 0:00:33.720 --> 0:00:38.640 multiple times, particularly in China, and the initial skyrocketing value 0:00:38.640 --> 0:00:43.680 of cryptocurrencies all had an effect on microchip supply. Meanwhile, 0:00:44.080 --> 0:00:47.920 multiple countries, including the United States, started looking into ways 0:00:47.920 --> 0:00:51.239 to shift away from depending so heavily on China for 0:00:51.360 --> 0:00:55.160 chip fabrication. And when we talk about chips like CPUs, 0:00:55.720 --> 0:00:59.840 we often will focus on two major factors. So first 0:01:00.320 --> 0:01:04.480 is the process used to actually fabricate discrete components on 0:01:04.520 --> 0:01:07.640 the chip. We typically reference this in terms of a 0:01:07.840 --> 0:01:13.160 nanometer process, and the fewer nanometers represents more advanced processes, 0:01:13.200 --> 0:01:19.080 so you're working backward in numbers. Secondly, is the chip's architecture, 0:01:19.360 --> 0:01:21.200 and that's really what we're going to focus on in 0:01:21.240 --> 0:01:23.039 this episode. But in order to do that, we also 0:01:23.080 --> 0:01:25.560 have to talk about the other stuff. So a quick 0:01:25.600 --> 0:01:29.720 word on the fabrication process part. You might hear that 0:01:29.880 --> 0:01:34.039 company used a seven nanometer process or a five or 0:01:34.080 --> 0:01:38.240 even a three nanometer process to make the chip. And 0:01:38.440 --> 0:01:42.440 you may know that a nanometer is one billionth of 0:01:42.680 --> 0:01:46.880 a meter. It's one scale up from the atomic scale. 0:01:47.319 --> 0:01:51.040 So a typical human hair measures between eighty thousand and 0:01:51.200 --> 0:01:55.800 one hundred thousand nanometers thick, as in, if you measured 0:01:55.800 --> 0:01:59.160 the diameter of the hair, that's the range you would 0:01:59.200 --> 0:02:02.520 be at. So when you're talking about seven or five 0:02:02.760 --> 0:02:08.320 or even three nanometers, that's super duper small, right, Well, 0:02:08.360 --> 0:02:11.480 it would be if the nanometer designation still referred to 0:02:11.720 --> 0:02:15.760 component size. Now, once upon a time the scale reference 0:02:15.800 --> 0:02:19.480 to a process actually did correspond with at least some 0:02:19.639 --> 0:02:23.320 component size on the chip itself, but that has not 0:02:23.480 --> 0:02:26.880 been the case for several generations. Now. Part of the 0:02:26.919 --> 0:02:29.800 reason for that comes down to the limitations of physics. 0:02:30.200 --> 0:02:32.720 As you shrink down to the bottom end of the 0:02:32.800 --> 0:02:36.400 nanoscale and into the atomic scale, you start to have 0:02:36.440 --> 0:02:40.799 to contend with quantum mechanics. Now, we don't encounter quantum 0:02:41.520 --> 0:02:45.760 mechanic effects on our normal scale like in our everyday lives. 0:02:46.440 --> 0:02:49.440 But at that tiny scale, things start to behave in 0:02:49.480 --> 0:02:53.880 a really wonky way, and relying on physical structures to 0:02:54.000 --> 0:02:58.079 rain in quantum silliness becomes a big challenge. I've done 0:02:58.080 --> 0:03:00.440 full episodes kind of about this. We're not going to 0:03:00.560 --> 0:03:05.120 dive too deeply into it, so instead, the scale really 0:03:05.160 --> 0:03:08.120 is more of a marketing strategy. When you hear it's 0:03:08.120 --> 0:03:10.760 a five nanometer process, it doesn't mean that anything on 0:03:10.800 --> 0:03:14.040 that chip actually measures five nanometers in size. It's a 0:03:14.040 --> 0:03:18.520 way of indicating this process is more advanced than the 0:03:18.520 --> 0:03:24.400 previous seven nanometer process. So it really means that when 0:03:24.440 --> 0:03:27.040 you get down to the process and the architecture, you 0:03:27.080 --> 0:03:30.600 start to converge on essentially the same meaning. So let's 0:03:30.680 --> 0:03:37.040 talk about that architecture. What does chip architecture actually mean. Well, 0:03:37.120 --> 0:03:40.160 we're going to stick with CPUs, also known as central 0:03:40.200 --> 0:03:43.280 processing units, and we can think of a CPU as 0:03:43.320 --> 0:03:49.600 having three major components. These are the registers, the arithmetic 0:03:49.680 --> 0:03:54.240 logic unit or ALU, and the control unit. So registers 0:03:54.280 --> 0:03:58.000 act kind of like memory, and that they hold information 0:03:58.080 --> 0:04:02.280 that the CPU needs in order to complete operations. Logic 0:04:02.320 --> 0:04:06.200 gates make up the quote unquote memory of registers, and 0:04:06.240 --> 0:04:09.000 a logic gate follows a specific rule. It creates an 0:04:09.000 --> 0:04:12.560 output based upon the input coming into the logic gate. 0:04:13.400 --> 0:04:15.560 I'll do a full episode just about logic gates in 0:04:15.560 --> 0:04:18.240 the future to kind of expand on that and explain 0:04:18.560 --> 0:04:22.000 how these logic gates work and how by combining logic 0:04:22.040 --> 0:04:26.440 gates you can create different outcomes. So registers operate faster 0:04:26.800 --> 0:04:31.240 than RAM, a random access memory, which I often at 0:04:31.320 --> 0:04:36.560 least will compare to short term memory in humans. RAM, 0:04:36.720 --> 0:04:40.520 in turn operates faster than a solid state drive or 0:04:40.520 --> 0:04:43.760 a hard drive, which I compare to long term memory 0:04:43.839 --> 0:04:47.839 with humans. So you've got registers, which are the fastest 0:04:47.920 --> 0:04:51.880 access of memory total, but it holds very little information. 0:04:52.000 --> 0:04:55.320 It's just tiny, tiny bits of information. Then you have RAM. 0:04:55.520 --> 0:04:59.600 Then you've got solid state drive or hard drive in registers. 0:04:59.640 --> 0:05:03.080 We actually we have five basic types, so let's list 0:05:03.080 --> 0:05:07.640 them off, shall we. The instruction register stores the address 0:05:07.680 --> 0:05:10.560 and random access memory of the instruction to be used 0:05:10.720 --> 0:05:14.159 in a given operation. So that instruction could be some 0:05:14.480 --> 0:05:19.600 basic arithmetic function for example like AD. Next, you've got 0:05:19.600 --> 0:05:23.320 the memory address register. This stores the address within RAM 0:05:23.720 --> 0:05:26.279 of the data that is to be processed. So this 0:05:26.400 --> 0:05:28.680 is the data that's going to be transformed by that 0:05:28.800 --> 0:05:32.400 instruction in some way. Your instruction register has the info 0:05:32.480 --> 0:05:35.240 on what operation to use. The memory address register has 0:05:35.279 --> 0:05:37.800 the info on which data is going to undergo that operation. 0:05:39.160 --> 0:05:42.560 Then you've got the memory data register. This stores the 0:05:42.640 --> 0:05:46.680 data that the CPU is actually processing at any given time. 0:05:46.800 --> 0:05:48.800 So while the other two registers are kind of like 0:05:48.880 --> 0:05:51.680 looking into the future like the next step, the memory 0:05:51.760 --> 0:05:54.839 data register is concerned with what's going on right now, 0:05:54.880 --> 0:05:58.760 gosh darn it. Then you've got the program counter. This 0:05:58.920 --> 0:06:02.480 stores the address and RAM of the next instruction coming up, 0:06:02.520 --> 0:06:05.960 so the next one down the line. Finally, you've got 0:06:05.960 --> 0:06:10.800 the accumulator that stores the results of the calculations that 0:06:10.839 --> 0:06:14.520 were just performed. So the registers are one part of 0:06:14.680 --> 0:06:19.919 CPU architecture. Now let's talk about the ALU or arithmetic 0:06:20.120 --> 0:06:25.960 logic unit. The ALU is the brains of the CPU. 0:06:26.440 --> 0:06:29.840 Within the ALU are logic circuits which actually carry out 0:06:29.880 --> 0:06:33.839 the operations on data. These operations span a wide range 0:06:33.839 --> 0:06:39.800 of arithmetic tasks like addition and subtraction, to things like 0:06:39.920 --> 0:06:44.080 incrementation and also comparison. So, for example, you might have 0:06:44.240 --> 0:06:47.480 a pair of operations that each produce a result and 0:06:47.520 --> 0:06:50.320 the ALU has to compare these results with one another 0:06:50.600 --> 0:06:54.039 to determine if they are the same or different. That's 0:06:54.080 --> 0:06:57.800 the kind of basic task the ALU handles, and it 0:06:57.800 --> 0:07:02.839 does this super fast. Finally, you have the control unit, which, 0:07:02.880 --> 0:07:07.120 as the name suggests, controls the process. The control unit 0:07:07.200 --> 0:07:10.520 receives instructions, decodes those to get to the meaning of 0:07:10.560 --> 0:07:13.920 the instructions, sends commands to the other components to carry 0:07:13.960 --> 0:07:17.360 out those instructions, et cetera. The control unit is kind 0:07:17.360 --> 0:07:19.880 of like a floor manager. It makes sure all the 0:07:19.920 --> 0:07:23.560 departments are responding appropriately given the program that's running at 0:07:23.600 --> 0:07:27.480 any given time. The control unit also has a clock, 0:07:27.840 --> 0:07:30.520 but that clock isn't meant to keep your computer's time 0:07:30.640 --> 0:07:35.040 accurate to local time. This clock oscillates a certain number 0:07:35.040 --> 0:07:37.920 of times per second, and we measure this in hurts, 0:07:38.040 --> 0:07:42.640 So one oscillation per second would be one hurts. Typically, 0:07:43.080 --> 0:07:46.960 with processors today, we're talking about the gigahertz range. A 0:07:47.000 --> 0:07:50.720 gigahertz would be a billion oscillations per second. So a 0:07:50.800 --> 0:07:54.480 three point two gigahertz CPU has a clock that in 0:07:54.520 --> 0:07:58.360 the control unit that oscillates three point two billion times 0:07:59.000 --> 0:08:02.880 every single So now the clock speed relates to how 0:08:03.000 --> 0:08:07.920 quickly the processor can actually complete these operations. Some operations 0:08:07.960 --> 0:08:11.560 require multiple oscillations, but that clock speed or frequency, if 0:08:11.600 --> 0:08:14.280 you prefer, gives you an idea of how fast or 0:08:14.320 --> 0:08:17.560 powerful your computer is. Now. Other factors also play into 0:08:17.640 --> 0:08:20.080 this too. It's not just clock speed, but that is 0:08:20.160 --> 0:08:23.000 one big component in it. If you're familiar with the 0:08:23.080 --> 0:08:26.280 term overclocking, then all of the stuff I'm talking to 0:08:26.320 --> 0:08:29.760 you about is old news to you, right. Overclocking is 0:08:29.800 --> 0:08:34.040 the practice of increasing that clock oscillation speed in the 0:08:34.080 --> 0:08:40.600 control unit beyond its default settings, which typically the manufacturer creates. 0:08:40.640 --> 0:08:44.079 Like they create default settings, they say this processor is 0:08:44.240 --> 0:08:48.000 rated at this particular clock speed, and going beyond that 0:08:48.120 --> 0:08:52.640 could potentially reduce the useful lifespan of the processor or 0:08:53.120 --> 0:08:57.600 cause it to overheat, et cetera. So elite gamers typically 0:08:57.679 --> 0:09:02.160 will use programs to boost the clock speed on CPUs 0:09:02.160 --> 0:09:05.840 to get past these limitations and to push it faster 0:09:05.960 --> 0:09:08.800 than what it was rated as in order to milk 0:09:08.840 --> 0:09:12.360 out higher performance in their gaming ricks. Doing this does 0:09:12.400 --> 0:09:14.520 come with some trade offs. I mean, it does mean 0:09:14.559 --> 0:09:16.800 that you might be burning through your CPU faster than 0:09:16.800 --> 0:09:19.840 you usually would. It also typically means that the computer's 0:09:19.880 --> 0:09:21.920 going to generate a lot more heat, so you need 0:09:21.960 --> 0:09:25.040 to have a good heat dispersal system in place to 0:09:25.120 --> 0:09:27.880 carry that heat away from the processor because, as we know, 0:09:29.080 --> 0:09:33.440 heat and electronics are not super friendly with one another. 0:09:34.520 --> 0:09:39.160 Connecting all these different components are wires called buses, So 0:09:39.280 --> 0:09:43.480 a bus might carry instructions, another bus might carry data. 0:09:44.120 --> 0:09:46.520 The capacity of buses also plays a part in how 0:09:46.559 --> 0:09:49.800 powerful a computer is. I'll have to do another episode 0:09:49.840 --> 0:09:52.960 to explain things like what is a thirty two bit 0:09:53.120 --> 0:09:56.920 machine versus a sixty four bit machine, or even with 0:09:57.000 --> 0:10:00.400 the old game consoles, an eight bit machine. We'll talk 0:10:00.440 --> 0:10:03.240 about bitwidth and that kind of stuff, but that kind 0:10:03.240 --> 0:10:06.040 of plays into things like buses. You can think of 0:10:06.040 --> 0:10:08.720 it sort of like roads. How wide is the road 0:10:09.040 --> 0:10:12.280 so how many vehicles can pass side by side at 0:10:12.280 --> 0:10:15.440 the same time. And one other thing that we will 0:10:15.440 --> 0:10:18.880 mention will be cores, and I'm going to get to 0:10:18.920 --> 0:10:32.640 that after we take this quick break. Okay, before the break, 0:10:32.760 --> 0:10:35.360 I teased that we're going to talk about cores. A 0:10:35.360 --> 0:10:38.760 CPU core is the smallest unit that can carry out 0:10:38.800 --> 0:10:42.560 all the jobs that a CPU does. So if you 0:10:42.600 --> 0:10:44.959 hear of a multicore CPU, that means each of those 0:10:45.000 --> 0:10:48.360 cores can do the job of a CPU, and they 0:10:48.400 --> 0:10:51.040 can have multiple cores. You'll hear things like dual core, 0:10:51.200 --> 0:10:53.920 which means there's two of them, or quad core, meaning 0:10:53.920 --> 0:10:57.880 there's four of them, and beyond, each core can carry 0:10:57.880 --> 0:11:00.680 out the duties of a CPU. So does that mean 0:11:00.720 --> 0:11:04.360 a dual core or quad core processor is automatically better 0:11:04.440 --> 0:11:10.120 than a single core processor. Not necessarily so. For some 0:11:10.520 --> 0:11:13.960 types of computational problems, you can actually divide up the 0:11:14.000 --> 0:11:18.640 problem into smaller tasks that could be completed simultaneously. So 0:11:18.800 --> 0:11:22.400 these are the types of problems that multicore processors are 0:11:22.600 --> 0:11:25.760 great at tackling because each core can tackle a different 0:11:25.800 --> 0:11:29.480 set of tasks and thus collectively they'll get to the 0:11:29.480 --> 0:11:33.760 answer faster. But if the problem cannot be broken down 0:11:33.800 --> 0:11:38.480 into smaller pieces, a very powerful single core processor might 0:11:38.559 --> 0:11:43.360 be better than a decently powerful multicore processor. And I 0:11:43.480 --> 0:11:46.760 use this analogy all the time. Longtime listeners are probably 0:11:47.360 --> 0:11:49.880 tired of it, and they've anticipated it, and yes, it's 0:11:49.920 --> 0:11:52.319 okay to skip ahead a little bit if you are 0:11:52.320 --> 0:11:55.080 one of those people. But I like to describe multicore 0:11:55.200 --> 0:11:58.800 processors versus a single core processor by talking about an 0:11:58.840 --> 0:12:01.839 advanced math class, And in this version of it, I'm 0:12:01.840 --> 0:12:04.400 going to say there are five students in this advanced 0:12:04.640 --> 0:12:08.320 math class. Now imagine four of those five students are 0:12:08.360 --> 0:12:12.160 all really good at math, right, they're gifted students. However, 0:12:12.200 --> 0:12:16.200 the fifth student is a genuine math genius. And the 0:12:16.280 --> 0:12:20.920 genius always completes any given problem faster than the other 0:12:21.000 --> 0:12:24.080 four students can. And one day the teacher presents a 0:12:24.200 --> 0:12:27.360 challenge to the class. It's a pop quiz that has 0:12:27.600 --> 0:12:31.319 four questions on the quiz. The genius has to try 0:12:31.360 --> 0:12:35.079 and complete all four problems, but the other four students 0:12:35.280 --> 0:12:38.040 can actually divide up the quiz and each student can 0:12:38.080 --> 0:12:41.560 tackle a single problem on there, so collectively they can 0:12:41.600 --> 0:12:46.400 solve the quiz together. So who is going to finish first? Well, 0:12:46.440 --> 0:12:50.560 if we assume that each problem is discreete and independent 0:12:50.920 --> 0:12:53.960 of the outcomes of the other problems, the four students 0:12:54.000 --> 0:12:57.280 are likely to finish their quiz collectively before the genius 0:12:57.280 --> 0:13:00.840 because each one's just doing one question, and the genius 0:13:00.920 --> 0:13:04.120 is still faster than all the individuals. But they have 0:13:04.200 --> 0:13:07.960 to do all four questions, whereas each smart student just 0:13:08.000 --> 0:13:11.120 has to do one. The multi core processor wins in 0:13:11.160 --> 0:13:16.800 that scenario, but Let's say you find out that problem 0:13:16.880 --> 0:13:19.679 two on the quiz actually depends upon the outcome of 0:13:19.720 --> 0:13:23.640 problem one, and you find out the problem three depends 0:13:23.720 --> 0:13:27.520 upon the outcome of problem two, and the problem four 0:13:27.600 --> 0:13:30.120 depends on the outcome of problem three. Well, now you 0:13:30.160 --> 0:13:32.560 can't just divide up the problems between the four students 0:13:33.040 --> 0:13:35.720 because the student working on problem two has to wait 0:13:36.080 --> 0:13:37.960 to find out what the answer to problem one is 0:13:38.000 --> 0:13:41.120 before they can get started. The genius in that case 0:13:41.280 --> 0:13:43.640 is going to win that race, right, because they're still 0:13:43.679 --> 0:13:47.160 faster than any individual is. So for certain types of 0:13:47.240 --> 0:13:52.000 computational problems and processes, multicore is the way to go, 0:13:52.600 --> 0:13:55.400 but not in every case, just in a lot of them. 0:13:55.679 --> 0:13:59.320 For a lot of computer users, it's more important to 0:13:59.320 --> 0:14:02.880 go multi corep because the typical uses that they rely 0:14:03.000 --> 0:14:06.680 upon with computers falls into that multi core set of problems. 0:14:06.679 --> 0:14:10.760 This includes gamers, So a multi core processor matched with 0:14:10.840 --> 0:14:15.360 a really good graphics processing unit that's more important than 0:14:15.400 --> 0:14:19.200 having just a single core super fast processor. But again, 0:14:19.280 --> 0:14:22.920 it all depends on how you can thread the computational problems. 0:14:23.560 --> 0:14:27.480 And that's the general description of what computer architecture means. 0:14:27.520 --> 0:14:30.760 The actual design and layout of these components is what 0:14:30.840 --> 0:14:34.600 sets one chip apart from another chip. Since it is 0:14:34.720 --> 0:14:39.000 increasingly challenging to shrink components down without getting into quantum 0:14:39.040 --> 0:14:42.360 effects or generating too much heat in a very small space, 0:14:42.760 --> 0:14:47.120 finding the best possible layout and orientation of components is critical. 0:14:48.200 --> 0:14:49.600 You know you're not going to be able to cram 0:14:49.720 --> 0:14:52.120 a whole lot more on, but you might be able 0:14:52.160 --> 0:14:55.800 to find an orientation that gets a little better performance 0:14:55.880 --> 0:14:58.920 out of the components you have now. Back in the day, 0:14:59.320 --> 0:15:02.080 Intel which which is one of two major companies behind 0:15:02.120 --> 0:15:05.800 the processors used in most computers these days, used a 0:15:05.880 --> 0:15:08.680 development approach and chip designed that the company referred to 0:15:08.760 --> 0:15:12.160 as the tick talk method. So you can think of 0:15:12.200 --> 0:15:16.080 the tick part of TikTok as taking the same chip 0:15:16.160 --> 0:15:20.200 layout design from the previous generation, but then shrinking everything 0:15:20.240 --> 0:15:22.320 down a little bit, which allows you to cram more 0:15:22.320 --> 0:15:26.040 components on the chip. So you're following the same architectural 0:15:26.160 --> 0:15:29.440 plan as the previous generation, but now all the components 0:15:29.440 --> 0:15:31.560 are slightly smaller so you can have more of them there. 0:15:32.080 --> 0:15:36.600 The talk sequence would involve creating a new architecture that 0:15:36.880 --> 0:15:40.800 better takes advantage of these smaller components, and then it 0:15:40.800 --> 0:15:44.440 would repeat tech talk Tick talk. So with tick you 0:15:44.480 --> 0:15:46.680 shrink stuff down, but you follow the same game plan 0:15:46.760 --> 0:15:49.240 as before. With talk, you create a new game plan, 0:15:49.720 --> 0:15:52.480 and then you do tick again. And so each generation 0:15:52.520 --> 0:15:59.040 of Intel fell into one of those two design principles, 0:15:59.560 --> 0:16:02.040 and in this way Intel would iterate its chip designs. 0:16:02.040 --> 0:16:05.160 Each generation would improve upon the last. At least that 0:16:05.280 --> 0:16:08.520 was the idea, either by adding more capability in the 0:16:08.560 --> 0:16:11.840 form of more components added to the chip, or finding 0:16:11.920 --> 0:16:15.359 a new way to arrange those components that improve performance. 0:16:15.640 --> 0:16:19.320 And by improved performance, I mean not just being faster 0:16:19.840 --> 0:16:24.320 or more capable, but also more power efficient or creating 0:16:24.400 --> 0:16:28.320 less heat, because these things do matter quite a bit. 0:16:29.000 --> 0:16:33.720 And that's our overview of chip architecture. I'll do more 0:16:33.760 --> 0:16:36.640 episodes about the basics of CPUs soon. Maybe i'll talk 0:16:36.680 --> 0:16:40.120 a bit about what makes an Intel chip different from say, 0:16:40.760 --> 0:16:44.280 an AMD chip. And you may know, if you've ever 0:16:44.360 --> 0:16:49.320 built a computer, the type of processor you want ends 0:16:49.400 --> 0:16:52.080 up mattering a big deal, because it will tell you 0:16:52.160 --> 0:16:54.920 what kind of motherboard you can use, for example, because 0:16:54.960 --> 0:16:57.160 a motherboard designed to work with an Intel chip is 0:16:57.200 --> 0:17:00.840 not going to work with an AMD chip. Thing. So 0:17:00.920 --> 0:17:03.840 we'll do another episode to talk a bit more about 0:17:03.840 --> 0:17:06.480 this in the future, and keep it nice and short 0:17:06.520 --> 0:17:10.040 and simple so that folks can listen, get a good understanding, 0:17:10.080 --> 0:17:11.800 and then know what to look for when they move 0:17:11.800 --> 0:17:14.040 forward if they ever decide to build their own computer. 0:17:14.440 --> 0:17:16.679 And I think we'll also, like I said, to an 0:17:16.720 --> 0:17:20.000 episode about things like logic gates to kind of understand 0:17:20.080 --> 0:17:23.119 at a very very very basic level, what is going 0:17:23.160 --> 0:17:27.040 on when a computer is processing information. That's it for 0:17:27.080 --> 0:17:30.520 this Tech Stuff Tidbits episode. I hope you are all well. 0:17:30.600 --> 0:17:34.600 Just a reminder next week, I am on vacation and 0:17:34.680 --> 0:17:36.680 I will be back the following week, so we will 0:17:36.720 --> 0:17:40.200 likely have some reruns playing next week, but I will 0:17:40.240 --> 0:17:43.520 be back and I'll talk to you again really soon. 0:17:49.720 --> 0:17:54.359 Tech Stuff is an iHeartRadio production. For more podcasts from iHeartRadio, 0:17:54.680 --> 0:17:58.400 visit the iHeartRadio app, Apple Podcasts, or wherever you listen 0:17:58.440 --> 0:18:03.040 to your favorite shows.