WEBVTT - Apple Is at the Cutting Edge of a Revolution in Chips 0:00:10.800 --> 0:00:14.920 Hello, and welcome to another episode of the Odd Lots Podcast. 0:00:14.960 --> 0:00:20.040 I'm Joe Wisnthal and I'm Tracy Halloway. Tracy, I had 0:00:20.079 --> 0:00:22.680 a thought, and it might be a little controversial, but 0:00:23.320 --> 0:00:27.200 what if um you with a controversial thought? Never No, 0:00:27.400 --> 0:00:29.160 it's like, I know, I'm just gonna throw this out 0:00:29.200 --> 0:00:31.240 there and tell me what you think. What if we 0:00:31.680 --> 0:00:35.960 made that Odd Lots wizon about market and finance anymore, 0:00:36.000 --> 0:00:41.200 but just about the semiconductor history? Could we maybe could 0:00:41.200 --> 0:00:45.480 we maybe start with a semiconductor series before we completely 0:00:45.560 --> 0:00:49.920 rebrand the podcast? All right, we'll we'll revisit the question 0:00:50.000 --> 0:00:52.440 of a total rebrand. But I don't know about you, 0:00:52.479 --> 0:00:55.800 but I found our episode a couple of weeks ago 0:00:56.360 --> 0:01:00.800 with Stacy Raskin from Bernstein about the kind of Intel 0:01:00.840 --> 0:01:03.120 to be fascinating, And after that, I was like, I 0:01:03.160 --> 0:01:05.440 just wanted to talk more chips, Like I just that 0:01:05.640 --> 0:01:08.479 like that really sort of like it really went my appetite. 0:01:09.040 --> 0:01:11.960 It was definitely a lot more interesting than I had 0:01:12.000 --> 0:01:15.240 been expecting. I think I mentioned when we recorded it 0:01:15.400 --> 0:01:18.440 that I hadn't been following all the ins and outs 0:01:18.600 --> 0:01:22.640 of the semiconductor industry, except of course, as it pertained 0:01:22.680 --> 0:01:26.399 to the US China trade tension. So I wasn't aware 0:01:27.319 --> 0:01:31.200 of some of the competition between the big makers, like 0:01:31.319 --> 0:01:34.399 just how bad things had gotten for Intel, and I 0:01:34.440 --> 0:01:37.160 wasn't aware of some of the strategic decisions that we're 0:01:37.200 --> 0:01:41.040 going into the making of their chips and the way 0:01:41.080 --> 0:01:44.679 that some other chip makers have proceeded in a very 0:01:44.920 --> 0:01:49.520 uh different direction. Yeah, right, it's that different direction that's 0:01:49.560 --> 0:01:52.760 interesting because, I mean, we talked with Stacy a lot 0:01:52.800 --> 0:01:56.880 about the sort of manufacturing issues that Intel has been 0:01:56.880 --> 0:02:01.279 facing as it tries to unveil it's seven animator chips. 0:02:01.680 --> 0:02:05.840 But there's more to the story than just faster and 0:02:05.960 --> 0:02:11.080 smaller chips. There's also a fundamental ship happening in what 0:02:11.200 --> 0:02:13.960 chips are and how they're used and how they're designed. 0:02:14.320 --> 0:02:19.040 It goes beyond just sort of manufacturing problems. Yeah, and 0:02:19.160 --> 0:02:22.680 I guess the big news on that front is Apple 0:02:22.840 --> 0:02:27.120 and its new series of Mac processors. Uh. That news 0:02:27.160 --> 0:02:29.480 came out this month, and everyone, as far as I 0:02:29.480 --> 0:02:33.280 can tell, all the technologists that have YouTube channels are 0:02:33.400 --> 0:02:37.640 very very excited about what exactly this means. I don't 0:02:37.720 --> 0:02:41.800 completely understand the technology, but I am interested in learning 0:02:41.800 --> 0:02:44.680 why everyone else um is so excited and why a 0:02:44.680 --> 0:02:46.600 lot of people are saying this is a game changer. 0:02:46.639 --> 0:02:49.280 I heard one there's one guy on YouTube who was 0:02:49.320 --> 0:02:53.200 talking about how this is going to make Apple computers 0:02:53.200 --> 0:02:57.119 basically on a different level to other types of computers. 0:02:57.160 --> 0:02:59.680 So in the way that we don't necessarily compare Apple 0:02:59.720 --> 0:03:03.800 phone with androids anymore, we will no longer be comparing 0:03:04.200 --> 0:03:07.480 Max with you know, your average PC. It's just like 0:03:07.760 --> 0:03:11.440 two different types of equipment. Yeah, there was like a 0:03:11.520 --> 0:03:14.480 medium article about the Apple chip then went viral on 0:03:14.560 --> 0:03:20.960 like articles about chips don't typically go viral, but that 0:03:21.120 --> 0:03:23.839 I think sort of gives you an indication of the enthusiasm, 0:03:23.960 --> 0:03:26.840 the excitement that people have for all this stuff. So 0:03:27.040 --> 0:03:28.880 definitely a lot to talk about. But I have to 0:03:28.880 --> 0:03:31.440 admit I still don't like really get it. Like I'm 0:03:31.480 --> 0:03:33.440 trying to read about this stuff and understand it, but 0:03:33.480 --> 0:03:37.560 I don't totally get Yeah, I'm in that same place. 0:03:37.920 --> 0:03:41.600 So today we are going to be talking more as 0:03:41.720 --> 0:03:44.760 anyone could have guessed about chips. And I'm very excited 0:03:44.800 --> 0:03:48.040 because in addition to being a sequel of sorts to 0:03:48.200 --> 0:03:52.240 that other chip episode, um our guest who is who 0:03:52.280 --> 0:03:55.040 writes on this stuff, and he's prominent on Twitter, etcetera. 0:03:55.320 --> 0:03:57.760 We're going to do something we've never done before. He's 0:03:57.760 --> 0:04:01.000 going to dos himself on this episod. So he's been 0:04:01.000 --> 0:04:05.200 writing under and tweeting under a pseudonyms, agreed to come 0:04:05.240 --> 0:04:08.800 on and with this episode actually tell us who he is. 0:04:08.840 --> 0:04:11.320 So this is a a totally new thing for us. 0:04:11.320 --> 0:04:14.480 I don't think we've done this before. No, we definitely haven't, 0:04:14.520 --> 0:04:18.279 and we should just stress this is a voluntary self 0:04:18.520 --> 0:04:21.080 docs scene. We haven't forced him into it or anything 0:04:21.120 --> 0:04:24.359 like that. That's exactly right. So we're going to be 0:04:24.440 --> 0:04:29.000 speaking with the author of the newsletter called Mules musing. 0:04:29.680 --> 0:04:32.920 He tweets under the handle at full all the time, 0:04:33.520 --> 0:04:37.080 great stuff on the matter of chips. But we're going 0:04:37.080 --> 0:04:39.520 to find out now who he is. So U Mule, 0:04:40.560 --> 0:04:43.960 thank you for joining the Outlaws podcast? Who are you 0:04:44.120 --> 0:04:48.360 and why are we listening to? Hey? I'm My name 0:04:48.520 --> 0:04:51.960 is Doug O Offlin, and I am just a huge 0:04:52.000 --> 0:04:56.200 nerd with an investing background. I guess I previously worked 0:04:56.200 --> 0:04:59.240 at a byeside firm in Dallas and kind of just 0:04:59.279 --> 0:05:02.560 taking a break tween things started writing a newsletter. Um, 0:05:02.640 --> 0:05:04.920 and clearly there's been a lot of interest in semi coonnucters, 0:05:04.960 --> 0:05:07.560 not just from you guys, but from the investing community 0:05:07.560 --> 0:05:09.240 at large, and it kind of just hit a note. 0:05:09.400 --> 0:05:13.040 So I'm I'm really obsessed with the whole you know, 0:05:13.080 --> 0:05:15.599 the whole revolution all the YouTubers are going crazy about. 0:05:15.920 --> 0:05:18.880 I'm obsessed, Like, it's not just um, it's not just 0:05:18.920 --> 0:05:20.960 all the technology stuff, but there's a lot of business 0:05:21.000 --> 0:05:24.400 and investing um things that could happen, and I think 0:05:24.400 --> 0:05:26.720 it's I think it's truly a revolution. Like it's it's 0:05:26.760 --> 0:05:29.120 about as exciting as it's ever been in semiconductors, which 0:05:29.160 --> 0:05:32.520 is frankly not that exciting, but it's still very exciting 0:05:32.560 --> 0:05:35.760 to me. UM. So yeah, that's that's what I've been 0:05:35.800 --> 0:05:39.080 doing recently. I have to say that while you were 0:05:39.400 --> 0:05:42.840 introducing yourself, I just had a look at your Twitter 0:05:42.880 --> 0:05:46.040 profile page and I saw the binary code on your 0:05:46.080 --> 0:05:50.919 profile and I looked it up and you must be 0:05:50.960 --> 0:05:53.000 the first All Thoughts guests that we've ever had who 0:05:53.120 --> 0:05:57.040 who has basically done like a practical joke in binary code. 0:05:57.120 --> 0:05:59.719 So um, kudos for that, and I guess I give 0:05:59.800 --> 0:06:03.840 some indication of of your interests. H and everyone else 0:06:03.880 --> 0:06:06.160 can go and check it out and translate it into 0:06:06.200 --> 0:06:09.120 English to see what it actually says. But before we begin, 0:06:09.200 --> 0:06:13.280 I'm curious, why did you decide to start the sub 0:06:13.320 --> 0:06:17.120 stack and why were you tweeting under the pseudonym to 0:06:17.200 --> 0:06:22.680 begin with? Well, um, especially in the past past, it's 0:06:22.800 --> 0:06:25.480 it's not you know, I was employed, and it's um, 0:06:25.480 --> 0:06:29.200 it's not you're not supposed to talk publicly about public 0:06:29.240 --> 0:06:31.919 markets if you're in the business. So that was that 0:06:32.040 --> 0:06:35.120 was the big reason for a pseudonym. And frankly, Finfoot 0:06:35.160 --> 0:06:37.800 is an amazing community. There's a lot of um, there's 0:06:38.000 --> 0:06:41.240 more information there than anywhere else, so it's it's an 0:06:41.279 --> 0:06:44.320 awesome place to learn. But obviously you have to be 0:06:44.839 --> 0:06:48.480 extremely private if youtuose to do so. So um that 0:06:48.480 --> 0:06:50.359 that was the reason why I under a pseudonym, and 0:06:50.360 --> 0:06:53.520 then the sub stack. I guess, um, I'm just kind 0:06:53.560 --> 0:06:56.440 of like in between things. Next year, if everything goes right, 0:06:56.440 --> 0:06:58.800 I'll take a little bit of a gap. But you know, 0:06:58.920 --> 0:07:03.960 I'm like born to do or keep researching and doing 0:07:04.000 --> 0:07:06.159 stuff like this, Like this was pretty much what I 0:07:06.200 --> 0:07:07.840 was doing in my free time, and I was like, heck, 0:07:08.080 --> 0:07:09.960 a lot of other people could really learn a lot 0:07:10.040 --> 0:07:12.120 from this. As well, and so then I started writing 0:07:12.120 --> 0:07:15.520 for a broader audience, and um, writing in general has 0:07:15.560 --> 0:07:18.680 been really fruitful. Like you, it's it's a crazy skill 0:07:18.720 --> 0:07:21.040 to have and I'm frankly not I'm getting a lot 0:07:21.080 --> 0:07:22.920 better at it, but it's still um, it's been a 0:07:23.000 --> 0:07:25.280 learning journey. So that's why I started the sub stack 0:07:25.440 --> 0:07:27.840 and I've just been kind of writing the wave. I 0:07:27.920 --> 0:07:33.000 have to say, I feel like a deep, emotional, resonant 0:07:33.160 --> 0:07:38.080 connection with your story. Because I got my start I 0:07:38.120 --> 0:07:42.040 worked for a small portfolio management company. I also started 0:07:42.080 --> 0:07:45.360 I started blogging in two thousand five under a blog 0:07:45.480 --> 0:07:48.200 called The Stalwart, which is now defunct, but that's why 0:07:48.200 --> 0:07:49.880 I got my Twitter handle, and it was kind of 0:07:49.920 --> 0:07:53.160 the same thing. I just wanted to practice writing for 0:07:53.320 --> 0:07:55.760 the skill of writing and keep up to date, and 0:07:55.760 --> 0:07:57.800 it was between jobs, so kind of like you. So 0:07:58.440 --> 0:08:00.760 I'm I'm a big fan of people with your art 0:08:00.800 --> 0:08:03.320 and I think big things. But let's let's move on 0:08:03.760 --> 0:08:08.200 and let's talk chips. I mentioned there's this a viral 0:08:08.320 --> 0:08:12.120 article about the Mac chip that came out, and articles 0:08:12.120 --> 0:08:15.640 about chips don't particularly go viral, but everyone's raving about 0:08:15.680 --> 0:08:18.960 this new chip from Apple, So why is the chip 0:08:19.040 --> 0:08:23.160 going vibral okay, so then yeah, now and I can 0:08:23.200 --> 0:08:26.120 talk ages about this. UM. So the the the M 0:08:26.160 --> 0:08:29.720 one chip is really exciting, so UM. The reason why 0:08:29.960 --> 0:08:32.400 is it's kind of it's kind of like a hybrid 0:08:32.520 --> 0:08:35.200 and we've we've seen we've seen it before, frankly, like 0:08:35.240 --> 0:08:37.559 your iPhone is powered by something that's similar to and 0:08:37.760 --> 0:08:39.880 M one chip, but we've never seen it for a 0:08:39.960 --> 0:08:42.960 Mac or a desktop or any other form factor other 0:08:43.000 --> 0:08:46.360 than iPhone. And part of it is because up until 0:08:46.360 --> 0:08:48.440 this point, all the laptops in the world have been 0:08:48.559 --> 0:08:50.920 mostly running x A D six. So x A D 0:08:51.000 --> 0:08:53.760 six isn't UM is like just think of it as 0:08:53.760 --> 0:08:57.600 like a language of processors. And for a long time 0:08:57.600 --> 0:09:00.800 that was kind of like, you know, the defen ACTO standard. 0:09:00.880 --> 0:09:03.360 But the iPhone getting a lot better has pretty much 0:09:03.400 --> 0:09:05.680 made a second language become a lot more popular, which 0:09:05.720 --> 0:09:09.480 is ARM. And now that it's become a lot more mature, 0:09:09.679 --> 0:09:12.640 it's actually kind of um. In the past, it never 0:09:12.760 --> 0:09:16.040 was fast enough to really compete in a actual computer, 0:09:16.559 --> 0:09:19.439 but now it seems to be that the M one 0:09:19.520 --> 0:09:23.520 chip is like supermature. The ARM infrastructure is ready and 0:09:23.600 --> 0:09:27.080 the laptop looks really compelling, and a big reason for 0:09:27.160 --> 0:09:29.880 what that is is it's not really a CPU like 0:09:29.960 --> 0:09:32.720 it's it's actually a lot of different dies on one 0:09:32.800 --> 0:09:35.440 giant die, and it looks like there's a lot of 0:09:35.480 --> 0:09:38.480 benefits of arm over x eight six that we didn't 0:09:38.480 --> 0:09:41.280 even know where possible up until the M one came out. 0:09:41.360 --> 0:09:45.160 So so, I mean, I think from what I understand, 0:09:45.200 --> 0:09:47.559 this is kind of the heart of it. So normally 0:09:47.559 --> 0:09:50.520 when we talk about chips, you know, certainly when we're 0:09:50.520 --> 0:09:53.280 talking about chips made by Intel or m D or 0:09:53.320 --> 0:09:58.320 something like that, we're talking about CPUs, these central processing units. 0:09:58.360 --> 0:10:02.720 But the Apple hip, the M one is not a CPU. 0:10:02.800 --> 0:10:05.560 It's it's something else, or only one component of it 0:10:05.600 --> 0:10:08.000 is a CPU. Can you can you explain that in 0:10:08.000 --> 0:10:10.800 a bit more detail. Yeah, So it's called a system 0:10:10.840 --> 0:10:13.720 on chip or sock, and what that means is that 0:10:14.040 --> 0:10:17.440 it's it has a traditional die like any other piece 0:10:17.480 --> 0:10:20.600 of silicon, but there are little pieces on the silicon 0:10:20.720 --> 0:10:23.839 that each have different jobs. And so instead of having 0:10:24.000 --> 0:10:27.520 one giant piece of silicon that does everything kind of 0:10:27.520 --> 0:10:31.000 well pretty much, this this silicon each has little parts 0:10:31.000 --> 0:10:34.199 that do their job really well, and then together it 0:10:34.240 --> 0:10:36.680 does a lot better as a system, because it's like, hey, 0:10:36.720 --> 0:10:39.079 instead of having a ton of random you know, instead 0:10:39.120 --> 0:10:41.560 of having a general computer that can do everything kind 0:10:41.600 --> 0:10:44.079 of well, they have like, you know, ten specials in 0:10:44.120 --> 0:10:46.560 the house. And when you look at a laptop, really 0:10:46.559 --> 0:10:49.840 you're only doing ten types of workloads or or however 0:10:49.880 --> 0:10:52.600 many number of workloads. And so they said, instead of 0:10:52.640 --> 0:10:55.440 trying to make just a faster CPU, why don't we 0:10:55.440 --> 0:10:58.160 specialize a little bit, and we'll break down all the 0:10:58.200 --> 0:11:01.920 things that the Mac the laptop has to do, and 0:11:01.920 --> 0:11:04.520 we're gonna make them do it really well and with 0:11:04.559 --> 0:11:07.720 these specialized ships. So that's pretty much what I'm what 0:11:08.080 --> 0:11:11.720 the differences, and so this is kind of in the 0:11:11.760 --> 0:11:15.280 picture of the broader things kind of been a revolution anyways, 0:11:15.320 --> 0:11:18.000 Like getting off the CPU has been happening for a 0:11:18.040 --> 0:11:20.800 long time. And part of the reason why now it's 0:11:20.800 --> 0:11:23.080 become such a big shift is because, you know, in 0:11:23.120 --> 0:11:25.480 the last episode with Stacy, guys talked about more laws 0:11:25.480 --> 0:11:27.679 slowing down or maybe it's even you know, they love 0:11:27.760 --> 0:11:29.920 to say it's dead. It's not quite dead, but it's 0:11:30.280 --> 0:11:33.079 it's slowing down quite a bit. And so now the 0:11:33.760 --> 0:11:36.080 road for it definitely seems to be specialization and so 0:11:36.160 --> 0:11:39.000 the M one did this really well. It's specialized in 0:11:39.040 --> 0:11:40.920 a lot of different ways, put it all onto one 0:11:40.960 --> 0:11:43.880 piece of one piece of silicon, and also coupled the 0:11:43.920 --> 0:11:46.120 memory really close together, which is a whole you know, 0:11:46.200 --> 0:11:48.440 kind of in the same vein but a little different. 0:11:48.760 --> 0:11:52.000 And now the M one just rocks. Frankly, it's it's 0:11:52.040 --> 0:11:55.280 not yeah, it's it's a little bit shaky as a 0:11:55.320 --> 0:11:57.400 first product, right like the Apple Watch one or the 0:11:57.440 --> 0:12:00.720 iPhone one wasn't the best product ever, but it's still 0:12:01.040 --> 0:12:21.000 it still is, um a miracle, quite frankly. So you know, 0:12:21.080 --> 0:12:24.200 you talk about how this sort of emerged out of 0:12:24.600 --> 0:12:27.720 Apple's chips for the iPhone and something I think, you know, 0:12:27.760 --> 0:12:31.800 like I buy a new iPhone every few years like 0:12:31.920 --> 0:12:34.800 anyone else. Um, I don't buy every single one that 0:12:34.840 --> 0:12:37.160 comes out, but you know, when my mine feels like 0:12:37.160 --> 0:12:39.079 it's getting old or whatever, I buy a new one. 0:12:39.559 --> 0:12:42.280 But honestly, the only thing I do on my iPhone 0:12:42.320 --> 0:12:46.080 besides tweeting is taking photos. And so even though like 0:12:46.120 --> 0:12:49.479 I have this like incredible computer in my pocket, basically 0:12:49.640 --> 0:12:52.920 is just a camera with Twitter attached or Twitter with 0:12:52.960 --> 0:12:56.839 a camera attached. And so I'm curious like how much 0:12:57.360 --> 0:13:01.120 has this sort of like shift tour words. Uh, These 0:13:01.160 --> 0:13:05.040 sort of focused tasks instead of generalist chips come out 0:13:05.080 --> 0:13:07.720 of this fact that yes, you have I have this 0:13:07.760 --> 0:13:11.280 whole computer. Well, there's one thing that it's really important 0:13:11.320 --> 0:13:13.679 for the iPhone to do for a lot of people, 0:13:13.720 --> 0:13:17.040 and that is take really clear can really uh really 0:13:17.120 --> 0:13:20.240 clear photos, and this sort of recognition that there are 0:13:20.280 --> 0:13:23.360 just a few simple tasks that on a phone you 0:13:23.400 --> 0:13:27.280 have to get done extremely well for most people. Yeah, 0:13:27.320 --> 0:13:30.320 that's actually the perfect that's like the perfect thing to 0:13:30.320 --> 0:13:33.240 bring up because on the Apple phone, right, the A 0:13:33.360 --> 0:13:36.160 fourteen or A fifteen or whatever iteration of the A chip, 0:13:36.400 --> 0:13:39.280 the biggest die, the biggest specialized die is something called 0:13:39.280 --> 0:13:43.880 an IPU or an image process Sorry can you die when? Uh? 0:13:44.280 --> 0:13:46.320 Die is just like a piece of silicon. Sorry, it's 0:13:46.320 --> 0:13:49.679 it's um. Yeah, just just think of it as like 0:13:49.720 --> 0:13:52.200 a unit of silicon, if that makes sense. And the 0:13:52.240 --> 0:13:55.200 biggest piece of like that's been taking share from everything 0:13:55.200 --> 0:13:58.640 else is an IPU, and the CPU has become a 0:13:58.679 --> 0:14:01.520 smaller and smaller part and similar in that same vein right, 0:14:01.559 --> 0:14:03.560 like computers, one of the things that we've been using 0:14:03.600 --> 0:14:06.719 the most is GPUs to play video games. So just 0:14:06.800 --> 0:14:09.720 like the CPUs kind of losing relevance or the you know, 0:14:09.880 --> 0:14:13.080 the CPS losing relevance in your phone to an IPU 0:14:13.280 --> 0:14:16.840 or for your images in in a desktop computer, for example, 0:14:16.840 --> 0:14:19.600 the CPU seems to be losing relevance to a GPU 0:14:19.680 --> 0:14:21.560 for a lot of people who are using it for games. 0:14:21.640 --> 0:14:24.480 So kind of that same that same energy is you know, 0:14:24.600 --> 0:14:28.600 disrupting one and the other. And I think you ask 0:14:28.640 --> 0:14:32.400 a question about like why Apple Apple in the iPhone 0:14:32.400 --> 0:14:35.880 specifically kind of like shifted over to the Mac. What's 0:14:35.920 --> 0:14:39.120 interesting about that is in the iPhone itself, they actually 0:14:39.160 --> 0:14:42.200 had a lot of requirements that were not that you 0:14:42.200 --> 0:14:44.680 didn't really need to use in a in a computer, right, Like, 0:14:45.040 --> 0:14:47.080 it has a lot smaller form factor, it has a 0:14:47.080 --> 0:14:49.760 battery life constraint. It has to also not have a 0:14:49.760 --> 0:14:52.760 fan to actively cool it, and so all these things. 0:14:52.960 --> 0:14:54.680 Um in the beginning, we're just like, you know, a 0:14:54.720 --> 0:14:58.040 pain pain for them to to manufacture around. But as 0:14:58.080 --> 0:15:00.000 they got better and better and better, they pretty much 0:15:00.080 --> 0:15:03.000 got really close to a laptop in an iPhone. And 0:15:03.040 --> 0:15:06.120 that's while having a smaller form factor, while having no 0:15:06.600 --> 0:15:09.200 active fan. And then they're like, wait, why don't we 0:15:09.240 --> 0:15:11.520 just do this for a laptop. And so that's kind 0:15:11.520 --> 0:15:13.560 of what they did. And the M one really is 0:15:13.640 --> 0:15:17.160 like a super souped up um Apple chip. And the 0:15:17.160 --> 0:15:19.760 the announcement that you guys just broke the story on 0:15:20.000 --> 0:15:22.040 is that they're going to actually be making even bigger, 0:15:22.080 --> 0:15:24.640 beefier versions of this because the M one is actually 0:15:24.680 --> 0:15:27.240 made to be put in a laptop with no active cooling. 0:15:27.400 --> 0:15:29.360 Like that's also a big part of it. There's like 0:15:29.400 --> 0:15:31.400 it's supposed to not heat up, it's supposed to not 0:15:31.400 --> 0:15:33.440 have a fan. But what if we what if we 0:15:33.520 --> 0:15:36.120 just like let it run loose and you know, actively 0:15:36.120 --> 0:15:38.600 cool to put a fan. See how hot, how hot 0:15:38.640 --> 0:15:40.920 we can get this thing going? And I think what's 0:15:40.920 --> 0:15:43.280 gonna happen is we're gonna be really shocked by the outcome. 0:15:43.720 --> 0:15:47.040 And so all the manufacturing and the constraints that they 0:15:47.040 --> 0:15:49.640 had to do to make the chip fit into a 0:15:49.680 --> 0:15:52.680 small phone pretty much when you actually scaled it up 0:15:52.680 --> 0:15:55.240 into a laptop, it made it a lot better. And 0:15:55.280 --> 0:15:57.160 that's kind of part of the M one, you know, 0:15:57.240 --> 0:16:00.600 the Apple silicon magic. And Apple isn't the only company 0:16:00.640 --> 0:16:04.080 pursuing this, like I think, you know, almost every company 0:16:04.200 --> 0:16:06.600 is kind of looking at Intel and saying like, okay, 0:16:06.640 --> 0:16:09.080 we need to transition off this vendor. And whether it's 0:16:09.080 --> 0:16:11.520 a data center or whether it's a phone, or whether 0:16:11.560 --> 0:16:14.600 if it's um anything else, Like everyone is kind of 0:16:14.640 --> 0:16:17.160 going this way. And that's what's exciting. Is this isn't 0:16:17.280 --> 0:16:20.040 just an Apple story. Qual Calm mentioned they might do 0:16:20.080 --> 0:16:23.680 heterogeneous chips, which is like one way to call this 0:16:23.720 --> 0:16:26.720 whole thing is heterogeneous, as in many different types of 0:16:26.800 --> 0:16:30.120 chips put together into a larger package. And so you know, 0:16:30.240 --> 0:16:33.640 Android might have a similar and one like chip very soon. 0:16:33.920 --> 0:16:36.160 That's probably you know, a few years away. But the 0:16:36.200 --> 0:16:39.320 writings on the wall that everyone has to shift into 0:16:39.640 --> 0:16:42.920 the this kind of M one style of chip, if 0:16:42.920 --> 0:16:46.640 that makes sense. So it does make sense. Um, this 0:16:46.720 --> 0:16:48.200 is actually what I want to ask you next. So 0:16:48.280 --> 0:16:52.400 I think I understand why Apple sort of struck upon 0:16:52.520 --> 0:16:56.680 this design given its experience with the iPhone, why it 0:16:56.840 --> 0:17:02.680 thought that specialized chips for specialized asks would be more efficient. 0:17:03.280 --> 0:17:07.879 But why was it able to actually produce the chip 0:17:08.320 --> 0:17:10.639 ahead of time? Or maybe another way of saying this 0:17:10.760 --> 0:17:14.680 question is, Um, why was Apple able to do this when, 0:17:14.720 --> 0:17:18.040 as you just mentioned, the entire industry seems to, you know, 0:17:18.240 --> 0:17:22.880 maybe at different speeds be moving in this direction. So 0:17:23.160 --> 0:17:27.639 Apple silicon has always been a leader. They've they've been 0:17:27.680 --> 0:17:30.080 they've been crushing it for a long time. But in 0:17:30.119 --> 0:17:32.640 particular this this kind of goes back to the last 0:17:32.640 --> 0:17:35.439 episode you guys had. T SMC really enabled them to 0:17:35.520 --> 0:17:38.840 do this, and Apple has been buying and building silicon 0:17:38.920 --> 0:17:41.959 expertise in house for years. This is not an overnight 0:17:42.000 --> 0:17:45.639 success story. The the original chip, like the A seven 0:17:45.720 --> 0:17:48.560 or whatever, was when they first started to um to 0:17:48.720 --> 0:17:51.639 really make this more heterogeneous, and they've been every year 0:17:52.440 --> 0:17:55.480 it adds more specialization, and it's gone to this point 0:17:55.480 --> 0:17:57.959 where now the CPU is like one of the not 0:17:58.040 --> 0:18:00.560 the least important, but it's just one aspect of the 0:18:00.640 --> 0:18:03.520 chip in the phone. And so Apple has been doing 0:18:03.520 --> 0:18:06.320 this for a long time. T s MC has been 0:18:06.359 --> 0:18:09.480 a huge part of it enabling this, and then other 0:18:09.560 --> 0:18:12.360 companies kind of look to Apple and are seeing like, okay, 0:18:12.400 --> 0:18:15.200 we can do this too. So I hope that kind 0:18:15.200 --> 0:18:18.159 of answers the question when it comes to manufacturing, and 0:18:18.160 --> 0:18:20.560 of course that was the big thing we talked about 0:18:20.640 --> 0:18:24.120 on our last chip episode, and this idea that it's 0:18:24.119 --> 0:18:27.200 like getting more and more difficult to sort of shrink 0:18:27.280 --> 0:18:31.720 these chips have the smaller an animator form factor. Is 0:18:31.760 --> 0:18:34.960 that an issue here? Or does sort of rethinking the 0:18:35.000 --> 0:18:39.800 fundamental architecture of the chips such that it's it is 0:18:40.080 --> 0:18:45.159 h heterogeneous chip. Does that sort of sidestep some of 0:18:45.200 --> 0:18:49.760 these issues? If that's does that that questions? Yeah? Yeah? 0:18:50.000 --> 0:18:52.359 Um actually, I think that's the huge driver of why 0:18:52.400 --> 0:18:55.480 this whole thing exists, right, Moore's law slowing down and 0:18:55.520 --> 0:18:58.440 the struggle to to shrink smaller and smaller pretty much 0:18:58.480 --> 0:19:02.359 made them think, um, hey, we have to improve this somehow. 0:19:02.600 --> 0:19:04.280 We can't do it through the ways we used to 0:19:04.280 --> 0:19:06.280 do it. Why don't we specialize a little bit. So 0:19:06.320 --> 0:19:08.359 it's kind of like a mix of both that's happening 0:19:08.359 --> 0:19:10.960 at the same time. I definitely think that the desire 0:19:11.080 --> 0:19:16.399 for better, faster chips was what helped the heterogeneous revolution begin. 0:19:16.960 --> 0:19:20.640 But um t SMC being there to essentially make it 0:19:21.040 --> 0:19:24.120 so that anyone who isn't because in the past they 0:19:24.119 --> 0:19:27.679 all until also owned the method to get smaller. Right, 0:19:27.720 --> 0:19:29.800 So it's like, you know, you can't compete against some 0:19:29.920 --> 0:19:32.960 a general chip that's getting smaller, getting faster, better, faster 0:19:33.040 --> 0:19:35.760 than you could specialize your chips faster. But now that 0:19:35.800 --> 0:19:38.480 t SMC is clearly in the lead and anyone can 0:19:38.640 --> 0:19:41.520 can reach out to them and work with them, now 0:19:41.560 --> 0:19:43.360 everyone's like, well, why don't we just make our own 0:19:43.400 --> 0:19:46.760 special little chips? And so that's really been um kind 0:19:46.800 --> 0:19:48.920 of all the things that's made this happen at this 0:19:49.040 --> 0:19:51.680 right moment right now. I mean, on that note, how 0:19:51.880 --> 0:19:56.000 easy is it going to be for competitors to make 0:19:56.119 --> 0:19:58.520 these specialized chips? Then are they going to be able 0:19:58.600 --> 0:20:00.919 to do the same thing that apples do? And in particular, 0:20:01.600 --> 0:20:04.480 is Intel going to be able to do it? Given that, 0:20:04.600 --> 0:20:08.080 as we discussed in the last episode, their business seems 0:20:08.160 --> 0:20:11.800 to be almost entirely based on this idea of selling 0:20:12.040 --> 0:20:16.760 you know, general purpose CPU which then can go into 0:20:16.800 --> 0:20:21.760 a large variety of desktop PCs or whatever. Yeah, so 0:20:21.920 --> 0:20:24.840 Intel is already pursuing this. I mean, the people in 0:20:24.920 --> 0:20:27.200 Intel are smart and they know what they're doing. Um, 0:20:27.200 --> 0:20:30.160 but they're obviously tied to a very large platform. It's 0:20:30.200 --> 0:20:33.439 something called Phobos, which is like kind of their method 0:20:33.520 --> 0:20:36.159 of stacking the chips together. They have this um, this 0:20:36.240 --> 0:20:38.920 new platform and their Intel architecture. Day They're like, okay, 0:20:38.960 --> 0:20:40.919 we're gonna go all in on chiplets, like we're going 0:20:40.960 --> 0:20:43.160 to do the same thing that everyone else is doing. 0:20:43.480 --> 0:20:45.880 But that kind of brings it back to the point 0:20:45.920 --> 0:20:47.480 like Intel is going to be doing the same thing 0:20:47.520 --> 0:20:49.320 that everyone else is doing. And if they don't have 0:20:49.359 --> 0:20:53.280 the manufacturing advantage like they used to, pretty much everyone's 0:20:53.280 --> 0:20:56.439 on the same level playing field and it's it's for 0:20:56.520 --> 0:20:58.879 real right now. Am D for example, does have like 0:20:59.160 --> 0:21:01.920 a big, big reason why they won was kind of 0:21:01.960 --> 0:21:04.800 like a process heater geneous thing where it's like, Okay, 0:21:04.880 --> 0:21:06.639 we don't have to be the fastest, you know, we 0:21:06.640 --> 0:21:09.080 don't have to have the smallest um chip. But what 0:21:09.119 --> 0:21:10.760 we're gonna do is we're just going to make sure 0:21:11.200 --> 0:21:14.280 this really important part is is super small and that way, 0:21:14.440 --> 0:21:17.320 you know, it's like getting across across the finish line 0:21:17.600 --> 0:21:19.520 um by having the best of the little pieces that 0:21:19.560 --> 0:21:21.959 really matter and having a lot of like not as 0:21:22.000 --> 0:21:24.760 good pieces that didn't really matter as much. And so 0:21:24.960 --> 0:21:28.720 they're everyone is kind of going to this like hybrid playbook, 0:21:29.080 --> 0:21:32.240 and every company to a certain extent is pursuing this. 0:21:32.440 --> 0:21:35.560 And yeah, I mean I can talk to individual companies 0:21:35.560 --> 0:21:37.159 in particular, but like I know, it's like I know, 0:21:37.240 --> 0:21:40.280