Samsung Introduces the Galaxy Z Tri-fold - DTNS 5156

0:04This is the Daily Tech News for Tuesday, December 2nd, 2025. We tell you what you need to know, give you some of the important context, and help each other understand. Now listen, AWS reInvent is happening, and we got a lot to say about it, but today the trifold smartphone era has officially begun. I mean, Huawei is like it already began, but Samsung entered the chat. That's big news. That's big news. I'm Jason Howell. I'm Tom Merritt. Let's start with what you need to know with a big story.

0:34that's right samsung has entered the room and unveiled the galaxy z trifold as it sounds it is a triple panel foldable smartphone or trifold phone that opens into a 10 inch tablet like display pretty much i mean a 10 inch tablet like that's huge when it's unfolded i can't wait to check it out myself. When it's all folded up, it works as a 6.5-inch smartphone. So it's kind of the best of both worlds, right? The device has two titanium-housed hinges of varied sizes. Why are they different sizes? Well, that accommodates the fact that each panel varies in thickness from 3.9 millimeters up to 4.2 millimeters at its thickest.

1:19I think the center panel is that thickest one. The side panels actually fold inward to cover the centerfold. So it doesn't fold like the letter Z. I think Huawei has a triple tri-fold that does that. In this case, both panels fold inward. And Samsung says that the main display, quote, undergoes a 200 ,000-cycle multi-folding test, equivalent to folding the device approximately 100 times a day for five years. So totally normal, right? That's going to cover everything. Yeah, that's what I do in a day. That's what I do in a single day. Yeah, every single day for the rest of my life. Although it is a nice fidget when you've got a foldable, so I wouldn't be surprised.

2:01You just did two right there, right? 98 more to go. There you go. When folded, the device measures a total of 12.99 millimeters thick. These are just numbers. It's easier to compare it to what we've seen before. Compare that against the Z Fold 7. When that's folded, it's 8.9 millimeters thick. But that's also a thinner foldable. You've got the Z Fold 6 that came one year before. That device, when it was folded, 12.1 millimeters. So actually, this tri-fold is not far off that benchmark. As for the rest of the specs, kind of the rattle of specs, IP48 rated. So water resistance, yes. Dust resistance, not so much.

2:44uh snapdragon 8 elite for galaxy processor 16 gig ram 512 gig storage a 5600 milliamp hour triple cell battery so that's how they can kind of keep you know the smartphone slim um not bad there trip i'm kind of surprised about this triple rear camera setup on a smartphone like this they were able to put a 200 megapixel main a 3x optical telephoto so you get all three cameras however However, even though this is a tablet, when you unfold it, it does not support Samsung's S Pen for stylus use. So that's going to disappoint some people. If you're wondering how much all this costs, I'll just tell you right now it's expensive.

3:25It's going to launch in South Korea on December 12th for$3 ,590 ,400 South Korean won. Let's convert that. Okay. Somewhere around the$2 ,500 U.S. ballpark. Samsung does plan a U.S. release for this sometime in the first quarter of 2026. We don't know any details. We don't know what the price is going to be, but that should give you an idea of where this might stand. Yeah, it's going to be a couple thousand dollars. Are you just itching for this? Yeah, oh, for sure. Easily a couple of thousand dollars, especially when you consider like their Z Fold. What is that? Like$1 ,800,$1 ,700,$1 ,800, somewhere in that ballpark.

4:03This is a cutting edge device, the next cutting edge foldable device. So guaranteed you're breaking$2 ,000 at least. I'm curious about the way it folds instead of doing the Z. That's interesting. It's different. And I'm curious if I would like that better or not. When I say this, this is not me expressing skepticism even. That's just like, huh, is that good? Because that's why you have to have the different thicknesses, right? Because you're just folding one over the top and then the other over the top. And then you're going to have it three thick anyway. It's just a matter of whether you're folding one behind or in front of.

4:43So, yeah. But if it folds flat when it's in 10 inches as a tablet, then I guess it doesn't really matter how you fold it. Screenshots that I've seen seem like it does fold flat. Those earlier foldables, for those who may remember, is like you could unfold it, but it didn't quite go completely flat. I think that is a relic of the past as far as what we're seeing in foldables now. This does seem like it would fold flat. One of my curiosities about the folding mechanism is if they're both folding inward, that essentially means that there's a order in which you have to remember to do that, right? One panel has to fold first so that the other panel reaches over the top and folds the other way.

5:27And I wonder from a durability standpoint, if people get that wrong or if that's even, you know, a non-issue. Yeah, if that's a factor or not. But it might just be a little bit of an annoyance when you like start to fold the wrong one, the wrong, and you're like, oh, wait, I have to do it the other way. Whereas the Z Fold, I guess, kind of makes more natural sense to be like, oh, I'm just going to squish it down. Like, you know, it'll naturally fold the way it's supposed to. 10-inch display is, you know, that's what I've got on my tablet sitting in front of me now. So that's real tablet size, you know, so that's cool.

5:59I love that it's got DeX, Samsung's system for using it as a computer. So you can attach a wireless keyboard and mouse, or I guess you could wire or wireless keyboard and mouse, and you don't need to attach it to a monitor because it is a 10-inch monitor, right? That's really cool. That seems pretty cool. You can just set that thing up and suddenly, you know, start computing away on it. 10 inches is not a huge monitor for that situation, but in a pinch on an airplane or traveling, that could be useful for folks. This is a bit of a proof of concept. This is definitely an early adopter model from where I sit, don't you think?

6:38Oh, 100%. I think still, even just the two-folding panel device is not a proof of concept. Obviously, it's been around for a while. But I haven't known many people that actually have one, that actually look at the two-panel foldables. And I'm not even talking about the clamshell. The clamshell is a little different. That, I think, has a wider commercial appeal. Like the Z Flip. Right, the Z Flip, exactly, or the Motorola Ultra. Yeah, when we're talking about the book style, I still feel like it's a real niche kind of group of people. People that love it absolutely love it, of course. I'm not saying that there aren't people that are willing to spend a large amount of money more for a phone that does that.

7:26Then now we're talking about a larger and even larger, more expensive device. I think that funnel gets even smaller. I completely agree. I think this is firmly in the kind of trying to prove itself range. change also just kind of like paving the way for you know maybe there is a time down the line where this is just more widely adopted makes more sense on a wider commercial scale and you got to start somewhere it's i mean to that end this it's really impressive to me that devices like this even exist right now you know yeah it's a technological marvel i think i think it's going to get us to a place where this becomes a commodity and gives you uh pardon the pun flexibility and design because I like my Pixel Fold.

8:13I don't love it. It's not like, oh, I wish every phone folded. I can never go back. But if every phone can work in the normal phone style and then unfold when I need it to, that would be a nice perk. I don't, at this point, want to pay extra for it now that I've tried it. I want that to be just a feature that is there with nice cameras and storage and all of that. And I think we are headed that way. Apple's going to join this, it sounds like, within the next year. And once they do that, then the trifold becomes the leading edge, and this starts to be commoditized. And you start to see medium-priced phones start to have foldables because economies of scale happen, and the technology gets cheaper to make.

9:00Yeah. Yeah. I think when I think about this, I realize my mind is kind of also going to the AI glasses kind of territory, which is – and hopefully my comparison makes sense. But in that regard, AI glasses don't make a whole lot of sense until the glasses are the same as other glasses. And then suddenly because there's no extra like, oh, well, I've got to commit to the fact that what I'm wearing on my face are these huge, chunky beasts in order to get that advantage. Now I'm just putting on a normal pair of glasses, and I get that advantage, so why wouldn't I? And I think it's the same with these foldables.

9:35This tri-fold is actually thinner than I kind of expected that it would be. The fact that it's only an inch thicker than last year's Z Fold really brings it into the territory to where it can be a smartphone 90 % of the time. And that 10 % of the time where you're like, man, I wish I had a tablet right now, you can unfold it. Did you say an inch thicker? Because an inch thicker is a lot. well is it well so what is it it's the i'm doing that off rattling off my uh my head here uh the 12 12.99 millimeters thick compared to the z fold 6 12.1 so not quite a full inch i mean yeah it is not an inch you're talking about millimeters right so it's it's a it's a half a centimeter oh dude okay yeah yeah sorry about that okay sorry just check it's been it's been a morning Thank you.

10:25No worries. If we would just all get on the metric system, none of this would be able. Thank you. Thank you, Tom. I appreciate that. But my point being, most people aren't going to sacrifice the fact that they have, let's say, if you put your phone in your pocket, are you okay with having a massive brick in there for the use or for the fact that you have a tablet in those small amount of times when you need it? Most people possibly or probably are not going to do that. But if it's just the same size or near the same size as a normal phone you'd put there and it does that extra thing, then we're talking more widespread appeal.

11:02Yeah. Agreed. Thank you for the correction. Yeah, no worries. I knew somebody was going to be like an itch there. Oh, yeah, for sure. You guys. For sure. And instead of doing that, now you can become a patron and support us because that's how DTNS is made by you, the listener. Big thanks to Philip Les, Howard Yermish, John Atwood, and our new patron, Rob. Welcome, Rob. Thank you for joining us. There is more we need to know today. Let's get to the briefs. Yes, indeed. AWS reInvent 2025 kicked off on Tuesday. A lot of this is relevant to cloud industry insiders, but there are a few notable things that the general consumer might want to be aware of.

11:45uh aws introduced the latest version of its llm training chip the three nanometer tranium three three nanometers it's the third one that they've made training is right in the name it all makes sense uh it will power a new system called the tranium three ultra server aws promises the system however will be four times faster than the current system with four times the memory available for training and inferencing. And you can link more Ultra servers together now than you could before. Each Ultra server can host 144 Tranium 3 chips at once, and you can link enough of those together to be able to use a million Tranium 3 chips for your app if you're an app provider.

12:30That means your apps are going to get better. The systems are also 40 % more energy efficient. That's AWS's way of trying to get into this race with NVIDIA and Google to say, hey, if you use us, we use less energy, which means they're cheaper. And they're cheaper in other ways, too. AWS says Anthropic, LLM, Karakuri, Splash Music, and Descartes have all been testing Tranium 3 systems. And overall, inferencing costs are down 50 % as a result. They didn't say for any single one of those it is, but they say in general, their costs fell 50%. The next chip in the line will be called, as you might guess, Tranium 4, and will support NVIDIA's NVLink Fusion high-speed chip interconnect technology.

13:15That means customers that build apps with NVIDIA GPUs in mind can use Tranium 4 more easily. But there's no timeline on when that will come. My guess is about this time next year at AWS reInvent, we'll be talking about Tranium 4. But they're letting you know what the roadmap is. So you were saying the 3 nanometer, Tranium 3, the tie-in. And as we know, these things, probably next step on that elevator is from 3 nanometer to 2 nanometer. Yeah, no, it's the third Tranium they've done. The previous one was Tranium 2. The 3 nanometer and the 3 in the name are just coincidence. So it lines up right now, but that's not an indication of the future.

13:54They're going to reinvent what the Tranium 2 actually means when they hit 2 nanometer down the line. Reinvent, I see what you mean. Nice. uh aws also offering a service called nova forge which costs the company a hundred thousand dollars a year to get access to amazon models early in the training process and if you know what this means as early as pre-training this is something you cannot do unless you're an ai company right now so it's kind of a big deal a hundred thousand dollars sounds like a lot to me but for a company that would want to customize a model to its own needs without having to spend tens to hundreds of millions of dollars, it can now train this stuff, not quite from scratch, which would be way more expensive, but almost as good as training from scratch for a lot less money.

14:42Nova Forge lets a customer supply its own data. It can refine open-weight Nova 2 Pro or Nova 2 Lite models and continue training on its own compute infrastructure. So you're not going to get AWS help training these. You're not going to get AWS infrastructure included in the price. This just lets you take an early pre-training model and make it yours for a lot less money than you would otherwise. Reddit did it and said we could not get the system to perform the way we wanted to when we were taking a fully trained model and then trying to get it to use our data. When we put Reddit data in from the beginning, the model really understood how Reddit worked, and they are going to use it to improve its auto-moderation performance, among other things.

15:29This is useful for Amazon because they're at the back of the pack for Enterprise LLMs with 5 % of the market. Anthropic, you may not realize, leads the way for Enterprise LLM with 32 % of the market. Amazon says the Nova 2 Pro model benchmarks at least as well as Claude Sonnet 5.5, GPT 5.1, and Gemini 3.0 preview and is available in early access to Nova Forge subscribers. Start now. Now, Amazon also announced the Nova 2 Omni model, which can process images, speech, text, and videos as input. I think it's maybe the only model that can do all of those as input at once. And then it can generate images and text as its output.

16:11And there's also a new real-time voice model called Nova Sonic, which I imagine will inform your Amazon voice assistant down the road. Yeah, this is all really interesting. I mean, it's such a wild time right now for the development of the major companies trying to kind of instill or kind of introduce their own major plays for the platform and the infrastructure of so many other companies. You know, they're all relying on each other. You've got Google doing its thing with its TPUs. You've got OpenAI, and it's really kind of tight in integration with the NVIDIA pipeline and Amazon doing what it's very capable with in the cloud and really kind of supercharging it with its own AI infrastructure.

16:58It's kind of confusing to kind of follow along, but everyone wants to be in control of these systems, right? They get so much value. I mean, if I understand it right, though, they don't expect to be in control. They don't expect everybody's going to use just one chip. NVIDIA chips are good for certain things that TPUs from Google are not. And so you still see people mixing those. Same thing for Tranium 3. They expect you to want to use NVIDIA chips. That's why they're going to make Tranium 4 use the Interconnect technology so that you can use Tranium 4 with NVIDIA chips easier. I don't see this stuff as, uh-oh, Google's coming for NVIDIA's lunch so much as Google is coming for side dishes and possibly will be the lunch with NVIDIA becoming the side dish.

17:44But you're going to use a lot of these things because, again, NVIDIA chips, you can buy yourself and build your own data center. You're not going to buy Tranium 3 chips and build your own data center. AWS is going to build that data center for you, at least at this point. Meta is going to lease the chips from Google. So it's a lot more complicated than just who, you know, like you and I, when we build a desktop, we pick a GPU and we put it in the desktop. You can mix and match a lot of this stuff. You can mix and match build versus buy versus rent, all that kind of thing. I think the most interesting part of this entire announcement, though, is the NovaForge thing, because I have been of the opinion.

18:22I keep saying these LLMs are going to get useful when you can customize them, when you can train them on your own data. And this makes it much more cost effective to do that. 100 % yeah being able to really focus it entirely that's that's the key and I think we're going to see a lot more of that in the coming couple of years a real uh double down on that so yeah you're right um also I just I had this vision of like all these companies at their own thanksgiving dinners um when you were talking about side dishes and stuff yeah I couldn't tell you what that vision looks like other than that there were just a bunch of cranberry sauce right now but they would really prefer to at least to be dressing.

19:02Yeah, yeah. Apple's AI chief, John Giandria, John Andrea, thank you very much. John John Andrea, there we go. Thank you. Set to retire next spring, shifting to an advisory role, at least until then, former Microsoft corporate VP on AI and, by the way, longtime Google engineering lead, I think with Gemini there, Amar Subramania takes over as vice president of AI at Apple. He's going to report directly to Craig Federighi and will lead Apple's foundation models, machine learning research, and AI safety efforts within the company. The remaining parts, the remaining pieces of John Andrea's organization will move to Sabi Khan.

19:46Did I get that one right, Tom? And Eddie Q. A lot of names that I'm having a hard time with today. Yeah, Eddie Q is a tough one. I get it. Yeah, I know. Seriously. As part of a broader leadership shuffle that shifts AI infrastructure and search knowledge to align with their existing teams. So definitely a lot of reorg happening here. Obviously, a lot of this comes on the heels of at least a few very public and key, I'd say, AI misses with the kind of failure to ship the Apple Intelligence Siri upgrade, among other things, some defections in the departments, that sort of stuff. I think some of that stuff does get exaggerated, but I'm not going to deny that it plays a big part in this.

20:30John Andrea was also involved in Apple's car development that never happened. So you can see sort of a pattern of like really good with AI, not really good with delivery. And as someone once said, good artistship. So I think John Andrea is in his 60s. It's not like he has to step down, but he's had a long story career. The guy worked on Netscape. He was on the web team for Netscape. And he may just be like, you know what? I want to go off and do a different thing anyway. So it does seem like it's an amicable parting of the ways. I don't know enough about Amar Sabramanya to say whether this is the right choice or not, certainly well qualified with the experience, right?

21:14Definitely seems qualified. I mean, a lot of direct AI leadership experience leading into this, you know, and knowledge, especially when we're talking about, you know, some of the reports that Apple might be utilizing Gemini for an upcoming Siri update and everything. So, I mean, if he knows some people, you know, for his old job, right? 100%. Although you could say like he was part of Gemini when it wasn't good. And now it is good after he left. Like probably correlation, not causation. Yeah, I'm guessing. Yeah, same. Big thanks to Motang on our DTNS subreddit for keeping us current on this next one.

21:51We mentioned briefly yesterday that India's telecoms minister had given all phone makers in the country 90 days to begin preloading the state-owned anti-theft Sanchar Sathi app. The order suggests that users would not be able to delete it. Phone makers have to put it on there and keep it on there. the app lets you report and lock lost or stolen devices based on your phone's IMEI number. So it's a good cause that they want to do here. Basically, it makes it easy to ask carriers to prevent that phone from connecting to networks. It works. There's 5 million installs out there. People swear by it, say, yes, it has stopped.

22:30It's reduced theft. It helped me out to prevent fraud because you can also report fraudulent phone calls. But Reuters reports that Apple will tell the Indian government that it does not comply with such orders in any market due to security and privacy risks. So Apple's going to say, look, you got rules in your country about what can be in the app store? Fine. You got rules on what we have to put on our phone from the beginning? Not fine because we don't control that code. So we are not going to be responsible for putting it basically as part of the operating system. Yeah, this is curious. I mean, India is such a big country for Apple for a number of reasons.

23:15Oh, huge, right? And I wonder how much of a choice Apple's going to have here in the end. Well, how much of a choice does India have too, right? Like this is a little bit of a staring contest. I think they can come to an agreement to say, we'll put a flow in phone setup where we offer to let people install it right away. Or, hey, we've got our own anti-theft stuff that's just as good. Maybe that counts. That was my other. Whether India buys that, it's just as good or not. I don't know. I don't understand why India is being so pushy with this, except that it works. And they're like, well, we can cut down thefts even more if we just tell everyone they have to have it on there.

24:02But Apple comes from the let the user decide school sometimes, not all the time. Basically, Apple comes from the like, if Apple decides it's good and Apple made it, then we'll tell the user what to do. Otherwise, no. Yep, exactly that. Yeah. Then it's not like it's a non-issue either, the thefts of these devices in India. It all makes sense. In any place, yeah. Yeah, in any place. Europe, UK, there's lots of places where theft of these devices is skyrocketing. I wonder if there's – is there something – is there a deeper kind of protection for users from the IMEI model versus what Apple offers with fine line locking?

24:47I mean, it's at the carrier level, so it's a little more ubiquitous, right? You can't just – there's fewer ways around it, but it seems like both are fairly effective. Yeah. You know, maybe we see a future where Apple bakes some sort of IMEI thing into Find My 2-Russes. Well, I'm not certain that they don't use the IMEI system in their flow as well. But I think the thing that's safe to say is Apple could say, we'll ask if they want to install it. We'll encourage them to install it, but we're not going to pre-install it. And we're going to see what India says about that. Yeah, we will see. OpenAI CEO Sam Altman has reportedly put the company on code red.

25:30This is according to a report by The Information. And all of this is apparently in a reaction to, at least to some degree, the latest Gemini release, along with Anthropics Momentum with Enterprise customers that we were just talking about a few minutes ago. The internal memo shares that newer company initiatives like ChatGPT ads, also shopping and health agents, also the Pulse personal assistant will be prioritized to make room for improvements to ChatGPT's speed, personalization, reliability, informational capability, all that kind of stuff. OpenAI is facing a lot right now. And Gemini 3, say what you will about it, it got a lot of positive press.

26:16It's had a lot of positive reaction. And so, yeah, sounds like a little bit of a reaction from OpenAI. Hey, Code Red works for Google. Totally. Exactly. Just a couple of years ago. And I feel like Sam Altman is rightly identifying that maybe they were getting a little ahead of themselves. Maybe they were resting on their laurels a little bit and doing some things that weren't exactly what the users wanted. I'll say, yesterday, I tried to get ChatGPT to do a spelling and formatting check, as I always do, and it stopped me. It said some weird thing about violating creator content stuff. And I'm like, this is all my own writing that I'm putting in here.

26:58So I went to Gemini. Gemini not only had no problem doing it, but it did it faster than chat GPT 5.1 usually does. So I mean, one point of data does not prove anything. But in my own experience yesterday, I was like, oh, Gemini is actually better at this, at least today. We'll see about today, but it was yesterday. One point of data might not mean something for the wider kind of scale of what this is all about. But on a user perspective, on a user basis, a user might only give something one try and realize that. And that's all they need to make their decision and be like, oh, well, Gemini is better.

27:33I'm just going to use that from now on. So yeah, it is important. Definitely matters. All right, let's finish up with some quick headlines that are good to know if you just want to understand the news in the future. Yeah, Mistral launched its Mistral 3 lineup, a 10-model release that combines an open-weight frontier model with compact, offline-capable models that enterprises can fully customize, kind of like what we were talking about. Keep an eye on Mistral. Apple Music Replay 2025 is here with sections that showcase user discovery of your new artist, loyalty to those artists, listeners came back to, comeback artists who returned to the listeners rotation.

28:08Your total minutes listened, number of artists listened to. I listened to 7 ,111 minutes of music last year. Wow. Yeah. Good work. I don't have Apple Music. Otherwise, I'd give you my stats, too. But I do have YouTube, and YouTube is doing the end-of-year recap thing as well in its main app, supposedly, highlighting users' top videos, channels, topics, YouTube music stats. So you can find all that there. So apparently you can find this in your You tab inside of the app or at YouTube.com slash recap, neither of which is giving me my stats. Otherwise, I'd read you some stats as well. Yeah, me either.

28:51This happens every year. We get the news and it takes like a day for it to show up for me. Amazon will test 30-minute deliveries in parts of Seattle and Philadelphia for select grocery store, drugstore-style items. Cost you$3.99 if you're a Prime member or$13.99 if you're a non-member. Amazon already does 15-minute delivery in the UAE. And man, India, Korea, Japan, they all have like super fast delivery. Southeast Asia, all over the world. China, super fast delivery is a thing in large parts of the world. to Amazon's kind of late to this game and wants to bring it to the US. So 30, 15 minute over there.

29:28Then we're going to see it like, at some point, it's going to be like 13 minute. Some of the India delivery places have been doing five minute deliveries. No joke. Oh, are you kidding me? No joke. Yeah, it's crazy. That is crazy. And finally, MKBHD, you might know him as YouTuber Marques Brownlee, is shutting down his panels wallpaper app on December 31st. He actually acknowledged the app's failed collaborative vision as a part of the reason. He's refunding subscribers, deleting user data, and got to love to see it, open sourcing the code on GitHub. Ah, very nice. What a gift to everyone. Thank you, Marquez.

30:05Speaking of gifts, if you need a physical gift like a mug or a pint glass or a poster or a shirt, maybe you want something holiday themed. We got Len Peralta to draw Techie Santa. Drones are pulling his sleigh. He's got a bag full of laptops and cell phones. You can get that on a greeting card or a hoodie or a bunch of other stuff. Go check it out, folks. We got good stuff in there, including my e-book, Synced, to help you know a little more about tech. It's all at DailyTechNewsShow.com.

30:39We end every episode of DTNS with some shared perspectives. And today, David has a great question. Yeah. He said, with the memory shortage and cost hike, I'm curious how this might affect the system on a chip PC market. Might we see an increase in these purchases for general consumers or OEMs who are feeling the price crunch? Push these models more in the marketing. I'm thinking ARM's system on a chip PCs where RAM is right on the chip. And I doubt those CPU style fabs have any overlap with the general RAM producers, since I imagine they're building the whole thing on die themselves. Of course, that would apply to Mac's M series as well, for that matter.

31:16Even if Apple didn't lock in prices on RAM, I'm not sure how it would affect them anyway. Now, come to think about it, with the M series, I imagine it's not an issue unless they have a PC with a hybrid approach with some RAM on the system on chip and an auxiliary DIMM slot. But I kind of doubt they have anything like that. I have heard of it in the PC ARM space with Snapdragon, but not in the Apple space. Anywho, if you all have any comments, I would love to hear them. Well, I looked into this because I thought, oh, wait, maybe that's true. Maybe those dies are made separately from the RAM procedures and then the RAM factory lines that are rising in price.

31:53And it turns out RAM is not generally on die. It's packaged with the rest of the components to shorten the distance. So when you see it displayed as being part of the M5 chip, it's actually packaged. It's not on die. I see this referred to as system in package or memory on package, sometimes even package on package. But the RAM component in all of them is usually DDR4 or DDR5, including in the M-series chips. Interesting. The more I know, I didn't know that. So just because it's RAM in a system in package does not mean it is immune from the price increase, unfortunately. Well, folks, if you've got a question like that, or you've got a thought, or maybe you've got an insight, if you're like, wait, actually, there is a difference with system and package, send it to us, feedback at dailytechnewsshow.com.

32:46Yeah, and big thanks to David for that email, for contributing to today's show. And thank you for being along for Daily Tech News Show. You can keep us in business and get a 12 % discount for new patrons. You just got to become a patron. Go to patreon.com slash DTNS. We'll see you tomorrow.

33:08the dtns family of podcasts helping each other understand diamond club hopes you have enjoyed this program