AI Made Everything Expensive (ft. GamersNexus)

In this episode, Ed Zitron is joined by Steve Burke of GamersNexus to talk about how demand for AI GPUs helped the RAM cartel make every single consumer electronic more expensive, and how bad it’ll be when things unwind. Save $10 off a year of my premium newsletter: https://edzitronswheresyouredatghostio.outpost.pub/public/promo-subscription/gzqwkv54e1 - I’d be so grateful! GamersNexus on the RAM cartel: https://www.youtube.com/watch?v=jVzeHTlWIDY Support GamersNexus’ AI Surveillance Dystopia Series: https://store.gamersnexus.net/ai-dystopia https://www.youtube.com/gamersnexus https://gamersnexus.net/ https://bsky.app/profile/gamersnexus.bsky.social https://x.com/GamersNexus YOU CAN NOW BUY BETTER OFFLINE MERCH! Go to https://cottonbureau.com/people/better-offline and use code FREE99 for free shipping on orders of $99 or more. --- LINKS: https://www.tinyurl.com/betterofflinelinks Newsletter: https://www.wheresyoured.at/ Reddit: https://www.reddit.com/r/BetterOffline/ Discord: chat.wheresyoured.at https://discord.com/invite/QUUQUP9szv Ed's Socials: https://twitter.com/edzitron https://www.instagram.com/edzitron https://bsky.app/profile/edzitron.com https://www.threads.net/@edzitron Email Me: [email protected] mailto:[email protected] See omnystudio.com/listener https://omnystudio.com/listener for privacy information.

• 0:00Speaker 00

  This is an iHeart Podcast. Guaranteed Human.

• 0:04Speaker 01

  Snoring, gasping for air during sleep, daytime sleepiness? I'm Shaquille O'Neal, and this shouldn't be anybody's experience. Ask your doctor about Zepbound Trisepotide, the first and only FDA-approved prescription medicine for moderate to severe obstructive sleep apnea in adults with obesity.

• 0:20Speaker 02

  Zepbound is a prescription medicine used with a reduced calorie diet and increased physical activity to help adults with moderate to severe obstructive sleep apnea, OSA, and obesity to improve their OSA. Zepbound is approved as a 2.5, 5, 7.5, 10, 12.5, or 15-milligram injection. Zepbound contains trisepotide and should not be used with other trisepotide-containing products or any GLP-1 receptor agonist medicines. It is not known if Zepbound is safe and effective for use in children. Don't share needles or pens, or reuse needles. Don't take if allergic to it, or if you or someone in your family had medullary thyroid cancer, or if you've had multiple endocrine neoplasia syndrome type 2. Tell your doctor if you get a lump or swelling in your neck. Stop Zepbound and call your doctor if you have severe stomach pain or a serious allergic reaction.

• 1:06Speaker 02

  Severe side effects may include inflamed pancreas or gallbladder problems. Tell your doctor if you experience vision changes before scheduled procedures with anesthesia, if you're nursing, pregnant, plant-to-be, or taking birth control pills. Taking Zepbound with a sulfonylurea or insulin may cause low blood sugar. Side effects include nausea, diarrhea, and vomiting, which can cause dehydration and worsen kidney problems.

• 1:27Speaker 01

  Talk to your doctor. Call 1-800-545-5979, or visit zepbound.lily.com.

• 1:38Speaker 03

  Every small business owner has that one moment that could have broken them, but remarkably, it didn't. Hi, I'm Kathleen Griffith, co-host of The Unshakeables, along with Stevie Barron, CEO of Chase for Business. We have more incredible stories of overcoming the impossible still to come in season three. Listen to The Unshakeables wherever you get your podcasts and learn more at chase.com/podcast. JPMorgan Chase Bank, N.A., member FDIC. Copyright 2026 JPMorgan Chase and Company.

• 2:14Speaker 04

  Jardiance has a big story to tell. Discover Jardiance: Empigliflozin 10 or 25-milligram tablets. Visit jardiance.com, call 1-888-968-6648, or talk to your doctor to see if Jardiance is right for you.

• 2:30Speaker 00

  Coca-Cola. For the big, for the small, the short, and the tall. Peacemakers, risk-takers. For the optimists, pessimists. For long-distance love. For introverts and extroverts. The thinkers and the doers. For old friends and new. Coca-Cola. For everyone. Pick up some Coca-Cola at a store near you.

• 3:06Speaker 05

  Hello and welcome to Better Offline. I'm your host, Ed Zitron.

• 3:21Speaker 05

  Subscribe to the newsletter, join the subreddit, buy a t-shirt. All the links are in the notes. And here's a joke to kick us off. What do you call the CEO of a French RAM company? That's right, Emmanuel Macron. And that's what we're here to talk to Steve Burke of Gamers Nexus about today. Steve, welcome back to the octagon.

• 3:37Speaker 06

  Emmanuel Macron.

• 3:39Speaker 05

  Yeah. That one in the chain.

• 3:41Speaker 06

  You didn't tell me that one before we started.

• 3:42Speaker 05

  Yeah, because you would have killed me before I read it on the... But look, we're here to talk about RAM. So we'll start with a simple question. Why is RAM so expensive right now? What is doing this?

• 3:56Speaker 06

  I mean, the reason it's so expensive is not grounded in reality, I think. Yeah, it's the... The surface of it is that all the data center buildout is consuming all the memory, and therefore, you know, the price higher because everyone wants it. But I do think there's a little bit more to it than that.

• 4:15Speaker 05

  Right. So from what... So when you build chips, we're going to really simplify this one for the audience because it's a pain in the ass to explain otherwise. But when you build chips, you have something called a wafer. You have the silicon out, and then you etch it using an lithography machine. But you can make less high... So there's the high bandwidth memory that they use inside GPUs, and then there's the regular DRAM that they use in everything else, right? Is that... It's a simplification, but is that fairly accurate?

• 4:44Speaker 06

  I think like the... The most base level stuff to know is that memory, some form of RAM or memory, is in almost everything. A common misconception, I'm not sure... I think your audience is like fairly technical, but a common layperson misconception is that storage is the same as memory. They'll use it interchangeably. Storage is like your drive that you, you know, you put data on, like photos or whatever. And then memory is a temporary form of storage, volatile storage, that gets wiped when it's not actively being used, when you restart or something like that.

• 5:17Speaker 05

  And when they put stuff in memory, that's where... That's like RAM and such, where they... Like, that's where you put the stuff that you're going to need quickly.

• 5:25Speaker 06

  Right, yeah. So, yeah, memory is much faster than storage, permanent storage. And so it acts as this sort of transactional buffer to help get things to you quickly. So memories in phones, refrigerators these days, cars, there's all kinds of different grades of memory. And it's used in pretty much everything to do some kind of quick... Like in a car, maybe you need to retrieve something related to, like, if there's a built-in map on the screen or something like that. Or in a refrigerator, who the fuck knows, honestly? But I think this... I think the...

• 5:59Speaker 05

  So that you can have a giant horrible screen on your Samsung fridge?

• 6:02Speaker 06

  That displays you ads occasionally.

• 6:04Speaker 05

  Yes, exactly.

• 6:05Speaker 06

  Yeah. And so, yeah, to go to your wafer con... Yeah, there's different types of sort of production for memory. And you'll find DRAM is used in all of it in some capacity, for the most part. So you'll find DRAM dies in HBM, like inside of an HBM package. But data center, high-end GPUs will use high bandwidth memory. And basically, every consumer device uses something called GDDR, which is just a lower bandwidth, kind of more normal solution. That's not as overboard on the performance for a consumer application. But they all kind of come from the same place, which is a wafer. They're all cut from...

• 6:48Speaker 05

  Right. So, right now, they're taking up more... They are using more of the... How much of that capacity are they actually using to build high bandwidth memory? So the stuff that's going in the GPUs?

• 7:01Speaker 06

  It appears to be an awful lot. I mean, so I know that the recent class action lawsuit, or complaint, I should say, that was filed, notes that the profitability... The plaintiff's attorneys claim that the profitability of traditional, like conventional DRAM is greater than the profitability of HBM. That was one of their claims. And yet, they sort of argue that, and it seems like their evidence is good, but they argue that, why would you shift more and more capacity to producing this lower-margin thing, HBM, right? If like the easier thing to make is potentially higher margin?

• 7:45Speaker 06

  And so it does seem like more of it's shifting over there. You know, OpenAI caused that sort of big kerfuffle previously when they claimed 40% of just all DRAM supply globally.

• 7:55Speaker 05

  But that didn't happen, though.

• 7:56Speaker 06

  Yeah, it didn't happen. It was all like, basically...

• 7:58Speaker 05

  One of the most insane... Just an insane story, by the way, just last year. For those of you who don't remember, open... A story came out that OpenAI was buying 35 to 40% of all the world's DRAM wafers. It never happened.

• 8:11Speaker 06

  Yeah.

• 8:12Speaker 05

  But we just talked about it like we did, like it did, but nevertheless...

• 8:16Speaker 06

  It did affect the price.

• 8:18Speaker 05

  It did. But here's the question. So why is everything so expensive? That like, is it... Is it the... Because are they just jacking the price up?

• 8:29Speaker 06

  I think so. I mean, so, yeah, there's a couple of things here. I'll lay out like some of the highlights, and we can... You can figure out how you want to go through them. But one of them is the memory industry has convictions, actually, in its past of price-fixing collusion, and as the DOJ dubbed it, cartel-like behavior. So this goes, the early 2000s, there was a bust. 2016, 2018, there was an attempt at a similar lawsuit. In the early 90s, the memory manufacturers were talking about how they thought their competitors were price-fixing and colluding. And so this is a long and storied history where they... And now there's an active lawsuit, punitive class action for price fixing.

• 9:12Speaker 06

  So there's that antitrust aspect, which that, obviously, you know, prices would be higher if you're just doing illegal things to make the prices higher. There's also the sort of physical manufacturing side, which is that to make memory, it's not quick to get another fab online. And the companies seem to sort of signal to each other through earnings reports whether they're going to expand capacity or maintain capacity of their fabs, which are factories for memory. So there's that aspect, which is if there's a real or not a sudden increase in demand, there's going to be a long latency to react to that with more supply, if they even want to.

• 9:55Speaker 06

  So the current projections are 2028 or 2030 for getting like more fabs online. The other reason it goes up, though, is really simple, which is that in a server that goes in a data center, some of Nvidia's newer solutions will use 1.5 terabytes of system memory. So that's DRAM. They'll use hundreds of...

• 10:15Speaker 05

  Don't they also use the same RAM as we find in smartphones as well?

• 10:21Speaker 06

  Yeah, LPDDR typically, yeah. Yeah, so Grace will use, I think it's 5X, I think is what they're on.

• 10:28Speaker 05

  Yeah, I pulled the numbers, and it's something like 20 terabytes for an NVL72, which is a 72 GPU rack.

• 10:36Speaker 06

  Yeah, so you they do the single blade for the big boy is one and a half terabytes of system memory alone. And you know, for perspective, a consumer desktop computer might have 16 gigabytes with Windows these days, you probably want like 32. And they're they're taking one and a half terabytes for one. You know, server solution, but that's just the system memory. They also have other processors on there like the Blue Point. Or Bluefield, I think it's called DPU, takes 128 gigabytes, if I remember correctly, the GPUs take hundreds of gigabytes to like close to half a terabyte per GPU. Of HBM. And the HBM.

• 11:20Speaker 06

  Has a there's kind of a yield factor at play where. You know, you can, when you're talking about manufacturing, that means you're also talking yields. How many good products do you get for every bad? And HBM is more difficult to make. If you have a failure in manufacturing, then you're also potentially losing more silicon dies that could have gone to more consumer devices.

• 11:44Speaker 05

  And you and HBM is made by stacking the memory, right? So it requires just physically more memory. Forgive me if that's kind of an ape's description.

• 11:53Speaker 06

  No, no, it's fine. Yeah. So HBM, high bandwidth memory is literally just really high bandwidth. It's kind of like if you were to. If you were a city and you wanted to solve a public transit problem and you just built a shitload of, I don't know, let's take it.

• 12:10Speaker 05

  A massive tunnel.

• 12:11Speaker 06

  You you build like a bunch of trains. You build a ton of trains for passengers. But you don't have enough tracks to put those trains on. They can't actually get on the track without blocking each other. You haven't really solved your problem. And so high bandwidth is trying to solve that by saying, okay, we've got all this capacity for passengers, which is the memory itself. Now we need something like a bigger road system to actually feed that capacity through to access it. And so HBM tries to solve that. Especially because you look at GPUs that might have like 12 plus times the amount of capacity that a consumer GPU would have.

• 12:48Speaker 06

  And so you don't want to starve the the GPU core of the data that sits in the memory by not being able to get it fast enough. And so HBM tries to solve that problem.

• 13:01Speaker 06

  So yeah, to go back to your point of stacking, high bandwidth memory is physically vertically stacked. As in they take silicon dies for memory and they stack them atop each other and then they connect them all together. As opposed to consumer memory where it's it's closer to, there's still layers, but it's closer to a flat plane.

• 13:23Speaker 05

  Right. But it just means that by virtue of that, it takes up more space in the place where they built the chips.

• 13:29Speaker 06

  Yeah, yeah. If you're on a consumer device, then the consumer memory, you'll need more physical area to put the same amount of capacity, whereas with the high end server stuff, you just stack that capacity vertically. And then it's also HBM's really significant advantage is that it's very low power comparatively.

• 13:52Speaker 05

  Right.

• 13:53Speaker 06

  Yeah. And so.

• 13:54Speaker 05

  But just to be clear, it takes up more space on the fabricator itself.

• 13:58Speaker 06

  Oh, yeah. Yeah. If you're talking about the fab space, then yes, HBM requires a lot more space for a few reasons. In terms of manufacturing area, in terms of like footprint on a product, it's less, but in a fab, it's more. And

• 14:13Speaker 05

  Why is that?

• 14:14Speaker 06

  In a in the fab, one of the reasons, so there's a few, but one is that HBM requires additional silicon pieces to be able to function. So with GDDR consumer GPU memory, you pretty much you have the memory module, it looks like a black rectangle. And then if you were to use like acid or something to etch away the black plastic part, you'd be left with pretty much a singular silicon die underneath. That's all it is for consumer. For HBM, you also need a logic die. So there's another piece of silicon that interfaces with the memory and the stack of memory. Because you need something to kind of

• 14:57Speaker 06

  to command and control and understand. You know, the vertical tower you've got of memory. And then you also need an interposer die, which sits underneath everything. So this is a solution that will sit under your GPU core, which is a very large piece of silicon and is shared with the HBM stacks. So all of your your core and all your HBM is stacked on top of the same. It's called an interposer. It looks like a PCB. And they all communicate through that. So that's another piece of silicon. And so for this, these reasons alone, there's also others, but you know, these alone, now you're involving more fabs, there's more logistics, shipping things around. And there's more fab or fabrication plant

• 15:43Speaker 06

  area allocated, even if you're just talking about making, someone's got to make the logic dies, TSMC makes the interposers.

• 15:49Speaker 05

  Right. Yeah. There's just more moving parts and it kind of takes up more space.

• 15:53Speaker 06

  Exactly. Yep.

• 15:55Speaker 05

  So why? So to to my earlier question, so this stuff is obviously lower margin, but they seem to be able to sell an absolute shit ton of it. So is that potentially why they're doing this? That just they know there is insane demand on a level that for a high, well, I guess they can charge what they want for it as well. I'm just having trouble trying to work out what it is they're doing other than just filling their boots.

• 16:20Speaker 06

  Have you tried asking the chat GPT?

• 16:23Speaker 05

  I know. I could ask my cat and get an answer. I'd respect more, but it's genuine, but it's genuinely because it seems that unless I'm misunderstanding here that these micro and SK Hynix and Samsung, they're just kind of cranking up the price of regular RAM.

• 16:40Speaker 06

  Yeah.

• 16:40Speaker 05

  Because they can?

• 16:43Speaker 06

  I mean, to some extent, yeah. So I think this is where the price fixing allegations come into.

• 16:49Speaker 06

  I don't know if you try, if you try to take their side, which I really don't want to.

• 16:53Speaker 05

  I'm trying. I'm trying.

• 16:54Speaker 06

  Yeah, if you try to, I think their argument, whether it's for truthful or not, but would be we've got this crazy high demand for HBM. These are guaranteed customers. I can sell terabytes of it to Jensen. And it's it's one point of contact. It all moves quickly and easily. And this, there's this data center demand right now. And we don't know if it's going to be here forever, but consumers, you know, they're they're always going to have to buy electronics or whatever. So I'm not exactly sure if that's, to me, it seems like maybe you make some sort of you you assign some value to the B to B relationship over B to C or something.

• 17:37Speaker 06

  I maybe, but also I guess if they.

• 17:43Speaker 06

  Yeah, I'm not sure if if you make.

• 17:44Speaker 05

  But why would that mean that they charge more for regular RAM?

• 17:48Speaker 06

  Oh, as far as why more for regular, that part's easy. That's just because the HBM itself takes the same, a lot of the same machines as consumer memory. And so if you're making more HBM, you're naturally making less consumer, which I think is where you start seeing the the supply squeeze and the price go up. Likewise, a single piece of HBM can use up to something. I think it's like, it might be like 32 dies max or something of DRAM. You'll commonly see like four referenced, but single piece of HBM will take more DRAM dies to build. And so one of the problems is

• 18:29Speaker 06

  if, let's just say four, if you have a piece of HBM that's got four DRAM dies in it and you have one failure somewhere in that stack or in packaging and assembly. Now you've destroyed. You know, four individual pieces rather than if you have a yield failure with a consumer device, you just try one.

• 18:50Speaker 05

  Sorry to ask a stupid question. What is a die exactly? Is that just the what we would think of as a stick of RAM?

• 18:57Speaker 06

  Uh, no, a die is so so a wafer. We'll start at the wafer level. A wafer is just a disk, right? It looks like a CD that typically eight or 12 inches.

• 19:11Speaker 06

  And they they process it and they pattern it where they literally project like a pattern or they lithographically etch a pattern onto the surface of this disk. Once that's done, you know, these these patterns, they're called patterns because it's a repeating sort of set of shapes that ultimately form the logic that makes the silicon. And the the word die comes from the process dicing. So with a wafer, when they're done producing it and patterning it, and it's basically ready to become a product, they need to cut that disk into a bunch of small squares.

• 19:53Speaker 05

  Right. So it's like a big slab that they cut up.

• 19:56Speaker 06

  Yeah, it's I mean, it's like making a pizza with a pizza cutter, you know, except they're they're doing a pattern that only psychopaths would do on a pizza, which is making them squares. And and so they they cut it into the squares. That's called dicing. And then each of those pieces, the squares that yield from that, those are the dies. And so if you were to pull up an image of any GPU, typically you'll see one large, it's called a monolithic die, which means one singular sort of piece of shiny reflective silicon mirror like finish. That's your die. And then for memory, same exact concept, just a lot smaller.

• 20:35Speaker 05

  Right. So, so right, okay, so there's less space to make regular RAMs, so they just crank up the price. That's really, that really is it. That's their entire thing. Right, but I saw Micron's earnings and they had eighty four point nine percent margin. So it just kind of feels like they're setting the price to whatever they want it to be.

• 20:55Speaker 06

  You don't think they're, you don't think that is, is that not enough? I think they need more, don't you?

• 21:01Speaker 05

  Maybe they should get ninety five percent. Well, it's just kind of.

• 21:04Speaker 06

  Ninety five, why are you setting the, setting it so low?

• 21:06Speaker 05

  They should get a hundred and ten percent margins. No, it's just, it's very frustrating as well because I've read now arguments in favor and against Micron where it's like, oh, Micron got screwed by Apple and consumer electronics companies during the last, during the last supply chain crisis because they, they kind of made a bunch more RAM than they needed to, from what I understand.

• 21:32Speaker 06

  Prime River, you know, like.

• 21:33Speaker 05

  No, it really, it really is. But what, and it seems to be there's this boom and bust cycle in memory where they get a bunch of demand. And then what? So do they build more capacity? Is it that they just ordered too much RAM and then people cancel their orders? Like what hap, what happened last time?

• 21:52Speaker 06

  It is cyclical. Um, yeah, so it's cyclical. I will say right now is nothing like I've ever seen. And we spoke to a bunch of companies purchasers of memory like G-Skill and Corsair, who sell the sticks of RAM that you put in a computer. And those guys were telling us on record that likewise, they've never, this is not like part of the cycle. This is special. And as for.

• 22:17Speaker 05

  How is it different?

• 22:19Speaker 06

  How, the difference is how much? The difference is the percentage increase it is this time. So, you know, you're looking at like seven hundred percent increase in price in some situations in a very short period of time. Normally cyclically, you'll see some ebbs and flows, but it might be like back in the day, a fifty dollar kit of DDR four became eighty five or ninety dollars, maybe even a hundred and twenty if it's a really bad year, you know, but you're still at like a little worse than double. Whereas now a two hundred dollar kit of memory is suddenly a thousand to twelve hundred dollars. And so it's just the, the amount of the price increase is what's different. And then also,

• 23:04Speaker 06

  this time, the sort of contest for the supply is, it's coming from within the own industry where the very companies that would be the cause for an increase in sales of consumer supply are now the cause of these like insane unprecedented volume of sales for AI components.

• 23:33Speaker 05

  All right, everyone, I'm back talking about Quince, one of my favorite clothing brands. And as we speak, I'm currently wearing one of their excellent cotton knit overshirts that wear like a jacket, nice pre fall piece to wear as things cool off a bit. I've also been eyeballing one of their heritage water resistant Harrington jackets too for later in the year to pair with their Mongolian cashmere sweaters I bought. And I have several of them and I really can't recommend Quince's stuff enough. I love Quince because they focus on high quality wardrobe staples made with premium materials, jackets, pants, fitted tees and polos, activewear and even travel gear. Everything at Quince is priced fifty to eighty percent less than similar brands. And because they work directly with ethical factories and cut out the middlemen, you're getting more for less. Find the four pieces

• 24:14Speaker 05

  you'll reach for most at Quince. Download the Quince app for app exclusive offers or go to quince.com. Get free shipping on your order and three hundred and sixty five day returns. Now available in Canada and the UK too. And when Quince asks where you heard about them, let them know it was on better offline. It's the best way to support the show.

• 24:33Speaker 07

  Snoring, gasping during sleep, feeling fatigued? Ask your doctor about Zepbound, terzepatide, the first and only FDA approved prescription medicine for moderate to severe obstructive sleep apnea, OSA, and adults with obesity. Zepbound is a prescription medicine used with a reduced calorie diet and increased physical activity to help adults with moderate to severe obstructive sleep apnea, OSA, and obesity to improve their OSA. Zepbound is approved as a two point five, five, seven point five, ten, twelve point five, or fifteen milligram injection. Zepbound contains terzepatide and should not be used with other terzepatide containing products or any GLP one receptor agonist medicines. It is not known if Zepbound

• 25:15Speaker 07

  is safe and effective for use in children. Don't share needles or pins or reuse needles. Don't take if allergic to it, or if you or someone in your family had medullary thyroid cancer, or if you've had multiple endocrine neoplasia syndrome type two. Tell your doctor if you get a lump or swelling in your neck. Stop Zepbound and call your doctor if you have severe stomach pain or a serious allergic reaction. Severe side effects may include inflamed pancreas or gallbladder problems. Tell your doctor if you experienced vision changes before scheduled procedures with anesthesia. If you're nursing, pregnant, gland A B, or taking birth control pills. Taking Zepbound with a sulphonal urea or insulin may cause low blood sugar. Side effects include nausea, diarrhea, and vomiting, which can cause dehydration and worsen kidney problems. Talk to your doctor. Call one eight hundred

• 25:58Speaker 07

  five four five five nine seven nine, or visit zepbound.lily.com.

• 26:04Speaker 08

  Financial growth begins long before the first investment. It comes from understanding what you're building toward, what's at stake, and what success looks like for you. At Oppenheimer, we bring bold thinking guided by the full strength of our expertise to put capital to work building and protecting wealth that lasts generations. Put the power of Oppenheimer thinking to work for you. Wealth management, capital markets, investment banking.

• 26:33Speaker 09

  I love football season, the games, the rivalries, the food, getting everyone together. No game day feels complete without a trip to Total Wine and More. I can find everything from cold beer and canned cocktails to wines and spirits for the whole crew. And if I want to bring something different to the watch party, the Total Wine team always has ideas. Plus, it's nice knowing I'm getting the lowest prices. Get your game day lineup ready at Total Wine and More. Spirits are not sold in Virginia and North Carolina. Drink responsibly. B21.

• 27:08Speaker 05

  But it isn't based like, so the twenty twenty one supply chain crisis, chip crisis, was because there was a flood of consumer spending after COVID.

• 27:15Speaker 06

  Yeah.

• 27:16Speaker 05

  And that led to a bunch of consumer demand for products and overall spending across the board, but like consumer demand drove it. This one seems different because it's not demand outside of the, outside of whatever like psychosis, the various hyperscans has. Like they're not spending a trillion dollars in CapEx really for like, it's not like consumers can't access whatever Microsoft.

• 27:42Speaker 06

  You know, what's interesting is the big three memory makers decided not, I think it was twenty twenty two or three, to bring on more capacity in some of these situations because they were talking about how they, they got burned by oversupply and sort of following COVID, right? But what's interesting is if you go and look at some of the small suppliers, not many of them, but there's a couple of small memory makers that make up like less than five percent of the market. Winbond, Nanya are two of them. Nanya is headquartered at least in Taiwan. I don't know where their fab is, but they, like these two companies, as far as I'm aware, I was just looking at this this morning, they increased

• 28:25Speaker 06

  their, their production capabilities at the time when the big three were saying that they weren't going to do it because, you know, they got burned.

• 28:34Speaker 05

  Right.

• 28:35Speaker 06

  Is it interesting to me? Because if you're, this is the argument the lawsuit makes too. If you're a small company, I would think you're at the biggest disadvantage, the biggest risk to increase your production capabilities with this massive amount of spend. If your biggest competitors are saying like this, it looks like bad times ahead, you know, we don't want to spend into this. But if you're a small guy, like you're at the greatest risk of going under if you fuck up your, your CapEx, right?

• 29:03Speaker 05

  Why didn't that, this, maybe this is a dumb question. Why didn't them increasing supply allow them to kind of undercut the Microns of the world?

• 29:11Speaker 06

  Yeah, that, that one is just because the, there's a, Micron does have a technology advantage, so they can make memory that these other guys can't. HBM is one of them. Maybe this is, I didn't think about this till now, but maybe there is a very, a small part, I think, but possibly part of an answer to your question of why HBM. That is something that's, that other companies are struggling to make successfully in current generations. So it's.

• 29:41Speaker 05

  We, I know only like SK Hynix, Samsung and Micron are the ones who can make HBM.

• 29:46Speaker 06

  Yeah, CXMT has also struggled with HBM and they're a new Chinese memory maker that's actually done really well with DDR five. And so there may be a little bit of a pulling the ladder up, you know, after them factor.

• 30:00Speaker 05

  Well, here's a question as well, when it comes to high bandwidth memory. Surely this is going to lead to an oversupply situation regardless, because HBM is used in what? Video cards? God, that's such an anachronistic term in the AI bubble. But like it's used in video cards and AI GPUs basically, right?

• 30:18Speaker 06

  Yeah, it's very limited. There's almost, there's effectively zero consumer use case.

• 30:24Speaker 05

  So when the AI bubble bursts, where does the high bandwidth memory go?

• 30:31Speaker 06

  Yeah.

• 30:34Speaker 05

  Oh, okay.

• 30:35Speaker 06

  Yeah.

• 30:36Speaker 05

  Because I saw.

• 30:36Speaker 06

  We don't talk about the bubble bursting, Ed.

• 30:39Speaker 05

  Well, that's the thing. I saw Samsung is talking about increasing their high bandwidth. They're spending like fifty something billion, five hundred billion dollars across an indeterminate period, but like also building an entire new high bandwidth memory, like factory.

• 30:55Speaker 06

  Yeah.

• 30:56Speaker 05

  And that, but I was looking, I think I seen you the graph for coming on. There was, I saw this chart where it's like, they were making what, a couple hundred million gigabytes of HBM? This in twenty twenty two, and they're now selling like seven or eight trillion a year?

• 31:12Speaker 06

  Yeah. So what's crazy is, um, the HBM bit demand, I think I, I think I first saw this on semi analysis, if I remember correctly, when I was at the sort of site, but the HBM bit demand from Nvidia was, uh, recently projected to exceed all HBM bit demand of all other companies combined from twenty twenty five.

• 31:38Speaker 05

  Good.

• 31:39Speaker 06

  And that's just Nvidia. And, uh, and again, you have more wafer area per bit than with DRAM consumed, that counts yield losses, right? And so like increasingly as they make more and more of this, uh, presumably for Nvidia's devices, you will continue to see consumers really get shafted if there's not new supply online soon. And likewise, if they're focusing all their efforts and resources on HBM, for whatever reason, uh, who's to say that the new fabs will even make consumer memory anyway, right? Like there's, they might just be more HBM.

• 32:19Speaker 05

  But the thing is, if they're all built to make, maybe I'm missing something. They're building this fact, these, these fabs specifically for high bandwidth memory. Behind bandwidth memory is only used really for the thing in the AI bubble.

• 32:32Speaker 06

  Right.

• 32:34Speaker 05

  What are they gonna do with all this capacity when it's gone?

• 32:37Speaker 06

  They would switch it over. They'd switch the line.

• 32:39Speaker 05

  I mean, they would, but it's like,

• 32:42Speaker 05

  I guess they'll just have, aren't they just creating a future nightmare? Uh, I mean, it's always good to have more capacity for them, I guess.

• 32:50Speaker 06

  I, yeah, I mean, I guess it's good to have more capacity. There's a lot of government money floating around too, right? Like Micron got a five billion dollar tax break in New York. Um, that as part of the, uh, I guess big beautiful bill.

• 33:04Speaker 05

  But five billion, like Micron said they're gonna spend two hundred and fifty billion.

• 33:09Speaker 06

  Right.

• 33:10Speaker 05

  Samsung and SK Hynix have made two point one trillion dollars worth of commitments. This just feels like it's, it's almost like they saw what happened in twenty twenty one and twenty twenty three, which ended with all of them doing at least a billion dollars of write offs. And when they were like, fuck that, we can go bigger.

• 33:28Speaker 06

  Yeah, well, I mean, I, I do think personally, I, I think that they are colluding again. I, I think they are price fixing. Um, it might be through like a, a parallelism rather than a literal behind closed doors, meaning like we're just gonna announce our changes in earnings reports and then our friendly competitors will kind of mirror us. And so.

• 33:56Speaker 05

  Yeah.

• 33:57Speaker 06

  Yeah.

• 33:57Speaker 05

  Like it's kind of a prisoner's dilemma. Like if they don't raise prices, the other guy will if they, and if they don't build capacity, the other guy will, they won't get the long-term deals and all that.

• 34:07Speaker 06

  Yeah. And I also, you know, the the wafers themselves, I have some interesting numbers I wrote down, but, um, I was looking up, when you were asking about HBM, I went and pulled up some of the original coverage I did of HBM, uh, which started to come online, or at least was being actively developed somewhere around, I don't know, twenty thirteen plus or minus a bit for AMD GPUs. So yeah, when AMD was working on Fiji and its Fury GPUs, they, they debuted HBM actually for consumer devices. They had HBM on Vega and on a couple of their architectures

• 34:50Speaker 06

  in like the twenty fifteen, maybe twenty eighteen area, plus or minus a year or two. And, uh, massively expensive, just like it is today. The power efficiency was great, which is really why they used it, although they were bandwidth starved too. But I, I pulled up, I had done some reporting at the time where I spoke to a bunch of the people in the know in the industry. And at the time, twenty seventeen or so, twenty sixteen area, the HBM two memory for one of the devices, um, should have cost around a hundred and fifty dollars cost for them to, to buy. And the interposer and the packaging should have been another twenty five dollars, one seventy five.

• 35:33Speaker 06

  Uh, and this is, these are, you know, this is old technology now. It's, I'm sure a lot different. But the point of bringing it up is that if you compare sort of time for time and at the same capacity, at the time it would cost AMD a hundred and seventy five bucks or so to put HBM on its device, whereas with GDDR memory, the non-HBM stuff, they would have been at somewhere around like sixty eight dollars.

• 36:01Speaker 05

  Uh, so they would be just, so I guess the thing is, they absolutely would make more money selling DRAM, or sorry, GDDR, like regular people RAM. It's just that right now they have the business to kind of, it's not just shitty, it's not these bullshit video cards for loser gamers, it's selling, um, AI GPUs to big, strong boys.

• 36:22Speaker 06

  Yeah, right. And, and yes, like Sam Altman and.

• 36:25Speaker 05

  Yes, the big, strong, hunky men like Dario Amade.

• 36:30Speaker 06

  Right, right. Yes. Yeah. He, I've never, by the way, I don't think I've ever seen someone spread so much fear so anxiously as him.

• 36:42Speaker 05

  It's, it's, yeah.

• 36:44Speaker 06

  Yeah.

• 36:44Speaker 05

  Pathetic. Well, yeah, you also, you have, here's a random thing as well. For this, a question to the listeners and to you as well, Steve, you ever see anyone on, you seen this weird habit of Silicon Valley people of using AI now to add muscles to guys?

• 37:00Speaker 06

  Yeah, I have seen.

• 37:01Speaker 05

  Seen this at all? What's that about? One of the weirdest things I've seen in the tech industry's history, and there's a lot of weird shit.

• 37:07Speaker 06

  I don't know if it's, well, they're, they're not used to working for intellectual property at this point. They just harvest it. So why work for the muscles either?

• 37:14Speaker 05

  Well, it's just, it's really strange. Like I, like, I think the, the TBPM, the techno, the blowjob brothers network, like they, um, they, um, they do it with like Dylan Patel, who's just like a regular sized guy. And it's just, it's, sorry, I've just realized that's weird, but then I saw someone do it for Dario Amade the other day. And this is like, your, your body's your body. I don't judge people for, for like how they look, but at the same time, Dario Amade is not like a like fifth course of steroids guy. Like it's just the weird, it's the weirdest thing I've ever seen.

• 37:47Speaker 06

  Well, I, I saw, did you see the story just recently of the, uh, copper thieves stealing shit from data centers too?

• 37:56Speaker 05

  No.

• 37:56Speaker 06

  That, that was.

• 37:58Speaker 05

  Average better offline listener there. Yep.

• 38:02Speaker 06

  Yeah, they probably wearing one of your T-shirts.

• 38:06Speaker 05

  Just, oh God, no, I disavow them.

• 38:09Speaker 06

  Yeah, the, uh, I saw there was a title of, um, of thieves stealing copper from data centers. And I was just like, oh, okay. I mean, it's really, I guess there's no honor among thieves, you know, it's just thieves stealing from thieves at this.

• 38:23Speaker 05

  It's also inevitable at this point because you make these giant things full of earth metals and then you put them in neighborhoods, usually with like, like disadvantaged farm neighborhoods or like, like distressed towns in Texas. You're like, oh, I wonder what will happen with the other business in the area.

• 38:41Speaker 06

  Yeah, I wonder what'll happen when we screw over all these people and.

• 38:45Speaker 05

  And like mock them via the media constantly. Anyway, back to, back to RAM. So is this something that's going to start affecting like storage? So hard drives and such?

• 38:58Speaker 06

  Yeah.

• 39:00Speaker 06

  Yeah. So you mean like actual, uh.

• 39:02Speaker 05

  Yeah, like nan, like, like solid state drives and the like.

• 39:06Speaker 06

  Yes. So it's already affecting it. Um. Storage is made at the same fabs for the most part as memory with basically the same tools and mostly by the same companies, although there's, there's like one or two differences. But for the most part, it's the same people and tools. And so definitely one affects the other anyway. But storage in general has, it hasn't gone up as violently as memory, but it has gone up. Some of the SSDs we had bought previously, I was just checking this actually for like 90 bucks, maybe about a year ago, are now closer to like 180, 190.

• 39:49Speaker 05

  Jesus Christ.

• 39:51Speaker 06

  Yeah. Let me pull up a chart here.

• 39:53Speaker 05

  That's, this is genuinely like, I think, I don't know if I'd say I do anything that's activism, but like, I think people just need to know this is AI's fault.

• 40:04Speaker 06

  Yes.

• 40:04Speaker 05

  And it, and the worst part is it's not even AI's fault because they need these data centers because they're taking forever to build. It's like you can access Claude and ChatGPT. You don't, there's no one, no one's like unable to log on because there's not enough data centers. Nothing's being stopped. They're just like, no, we need to do, we need to buy, buy the chips, need to build the data centers for some reason.

• 40:25Speaker 06

  Yeah, it's, I mean, yeah, there's not like a queue. I mean, I guess there, maybe they make the argument of like, we're trying to make it better and the models will be more complicated and that required.

• 40:35Speaker 05

  When's that gonna happen?

• 40:36Speaker 06

  But yeah, the, uh, storage though. So looking at an average price chart for four terabytes of solid state drive storage on VME on PC part picker, which averages it. They have the pricing at around, it looks like maybe about $270 in January last year. And now it's somewhere around, somewhere around $800, $900.

• 41:06Speaker 05

  And that's for what kind of storage?

• 41:07Speaker 06

  That is for four terabytes. Uh, so relatively large capacity, which is going to be affected the most.

• 41:14Speaker 05

  Jesus Christ.

• 41:15Speaker 06

  Yeah. And this is where, this is interesting too, because one of the ways you'll see consumers getting shafted, that's not as obvious, is some of the modern devices that should be your next sort of, you know, improvement upgrade on past generations, they're actually coming down in capacity. So I haven't looked at the exact device names for a while. We ran a report a while ago talking about some of the phones and laptops that have shifted from, say a prior model of 12 or 16 gigabytes down towards eight for memory. There's decreasing the storage capacity as well. Yeah.

• 41:52Speaker 05

  That's fucking, so everyone just suffers here. And it's pretty, like there's Microns, SK Hynix, Samsung, they could just, they could have not done this, just to be clear. Just so they could have sat with like 70% margins.

• 42:06Speaker 06

  They're already colluding anyway, in my opinion. So yeah, they could, they could collude in a different way.

• 42:12Speaker 05

  It's just, it's so fucking straight. Well, I mean, it's actually, no, that's not, that's not fair. This isn't strange at all. This is exactly, like, this is exactly what companies like this do.

• 42:23Speaker 06

  Yeah.

• 42:23Speaker 05

  Like this, it's this kind of vulgar, kind of chancerism. And their argument, from what I understand is, well, memory is cyclical, so we need to fill our boots when we can.

• 42:34Speaker 06

  And if you're calling something vulgar, then it's pretty vulgar.

• 42:38Speaker 05

  Yes. Just to be clear, I'm as the vulgarest man in tech, and that is a word now. But the thing is though, just to be clear, that's their argument though. The, because things are good now, they can charge what they want. So they can fill their coffers for the future.

• 42:52Speaker 06

  Yeah, I think so. And I mean, you see it affecting, I think that's, that's probably their argument. I think on the consumer side too, even a company like Lenovo, which is not a manufacturer of memory, but they buy it and they put it in stuff. Uh, I'm not sure if there, there may be, this I don't know for fact, but there may be a benefit for some of these companies as well, for a few reasons. One is, uh, there's a lot fewer, it kind of reminds me a little bit of COVID when all the stores that were 24/7, all the, the grocery stores and whatever, suddenly stopped being 24/7. And at least where I

• 43:32Speaker 06

  am, not one of them has gone back. And I think that the memory capacity kind of equivalent here is this offers a little bit of a reset. Uh, and so with.

• 43:43Speaker 05

  Lower that, lower those expectations.

• 43:45Speaker 06

  Exactly. Yeah. And now your next upgrade is more exciting. And what I noticed with Lenovo when I was buying a laptop was, uh, they had quietly snapped out of existence, their 64 gigabyte model memory, system memory of the device I was looking at, and they only had 32. And I think.

• 44:04Speaker 05

  Is the memory hard soldered onto it as well?

• 44:07Speaker 06

  It is. Yeah, you can.

• 44:08Speaker 05

  Oh, so you just can't upgrade? You're just stuck? That's insane.

• 44:11Speaker 06

  Yeah. And so I think what happens there for them, you know, in their favor, I might argue, well, they can't, they're having trouble getting supply, just like everyone else. That's probably true. Uh, but against them, if they've got 64 gigabytes of memory, because that's just one part of the, the total bomb for a laptop they sell, they might rather sell two people, two laptops at 32 gigabytes, right? Where they can get the markup on the whole thing.

• 44:39Speaker 05

  Right.

• 44:40Speaker 06

  Rather than one guy, a 64 gigabyte laptop where they're only marking it up, whatever the, hopefully, whatever the difference in the memory is. Um, so I think you see some of that here too, where there's.

• 44:49Speaker 05

  Well, they're just gonna use it as a chance to reset what people expect and just, also prices are probably never coming down, right?

• 44:56Speaker 06

  Lenovo itself, funny enough, uh, that I was using them in this example, just like two weeks ago, um, at ISC said that prices will probably, the word they used was never come back down to what they were, uh, last year. And, uh, so yeah, I think.

• 45:14Speaker 05

  Truly evil era. Just like a, like a disgusting era full of horrible perverts. Cause it's just, it's so frustrating as well because America has price fixing laws.

• 45:27Speaker 06

  Yeah.

• 45:27Speaker 05

  Like they, like we have price gouging laws, like we have means of doing this. We just don't do it because obviously of capitalism.

• 45:34Speaker 06

  And they've been enacted, you know, like the DOJ itself busted these same companies in the early 2000s.

• 45:41Speaker 05

  Yeah, that was before, that's before we had social media for psychopaths to talk about stocks all day. Now we need to get, now we need to, we need just borderline corruption or real corruption in everything. That's the only way to get things done these days.

• 45:54Speaker 06

  And I think also there's probably incentive to, you know, if, if you're, no administration of any party wants the economy to fail on their watch. And I think if you're worried about a bubble pop, and especially if maybe spooking investors with DOJ investigation or something might, you know, trigger part of that. Uh, it just seems like you wouldn't be incentivized to not probe too much.

• 46:32Speaker 07

  Snoring, gasping during sleep, feeling fatigue? Ask your doctor about Zepbound terzepatide, the first and only FDA approved prescription medicine for moderate to severe obstructive sleep apnea, OSA, in adults with obesity. Zepbound is a prescription medicine used with a reduced calorie diet and increased physical activity to help adults with moderate to severe obstructive sleep apnea, OSA, and obesity to improve their OSA. Zepbound is approved as a 2.5, 5, 7.5, 10, 12.5, or 15 milligram injection. Zepbound contains terzepatide and should not be used with other terzepatide containing products or any GLP-1 receptor agonist medicines. It is not known if Zepbound is safe and effective for use in children.

• 47:16Speaker 07

  Don't share needles or pins or reuse needles. Don't take if allergic to it, or if you or someone in your family had medullary thyroid cancer, or if you've had multiple endocrine neoplasia syndrome type two. Tell your doctor if you get a lump or swelling in your neck. Stop Zepbound and call your doctor if you have severe stomach pain or a serious allergic reaction. Severe side effects may include inflamed pancreas or gallbladder problems. Tell your doctor if you experienced vision changes before scheduled procedures with anesthesia. If you're nursing, pregnant, glandebi, or taking birth control pills. Taking Zepbound with a sulphonal urea or insulin may cause low blood sugar. Side effects include nausea, diarrhea, and vomiting, which can cause dehydration and worsen kidney problems. Talk to your doctor. Call 1-800-545-5979 or visit zepbound.lily.com.

• 48:03Speaker 10

  At Oppenheimer, we're proven because we're grounded in discipline. For 145 years, we've been building and protecting wealth through every market cycle with precision, clarity, and the courage to think boldly beyond the moment. This is what market-tested legacy looks like for this generation and the next. Put the power of Oppenheimer thinking to work for you. Wealth Management, Capital Markets, Investment Banking.

• 48:32Speaker 11

  Football season is back. Stop by Total Wine & More before kickoff for cold beer, craft bourbon, and ready to drink cocktails. Get the lowest prices and get your game day lineup ready at Total Wine & More. Spirits are not sold in Virginia and North Carolina. Drink responsibly. Be 21. Right now, you can save 60% on SimpliSafe home security. Traditional security systems only alert you after a break in has already happened, and that's too late. SimpliSafe is changing that.

• 49:01Speaker 12

  Stop. This is SimpliSafe. You're being monitored and recorded. I'm requesting police dispatch.

• 49:06Speaker 11

  Using the Outdoor Camera Series 2 and Advanced AI Alerts, our live US-based agents identify potential threats on your property and help deter them. So the intruder never gets into your house. The break in never happens. Because we don't just alert, we stop crime before it starts. SimpliSafe monitoring and deterrence plans start around a dollar a day with no long-term contracts. Order today to get 60% off a new system with professional monitoring. Save now at simplisafe.com/iheartradio. SimpliSafe, ahead of crime in real time. Outdoor deterrence plans start at $49.99 a month. Visit simplisafe.com/licenses for alarm license information.

• 49:51Speaker 05

  But isn't this going to cause just, or already is causing hyperinflation within all electronics-related products? Because it seems that these companies are fully dedicated to the graveyard smash of making everything more expensive forever. And at some point, that will start crushing all consumer electronics.

• 50:08Speaker 06

  Yeah, I think it will, you know, for some things that can migrate more to software as a service and hardware as a service. I think there's an end game there for some of these companies where.

• 50:20Speaker 05

  Tell me more.

• 50:22Speaker 06

  Think if it's, okay, we crushed demand for the gaming GPUs market. Because filthy, poor, casual gamers can't afford our big, strong GPUs that we need for data centers. You know, I think if you crush that market, there is potentially an out in the form of cloud gaming. Or maybe a better example, we don't have latency as an issue would just be if you, as a company, if you're worried about consumers getting systems that are powerful in their hands and doing, using LLMs at home or whatever.

• 51:00Speaker 06

  If that is suddenly out of reach because the memory is too expensive and the GPUs are too expensive. All right, I guess they come crawling back to Claude or ChatGPT. So I do think not everyone's going to be able to grab that parachute, but I think that moving services to cloud-oriented services that you pay for every month

• 51:23Speaker 06

  is a way that they can capture the people, which is going to be most of them at some point, who can't afford to actually own the full device to do stuff.

• 51:32Speaker 05

  Cloud gaming sucks.

• 51:34Speaker 06

  Cloud gaming sucks, and I, but I also, I don't think that's the only place where, you know, I think one of the, the real, if anyone here is genuine about LLMs and AI being a future, I would think at least companies like OpenAI would be worried about that about LLMs getting too good if they're locally hosted for like, let's just say the vast majority of basic tasks someone might want.

• 52:03Speaker 06

  And by, by all the prices of hardware remaining elevated, it does kind of keep that at bay a little longer.

• 52:10Speaker 05

  I just, I, because I understand that, and you've said this before, the idea that they want to push people towards not owning anything. You'll own nothing and you'll love it, I think is the phrase. And, but at the same time, do you think that that's actually their plan? Or do you think that my, my version of this is much simpler, which is, do you think they have one? Or is this just incredible short-termist thinking?

• 52:34Speaker 06

  Yeah, I don't think it's the plan. I think it's like the fallback for the few.

• 52:38Speaker 05

  Google Stadia, we all know, was a huge success.

• 52:44Speaker 06

  I contributed to the failure of that one. Yeah, that one.

• 52:49Speaker 05

  There are 10 people who must be so angry at you, man.

• 52:52Speaker 06

  I'm sure.

• 52:53Speaker 05

  There are dozens of us.

• 52:55Speaker 06

  GeForce now, I had numbers at one point, I don't have them currently, but had a lot of, actually, I think we talked about it on your show. Cause I pulled up their website, however many millions of users and both of us were kind of surprised at it.

• 53:11Speaker 06

  But yeah, I think, you know, I think gaming is sort of the easy one to point to. I think for a lot of casual users, they might be okay with cloud-connected gaming. But I do think there's more services than just gaming, you know, that could be sucked into the cloud, even just stuff like QuickBooks has been the one I've been ranting about recently.

• 53:33Speaker 05

  I'm surprised that that isn't already sassed.

• 53:36Speaker 06

  Well, basically is because they, the desktop version, it used to be a buy it once for $300, then it went to 700. And then they went to $1,500 annually for the desktop version. And you have to log in when you launch it, just like Adobe software.

• 53:54Speaker 05

  Oh, good.

• 53:54Speaker 06

  And now they're killing desktop entirely and you have to go to cloud.

• 54:00Speaker 06

  Which I actually, I just abandoned it and went to a free tool instead, but like the point being, yeah, I think a lot of these, you know, these services they do just increasingly go cloud-connected.

• 54:12Speaker 05

  Well, the problem is, is cloud-connected services still require RAM. I mean, just anyone who's ever used an AI tool that's still in your browser, well, I mean, Chrome itself is a hog. As we all know, and so it's just kind of like, it feels that there's just going to be a point that this escalates to the point that it kills large parts of the consumer electronics industry. I mean, Sony's delaying the next PlayStation, Microsoft is increasing the cost of the next Xbox while firing all the people who know how it works. Just an evil company. I mean, it's.

• 54:45Speaker 06

  And the steam machine is a great example too.

• 54:48Speaker 05

  I love that. I just got one. I fucking love it.

• 54:50Speaker 06

  Oh, did you really? Wow.

• 54:51Speaker 05

  I really love it. It's just what I, but the thing is, it's also insanely expensive.

• 54:55Speaker 06

  Yeah. It's unbelievable. Yeah, it's, um, cause the steam machine, we had talked to them and the quote they gave us was amazing when, I'll paraphrase it to the best of my memory, but they were talking about how, I asked them,

• 55:14Speaker 06

  how did it work buying from the memory manufacturer? Like, did you get a contract price or how did, you know, how did you guarantee your memory supply? And the answer was, there are no contracts and the, the Valve guy told us

• 55:27Speaker 06

  they're offered a price by the memory supplier, in this case, the one we got was Hynek's. And if they say no to that price, then the quote was, they never talk to us again.

• 55:39Speaker 06

  Which is why I feel like it is cartel-like behavior.

• 55:44Speaker 05

  I got some, I got some thoughts I don't think I want to share on the record with, on that one. That is just, you know what? I'm right back around to just, I hope these companies suffer. Like, I think the memory companies are evil. Like, that's just an evil way of doing business.

• 55:58Speaker 06

  Yeah, if it's coercion, right? Cause it's like, if, if the play is, I offer you a price, you say no. Fine, I'll take my supply and sell it to someone. I'll never even talk to you again. Like, you're now down to two companies you can buy from, right?

• 56:15Speaker 05

  And it's not like, and it's not, and if they'll never talk to you again, it doesn't matter if they build more supply.

• 56:21Speaker 06

  Right. Yeah. If, if you're on their blacklist.

• 56:23Speaker 05

  Valve as well, who were like the, like royalty within the PC. It's just, this era is just about the companies shitting on the people that made them rich.

• 56:35Speaker 06

  And it's, you know, the crazy part to me is it's like

• 56:41Speaker 06

  Valve is not a poor company, you know, they're a private, so we don't know how much they have, but they're definitely a multi-billion dollar company. And Valve is getting bullied by the trillion-dollar companies.

• 56:55Speaker 05

  Which are only the trillion-dollar companies because of the stock speculation.

• 56:59Speaker 06

  Right. Yeah. Not like actually, but yeah. It's just, it's insane to me because it's

• 57:07Speaker 06

  of course, there's same thing, Corsair has no real power in the purchasing and they're almost a billion-dollar market cap, not quite. I don't know if they've, how far they've fallen, but they were at one point a billion-dollar market cap. And at point being like, these aren't nobodies who want the memory and they're struggling to, to get the memory suppliers to talk to them. They're like, they're serious players. And they need a lot of it, you know, and the response is take it at this price or go fuck yourself, which is crazy to me.

• 57:36Speaker 05

  I was reading about the long-term agreements that SK Hynix is presenting. And apparently they're considering removing price caps because apparently they do the long-term agreements.

• 57:46Speaker 06

  Yeah, they do.

• 57:46Speaker 05

  And then they say, okay, we like, they won't go above this bounds so that you don't just get rogered. They're just not doing that anymore, I guess.

• 57:54Speaker 06

  Rogered is, I don't think I've seen Rogered.

• 57:57Speaker 05

  Classic British, classic British, beautiful British phrase.

• 58:00Speaker 06

  I was gonna say, like, when I took British literature and before I dropped out of college is the last time I read that. Yeah, I think the, the, in the memory supply, typically contracts, they do set a ceiling, then my understanding is the way the contract works, if you're lucky enough to get one, is that you're offered the supply and you have to buy it or you have to pay a penalty, basically.

• 58:28Speaker 05

  Yeah, take or pay, right?

• 58:29Speaker 06

  Yeah, yeah.

• 58:31Speaker 05

  That's, it just feels like the natural end of this is going to be horrifying. It's just going to be, it's going to be because they're going to be sitting on all this high bandwidth memory. And I was having a conversation with a friend of the show, Casey Kugawa, yesterday, and he was, he was saying that like, well, they'll sell it to someone. And while I agree on that, in theory, we're talking about.

• 58:54Speaker 05

  Going from a few hundred million gigabytes of HBM to trillions of it. So someone is going to be left holding the bag. Maybe it won't be the memory companies, but there's a situation where Nvidia or someone else is just sitting on the stack of this crap.

• 59:09Speaker 06

  Yeah. If, if they can't find devices to put it in and they oversupply, then it'll be the same type of apocalypse we've seen in the past. I mean, in 2016, like typically though, this benefits consumers on the crashing side of the wave. Cause you can, it becomes commoditized, you can get it super cheap. Um, I'm not sure what happens here because one, we've already reset the expectations like we were talking about. So there's always already been a regression in the technology. That'll take a few years to kind of get back to where we were. Uh, but then also whatever crash happens next seems like it's not going to be isolated to the memory industry.

• 59:50Speaker 06

  Right. It seems like it'll be a, like I sent you a photo during Computex of Nvidia's advertisement in Taipei. That said, it all starts here. Do you remember what you replied with?

• 1:00:02Speaker 05

  No, I do not.

• 1:00:03Speaker 06

  You replied with it parentheses, next global financial crisis.

• 1:00:09Speaker 05

  Yes.

• 1:00:10Speaker 06

  And all starts here.

• 1:00:12Speaker 05

  It really is though, because it's like, think about it. If we have just this glut of high bandwidth memory, how, how many graphics cards do they realistically make? Because when I was saying hundreds of millions of gigabytes, that was back in 2022. So this was the demand for these things for every games console manufacturer, every PC manufacturer, everyone. And as of 2022, when the economy was doing a lot better.

• 1:00:38Speaker 06

  Right.

• 1:00:40Speaker 05

  And they only needed a few hundred million.

• 1:00:44Speaker 05

  We could be sitting on arguably the largest supply glut in memory history.

• 1:00:49Speaker 06

  Oh, it's crazy.

• 1:00:50Speaker 05

  For like, like you can't unbuilt that memory, right? You can't pass it out.

• 1:00:57Speaker 06

  You can't really, there's no way to, you would need an HBM device to put it on. There's no real consumer HBM devices these days and it's not something you'd spin up fast. So yeah, if there's like too much HBM, you don't really have a lot you can do with it. Other than I guess sell it for cheaper or something. But like at the end of the day, you need something to attach it to. And, um, I mean, one thing to think about is part of those numbers, although I don't think all of it, but part of it is that the capacity per device has increased a lot for non-consumer devices since 2022. Uh, so some of it is accounted for there, uh, where you just put more on a device.

• 1:01:39Speaker 06

  But then you look at consumer and like somewhat notoriously, it has remained stuck at a like eight gigabytes to maybe 16 gigabytes for a consumer GPU, which eight is barely even enough to, to play modern games with any kind of reasonably, you know, high texture settings.

• 1:02:00Speaker 06

  And so you don't have that multiplying effect in consumer. It's just in data center, which all that to say, if more supply capacity goes online, if they're able to produce more, uh, but the data center demand slows down or the data center GPU demand slows, then and they switch back to just consumer DRAM, they're still going to have too much because the consumer devices are, they don't use that much memory. Um, so you, I don't know, you, you finally spec them up maybe, but.

• 1:02:33Speaker 05

  They've been, but you'll just have all of this insane amounts of high bandwidth memory to go. We're just going to get graphics cards with 128 gigabytes of high bandwidth RAM, just.

• 1:02:45Speaker 06

  Just because they need to put it somewhere, right?

• 1:02:47Speaker 05

  Yeah, we've got to do something with it. We're just gonna have a laptop and you'll be able to open 17 or 18 Chrome tabs at once. It's going to be incredible. No, I think, I wonder if there's also going to be a knock on effect of all of this on the used gaming market, like the used PC market.

• 1:03:04Speaker 06

  Well, that's a really common, um, common, I see people asking about like, oh, surely this is going to be good for secondhand in a few years. When they retire all these devices or the bubble pops.

• 1:03:15Speaker 05

  Oh, I just, I mean, just like right now, I mean, just people won't be buying new devices.

• 1:03:20Speaker 06

  Yeah. Well, I mean, on both of those, right now, it means that if you go look up RTX 30 series cards from, I don't know, like shit, I think it might be four or five years ago now on eBay, some of them are like within 20% of what they sold for new, you know, and so it's.

• 1:03:37Speaker 05

  Yeah.

• 1:03:37Speaker 06

  It's wrecking the used market because it's just like, it's not, you're paying close to new prices for something that's four years old. So that's not good. And then on the other side of it, none of the data center hardware is useful in consumer. There is no secondhand use case.

• 1:03:54Speaker 05

  Yeah.

• 1:03:54Speaker 06

  Yeah. LPDDR5X is either on some form of server specific cam, uh, card or is soldered to the board, so you can't use that. And maybe in China, they'll desolder it, you know, and solder it onto new sticks and sell Frankenstein thing. But, um, and then the GPUs, it's HBM, so it's, it's attached to the GPU. You can't just, you can't just like take it off. So.

• 1:04:21Speaker 05

  This is the thing as well. When it comes to like this post.com bubble recovery story, this is, I'm glad you brought that up because it's like, people like, oh yeah, they'll just use it for something else. It'll be useful infrastructure. It's like, what useful infrastructure? What are we going to do with all these GP? It's not, you can't run a black well, like a GB 300 is, well, it's a Grace and two black wells. You've got over a terabyte of different kinds of fucking Rams on that. I love saying Rams, um, and what are you going to do with that? It's not got, it's not got display port.

• 1:04:55Speaker 06

  Yeah.

• 1:04:56Speaker 05

  It's not going to do anything. It's still, it's built for one thing.

• 1:04:58Speaker 06

  Right. Maybe a local AI model or whatever, but anything reasonable.

• 1:05:03Speaker 05

  You couldn't do that unless it was in a consumer hardware shell. And I assume you need like a dishwasher plug for it or a washing machine plug, even like what kind of.

• 1:05:13Speaker 06

  You probably need 240 volt, um.

• 1:05:15Speaker 05

  Jesus Christ.

• 1:05:18Speaker 05

  People really, people need to do some, do some actually like reading about this because it's like, they need to realize that this can't go anywhere else. Like this, just, we're just going to have racks of these things. And it's not like in the future, they'll be cheaper to use.

• 1:05:33Speaker 06

  I think the, the, you know, if the dot com era argument of infrastructure was dark fiber or something where the, the fiber's in the ground. It's just, it's not used because all the companies went belly up. Eventually they used it. Uh, maybe that's your argument there. I, I really, I agree with you. I don't think it works for this one because the closest thing we have to an infrastructure gain maybe is power. It's the problem of they're not really bringing more online fast enough. Uh, and it's, it's at least currently it's not helping the prices for the rest of us. Like almost everyone's had their power costs go up lately. So maybe that would be the advantage.

• 1:06:14Speaker 06

  If you're like, because of data centers, we brought back nuclear and we have shitloads of solar and wind and whatever, but that's not what we're doing. We're putting master lines on the roof, you know?

• 1:06:24Speaker 05

  And I don't know, it's also what if the power doesn't, what if the power gets built and the demand isn't there? It's just, there's so many things that could go wrong here. And I think people just kind of, I think the reason that this is all coming to a head is because we're finally facing the physical problems. Yeah. Zero, like there's only so much capacity, there's only so, there's only so much money, there's only so much power, there's only so much silicon. I, I, at some point this all has to break and it might break everything.

• 1:06:56Speaker 06

  Yeah. I think, um,

• 1:06:59Speaker 06

  I don't know what the, I don't know what the Enron moment will be, you know? I'm not sure where it'll tip.

• 1:07:09Speaker 06

  But it does seem inevitable, but I don't know, it could be a year, it could be several years, right? Like, I have no idea.

• 1:07:16Speaker 05

  It could just, it feels like the moment the, the vibes shift. Because the trigger I'm looking for is that CapEx pullback. But what scares me somewhat is the Nvidia is the, I think, what 55, 65% of all high bandwidth memory demand, all for AI GPUs. That just feels like a single point of failure.

• 1:07:39Speaker 06

  Yeah. I think Nvidia, uh, I mean, Nvidia itself. Jensen Huang himself said of CoreWeave that without Nvidia, there is no CoreWeave. You know, like if I remember correctly, that was, I'll go dig that up. But they, you know, they create the circle, Nvidia is part of the circle and it's.

• 1:08:09Speaker 06

  Yeah, if anything happens to them, then all of it falters.

• 1:08:14Speaker 06

  And they are a kin maker.

• 1:08:17Speaker 05

  And yeah, but they're also going to be the king loser, I think, by the end of this. Just it kind, it's just so fucking.

• 1:08:23Speaker 06

  Just, I guess.

• 1:08:24Speaker 05

  Yeah, it's so fucking frustrating as well, because I feel like if we'd have slowed this down in 2024, we could have avoided this. But everyone was like, no, we need to go further and harder and faster and spend even bigger and more amounts. But one final question. Surely with the increase in high bandwidth, like, actually, maybe it's two questions in one. The memory companies are kind of have Nvidia over a barrel, too, don't they?

• 1:08:50Speaker 06

  Uh, in the sense that Nvidia doesn't really have alternatives, but Nvidia is also heavily involved in the R and D for all this new memory. So like, they were involved in the R and D for GDR seven, G five X and for the new HBM options. And so like, over a barrel, I feel like, I think you could make that argument. But I also think that Nvidia is more likely pulling a lot of those strains despite being the customer to very few suppliers.

• 1:09:26Speaker 05

  Right. But at the same time, doesn't this just mean that Nvidia's prices are going to get linearly more expensive? The longer this goes on.

• 1:09:34Speaker 06

  I, yeah, I do think, I do think the prices go up. I mean, we're, you know, with Nvidia on the consumer side, too, you can look at that as a gauge. They were supposed to launch their refresh of consumer GPUs like a year ago or something almost and

• 1:09:50Speaker 06

  not quite a year ago, it close. And like rumors were end of 2025, then it was first quarter 2026, and then the rumors were they were all canceled. And the reason for that is because this particular type of refresh, they call the super series. Normally you see an increase in memory capacity at roughly the same price, or you see a decrease in price at the same capacity. And right now, neither of those things is possible. And so Nvidia hasn't shipped the card. It may come out at some point, but for now they've delayed that indefinitely. And so to your point, it does start to self cannibalize, right? We're like, they are the reason that their own refresh for a consumer can't come out.

• 1:10:31Speaker 05

  And also, but regular, the cost of the things they're already making, the current generation will go up.

• 1:10:37Speaker 06

  Yeah. Oh, the current generation has already gone up in terms of like real world pricing. And officially, they haven't changed their MSRP from launch, you know, but it's, I mean, a lot of this stuff. The 5090 is, I guess technically, sometimes you can get them at MSRP, but more often they're like two to.

• 1:11:00Speaker 05

  But I, I mean, like the AI GPUs.

• 1:11:03Speaker 06

  Oh yeah. I mean, that I would, um.

• 1:11:06Speaker 05

  Just like, yeah, all of them will go, it just feels that at some point the money's going to run out and it's the longer it goes, the worse it will get.

• 1:11:14Speaker 06

  Well, and then you get the weird leaseback stuff, right? Like with Meta and the GPU.

• 1:11:18Speaker 05

  Oh, yeah. And all the SPV share. What a horrible era. Truly awful. I think that's a, that's a great place to end it for us. Steve, where can people find you? What are you working on at the moment? I know you have some sort of campaign, campaign, a dystopian campaign you want to?

• 1:11:33Speaker 06

  Yeah, we do have our AI dystopia coverage we're doing. We have a, we're trying something new. We do these documentaries that are like three hours long and we're trying something new and doing a short one. So we're doing two 30 minute ones coming up. And one of them is on the hard drive cartel, which is pretty interesting.

• 1:11:49Speaker 05

  Nice.

• 1:11:50Speaker 06

  And the, the other one is on this memory class action lawsuit. We read through all 118 pages of the filing and, and did a bunch of history research. So those are coming up within the next week on Gamers Nexus.

• 1:12:03Speaker 05

  And we'll get the links in there. I don't know what I'm going to be doing next week on the show because I'm recording this a little early. I'll have a monologue for you on Friday, though. I have been Ed Zitron. This is Better Offline. You know where to find all my stuff will be in the links. Thank you so much for listening.

• 1:12:25Speaker 05

  Thank you for listening to Better Offline. The editor and composer of the Better Offline theme song is Matt Osowski. You can check out more of his music and audio projects at mattosowski.com. M A T T O S O W S K I dot com. You can email me at [email protected] or visit betteroffline.com to find more podcast links and of course my newsletter. I also really recommend you go to chat dot where's your ed dot at to visit the Discord and go to r slash better offline to check out our Reddit. Thank you so much for listening.

• 1:12:57Speaker 13

  Better Offline is a production of Cool Zone Media. For more from Cool Zone Media, visit our website, coolzonemedia.com or check us out on the iHeart Radio app, Apple Podcasts, or wherever you get your podcasts.

• 1:13:30Speaker 07

  Snoring, gasping during sleep, feeling fatigue? Ask your doctor about Zepbound Terzepatide, the first and only FDA approved prescription medicine for moderate to severe obstructive sleep apnea, OSA, in adults with obesity. Zepbound is a prescription medicine used with a reduced calorie diet and increased physical activity to help adults with moderate to severe obstructive sleep apnea, OSA, and obesity to improve their OSA. Zepbound is approved as a 2.5, 5, 7.5, 10, 12.5, or 15 milligram injection. Zepbound contains Terzepatide and should not be used with other Terzepatide containing products or any GLP-1 receptor agonist medicines. It is not known if Zepbound is safe and effective for use in children.

• 1:14:14Speaker 07

  Don't share needles or pins or reuse needles. Don't take if allergic to it, or if you or someone in your family had medullary thyroid cancer, or if you've had multiple endocrine neoplasia syndrome type two. Tell your doctor if you get a lump or swelling in your neck. Stop Zepbound and call your doctor if you have severe stomach pain or a serious allergic reaction. Severe side effects may include inflamed pancreas or gallbladder problems. Tell your doctor if you experienced vision changes before scheduled procedures with anesthesia. If you're nursing, pregnant, plan to be, or taking birth control pills. Taking Zepbound with a sulphonal urea or insulin may cause low blood sugar. Side effects include nausea, diarrhea, and vomiting, which can cause dehydration and worsen kidney problems.

• 1:14:53Speaker 07

  Talk to your doctor. Call 1-800-545-5979 or visit zepbound.lily.com.

• 1:15:00Speaker 14

  Summer's still going strong. Make a stop at Total Wine & More for white wines, beer, and ready to drink cocktails. Just know you're getting the lowest prices. Keep the summer celebrations going at Total Wine & More. Spirits are not sold in Virginia and North Carolina. Drink responsibly. B21.

• 1:15:16Speaker 15

  Small businesses are the pulse of every community. They bring people together, create opportunities and drive growth. That's why Chase for Business helps small business owners like you by offering personalized guidance and digital tools that can help your business grow. With our guidance and your determination, you can take your business farther and help build a brighter future for your community. Learn more at chase.com/business. Chase for Business. Make more of what's yours. The Chase mobile app is available for select mobile devices. Message and data rates may apply. JP Morgan Chase Bank, N A, Member FDIC. Copyright 2026 JP Morgan Chase and Company.

• 1:15:50Speaker 16

  What if you could have even more and more and more help to pursue your goals? At LPL Financial, we offer more ways for advisors and their clients to thrive. So what if you could?

• 1:16:00Speaker 17

  Paid advertisement. Investing involves risk, including potential loss of principal. LPL Financial, LSE member FINRA, SIPC. This is an iHeart Podcast. Guaranteed Human.