메모리 카르텔, 가격 결정력을 상실했는가

THE DAILY H.E.A.T. Torching Wall Street’s Obsolete Playbook. Delivered straight to your inbox 5 days a week. Subscribe Home Posts Did The Memory Cartel Just Lose Its Pricing Power? Did The Memory Cartel Just Lose Its Pricing Power? Three companies have set the marginal price of memory for twenty years. Last week, a single Corsair stick with a Chinese DRAM chip inside showed where the first leak opened — not in HBM, but in the consumer DDR5 floor the entire memory trade has been resting on. May 27, 2026 I’ve been a trader and investor for 44 years. I left Wall Street long ago—-once I understood that their obsolete advice is designed to profit them, not you. Today, my firm manages around $5 billion in ETFs, and I don’t answer to anybody. I tell the truth because trying to fool investors doesn’t help them, or me. In Daily H.E.A.T. , I show you how to Hedge against disaster, find your Edge, exploit Asymmetric opportunities, and ride major Themes before Wall Street catches on. Table of Contents H.E.A.T. News vs. Noise: What’s Moving Markets Today ETF News In Case You Missed It H.E.A.T. Wall Street spent the last eight months pricing memory like it was the new oil. DRAM contract prices doubled. TrendForce now expects 1Q26 conventional DRAM contracts to rise 90–95% quarter over quarter, with PC DRAM at least doubling and NAND Flash up 55–60%. Micron’s forward P/E hit a multi-cycle high. Sell-side analysts upgraded the whole complex on a single thesis: the Big Three cannot build fast enough. Seagate’s CEO said the quiet part out loud to JPMorgan last Monday — “factories take too long” — and the stock fell 6% because investors heard supply discipline forever. Then a tipster on X posted a CPU-Z screenshot of a Corsair Vengeance DDR5 16GB kit sold in China, part number CMK5X16G3E60C36A2-CN, running at 6,000 MT/s. Underneath the heatspreader: not Samsung. Not Micron. Not SK Hynix. ChangXin Memory Technologies — a Hefei-based DRAM maker that, until eighteen months ago, didn’t ship DDR5 to anyone outside China. Tom’s Hardware and VideoCardz both picked it up. Corsair hasn’t formally announced the supplier change. The screenshot is the story. That one image does not break HBM. It does something more useful for our purposes: it identifies where the first crack opened. This rule applies here, and it is ironclad — the cure for high prices is high prices, and the cure arrives twice as fast when the supplier is Chinese. The question is no longer whether the cure is coming. It is which leg of the curve breaks first. CXMT Q1 2026 revenue 50.8B yuan ($7.4B), +719% YoY CXMT Q1 2026 net profit 33.0B yuan CXMT Q1 net profit attributable to parent 24.76B yuan, +1,688% YoY CXMT 1a-node (16nm) DDR5/LPDDR5X yield 80%+ (per prospectus) CXMT current share of global DRAM ~7.7% TrendForce 1Q26 PC DRAM contract price forecast At least 2x QoQ TrendForce 1Q26 NAND Flash contract forecast +55–60% QoQ YMTC NAND wafer capacity (2 fabs, today) 200,000 wafers/month YMTC Fab 3 capacity online by 2027 +50,000 wafers/month Sources: CXMT IPO prospectus filed 17 May 2026 (via Eastmoney / Hongxing News, 18 May); TrendForce, 2 Feb 2026; Reuters, 19 May 2026. The setup the Big Three just lost Here is how the memory thesis was supposed to work — and why every long-only fund in the country bought it. HBM, the high-bandwidth memory stacked next to every Nvidia GPU, sucked the wafer capacity out of the rest of the industry. Samsung, SK Hynix, and Micron — the Big Three — booked every fab they had to hyperscalers at prices that hadn’t been seen since the 2017 supercycle. PC memory, server RDIMMs, and consumer NAND all got starved. Prices ripped. TrendForce’s 1Q26 forecast captured the moment: PC DRAM at least doubling QoQ, NAND up 55–60%. The Big Three printed cash. Memory ETF flows ripped. The thesis was clean: AI scarcity is structural, the Big Three are the only games in town, and they will not over-invest the way they did in 2018. Every part of that sentence was correct in October. Every part of it has a hole in it today — but the holes are not in the part of the curve readers will assume. THE PROSPECTUS IS THE STORY CXMT filed its updated IPO prospectus on Sunday, May 17. The numbers inside have not been properly priced anywhere on the Street. Q1 revenue of 50.8 billion yuan — up 719% year over year — is not a ramp. It is a regime change. Net profit of 33.0 billion yuan (24.76 billion yuan attributable to parent) against a 2.83 billion yuan loss the prior-year quarter is the mathematical signature of a company that has already crossed the cost curve. Goldman’s 21 May note put it carefully — “scaling faster and materially more profitable than the market had appreciated.” In plain English: the Street modeled CXMT as a 2028 problem. The prospectus says it is a 2026 problem. “Once China starts producing the commodity, what was formerly price euphoria quickly turns to collapse.”  — The maxim that broke DDR3, then DDR4 — now headed for the consumer leg of DDR5 How the bottleneck migrates — again Long-time readers know we have been mapping the AI buildout as a sequence of bottlenecks, not a single trade. Compute came first. Memory was the second leg, and it ripped. Networking and interconnects are the third. Power is the fourth. Each leg lasts until either capacity catches up or a non-cartel player walks through the side door. Memory is now the first leg to have its side door cracked open. The mechanism is worth being precise about, because it is not the obvious one. The distinction is HBM versus commodity DRAM. CXMT is not breaking Nvidia’s HBM supply chain today. It is attacking the lower-margin DDR5 and LPDDR5 demand that Samsung, SK Hynix, and Micron deliberately deprioritized while chasing HBM margins. That is where the price floor was hiding. If CXMT absorbs the abandoned consumer and second-tier server demand, spot DDR5 weakens first. Contract pricing follows with the usual one-quarter lag. The HBM boom can remain entirely real while the broader memory multiple de-rates underneath it — because the multiple was never just about HBM. It was about the whole curve. CXMT is not yet selling LPDDR5X into hyperscaler racks. ASML lithography constraints still keep it a node behind. What CXMT is doing — and this is the part the Street did not price — is vacuuming up the consumer and second-tier datacenter demand the Big Three walked away from. Corsair is the canary. Per Nikkei via Tom’s Hardware, HP and Dell are reportedly qualifying CXMT DRAM, while Acer and Asus are asking Chinese partners to source local memory. Qualcomm has reportedly explored custom DRAM work. None of that proves global share loss yet. It proves the OEM pipeline is no longer theoretical. Pull the consumer floor out, and the spot-to-contract spread that has been driving the headline narrative collapses. The Big Three then face a binary choice: cut HBM allocations to refill consumer (margin-destructive), or watch consumer share migrate permanently to a Chinese vendor that just IPO’d with a war chest (share-destructive). Both outcomes break the long memory multiple, even if HBM scarcity holds. Where the money actually moves The reflex trade is to short Micron. We think that is the wrong trade — Micron has the best HBM3E execution in the West, and HBM is a different commodity from what CXMT is touching. The cleaner positioning lives one layer up (the tools building CXMT’s capacity) and one layer down (the consumer leg of the curve). Here is the tier sheet: TIER TICKER / NAME WHY IT WINS KEY METRIC TO WATCH TIER 1 Structural NAURA (002371.SZ) AMEC (688012.SH) Chinese semicap. Beijing policy mandate to source domestically. Every CXMT and YMTC wafer expansion goes through them first. CXMT and YMTC capex run-rate; Chinese domestic tool-share gains at the 1a/1b node TIER 1 Structural SANDISK (SNDK) Pure-play NAND. Different thesis from the DRAM call: SNDK wins if YMTC stays export-constrained while global NAND stays tight — “China supply trapped domestically.” Not a CXMT proxy. Enterprise NAND ASPs; YMTC export licensing posture; U.S. Entity List news flow TIER 2 Catalyst CXMT IPO (Shanghai STAR Market, expected H2 2026) Direct exposure if you can access STAR Market. Multi-billion-dollar listing with a profitability story already in print. Forces sell-side to publish models that do not exist today. IPO pricing range; STAR Market sentiment; A-share access via Stock Connect TIER 2 Catalyst CORSAIR (CRSR) Margin relief catalyst. Cheaper inputs without lowering shelf prices = gross margin lift through the back half. Smallest expression of the thesis, cleanest read on its timing. Q3/Q4 component-cost commentary; mix of CXMT-sourced SKUs beyond the China-only kit TIER 3 Derivative ASML (ASML) Nuanced. Loses Chinese lithography upside (sanctioned), keeps non-China HBM/DRAM tool demand. Net: range-bound on this catalyst, not directional. China revenue mix; EUV utilization at non-Chinese fabs TIER 3 Derivative SMIC (0981.HK) Indirect — not a memory maker, but the policy umbrella protecting CXMT also protects SMIC. Sentiment vehicle for the broader domestic-substitution trade. Beijing semiconductor policy announcements; export-control news flow Pressure points: where the Big Three feel it first TICKER / NAME STRUCTURAL PRESSURE WHAT BREAKS THE THESIS MICRON (MU) Largest U.S.-listed pure-play exposure to the consumer/PC DRAM segment CXMT is targeting first. HBM3E is real, but cannot offset a structural ASP cut in legacy DDR5. If HBM4 ramps faster than expected and Micron locks long-dated hyperscaler contracts at fixed prices, the consumer leak matters less. SK HYNIX (000660.KS) Most levered to HBM, but also the largest seller of DDR5 into the PC OEM channel where CXMT is now landing. Korean trade-policy response — if Seoul matches U.S. blacklist scope on CXMT, the share migration slows materially. SAMSUNG (005930.KS) Diversified enough to absorb a DRAM hit, but memory is where the operating leverage lives. A CXMT-led ASP reset hits earnings revisions hardest here. Foundry recovery picking up the slack, or a successful HBM3E qualification at Nvidia. SEAGATE (STX) Narrative pressure, not direct unit-economics pressure. YMTC NAND does not automatically break enterprise HDD pricing — but the “supply tightness forever” story the CEO told JPMorgan is now hostage to a YMTC ramp that does not require new fabs, just utilization. The stock is trading the narrative. Enterprise HDD/SSD bifurcation holds; YMTC stays domestic-only; sell-side stops citing the “factories take too long” quote. MEMORY ETFs (broad complex) The trade is crowded. Positioning is the risk now, not fundamentals. A 10% consumer ASP cut from cartel members defending share takes the whole basket down even if HBM holds. Big Three explicitly cut capex in response, restoring scarcity narrative through 2027. Credibility Firewall Every issue, we separate what we know from what we think. Hold us to the left column. CONFIRMED (sourced, dated) Directional (our read) CXMT Q1 2026 revenue: 50.8B yuan, +719.13% YoY (CXMT prospectus, 17 May 2026; via Eastmoney / Hongxing News) Contract DRAM ASPs follow spot down within ~1 quarter; we expect visible cuts in consumer DDR5 by Q3 prints CXMT Q1 net profit: 33.0B yuan total; 24.76B yuan attributable to parent (same source) DDR5 follows the DDR3/DDR4 historical pattern of ~30–40% Chinese-led price compression once a credible alt-supplier scales CXMT 1a-node DDR5/LPDDR5X yield: 80%+ (prospectus disclosure) Big Three will defend HBM margins before defending DDR5 share — meaning the share loss is real, not transitory Corsair Vengeance DDR5 SKU CMK5X16G3E60C36A2-CN reportedly using CXMT DRAM via CPU-Z screenshots — Tom’s Hardware and VideoCardz coverage, 25 May 2026. Kit appears China-exclusive. Corsair has not formally announced CXMT as a supplier. Qualification ≠ ordering, but the pipeline is no longer hypothetical; first material orders within 2–3 quarters TrendForce 1Q26 forecast (2 Feb 2026): conventional DRAM contracts +90–95% QoQ; PC DRAM at least 2x QoQ; NAND Flash +55–60% QoQ European OEMs adopt CXMT faster than U.S. OEMs given lower political friction (CE/UKCA already certified) YMTC has hired CITIC Securities for pre-IPO tutoring (regulatory filing, per Reuters 19 May 2026) CXMT IPO prices in H2 2026 above prospectus midpoint; A-share liquidity amplifies the signal YMTC current capacity: 200,000 wafers/month across two fabs; Fab 3 adds 50,000/mo by 2027 (Reuters) HBM scarcity remains intact; the multiple compression comes from the consumer leg, not the AI leg Per Nikkei via Tom’s Hardware: HP and Dell reportedly qualifying CXMT DRAM; Acer and Asus reportedly asking Chinese partners to source locally; Qualcomm has reportedly explored custom DRAM work Sandisk re-rates higher on NAND scarcity only if YMTC export status does not change Goldman characterization: CXMT “scaling faster and materially more profitable than the market had appreciated” (Goldman note, 21 May 2026) Memory ETF positioning unwinds before fundamental ASP data confirms the thesis Why now THE CATALYST WINDOW Three things converge in the next 90 days. First, CXMT’s IPO pricing window — a multi-billion-dollar Shanghai listing forces the Street to publish models that have not existed before. Second, Computex 2026, where Chinese DRAM solutions are now expected to appear inside well-known brand kits, removing the “niche” excuse. Third, Big Three Q3 prints, where consumer DDR5 ASP guidance stops being a sideshow and becomes the print. Any one of these moves the narrative. All three together move the multiple. Bear case — where we are wrong WHAT KILLS THIS THESIS The cleanest bear case is regulatory. If the U.S. adds CXMT to the Entity List — the same treatment YMTC received in late 2022 — the international OEM pipeline (Dell, HP, Asus, Acer) freezes overnight. European certification (CE/UKCA) does not protect against a U.S. export-control action that scares Western OEMs out of the supply chain. CXMT becomes a domestic-only story. The cartel’s pricing power survives another two years. The second bear case is HBM dominance bleeding upward into multiples. If HBM4 demand from hyperscalers grows fast enough that Big Three group earnings keep climbing regardless of consumer ASP erosion, the multiple may not compress even as the consumer floor leaks. The thesis becomes a slow-bleed instead of a re-rating. The third is yield reality. Prospectus disclosures are not the same as third-party verification. If CXMT’s 80%+ yield claim turns out to include only a narrow product mix, the ramp slope flattens and the supply shock is delayed. Five takeaways 1.     The memory cycle just rhymed with DDR3 and DDR4. Both prior generations saw Chinese-led price compression once a credible alt-supplier scaled. DDR5 was supposed to be different because of HBM. The consumer leg of the curve is not different — it behaves the same way every cycle. 2.    HBM scarcity and DDR5 floor erosion can be true at the same time. This is the part most readers will get wrong. The trade is not “AI memory is over.” The trade is that the multiple was resting on the whole curve, and the consumer end of the curve just sprung a leak. 3.    The trade is not “short Micron.” Micron has the best HBM3E execution in the West and the most defensible long-dated contracts. The trade is positioning around the Big Three’s consumer/PC exposure, not their HBM franchise. 4.    The cleanest longs are the Chinese semicap tools. Naura and AMEC sit upstream of every CXMT and YMTC wafer expansion. Beijing’s domestic-tool mandate makes them structural beneficiaries regardless of whether U.S. sanctions tighten or loosen. 5.    The catalyst window is 90 days. CXMT IPO pricing, Computex product launches, and Big Three Q3 ASP commentary all land in the same compressed period. This is not a multi-year wait. The cartel ran the table for twenty years. A $40 stick of RAM did not end it. It showed where the first leak is. Space Exposure Designed For Potential Weekly Payouts The space economy is no longer just a government program — it’s becoming part of a global industry that is gradually advancing. But why view long-term growth as the only potential path forward? SPCI seeks to offer approximately 100% concentrated exposure to the space industry while striving to generate weekly income through a disciplined put credit spread options strategy. Gain exposure to the “final frontier” and a potential weekly distribution while you wait for the theme to play out. Discover SPCI. Distributor: Foreside Fund Services | Investing involves risk including possible loss of principle. News vs. Noise: What’s Moving Markets Today My thoughts on rates…. What Iran Tells Us About UFO Disclosure When governments confront unknown threats in their airspace, defense budgets surge and the same aerospace and surveillance companies move hardest. On March 2nd, Northrop jumped 6% and Lockheed 3.3% on the Iran news — and President Trump has since ordered the formal release of government UAP files, with the Pentagon confirming compliance. So if a conventional conflict can move these stocks this fast, what happens when the bigger story breaks? See the UFOD holdings: [thetruthisoutthereufod.com ETF News New HALX ETF Targets Energy, Utilities And Industrials Over Big Tech - Tuttle Capital Heavy Assets Low Ob Tuttle Capital's HALX ETF targets energy, industrial, and utility stocks as investors seek diversification beyond Nvidia and AI-driven trades. Benzinga • Chandrima Sanyal wsj.com How Weird Are ETFs Getting? Try UFO and Midnight-Bitcoin Funds www.wsj.com/finance/investing/how-weird-are-etfs-getting-try-ufo-and-midnight-bitcoin-funds-d539f7e4?mod=finance_lead_pos3 Wall Street Now Has An Alien ETF—And Traders Have Bet $42 Million On It On Polymarket - Palantir Technolo A new ETF bets the government will confirm aliens are real. Polymarket traders have staked $42 million on whether it ever happens. Benzinga • Daragh Thomas A Stock I’m Watching $OUST ( ▲ 3.58% ) OUST (Ouster) is one I'm watching as a pure-play on the LiDAR commercialization curve finally bending the right way. After the Velodyne merger, they've consolidated share in a fragmented industry and are leaning into higher-margin software-attached deployments rather than just selling sensors as a commodity. The digital LiDAR architecture is genuinely differentiated, costs keep coming down, and the end markets — industrial automation, smart infrastructure, robotics, automotive — are all moving from pilot to deployment at the same time. Revenue is growing double-digits, gross margins are expanding, and management has been disciplined on cash burn, which puts them in a much better spot than the SPAC-era LiDAR cohort that's already been culled. It's still a show-me story, but the setup as the survivor in a thinning field with improving unit economics is exactly the kind of asymmetry I like. In Case You Missed It Great talk on with the Acquirers Podcast on markets, value investing, inverse Cramer, and Michael Gayed joins me to talk about taking income from your portfolio and how to get more than 4%…… The H.E.A.T. (Hedge, Edge, Asymmetry and Theme) Formula is designed to empower investors to spot opportunities, think independently, make smarter (often contrarian) moves, and build real wealth. The views and opinions expressed herein are those of the Chief Executive Officer and Portfolio Manager for Tuttle Capital Management (TCM) and are subject to change without notice. The data and information provided is derived from sources deemed to be reliable but we cannot guarantee its accuracy. Investing in securities is subject to risk including the possible loss of principal. Trade notifications are for informational purposes only. TCM offers fully transparent ETFs and provides trade information for all actively managed ETFs. TCM's statements are not an endorsement of any company or a recommendation to buy, sell or hold any security. Trade notification files are not provided until full trade execution at the end of a trading day. The time stamp of the email is the time of file upload and not necessarily the exact time of the trades. TCM is not a commodity trading advisor and content provided regarding commodity interests is for informational purposes only and should not be construed as a recommendation. Investment recommendations for any securities or product may be made only after a comprehensive suitability review of the investor’s financial situation.© 2026 Tuttle Capital Management, LLC (TCM). TCM is a SEC-Registered Investment Adviser. All rights reserved. Keep Reading May 19, 2026 • 2 min read I just launched something This is the trade I've been building toward for 45 years. May 26, 2026 • 10 min read Your Portfolio Isn't Up. The Ruler Shrank. Wall Street keeps repricing assets in a currency that no longer holds its shape — and the trades that win from here aren't the ones that benefit from inflation. They're the ones that get paid in something other than the dollar. May 27, 2026 • 2 min read Low P/E used to mean bargain. I'm not sure that's true anymore. View more 4