My conviction is lower, but not broken. CXMT changes the long-term risk premium for memory. If it can turn 4F² into competitive, high-yield mass production, Chinese supply could eventually pressure conventional DRAM pricing and challenge the industry's current structure. But announcing an architecture is very different from producing server-grade DRAM economically at scale. Yield, reliability, customer qualification and volume are the real tests. Until CXMT demonstrates those, I see the sell-off as the market pricing in a credible new risk rather than proof that the memory cycle has fundamentally changed. Hock Tan's warning matters too. If capital becomes the constraint on AI build-outs, demand expectations could cool even before new Chinese supply arrives. So I would still hold quality me

The AI hardware trade just got hit with two reality checks simultaneously: a structural supply-side threat out of China and a demand-side paradigm shift from Broadcom. Here is the breakdown of what actually changed, what’s noise, and whether memory conviction should hold. 1. The Supply Shock: CXMT’s 4F2 DRAM Breakthrough The Announcement ChangXin Memory Technologies (CXMT) unveiled progress on a 4F2 DRAM architecture, aimed at server/datacenter memory by end-2026. Standard 6F2 Architecture: [ Cell Area: ~6F² ] --> Scaling limit nearing CXMT 4F2 Target: [ Cell Area: ~4F² ] --> ~30-50% die size reduction Why Wall Street Freaked Out * Density Advantage: A shift from 6F2 to 4F2 design theoretically reduces cell area by roughly 30% to 50%, yielding significantly more dies per wafer withou

Hock Tan said the truth and yet Zuckerberg announced to spent even more money on AI. While Meta and Google are making good money from their AI optimized ads, OpenAI and Anthropic haven’t found a profitable business model. The cheaper open Chinese models are forcing them to lower the prices even further while newer models get better. It’s a race to the bottom, AI is already a commodity!

$SanDisk Corp.(SNDK)$ sell. Bubble popping

The indices: the Dow outran the Nasdaq for the third session in a row All three indices split again on Monday: the $Dow Jones(.DJI)$ rose 0.38 per cent to 51,850.55; the $S&P 500(.SPX)$ fell 0.06 per cent to 7,807.24, essentially flat; and the $NASDAQ(.IXIC)$ fell 0.44 per cent to 27,245.90. The 10-year Treasury yield stayed at 5.24 per cent, unchanged from Friday. Monday was the third consecutive session in which the Dow outperformed the Nasdaq. Nvidia fell 1.29 per cent to US$226.32 and AMD fell 2.28 per cent to US$594.24, extending the declines that began last Thursday on the OpenAI revenue report. Memory st

One technology announcement. Four stocks under pressure. On Monday, October 12, $SanDisk(SNDK)$ fell 5.97%, $Lumentum(LITE)$ dropped 9.78%, $Micron Technology(MU)$ lost 3.88%, and $SK Hynix(000660)$ declined 4.02%. The catalyst? China’s memory-chip maker ChangXin Memory Technologies (CXMT) announced progress on a next-generation DRAM architecture that could intensify competition in the global memory market.  But has the competitive threat changed overnight, or is the market pricing in a risk that still needs to be proven? 🔬 1. What has CXMT actually achieved? CXMT says it has made progress on a 4F² DRAM cell architecture, designed to fit more memory cells into a smaller area. Mainstream DRAM has traditionally used a 6F² cell design. Moving to 4F² could improve memory density and potential

The memory trade just got a new risk to consider. On Monday, $SanDisk(SNDK)$ fell 5.97%, $Lumentum(LITE)$ dropped 9.78%, $Micron Technology(MU)$ lost 3.88%, and $SK Hynix(000660)$ declined 4.02%. The trigger? China’s ChangXin Memory Technologies (CXMT) announced progress on a new DRAM architecture that could increase memory density and improve manufacturing efficiency. The company plans to introduce a server-memory product using its 4F² design by the end of 2026.  But I think there are two separate stories here. 🔍 1. The threat is pricing power, not just market share The 4F² architecture aims to fit more memory cells into a smaller area than the conventional 6F² design. If CXMT can successfully scale this technology, it could eventually become a more serious competitor in server DRAM, put

$07709$ fck This piece of shit 

News and my thoughts from the past week (28Sep2026) Benson's selected items (drafted from his inputs, his fact-check notes applied): 1. Weather and natural disasters across Asia. Typhoon Dujuan swept Japan's Kanto region around 21–23 September, bringing record rainfall, landslides and flooding with a rising death toll, and another storm was intensifying near Okinawa by week's end. Vietnam is still recovering from heavy rain and flooding across its northern and north-central provinces from mid-September, with homes inundated and crops damaged. In Thailand, all 50 Bangkok districts were declared disaster areas after severe rainfall, school closures and disruption to government travel — continued rain raises the risk of worse. In the Pacific, Hurricane Polo is real but its near-term threat is

Two-Thirds of Future Capacity Sold — Is Memory Becoming Less Cyclical? Memory stocks surged Tuesday, but the more interesting story may be happening underneath the share prices. $SanDisk Corp.(SNDK)$   SanDisk says roughly two-thirds of its FY2028 bits are already covered by new business-model agreements. That is significant because memory has historically been one of the most cyclical parts of semiconductors. What changed? SanDisk says its new agreements with eight customers cover approximately 50% of FY2027 bits and two-thirds of FY2028 bits. The contracts include committed volumes, minimum financial guarantees and pricing mechanisms.  That gives SanDisk something memory companies traditionally haven’t had: greater visibility into

Preview of the week starting 28Sep2026 - Payrolls, PCE and a coin-flip in Sydney Economic Calendar (28Sep2026) Economic calendar for the week (source: Investing.com, via Benson) A genuinely live week: two US prints that can move the rates path, one central-bank decision that is too close to call, and China going quiet for Golden Week from Thursday. 1. The RBA decides — and the market has not made up its mind (Tuesday 29 September, 2:30 pm AEST). The cash rate sits at 4.35%. Pricing is roughly 55/45 between a 25 bp rise to 4.60% and a hold; NAB expects the rise this week, ANZ, CBA and Westpac lean to November. The awkward part: August CPI lands on Wednesday, after the vote, so the board decides with July's 3.5% headline (3.6% trimmed mean) as its freshest read. Whichever way it goes, the st

The global semiconductor memory market in 2026 is experiencing an unprecedented structural squeeze. While Big Tech hyperscalers, GPU titans, and cloud providers aggressively absorb massive volumes of High-Bandwidth Memory (HBM) and enterprise-grade DDR5, persistent shortages continue to plague the electronics ecosystem. In this article, we would like to offer a comprehensive financial analysis of High-Bandwidth Memory deficits, supply chain bottlenecks, and equity options execution. 1. The Paradox of Abundance: Why Memory Remains Constrained The contemporary artificial intelligence infrastructure boom has fundamentally re-architected the semiconductor demand curve. Hyperscale cloud providers and AI accelerator developers are procuring record volumes of memory silicon. Yet, persistent suppl

I don't think the market suddenly discovered dilution a day late. The shares were being sold through an ATM programme, so the dilution was happening gradually and was already known. What changed when the raise was completed was certainty: 29.3 million new shares had actually been issued, raising US$1.94B. That's roughly 5% dilution against Rocket Lab's previous common-share count. Thursday's rally may therefore have been relief that the Iridium financing was secured and the US$3.6B bridge facility could be cancelled. Friday then looked more like the market digesting the other side of that trade: a stronger balance sheet, but more shares dividing the future upside. So I wouldn't say dilution was priced exactly one day late. The market first rewarded financing certainty, then reconsidered w

I find the memory strength interesting because $SanDisk Corp.(SNDK)$ , $Micron Technology(MU)$ and $SK hynix(SKHY)$ all moved higher while the broader chip chain also remained positive. To me, this looks more like money staying within the AI semiconductor theme rather than a simple rotation away from chips. For Micron, the September 30 earnings will be important. I want to see whether margins and guidance can support the current memory-cycle optimism. A strong report could reinforce the thesis, while weaker guidance would make me more cautious. I am still bullish on semiconductors over the longer term, but I prefer to accumulate gradually during pullbacks rather t

I am leaning toward the view that AI has strengthened the memory cycle, but it has not eliminated the cycle completely. HBM and server DRAM demand are structurally stronger because AI servers are consuming much more memory, so I think this upcycle can last longer than a traditional cycle. At the same time, I understand Burry’s argument. Strong pricing will attract more capacity, and if supply catches up with AI demand, memory margins can compress quickly. For me, the key risk is the timing of the supply response, especially from new capacity and improving technology. I am still constructive on $Micron Technology(MU)$ for the mid to long term, but I prefer watching pricing, inventory and supply data rather than simply following the bullish narrative

I’d choose ③ — DRAM can stay strong, but NAND may peak first. AI is changing memory demand, but not every segment benefits equally. HBM and server DRAM remain closely tied to AI infrastructure, with rising memory content per server helping support pricing. NAND is different. Enterprise SSD demand is strong, but NAND still has greater exposure to consumer electronics. If new capacity ramps faster than demand, NAND pricing could weaken earlier. That’s why I wouldn’t ask whether the entire memory cycle has peaked. The more important question is which segment turns first. Burry’s warning still matters: high margins eventually attract supply. But timing is everything. For MU, SNDK and SKHY, I’d watch pricing, inventories and 2027 capacity growth closely. The memory trade may not be simply bulli