Market snapshotView all

Gold Index57Neutral · Updated 2026-07-22

China A-Share Breadth21.82%Cool · 12/55 industries above MA20

Coming nextComing soon

Coming nextComing soon

Latest articlesView all

• Was July's Selloff Already Written in April?2026-07-18A line-by-line review of two signals identified in April: the M1 turn and the market's final rally. The abrupt reversal after the growth-to-value ratio reached a historic 2.43 extreme was not a surprise; it was the script playing out.
• The Final Rally in China A-Shares2026-04-15The rebound in M1 growth explained the previous leg of the A-share rally and also suggests that the current advance is closer to a final liquidity-driven push.

Daily X picksView all

• 

  Paradis Labs@ParadisLabs7/21, 22:23Memory market update: - Counterpoint forecast a 20% QoQ increase for DRAM. - Trendforce are more conservative at 13-18% for conventional DRAM and 10-15% for NAND. - Morgan Stanley raised Q3 PC DRAM forecast to 15-20% increases QoQ from 3-8%. These aren't perfectly comparable baskets but the main conclusion is that tightness still persists. SK Hynix CEO and Chairman both confirmed supply tightness post-US IPO, potentially into the 2030s. - Samsung is reportedly seeking a Q3 DRAM ASP increase of up to 20% QoQ, with LPDDR increases potentially exceeding 20%. That would be their third consecutive quarterly hike after +90% in Q1 and +60% in Q2. This is above TrendForce's forecast but if price increases stick, $MU and SK hynix would likely follow. - TrendForce expects Q3 server DRAM contract prices to rise 13-18%. Multi year supply agreements limit price increases for some large volume US cloud customers. Which means customers without LTAs and incremental volumes sold outside those agreements would absorb the largest hikes. - SK hynix reportedly plan to scrap price caps in its new LTAs so that spot upside flows straight through, while $MU's SCAs carry floors and ceilings. - TrendForce estimates HBM will consume 22% of the top three suppliers' DRAM wafer input in 2026 while producing only 9% of DRAM bits. Those figures rise to 30% & 13% in 2027. This disproportionate wafer consumption is the main structural reason why conventional / server DRAM can remain scarce, even as capex increases. - On HBM4 itself, 2027 contract negotiations land around Q4'26 and Digitimes model HBM4 moving from ~$2/Gb in 2H26 toward $4-5/Gb next year. - Morgan Stanley suggest NAND remains undersupplied through 2027. - TrendForce expects mature SLC NAND pricing to rise 120-170% in 2H26, driven by shrinking mature-node capacity & MLC -> SLC migration. Positive read-through for $SNDK, Kioxia and $MU, while $SIMO offers controller leverage. - Worth noting Kioxia has filed for a US ADS listing, and $SNDK has raised NAND product pricing double digits while extending the JV with Kioxia to 2034. - $MU also recently broke ground on a ¥1.5 trillion Hiroshima expansion, targeting advanced DRAM & HBM shipments around summer 2028.
• 

  Paradis Labs@ParadisLabs7/21, 17:55$IQE up 18.4% today after reporting "H1 performance ahead of expectations." This is why sticking to a thesis is so important... IQE have also upgraded their 2026 revenue growth guidance from 20% -> over 30%. Driven by InP demand for AI and data centre applications. - H1'26 revenue: £64m vs. £45m H1'25 - Strengthened balance sheet via $MTSI investment leading to debt-free position + bolstered cash position at £42m. I personally think their new ~30% rev growth target is slightly on the lower end - probably conservative estimates on purpose after some recent history of over-estimating growth too soon. But I do believe that we're in the midst of their financial inflection point with IQE forecasting "low-teens £m adjusted EBITDA." Hope they can continue growing sustainably. Looks like they're doing everything right currently, rather than going all guns blazing and absolutely trashing the balance sheet.
• 

  Serenity@aleabitoreddit7/21, 17:14It's slightly more nuanced. Basically, $AMD SF event is tomorrow, Wed/Thurs (July 22, 23). From Jefferies client note, AMD is expected to provide additional disclosures around their likely scale-up CPO roadmap + MI500. If they confirm CPO route, would positive thematically since it's not just $NVDA forcing the transition. As for $SIVE, we'll get a better read if they're directly in AMD supply chains. Since right now, it's possible to map it through $GFS (reportedly involved in AMD MI500 CPO work) or AYAR (which AMD invested in). But no formal confirmations. So if you're turning in: - CPO thematic positive: MI500 explicitly commits to CPO/optical UALink. - AMD gives details using Ayar Labs or GF SCALE for MI500 CPO: That would be extremely positive for $SIVE, given they're the upstream laser supplier / reference laser. Currently, Blayne Curtis from Jefferies expect AMD to work with $ALAB on UAL and $AVGO on ESUN... but juicy details are here: "We are looking for whether AMD commits to a CPO based scale up solution and who supplies the optical engine" We'll get a broader read through on the supply chain come Wednesday/Thursday, too early to tell today.
• 

  Jukan@jukan057/21, 17:17Interesting. Hybrid bonding is expected to be used for Feynman, which will feature custom HBM (cHBM). It appears the technology has been pushed back from conventional HBM4E. https://t.co/YgXE20bxn6
• 

  Serenity@aleabitoreddit7/21, 03:36TBH, casually posting novel information discovery to X is how research should be done. Apparently this was #16 on the Smale's list of Mathematical Problems for the Next Century? Kinda shows that the value generated from AI does not scale linearly with token costs... After seeing this, bears are still probably going to say: "B-but where's the profit coming from after AI usage?" AI + capex is likely going to keep going brrr.

Daily research picksView all

• English research summaries are being prepared.