Orbit Post Sitemap
#台积电Q3净利润超预期，盈利创历史新高
台积电最新财报利润大涨将近七成，为什么股价反而跌了？ $TSM
聊到台积电刚刚公布的三季度财报，营收突破1.49万亿新台币，净利润飙到7631亿新台币，同比激增近69%，各项指标都把市场预期甩在身后。2nm和先进制程占比极高，AI芯片的需求依旧硬核
有意思的是，业绩利好公布后市场却小幅跳水。市场并不是担心它赚不到钱，而是开始对它极具扩张野心的资本支出打问号
今年台积电资本开支直接拉到640亿美元，未来几年甚至可能冲到800亿上下的天文数字
全球同时铺开近20座工厂，这种前所未有的扩产速度，意味着极高折旧成本和海外建厂的管理摊薄。而且毛利率指引出现微调，市场开始警惕产能过剩以及AI投入产出比的边际效益下降
预计台积电短期的逻辑是垄断溢价，没人能替代它，但中长期的逻辑将从AI红利期转向产能消化期。一旦各大科技巨头的AI投入转入精细化精算阶段，巨大的资本开支就会变成双刃剑
业绩是顶峰，但狂奔的警钟也敲响了
DYOR Last night the CPI was released, and the market reacted very strongly. The US stock market dropped by 1.8 points, then the night session fell another 1 point, and during the day the Korean stock market crashed again, with Hynix, SanDisk, Micron all plunging. Even crypto took a big hit, except gold which might still be okay, since gold is known as a safe-haven asset and many countries are actively buying it. So in the short term, gold is relatively hard to fall.
Last night's CPI release was very poor; the expected value was 3.6, but the actual release was 3.7, higher than expected. So the market thinks the Federal Reserve might raise interest rates in October. Currently, the market's predicted probability of a Fed rate hike has increased significantly. When we looked yesterday, the probability of a 25 basis point hike was only 19%, but today it has risen to 27%. It is expected to continue rising, but as long as before the FOMC meeting at the end of the month this data does not rise above 50%, and crude oil prices can fall back somewhat, the probability of a rate hike is not high.
However, a rate hike is like a double-edged sword; as long as there is an expectation, it will keep pressuring the market. It might be better to just raise rates in October. If rates are raised and inflation can be controlled by November, that would be good. Also, after the midterm elections in November, Trump will definitely intervene in the Iran issue, since oil prices have been high for half a year. They can't stay this high forever!
But if rates are raised in October and inflation is not controlled, then rate hikes will likely continue, and the market will be doomed, leading to a worse outcome.
So currently, the US... The market is no longer betting on when the interest rate will be cut, but on how many more times it will be raised.
Once the CPI came out, the bets changed immediately. The probability of no change in October was nearly 80%, but now the bets on rate hikes have piled up, with three more expected by the end of 2027.
But this time, the reason for the rate hikes is oil.
Rate hikes suppress demand, but they can't control oil prices. Saudi Arabia won't pump out two more barrels just because the Fed raised rates once.
So this medicine can't cure this illness; it can only suppress the economy first. Prices that should rise will still rise, and life will still be tight.
For us here, the meaning is very direct: money will be tightened, and risk assets won't have fresh liquidity. Most rebounds in this environment are just for people to sell, don't mistake it for a market rally.
BTC is at 81,000, keep an eye on the 80,000 line; if it breaks, it will continue downward. I'm not buying more.
With the market this bad, having to watch it every day, what's the point?
#TradersBetOnFedHikingRatesThreeMoreTimes #SeptemberCPIIncoming HowWillCrudeOilAndUSTreasuriesMove $BTC 【Ondo locked assets have reached $4 billion, but $ONDO has yet to take off 】
Ondo Finance's total locked assets have surpassed $4 billion, with tokenized stocks and ETFs accounting for over $1.3 billion, and the cumulative number of holders exceeding 1 million. The story of Wall Street going on-chain is indeed growing bigger with Ondo.
But ONDO hasn't taken off yet. TVL represents the platform's asset scale, not $4 billion directly buying ONDO. Treasury bond products provide yield, stocks and ETFs expand trading scenarios, and Ondo is still expanding into private markets; the business footprint is indeed growing.
ONDO has actually been falling these past two days, around $0.47 on October 12, then continuously dropping to about $0.43 on October 15. Business is hitting new highs, but the token is falling; clearly, the funds are not trading according to the same script.
I am optimistic about the long-term potential of the RWA sector, but right now, Ondo's growth cannot be directly equated with ONDO's value growth. ONDO is currently mainly a governance token; more platform assets do not mean token holders can directly share more income.
What to watch going forward is whether new assets can continue to grow, whether trading activity can be realized, and whether the token has a clearer value capture mechanism. The story has grown big, but whether the token can get a piece of this cake remains to be seen.Bitcoin fell below 82,000, and inflation data crushed stocks, bonds, gold, and crypto across the board. Mastercard spent $1.8 billion to acquire BVNK, with the stablecoin payment sector continuing to be swallowed by traditional giants. Chainlink feeds prices on Stellar, and the RWA narrative is still unfolding. Richard Heart was arrested in Turkey; it's an old scheme, stay away.
Just closed the security booth window; the wind outside is a bit strong.
TRUTH current price is 0.0172380. The candlestick broke below the short-term moving average, MACD formed a death cross, volume at the top stalled, and the trend has weakened. Looking at the liquidation chart, there are many short stop-losses piled above 0.02, while the long support below is thin. Short-term weak correction, with a liquidation lure above that could trap bulls. If the rebound fails to surpass 0.018, it will continue to probe down, testing the 0.015 range.
Trading strategy: short position mindset. Entry zone between 0.0175 and 0.018, take profit first target at 0.0155, second target at 0.015. Place stop loss above 0.0188; if broken, accept the loss. Avoid blindly chasing longs; buying at this level just provides liquidity to the market makers.
Have some tea and keep watching the market.
$TRUTH
#Anthropic reports Q3 revenue rose to $16.8 billion and turned profitable
@OKX Planet Lessons from the South Korean stock market are replaying in the crypto market
South Korea's KOSPI fell from a high of 9114 in June to 6035, a drop of over 30%. It went from being the best-performing index globally in the first half of the year to the worst-performing index globally in the second half, all within just over three months.
The mechanism behind this crash is worth every crypto trader's close attention. Foreign investors withdrew $131 billion from the South Korean stock market this year, and foreign ownership of Samsung Electronics dropped from 52.4% at the start of the year to 46.38%, the lowest since 2008. When foreign investors concentrate their withdrawal on two stocks that together account for more than half the index's weight, the index has no second support point.
The crypto market has similar structural risks. BTC accounts for 58% of the entire crypto market capitalization, and ETF net assets account for 6.38% of BTC's total market cap. When institutional funds start withdrawing from ETFs, there is no second "buyer group" to catch the fall. Retail investors will only sell in panic, not buy.
South Korean retail investors net sold 14.197 trillion KRW in Q3, wiping out all the buying in July and August. Retail investors fled when they were most needed to hold. The retail structure in the crypto market is even more fragile due to higher leverage, 24-hour trading, and no price limit protections.
The only thing I learned from this crash in the South Korean stock market is: when liquidity recedes, don't believe any "this time is different" claims. $BTC $ETH $ZEC #美国CPI超预期升至3.7%，今晚PPI再迎考验 1. Inflation: Sticky and Confirmed Last night, CPI broadly exceeded expectations: overall 3.7% (expected 3.6%), core 2.6% (expected 2.5%). Tonight's PPI is a leading indicator for CPI, checking if production costs have been passed on. Expectations: year-on-year 5.3% (previous 5.4%), core year-on-year 4.4% (previous 4.6%). Focus on core year-on-year; 4.4% is the market's bet on cooling, exceeding it would be a double confirmation. 2. US Dollar and Treasury Bonds: 10-year at 5.3% DXY surged to 102.86, a one-and-a-half-year high; 10-year US Treasury at 5.28%, 2-year at 4.87%. The implication is straightforward: global liquidity continues to tighten, and the pricing anchors for all risk assets are rising. BTC can't rally, and this is the root cause. 3. Oil and Geopolitics: Inflation's upside option Brent at $105, WTI at $94.6. In September, US-Iran military tensions escalated, Brent crude rose 14% in a single month breaking $100, and the Strait of Hormuz shipping volume remains 30% below pre-conflict levels. Oil prices are the direct driver behind this round of CPI surprises, and energy components in PPI will also be pushed higher. As long as geopolitical tensions persist, inflation has an additional upside option. 4. Global Markets: Broad declines, only gold benefited from safe-haven demand Nasdaq -1.88%, Nikkei -4.38%, South Korea's KOSPI down 7 consecutive days totaling -20%. Gold fell for 3 consecutive days after CPI to 4083, and the dollar weakened this morning #TSMCQ3RecordProfit Palantir Q3 (Nov 2) guided $2.16B revenue but street wants $2.25B-$2.30B and $0.41 EPS, so it must beat its own guide by 6%+ again. Valuation is extreme at $500B+ market cap, 161x P/E and 73x sales vs Nvidia's 28x. Growth is real (+93% Q2) but price needs perfection - that's why it fell after 2 of last 5 beats despite beating.$BTC 81,000 is not support; it’s the ETF cost line holding firm. On October 15, BTC dipped to a low of 80,985, marking the third time it tested this area. The average ETF holding cost is about 81,500, just right underfoot. Once it closes below this, ordinary ETF holders collectively turn to losses, accelerating redemption pressure.$BTC
Bitcoin ETF net outflow is $261 million, with Fidelity FBTC alone running off $153 million, and BlackRock IBIT outflowing $111 million. In the past 24 hours, the entire network liquidated $452 million, with long positions liquidated at $394 million and shorts only $58 million. The bulls are getting hit, the bears are collecting money.
The rebound rejection at 82,500-83,000 is the entry point, with stop loss set above 83,500. The target is first 80,200, and if broken, then 78,000.
ETF funds are not buying; the rebound is an opportunity to short. $BTC
With CPI released, continuous ETF outflows, and long liquidations far exceeding shorts—the data has clearly set the direction. But especially at times like this, control your impulses; don’t chase shorts at support levels. Wait for the rebound to be rejected at resistance before entering. The biggest taboo in shorting is rushing; wait for the market to give you signals, which is more reliable than finding your own signals. #美国CPI超预期升至3.7%，今晚PPI再迎考验 #台积电Q3净利润超预期，盈利创历史新高 Currently, this wave has dropped from 116 to 100, a retracement of 13.9%, which falls into the shallow category. To know how long it historically takes to recover from this level, just review the data from the past 1000 days.
The method is simple: isolate each drop from a stage high, categorize by the percentage drop, then count how many days it took to return to the previous high. Returning to the previous high is calculated by closing price, not just touching it intraday.
For the 10% to 20% drop category, it occurred 53 times in the past two years, with 48 times returning to the previous high. The median time was 8 days, the fastest was 1 day, and the slowest took 219 days.
For the 20% to 35% drop category, it happened 24 times, with a median return time of 46 days. Moreover, 5 times have not returned yet, with the longest wait exceeding 300 days.
A deeper category, 35% to 50% drops, occurred only 5 times, with a median return time of 86 days, the fastest 37 days, and the slowest 202 days.
There is also a longer reference. $SOL's historical highest close was in January 2025 at $261.97, and it has not returned in the past 21 months, currently priced 60% lower.
Therefore, the deeper the retracement, the longer the wait to recover, and some in the deeper two categories never recover. This current 13.9% drop falls into the most common category, with a reference median of 8 days, slower cases taking a few months, and 48 out of 53 times it has recovered.#Netflix growth may slow, advertising business becomes new highlight
I actually don't care much about Netflix's revenue in this earnings report
Subscription plus advertising achieving about 12% growth is basically within expectations
What I'm watching is the profit margin guidance of 33.2%, compared to only 28.2% in the same period last year
This company has shifted from competing for users to making more money from each user
With expectations set so high, even a good earnings report might cause a drop
If the advertising part is described vaguely, some will sell after hours
Tech stock sentiment has been so fragile lately, don't be too confident on the night of the 20th #PalantirQ3Valuation
Palantir Q3 (Nov 2) guided $2.16B revenue but street wants $2.25B-$2.30B and $0.41 EPS, so it must beat its own guide by 6%+ again. Valuation is extreme at $500B+ market cap, 161x P/E and 73x sales vs Nvidia's 28x. Growth is real (+93% Q2) but price needs perfection - that's why it fell after 2 of last 5 beats despite beating.一天流入6.56亿，一周却流出36.23亿
$BTC 24小时净流入6.56亿美元。
同一周，净流出36.23亿美元。
这个数怎么算的：
一天的钱是短线进的，一周的钱是中期撤的。
两个方向不冲突，只是时间尺度不同。
谁在这儿挂单：
$ETH 一天进7亿，一周出35.59亿。
$SOL 一天进1.25亿，30天出20.41亿。
$ZEC 一天进1.46亿，承接最强。
单日流入说明有人试探。
周线还在流出，说明大钱没回来。
试探和总攻是两件事。
连续三天净流入，才算资金真回来了。
#渣打开放BTC与ETH交易，银行加密业务扩展
#$BTC $ETH $SNDK $SKHYNIX Hynix starts to consolidate while SanDisk plunges sharply!!!
Institutions begin to support Hynix, so Chai didn't dare to short it, but this time Chai plans to short SanDisk hard.
Recently, Chinese manufacturers have substantially overtaken; Yangtze Memory has completed overtaking in the NAND market, and Changxin Memory's breakthrough in the DRAM field has caused panic across the entire sector, coupled with many institutions lowering target prices, leading to a brief loss of market confidence in SanDisk.
Additionally, continuous insider selling at SanDisk, along with CPI data exceeding market expectations, dollar appreciation, and tech stocks collectively under pressure. In the short term, Hynix has triggered a protective cushion mechanism; institutions start to support the stock, and retail investors and whales begin to look forward to extra shareholder returns after the Q3 earnings release.
SNDK current price is suitable for entering short positions: 1348, first target 1322, stop loss 1400. The above is personal sharing only and does not constitute investment advice.
#SanDiskMSCIRebalancingEffective, NANDValuationInFocus #USCPIExceedsExpectationsRisesTo3.7%, TonightPPIFacesTestAgain #HynixRespondsToUSExpansionRumors Core DAO Business Truth on the London Stock Exchange (LSE) The $CORE token itself is not listed on the London Stock Exchange. What is listed is a BTC staking ETP product (1VBS) issued by a third-party issuer Valour (under DeFi Technologies), with the underlying staking technology supported by Core. Many community promotions simplify this as "Core listed on the London Stock Exchange," which is a promotional statement and not a listing of the CORE coin for trading. Product: 1Valour Bitcoin Physical Staking (1VBS) 1. What it is: An ETP (Exchange Traded Product, similar to an ETF), publicly traded on the London Stock Exchange, regulated by the UK FCA, with physical Bitcoin as the underlying asset. Bitcoin enters the Core network for non-custodial staking to generate yields. 2. Business logic - Valour holds real BTC, stored in institutional cold storage; - BTC is delegated to Core network validators for staking, generating staking rewards (nominal annualized about 1.4%); - Staking rewards are included in the product's net asset value, so investors buying this London Stock Exchange security indirectly receive "BTC price appreciation + staking rewards"; - Open to professional investors in 2025-09; FCA license obtained in 2026-01, opening trading to UK retail investors. 3. Core's role here: underlying technology service provider - providing Satoshi-PIf the dips before the CPI were all about onboarding the same group of people, then this time BTC returning to the support level is not a coincidence but a prearranged script. Have you noticed that in the last four CPI releases, the price has chosen to continue upward? Watching the rhythm around these data points, it feels like the market is actually trading on a "bad news fully priced in" inertia. Each time before the data, it sweeps a round of lows to wash out the unsteady positions, then after the release, it moves forward lightly. This time BTC once again stands firm near support; emotionally, everyone is still shouting 75k, but the more unified the shout, the more I feel the price might not obediently go there to pick people up. The bullish logic is clear: if the historical rhythm repeats, CPI is a ready catalyst, and if support holds, the target can look toward previous highs or even new highs. ETH and SOL will also warm up with risk appetite, especially SOL, a high-beta asset that often runs faster once BTC confirms direction. But the risks must be clarified: if support is clearly broken, this judgment is invalid. Also, the market may have already priced in the "post-CPI rise" expectation in advance; if the data just meets expectations, it can easily turn into a case of good news fully priced in. Another easily overlooked point is that capital preference now leans more toward BTC dominance; altcoins need volume to follow the rally, otherwise it's just a false fire. My own rhythm is to first see if support can hold, then watch how the first 4-hour candle after the data release closes. If it closes up on volume, it means the buying is real; if it spikes then falls back on low volume, then be cautious of another fakeout Is $SOXL at a low point now?
The price looks very low, but that doesn't necessarily mean it's the bottom.
To judge whether it's the bottom, you can't just look at how much the price has dropped; you need to look at two core signals:
✅ U.S. Treasury yields stop rising and start to fall
✅ Heavy holdings like storage and Nvidia stabilize, with earnings reports and guidance not continuing to disappoint
As long as U.S. Treasury yields remain high, even if this is a temporary low point, it's likely just a consolidation phase, making it difficult to start a big rally directly.
Even if there is a rebound, treat it primarily as a corrective bounce, not as the start of a new bull market.
This crash is due to macro interest rates killing valuations + negative news in the storage sector, compounded by 3x leverage amplifying losses. The price looks cheap, but bottom-fishing leveraged ETFs carries huge risks. The bottom is formed by a process, not just by falling prices. #USCPI unexpectedly rose to 3.7%, and tonight PPI faces another test $ZEC
I don't dare to buy in; 1120 was directly smashed through, with a low down to 1070
Brothers, I've been watching ZEC's trend for a while, and my advice is simple: don't reach out. It held at 1120 for a long time, and many thought it was the bottom, but this wave smashed right through it, hitting a low of 1070. Look at those big bearish candles with huge volume—it shows it's not a slow decline but bulls giving up one after another, forced liquidations triggering a chain reaction. Whoever is slow to exit gets buried. This kind of stampede-style drop means the bottom isn't something you can just guess.
I checked around, and $BTC and $ETH are also dropping simultaneously, indicating it's not just ZEC's problem but the whole market releasing risk. At times like this, trying to bottom-fish a single coin is going against the market.
Technically, without going into jargon, 1120 was a psychological bottom before. Now that it's broken, it becomes resistance above. If it rebounds back near 1120, there's a high chance people will sell because too many are trapped there.
My thinking is simple: I absolutely won't bottom-fish near 1112 now. I'll wait for a rebound near 1120; if it can't break through and starts turning down, that's where I'll short. Below, I'll first watch 1060, and if that breaks, then the 1000 round number.
#US CPI unexpectedly rose to 3.7%, and tonight PPI faces another test Hayes was still calling for a market rescue in the morning, but in the afternoon he went OTC to sweep through some altcoins.
Talking macro with his mouth, but buying old DeFi faces like AERO and PENDLE with his hands.
I can only take celebrity buys as a reference; his cost and position size are in a different league from retail investors.
He can afford to wait three months, but many people can't wait even three days.
$PENDLE #South Korean stocks fell nearly 8% intraday, global tech stocks collectively under pressure $SAMSUNG
Circuit breakers have already triggered 10 times this year.
▪️ On 10/15 at 11:24 AM, KOSPI dropped over 8%, triggering a circuit breaker and pausing trading for 20 minutes, marking the 10th time this year
▪️ Intraday reported at 5,529.61 points, down 7.97%; KOSDAQ 150 down 8.83%
▪️ Samsung Electronics −10%, SK Hynix −11%, Hanmi Semiconductor −10%
▪️ Nikkei 225 down 5%; Kioxia −12%, Advantest −14%
▪️ The trigger was attributed to the US CPI overnight exceeding expectations, with the probability of a rate hike in October rising to 43.7%
The disagreement is not about how much it fell this time, but that a tool meant to prevent panic has become routine.
Circuit breakers limit speed, not the magnitude of the drop — the same drop completed in one day does not trigger it, but if it happens within half an hour, it does. The 10th time this year indicates an increase in volatility frequency. The decline is not limited to Korean stocks: semiconductors in Japan and the US also fell simultaneously during the same period.
The direction is to wait and see — stability is only confirmed when the interval between circuit breakers lengthens and foreign net selling turns positive; if the frequency continues to increase, the volatility center will still be rising.#TSMC Q3 Net Profit Exceeds Expectations, Profit Hits Record High The AI industry chain's performance is still growing, but the market is starting to pay more attention: How long can this round of AI investment really last? On October 15, TSMC announced its Q3 2026 results, with net profit reaching NT$763.1 billion, a year-on-year increase of about 69%, exceeding the market expectation of NT$746.3 billion, setting a new quarterly profit record. This is also TSMC's sixth consecutive quarter of record-breaking profits. Q3 revenue was about NT$1.494 trillion, up approximately 51% year-on-year. This financial report sends three important signals. First, AI computing power demand remains strong. Major customers like Nvidia and Apple continue to require advanced process chips; AI servers and data center construction remain key drivers of TSMC's performance growth. TSMC's results indicate that AI demand has not visibly cooled on the performance side. Second, advanced process technology remains the core competitive advantage. TSMC benefits not only from chip demand growth but also from the technical barriers of advanced processes and advanced packaging. However, as the 2nm process gradually scales up, gross margins may be affected by initial costs, and it will be necessary to observe whether the new process can successfully translate into profits. Third, good performance does not necessarily mean the stock price will rise. The current market concerns are not only about whether AI demand exists but also whether companies can continue to bear high capital expenditures and whether future revenues can cover the increasing investments. TSMC remains cautious about the outlook for 2027, while the US AI stock sector is also affected by rising interest rates and financing pressures. There is another big move at 8:30 tonight, PPI
After CPI exploded to 3.7% last night, today's market hasn't caught a breath. BTC at 80168, just one step away from 80,000.
ETH at 2315, down nearly 10% in three days.
ZEC is even worse, at 1080, dropping from 1695 to 1080 within a week, nearly a 40% decline.
No need to look at liquidations either, this week's 1.19 billion list just got digested, and waves keep coming these past two days. Bulls just got buried, and those trying to bottom-fish got buried again. This market now specifically targets the fast movers.
Why is tonight's PPI so critical?
CPI and PPI are conjoined twins, one upstream and one downstream. CPI beating expectations means costs have already passed from factories to shelves.
If PPI also beats expectations, it means the next round of CPI will still rise, confirming the 50 basis point rate hike priced in for December.
My view remains the same: the first spike in a downtrend is not the bottom, and the second one may not be either. Today's spike at 80168 is the second probe of 80,000, a bit more solid than the first, but as long as PPI hasn't landed, any spike is temporary.
Once the data lands, watch if BTC 80000 holds. If it holds, this week's dip is over. If not, 78000 is next.
$ZEC
What do you all think, will tonight's PPI cause another explosion or give a breather?
#美国CPI超预期升至3.7%，今晚PPI再迎考验 $BTC $ETH From the risk premium perspective, distinguish between low-risk mainstream markets and high Beta sector markets.
When market risk appetite is high, capital is willing to speculate on high-volatility sectors; when risk appetite contracts, capital flows back to core assets like BTC and ETH for hedging.
BTC reflects capital's hedging willingness, ETH measures market risk premium. Layered screening: when risk premium rises, high Beta assets experience much larger drawdowns than mainstream ones.
Tracked assets list:
🟠BTC｜Benchmark for hedging assets
🔵ETH｜Gauge for risk premium
🟣SOL｜High Beta public chain
🟢AVAX｜Ecosystem public chain
🔷INJ｜On-chain finance
Key focus on the relative strength of BTC and ETH:
When risk appetite rises, ETH/BTC rises, and high Beta assets like $SOL and AVAX show greater elasticity; when risk appetite falls, capital flows back to BTC, and high Beta sectors experience significant drawdowns. Bitcoin needs to return to 86000; what's missing is not time, but a catalyst. The current price is about $80,985. 86000 is not a price issue, it's a confidence issue.
Why is 86000 so strong? This is the average cost zone for investors who entered in the past two years. The cost for holders of 6-12 months is about 88000, and for holders of 1-2 years, it's close to 100000. Every rebound near 86000 triggers selling to break even. Glassnode clearly lists 88000 and 100000 as major resistance levels above.
What conditions could push it back? It's not a technical pattern, but a macro shift. Currently, CPI is exceeding expectations, and market concerns about a Fed rate hike in October have dropped from 50% to about 20% probability. Hawkish expectations are suppressing all risk assets. KCM Trade analysts point out that weaker-than-expected inflation data is needed to reduce the likelihood of further tightening; this is the real short-term catalyst for an upward move.
On-chain data offers hope. Whale holdings have risen to 3.03 million BTC, with a 30-day growth rate the fastest since February this year. ETFs are selling, but large holders are buying. This signals a shift of chips from weak hands to strong hands, but until this turnover is complete, the price will struggle to effectively break through 86000.
The next observation point is very specific: this week's CPI and PPI data, as well as the FOMC meeting minutes. If inflation data unexpectedly weakens, a test of 86000 could come soon; if it remains strong, the 400-day moving average support at 80000 will face a real stress test.
$BTC 今早这一记向下的插针，狠狠带走了一大批高杠杆多头。
这一次震动之中，最出圈的就是大家熟知的麻吉大哥 除了他之外，链上还监测到一位巨鲸，手握价值3.52亿的ETH多单。当下浮亏已经接近1900万，距离清算线仅仅只差60美元，价格再往下轻轻踩一脚，就会迎来爆破时刻。
一幕幕惊心动魄的清算，时时刻刻都在提醒所有人：杠杆可以制造奇迹，也可以一瞬间抹平一切。但是抛开情绪，我们理性看待这一根长下影。
虽然这一针杀伤力十足，却未必是纯粹的坏事。
从盘面来看，这已经是BTC第三次回踩81000附近的支撑区间。一个值得留意的信号：每一次下跌，引发的恐慌性砸盘规模正在逐步递减。
这说明无脑恐慌出逃的筹码变少了，此刻离场的，大多是心里摇摆、扛不住震荡的浮动筹码。市场正在用剧烈的震荡，完成一轮痛苦的洗筹过程。
再看宏观层面：美国通胀数据超预期，依然持续压制着全球风险资产的估值，大环境的压力还没有完全解除。
不过，加密市场反复去触碰底部支撑、反复去测试底线的行为，从技术逻辑来讲，就是低位磨底的典型特征。Don't just remember Big Brother Maji as a whale; in these ten days, he has collapsed to nothing but a sigh. The ledger reveals three facts: 1. Heavy losses — a total loss of $36,572,500, with an additional $4,435,600 lost in the past 7 days, leaving total account assets at only $279,800. 2. Heavy positions — 2100 ETH fully leveraged long positions, valued at $5,074,400, 25x full leverage (account shows 36.17x), opening average price 2515.81, current unrealized loss $208,900, cumulative funding fees burned $1,362,700. 3. No way out — liquidation price 2397.51, available margin 0. What does stacking these three mean? Less than $280,000 left in the pocket, holding $5,074,400 in positions, margin cleared to zero, only a needle's distance from the liquidation price, no bullets left to top up. In ten days, the whale has turned into a small fish. Leverage is like this: as great as it feels when it feeds you, it is just as ruthless when it collects. So don't just focus on whales to catch the bottom; those who survive long in contracts are always the ones who control their positions.The storage trio also fell sharply: Micron dropped 7.4%, SanDisk fell 7.8%, and SK Hynix directly crashed 12.8%. The Philadelphia Semiconductor Index has also declined for six consecutive days.
But the price drop isn't the scariest part; the long-short ratios are absurd: $MU at 4.88, $SNDK at 5.01, $SKHYNIX at 5.91—each more extreme than the last. And the positions are still increasing: Micron surged from 33 million to 38 million, SanDisk from 70 million to 110 million, SK Hynix doubled from 40 million to 84 million. What does this mean? The bulls are buying more as prices fall, stubbornly holding on without selling.
The fundamentals haven't actually collapsed; HBM orders are still there, and long-term contracts remain intact. But the current problem isn't fundamentals—it's that the chip structure is too tight. After such a big drop today, the bulls still haven't given up, which itself is a risk. Let the dust settle for a while; once liquidation data stops spiking and market volatility narrows, then consider testing with a small position.
Key support levels below: Micron at 850, if broken then 800; SanDisk at 1300, if broken then 1250; SK Hynix at 900, if broken then 850.
Do you think this storage wave is a correction or a bearish reversal?
#KoreanStocksMiddayDropNearly8%,GlobalTechStocksUnderPressure
#MicronPlunge:BottomOrMidSlope?
#CanHBM4MassProductionDriveSKHynixProfitsToNewHighs? AI computing power demand has shifted from being "expected" to a quantifiable, structural source of revenue, and the rapid commercialization of 2nm technology has further widened the technological gap with competitors. The real test is not on the demand side—where there is virtually no ceiling—but whether TSMC can maintain its profit margin floor while simultaneously advancing 2nm capacity expansion, overseas factory construction, and next-generation process development. The current gross margin level of 66.4% remains absolutely leading in the global semiconductor manufacturing sector, but the trend of sequential narrowing is worth continuous monitoring. #TSMC Q3 net profit exceeds expectations, profits hit record high 📊 Storage Giants Plunge | SanDisk, SK Hynix: Is It a Valuation or Fundamentals Crash?
Brothers, the storage sector was bloodied overnight. In one sentence: South Korea triggered a circuit breaker, SanDisk broke key support, but fundamentals and stock prices are at odds.
Let's look at the data first——
Ticker Latest Price 24H Drop Key Level
SanDisk SNDK $1,403.34 -5.33% Continued to fall to $1,339 in after-hours
SK Hynix (Korean stock) ₩1,277,000 -10.01% Intraday drop exceeded 12%
KOSPI Index 5,554.22 -7.57% Triggered the 10th circuit breaker of the year
SanDisk slid from $1,787 at the beginning of October to $1,403, a decline of over 20% in two weeks. SK Hynix fared worse, with intraday losses expanding to 12%, hitting the lowest price since April 27, forcing the Korea Exchange to halt trading for 20 minutes.
Why did this happen?
The trigger was the US CPI exceeding expectations overnight, with high interest rate expectations suppressing all high-valuation assets. But the storage sector fell harder than other tech stocks due to an additional reason: prior leveraged positions being liquidated. Hedge funds' leveraged positions on Korean memory chips were forced into cascading liquidations, and retail investors' double-leveraged single-stock ETFs were simultaneously liquidated, mechanically amplifying the selling pressure.
But here a key divergence appears.
On October 14, Samsung Securities just released a report with a core view in one sentence: "The stock price decline excessively reflects profit concerns." They maintain a buy rating on Samsung and SK Hynix, reasoning that HBM production occupies existing capacity, new production lines require time, and the current profit cycle in the memory industry will last longer than before. "Peak margin does not equal peak profit," so the stock price drop is a valuation cut, not a fundamentals cut.
Data from the NAND side also supports this logic. Although the 512Gb NAND trade price fell to $0.30/Gb, this is due to a severe inversion between trade prices and official factory prices. Consumer-grade NAND contract prices are actually rising—about $0.33 per Gb, up 30% to 40% quarter-over-quarter. Manufacturers prioritize advanced capacity for HBM and server DRAM, systematically squeezing supply of consumer-grade NAND.
So the current situation is: stock prices are trading on "the AI bubble is bursting," but industry data says "capacity is insufficient and structural."
🎯 Trading strategy, key points:
SanDisk: $1,400 is the current dividing line between bulls and bears. After-hours already touched $1,339; if this level does not hold on the daily close, the next support is $1,250-$1,280. But note—this round of SanDisk's decline follows the sector's valuation cut, not its own performance issues. If the storage sector sentiment stabilizes later, SanDisk's rebound elasticity will be greater than the broader market. Do not short below $1,400.
SK Hynix: Korean stock sentiment is extreme; the first trading day after a circuit breaker often has aftershocks. But Samsung Securities' "buy" rating is not given lightly; institutions are starting to turn bullish at this level. Don't cut losses in panic after the circuit breaker, nor rush to bottom fish; wait until Korean stock trading volume normalizes.
Overall judgment: The storage sector is currently "macroeconomic sentiment suppressing fundamentals." CPI and interest rate expectations are short-term dominant variables; the long-term demand logic for HBM and AI data centers remains intact. Keep positions below half and hold bullets for sentiment clearance.
📌 The above does not constitute investment advice. Leverage carries risks; don't bet on direction during circuit breakers. TSMC's revenue has already exceeded expectations, so what is the Thursday earnings call really about?
TSMC's Q3 revenue was actually announced early, about $46.7 billion, higher than the company's own forecast of $44.6 to $45.8 billion. The Thursday earnings call isn't about revenue, but about the costs behind that revenue.
closed at $485.80 on 10/6, with this year's high at $487.47; the latest available close is $472.20 on 10/8.
#DailyOrbit Bitcoin briefly dropped below $81,000 early this morning, with $935 million liquidated across the network in 24 hours, bulls accounting for $860 million.
What happened? Spot ETFs saw a net outflow of $261 million in a single day, with Fidelity's FBTC outflowing $153 million and BlackRock's IBIT outflowing $111 million. However, on-chain whales acted contrarily: large holders excluding exchanges and mining pools have increased their holdings to 3.03 million BTC, with a 30-day growth rate being the fastest since February this year, shifting from distribution to accumulation.
Why is this noteworthy? Retail investors are withdrawing from ETFs, while large on-chain funds are accumulating. Analyst Murphy pointed out that Bitcoin has tested the $81,000 level three times, with on-chain realized losses decreasing sequentially, indicating a clearing of swing chips rather than widespread panic.
Which observation does this support? It supports the "chip turnover" logic, not a trend reversal. Leveraged long positions are being liquidated, but large on-chain holders are absorbing.
Next, watch for one signal: when ETF outflows slow down, whether whale accumulation accelerates simultaneously. The day these two directions converge will be the true turning point. $BTC Bitcoin fell below $81,000, but the most important thing to watch is not the price.
What has changed? ETFs are selling, whales are buying. Bitcoin spot ETFs saw a net outflow of $261 million in one day, with Fidelity's FBTC alone outflowing $153 million. But on-chain data shows the opposite: large holders' total positions have increased to about 3.03 million BTC, and CryptoQuant confirms this is the fastest monthly accumulation rate since February.
Why is this worth attention? This is not an emotional sell-off; it’s a shift in holding structure. Retail investors and ETF funds are withdrawing, while large on-chain holders are accumulating.
Which market observation does this support? It supports "chip turnover" rather than "trend reversal." This round of decline was driven by CPI exceeding expectations and leverage liquidations, with nearly $1 billion liquidated across the market in 24 hours, $860 million of which were longs.
Next, watch for one signal: when ETF outflows slow down, does whale accumulation accelerate simultaneously? If both move in the same direction, that is the real turning point. $BTC $ETH #TSMC Q3 net profit exceeds expectations, earnings hit a record high. Three signals to watch
As the proportion of 2nm continues to rise, whether the gross margin can hold above 65% is the most direct verification point.
The cash flow and financing situation of cloud providers—no matter how full upstream orders are, if downstream payments are delayed, it will backfire.
In the mature process price war, if mainland wafer fabs continue to expand capacity, TSMC can only maintain high gross margins by relying on more advanced nodes.Nearly $1B in crypto liquidations in 24 hours. $ETH took the worst hit — $286M wiped, down 6% to $2,350, worse than $BTC $197M. Leveraged longs got destroyed after hot CPI. The twist: Santiment shows ETH sentiment 2.05 bullish while BTC sits at 0.82 bearish.
My levels:
Level 1 → PPI cool at 5:30 PM = short squeeze, ETH reclaims $2,500
Level 2 → PPI hot = $2,200 comes fast
Level 3 → BTC dominance hits 60% = alts bleed more
Everyone's bullish ETH while it dumps hardest?
#DailyOrbit The first hurdle is gross margin. The Q3 guidance is 65% to 67%, with a midpoint of 66%, which is 1.7 percentage points lower than Q2's 67.7%, due to the 2nm ramp diluting about 3 to 4 percentage points. Market consensus is around 66.5%, so holding near the upper end of the range would be good news.
#DailyOrbit 2026 is not a harvest year for the Democrats: Looking back at the 2020 election map, the Democrats have far more Senate seats to defend than the Republicans. Many swing states voted blue in the presidential election but often experience a backlash effect in the midterms.
Weight of rural votes: The design of the U.S. Senate gives red states with sparse populations disproportionate weight. Even if the Democrats lead in the nationwide popular vote, losing a few key agricultural states means they cannot secure a Senate majority.
2. The double-edged sword effect of inflation and recession
Although high oil prices and inflation anger voters, this fire may burn everyone, not just the ruling party's allies.
Economic panic leads to risk aversion: If the economy truly enters a deep recession due to a collapse in oil prices, voters tend to favor divided government for stability rather than handing full power to one party to cause turmoil.
Republicans’ fiscal conservative image: Under severe inflation, some independent voters may believe the Republicans are better at controlling government spending and curbing inflation, thus flipping key districts to the GOP and preventing a Democratic sweep.
3. Historical pattern: the midterm curse for the president’s party
Data does not support a sweep: In U.S. political history, the president’s party almost always loses seats in midterm elections. Although many think Trump is too weak this time and that could backfire, the opposition wave is usually limited.
Base consolidation: The Republican strongholds are extremely solid. Even if Trump’s approval rating is low, these voters will still turn out to block the Democrats. As long as the Republicans hold onto a few key red state seats, the Democrats’ dream of a sweep will be shattered #Tesla delivery exceeds expectations, profit recovery becomes the focus $TSLA
Delivered 486,500 vehicles, 22,000 more than it produced itself.
▪️ Q3 deliveries 486,532 units, down 2.1% year-on-year, up 1.3% quarter-on-quarter, about 25,000 units above expectations
▪️ Production during the same period 464,391 units, deliveries exceeded production by about 22,000 units
▪️ Model 3/Y deliveries 478,200 units, down 0.6% year-on-year
▪️ Cumulative deliveries for the first three quarters about 1.32 million units, Q4 needs about 311,000 units to avoid full-year negative growth
▪️ Stock price rose more than 5% on the day the delivery report was released, then gave back gains
The disagreement is not about whether expectations were exceeded, but where the "excess" came from.
The excess is over analyst expectations, not last year’s same period — year-on-year is still negative. Deliveries exceeding production by 22,000 units comes from inventory, not new orders; it needs to be replenished next quarter. What really determines profit recovery is the gross profit per vehicle.
The direction is to wait and see — Q4 deliveries exceeding 311,000 units and a quarter-on-quarter rebound in gross margin will count as recovery; relying on inventory reduction is just moving stock around. Once CPI heats up, the whole market gets hammered. ZEC, SOL, and BTC are all falling, but the reasons behind their drops differ.
$ZEC is falling the hardest. 24h -12%, current price 1096, position volume down 13.6%, 19 million liquidations are all longs. Leverage is maxed out: funding rate annualized near 10%, a sudden reversal triggers a stampede. 1,000 is the critical support.
$SOL is dragged down by funding but the fundamentals are solid. Current price 102.3, 24h -5%, ETF has been selling for 7 consecutive days. Q3 application revenue 365 million (+42.6%), on October 21 Alpenglow mainnet launch, confirmation time cut from 12.8 seconds to 150 milliseconds. Breaking 100, watch for 90.
$BTC is the most resilient. Current price 80,927, down 10 points less than ZEC, whale holdings at 3.03 million coins hit a near one-year high. But USD spot is withdrawing, ETF ran 680 million last week. Breaking 80,000, watch for 78,000.
Among these three, ZEC is leveraging to pay debts, SOL is money waiting, BTC is money not fully withdrawn yet. There's another event tonight; will the decline continue? Stay tuned. #BTC spot ETF net outflow again, can institutional buying warm up? $AKE #US CPI unexpectedly rose to 3.7%, tonight PPI faces another test. This wave of AKE is not simply a “14% increase.”
24-hour low at 0.02413, high at 0.02954, close to a one-sided rebound. In 1 hour, active buying was 32.979 million tokens, active selling was 23.4506 million tokens, with active buy volume about 1.41 times the active sell volume. Open interest simultaneously rose to 122 million tokens, but the funding rate was only 0.005%, bulls dominate but no obvious crowding.
What does this indicate?
The market has indeed strengthened, but has not lost rationality yet.
Look at two other details:
Long accounts account for 62.88%, long-short account ratio is 1.69. This data only shows a higher number of accounts, it does not represent capital strength. A small holder with 100 USDT and a large holder with 1 million USDT have completely different weights in the account ratio.
At 16:00, the basis was about -0.00001 USDT, the price spread rate -0.034%, basically close to neutral. There is no uncontrolled price spread between futures and spot, nor obvious independent futures overheating.
So the most accurate current characterization is:
The bullish rebound has been established, but the trend reversal is still awaiting confirmation. $BTC $ETH I looked into it, and the root cause of this decline is still the US inflation data exceeding expectations. The market is starting to think that the Federal Reserve will delay rate cuts, and even doesn't rule out further rate hikes. As US Treasury yields keep rising, all high-risk assets are under pressure, with US tech stocks, storage, and cryptocurrencies all dropping together.
First, let's talk about US stocks, especially the storage sector we've been watching. It had risen too much before, and the stock prices had already priced in the AI benefits in advance. Now, investors have become very picky; if earnings reports are slightly below expectations, the stock gets sold off immediately.
Take SanDisk, for example, stuck at the 1500 resistance level for a long time, with many profit-taking positions waiting to be closed. Plus, Musk's tweet supporting Micron made the market worry that SanDisk's price hikes might be too aggressive and lose customers, so bulls dare not push the price up forcefully.
In the short term, storage stocks are unlikely to rally straight back to new highs; they will probably oscillate repeatedly, bouncing on good news but getting pushed down again when macroeconomic negatives appear.
Now, about the crypto space, like ETH and BTC.
They are most sensitive to interest rates; as long as the market expects rates to stay high, funds are unwilling to stay in crypto.
Once the market goes down, leveraged positions explode in a chain reaction, causing the drop to be much sharper than stocks. Even if there is a rebound in between, it looks more like a correction, making it hard to start a new big bull market immediately.
Next, let's look at two things:
US Treasury yields and statements from US officials.
As long as Treasury yields keep rising, it's hard for risk assets to have a big rally.
Only when inflation falls and the market bets again on rate cuts will funds return on a large scale.
In summary:
We are currently in a weak sentiment consolidation phase, so don't rush to bottom-fish Yesterday the market signaled: inflation is more stubborn than expected, and capital is beginning to recalculate the valuation of risk assets. The US September CPI rose 3.7% year-on-year, higher than the expected 3.6%; core CPI rose 2.6% year-on-year, also above expectations. The Nasdaq fell 1.88%, and BTC briefly dropped below $81,000. 👇 Next, I will focus on 1️⃣ Interest rate expectations, the key to this round of decline Inflation exceeding expectations means the Federal Reserve may need to maintain high interest rates for a longer period, and a rate hike again cannot be ruled out. High interest rates will suppress tech stock valuations and reduce the willingness of capital to hold high-risk assets. However, today US Treasury yields have fallen back from their highs, indicating the market is not simply trading along the logic of "high inflation = continuous rate hikes." We still need to watch PPI and statements from Federal Reserve officials. 2️⃣ Issues in the AI sector are shifting from growth to returns This time, the impact is not only on chip stocks; some AI infrastructure companies have fallen even more. The market worries: if financing costs continue to rise, can the huge AI investments really deliver the expected returns? TSMC's performance and AI demand still provide fundamental support, but good performance does not mean the stock price will immediately rebound. The key is whether future guidance can continue to exceed market expectations. 3️⃣ BTC needs to be wary of leverage liquidations and also observe whether capital is flowing back After BTC fell below $81,000, market sentiment clearly weakened. In the latest market, BTC is still under pressure around $80,000, ETF #TSMCQ3RecordProfit https://okx.ac/ul/8Ma1r6Q
TSMC's revenue has already exceeded expectations, so what is the Thursday earnings call really about?
TSMC's Q3 revenue was actually announced early, about $46.7 billion, higher than the company's own forecast of $44.6 to $45.8 billion. The Thursday earnings call isn't about revenue, but about the costs behind that revenue.
upward, but the current ceiling. Management previously said capital expenditure would be "significantly higher" over
#TSMCOn October 13, Anthropic disclosed in an official engineering article that their internal daily code merges increased from about 2,500 times in May to over 30,000 times in September. Seven engineers spent five months developing the version control system Scion, the code review tool Antique, and several related systems, completing the migration of the main repository. Against the backdrop of accelerated AI coding, this company chose to rebuild the infrastructure that supports code. The question worth asking is: what constraints did the original system face, and why was it worth the cost to migrate the entire development process? Merging 30,000 times daily is accompanied by 25 million checkouts behind the scenes. Version control is responsible for recording code changes, supporting multi-person collaboration, and integrating individual modifications into a shared version. The faster code is generated, the more work may be needed afterward to move, check, and integrate it. Anthropic revealed that their main branch history is about 100GB, and the compressed working copy still exceeds 25GB. Meanwhile, the continuous integration system, i.e., the CI that automatically builds and tests code, performs about 25 million code checkouts daily. These two counts should not be confused. Merging incorporates changes into the shared version; checkout prepares the corresponding version of the code for a specific task. The same code may be checked out repeatedly by different tests and jobs, so the checkout volume can be much higher than the merge volume. This explains why infrastructure has become a focus: even saving a little resource each time code is prepared, multiplied by tens of millions of operations daily, can create a considerable difference. Shortening the time needed to write a piece of code does not necessarily mean the entireAt 8:30 tonight, everyone's eyes will be fixed on the same number.
But before that number lands, the funding rates have already "spoken" in advance.
📊 Comparison of funding rate data from two major platforms
Platform A (cumulative funding rates):
▪ $BTC: 0.0039%
▪ $ETH: 0.0082%
▪ $SOL: 0.0163%
Platform B (current funding rates):
▪ $BTC: 0.0035% (slightly positive)
▪ $ETH: -0.0043% (rate turned negative)
▪ $SOL: 0.0033% (slightly positive) Supply is ready, but where are the customers?
Regarding $TAO, I want to ask further: Are participants competing for network rewards, or have they already found customers willing to pay?
Bittensor incentivizes mining, validation, and other participation activities by issuing TAO and subnet tokens.
Rewards can attract people to provide services, but whether there is demand for those services still needs to be verified separately.
If a subnet offers good rewards but has very few external customers, it looks more like building supply. To improve long-term expectations, I need to see service quality, repeat usage, and payment demand gradually catching up; enthusiasm for participation cannot be directly counted as revenue.
$FIL also cannot be judged by just counting hard drives.
Storage capacity means how much business can be accepted, but how much is actually stored and how much customers are willing to pay are two different questions.
I pay special attention to renewals: the first use might be a trial, but willingness to continue paying to store data better indicates the service’s value.
Capacity expansion can be planned in advance, but if orders don’t come in over the long term, equipment and maintenance costs still have to be borne.
For $ARB, the question is how application growth affects token demand.
Arbitrum One uses ETH to pay fees, ARB plays a governance role, and increased transactions do not require users to buy ARB proportionally.
So when I see an active ecosystem, I continue to investigate: what income growth generates, and how governance arranges these resources. Network development is worthy of recognition, but reasons to hold tokens still need specific mechanisms to support them.TSMC's Q3 record profit is notable not just for the 68.7% year-on-year increase, but for the mix behind it: advanced nodes supplied 79% of wafer revenue while 2nm was still only 8%. That suggests AI demand is already monetising at scale, although the durability of that demand remains the open question.
#TSMCQ3RecordProfit During the market pullback, PYTH rose against the trend by 5.74% to touch $0.08772, entering OKX's altcoin movement list.
The market dropped 4.85% in a single day, but PYTH on OKX rose 5.74% against the trend, reaching $0.08772. If you hold spot, there's no need to rush into contracts. I checked overseas news and reports this afternoon; Bloomberg revealed that mainstream derivatives positions shrank by $14 billion, but RWA perpetual positions linked to stocks and gold rose to $9 billion. Pyth accounts for 94.1% of this type of data pricing. Plus, the community previously decided to use all external income to buy back tokens, so short-term buying surged today on this news.
Switching to the OKX market page to check the market. BTC dropped 3.35% to $81,271.3, ETH fell 6.21% to $2,369.11. OKX perpetual total positions stand at $7.673 billion, with BTC holding $3.096 billion, altcoin contracts at $2.878 billion, and the altcoin-to-BTC position ratio at 0.929. The fear and greed index is neutral at 53. The total network market cap fell 4.85% to $2.73 trillion, with BTC market dominance rising to 59.52%. PYTH recorded a trading volume of 14.1 million USDT, making it one of the few altcoins moving green against the trend today. TSMC announced its Q3 financial report and held an earnings call today
but the revenue part
was actually already known by the market.
September revenue was 511.86 billion yuan,
a 54.6% increase compared to the same period last year;
Q3 total revenue was about 1.49 trillion yuan,
setting a new historical high,
and exceeding the upper range of the company’s own financial forecast given in July.
Cumulative revenue for the first three quarters grew by over 40%.
So the real focus today is not the numbers already announced,
but three things that have not yet been disclosed:
Q4 financial forecast
Whether the momentum of AI demand can continue until the end of the year.
Gross margin guidance
Whether the cost dilution in the early stage of advanced processes
will converge as expected.
Next year’s capital expenditure plan
This is the most important.
Capital expenditure is a bet on demand for the next two to three years.
It does not represent TSMC’s own view,
but what it sees from customer orders.
The financial report talks about the last quarter,
capital expenditure talks about the next cycle.#PalantirQ3Valuation TSMC's third-quarter sales exceeded the original guidance, and the fourth quarter is expected to continue growing. I am more inclined to believe that US stock TSM can rebound in the coming week. The revenue scale is already so large and still growing, which makes me happy. The third-quarter revenue announced today is $46.72 billion, exceeding the company's original guidance upper limit of $45.8 billion. The fourth quarter is expected to be between $52.3 billion and $54.3 billion, with a midpoint of $53.3 billion, about 14% more than the third quarter. This is still the company's estimate; actual revenue depends on subsequent deliveries. Before the earnings report, US stock TSM had already retreated to around $430. Today's earnings report has not yet been tested in the regular US trading session. At 8:30 tonight, the US September PPI will be released. If the data makes the market more worried that interest rates won't come down, it could also weigh on tech stocks. If the stock continues to fall after the earnings report, I will have to reconsider my judgment about a rebound this week. The company said that strong demand for advanced processes will still support the fourth quarter, and 2nm will continue to ramp up. In the third quarter, 2nm already accounted for 8% of wafer revenue, up from 3% last quarter, with new processes starting to ship more. However, the gross margin actually dropped from 67.7% last quarter to 66.4% in the third quarter. The company explained that this was mainly due to the initial ramp-up of 2nm, so a drop in gross margin alone should not be taken as a sign of weakening demand. The fourth-quarter gross margin guidance is 65% to 67%, still below last quarter's 67.7%. Revenue is going up, but gross margin hasn't caught up yet, which is indeed a bit unfortunate. #TSMC #USStocks #EarningsReport