Orbit Post Sitemap
#台积电Q3净利润超预期，盈利创历史新高
Brothers, TSMC's Q3 financial report just came out. As usual, don't just watch the hype, look at the essentials.
Single-quarter net profit reached 763 billion New Taiwan Dollars, a year-on-year surge of 68.7%, directly surpassing the market expectation of 746.3 billion, marking the sixth consecutive quarter of record-breaking results. The 3nm process accounted for 33% of wafer revenue, and the 2nm process, just starting mass production, contributed 8%. This ramp-up speed is much faster than expected. Advanced processes make up 79% of total revenue. In plain terms, global AI computing power orders are pouring crazily into TSMC's production lines.
What does this have to do with our crypto circle? A lot. Recently, Bitcoin dropped to $80,230 due to CPI data missing expectations, triggering $484 million in liquidations across the network within an hour, with long positions accounting for $464 million. But on the other hand, Bitcoin mining companies are accelerating the shift of their computing power contracts towards AI and HPC, with a cumulative contract scale of about $100 billion. Miners are no longer mining coins but selling computing power to AI companies; Genesis Digital Assets, IREN, and others are making this transition. This is not a small trend.
On the funding side, Wintermute's recent analysis also confirms this—when mainstream coins are under pressure, money hasn't left the market but moved from DeFi to AI-related tokens, with NEAR considered a typical beneficiary. AI and crypto now share the same computing power resources: Nvidia needs chips, TSMC manufactures chips, miners supply power and facilities, and the tokens we trade act as intermediaries for capital rotation.The battle to defend 80,000 has begun, and 4.8 billion has already withdrawn 😱
Looking at BTC, 81,000 didn't hold, and it broke through directly this afternoon.
ETH (-5.94%), "If 2400 breaks, watch 2350"—just one breath away.
SOL (-4.83%), all three brothers are in the red today, each worse than the last.
Money matters are the most telling.
In 24 hours, BTC contracts saw 1.849 billion exit, spot saw 452 million exit, totaling 2.3 billion.
ETH is even worse, contracts 1.973 billion, spot 570 million, totaling 2.543 billion.
Together, the two brothers total over 4.8 billion, and "running first as a courtesy" has become the mainstream.
But the last hour is interesting.
BTC contracts had a net inflow of 21.07 million in 1 hour, ETH contracts net inflow of 18.01 million.
Regarding fees, BTC is at 0.0075%, higher than at noon. After breaking 81,000, bulls are still paying premiums to hold positions. ETH fees at 0.0022%, just a step away from turning negative.
My view is just two points.
① It's normal to run money before the announcement; with CPI leaning hawkish and tonight's 20:30 event pending, everyone fears a rate hike repricing.
② The 1-hour inflow is the first batch betting on "bad news fully priced in," with light positions.
If PPI really exceeds expectations, these people will be tomorrow's liquidations 📉
Right now, I'm only watching one number: the capital flow 1 hour after the announcement.
👉 If it turns positive, it means bottom-fishers are adding positions, and I'll watch for a rebound.
👉 If it turns negative, even the bottom-fishers are wrong, so 80,000 will likely be seen 👀 #TSMCQ3RecordProfit
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.Over the past three days, XAUT has been rising, while xSKHY and xSNDK have been falling. This is not mysticism; it's the same money moving between two ends.
$XAUT is rising because safe-haven money is coming back. On the morning of October 15, gold prices broke through 4120~4150 consecutively, rising over 1% intraday, with XAUT following up to 4140. The US dollar index dropped by 0.42%, US Treasury yields fell, lowering the cost of holding gold; central banks purchased 63 tons of gold in September, the highest in nearly two years. But it is still 25% below the January ATH of 5505. This is a deep correction rebound, not a new high. 4100 is the bottom of this round.
$xSKHY is falling because the most crowded position in AI is deflating a bubble. ADR is reported at 138~139, down 9.5%~11% in 24 hours, KOSPI dropped 7%, triggering a sidecar. Four straws: Changxin 4F² breakthrough, Toshiba capacity expansion, HBM4 equipment delay, and $40 billion buyback ended early. Foreign investors have sold for 7 consecutive days. Demand is not bad; what’s bad is the valuation.
$xSNDK is falling with the same story, and SanDisk is even worse. Down 10.8% in a week, from the September 22 high of 1909 to the October 13 low of 1401, a 27% retracement. Even earnings beating expectations didn’t stop the sell-off.
XAUT’s rise means money is looking for a place to hide, while the other two fall because money is withdrawing from the most crowded spots. The industry logic hasn’t collapsed; what collapsed are valuations and positions. #GoldETFBigInflow, how safe-haven funds are reallocating TSMC’s revenue has already surpassed expectations, so what should investors focus on during Thursday’s earnings call?
The company previously reported Q3 revenue of around $46.7 billion, beating its own forecast of $44.6 billion to $45.8 billion. With revenue figures already out, the real focus of Thursday’s call shifts to the cost side of the business, profit margins, and how efficiently TSMC is turning strong demand into earnings.
#DailyOrbit #苹果新CEO迎来iPhone18首份成绩单 这份成绩单争议不小。第一，砍掉标准版，等于主动放弃走量市场，依赖高端，需求结构变窄；第二，售价抬升，普通用户感知升级有限，渠道已经出现分化，Pro部分机型破发，只有Max和折叠保持溢价；第三，折叠屏iPhone Duo成本高、定价贵，属于长期布局，短期很难贡献销量。CPI came out overnight, and something strange happened.
The US Treasury yield surged to 5.33% last night, then dropped straight back to 5.11% this morning. Money got cheaper, so BTC should have caught a breather.
But BTC didn’t rise; it kept falling, now at 81,000.
The key is how the US Treasury yield dropped.
Inflation fell, everyone thought no more rate hikes, so Treasuries went down, which is good—money dares to come out and play, and BTC follows up.
Today is not like that. Today the stock market crashed, and everyone got scared and fled to Treasuries. When money floods in, yields get pushed down.
Nasdaq fell 1.9% last night, the worst day since July. Asia continued to crash today; the South Korean stock market fell up to 8% intraday, Samsung and SK Hynix both dropped over 10%. The AI sector got smashed badly.
AI companies build data centers, mostly borrowing money. When rates rise, borrowing gets more expensive, and they are the first to buckle.
This time BTC wasn’t treated as a safe haven but sold off along with tech stocks. Those mining companies switching to AI computing power fell up to 13% last night.
So today’s drop in Treasuries is not good news for BTC; it shows the market is scared.
Oil hasn’t eased either. Brent is still around $105, and fighting broke out again near the Strait of Hormuz. The International Energy Agency is meeting in Paris today to discuss whether to release another 100 million barrels from emergency reserves. If they do, oil prices could ease a bit; that’s the only good news we can hope for right now.
I said before that if US Treasury yields fall back below 5.2%, it’s a signal for BTC to go up. The timing is right; BTC first looks at 83,200, then ETH at 2,410! $BTC $ETH US inflation data exceeded expectations, and Bitcoin directly fell below 81,000, with billions of dollars in liquidations hitting the market. AVAX, SEI, and INJ all followed the drop; none of these projects had internal issues. APT rose 51% over four weeks, but on-chain stablecoins shrank by 17.5%. The rally relies on sentiment rather than capital, and this structure won't hold for long. With macro pressure, rebounds are just brief escape windows.
I just opened the security booth window for some fresh air. A car came in, registered, and then I continued watching the market.
WAL current price is 0.0582, stuck above a previous dense zone. MACD shows a bearish crossover below zero, RSI peaked and is falling back, indicating weak short-term momentum. There is dense liquidation around 0.0571, which is a key support; breaking it would trigger a chain liquidation. Above, liquidation pressure at 0.06 is significantly reduced, and selling pressure is lighter. Short-term mainly sideways movement, but support below is clearer than resistance above.
Trading strategy: mainly short positions. Enter in batches between 0.0585 and 0.0592, take profit first target at 0.0572, second target at 0.0558. Stop loss at 0.0605; if price rises above this, admit the mistake and exit. Avoid long positions; consider catching a rebound only if 0.0571 truly holds, and keep position size light. In this market, staying alive is more important than making quick profits.
$WAL
#Tesla delivery exceeds expectations, profit recovery becomes the focus
@OKX Planet 上方空头像三块磁铁，但磁铁不一定吸得住价格 🌙 真要冲上去，是爆空行情还是又一次假动作？ 刚盯完12小时清算图，心里有种熟悉的小紧张。BTC、ETH、SOL上方都堆着空头仓位，像三根被拉到极限的橡皮筋。 - BTC：84375.7附近，约3218万U清算，距现价+0.74% - ETH：2545.16附近，约1398万U清算，距现价+1.3% - SOL：108.85附近，约200万U清算，距现价+1.03% 数字看着都不远，对吧？但跨市场联动这个镜头下，我更在意的是三件事。 第一，这三个触发点几乎挤在同一个窄幅区间里。意味着一旦风险偏好回暖，BTC带头往上戳，ETH和SOL的清算会形成共振，短线买盘会被动放大，山寨情绪也会跟着松一口气。这是偏多的路径。 第二，但共振也可能是反向的。如果BTC先假突破再收回，ETH和SOL上方的挤压就会变成诱多陷阱，追高的人反而成为下一批燃料。跨市场同步拉升需要的是一致预期，而现在宏观数据没落地，这种一致性很脆。 第三，我真正盯的是资金费率有没有同步转正、未平仓量有没有在拉升中继续堆。如果价格冲、费率不热、OI不增，那多半是脉冲，不是趋势。 偏空的风长鑫说“每片晶圆裸片数量提升至少50%”，先别替它改写成成本下降50%。
我翻了公司G5平台的原始公告，脚注写得很清楚：比较对象是自家第四代技术平台，裸片数量按8Gb颗粒基准换算。它提供了一个具体的工艺进步指标，不是与所有竞争对手成本对比的结论。
每片晶圆能容纳更多裸片，确实有利于分摊成本。但最终有多少合格产品能卖出去，还取决于良率、加工费用与封装测试。裸片数量增加，和每颗可销售产品的成本降低，中间还有一段生产过程，不能把百分比直接搬过去。
这轮存储股波动里，两种情绪都让我觉得太急。看到技术进展，就认定老厂商利润要被清空；或是觉得新玩家短期出货有限，完全不用在意。竞争影响可能先体现在客户议价。只要买家开始认真比较替代方案，哪怕份额还没大幅变化，报价谈判也可能变难。
对长鑫的进步，我愿意认真看，也希望讨论能落在同一个比较基准上。后面更想看到可销售产品的良率和客户交付证据。技术公告里连脚注都没读完，就给全球存储利润重新排座次，实在有点太着急。
#闪迪美光下跌，长鑫技术进展受关注 #TSMCQ3RecordProfit
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 $45.8 billion. The Thursday earnings call isn't about revenue, but about the costs behind that revenue.$SNDK SanDisk, Hynix, Samsung, and all those messy storage chip makers, just short and be done with it.
Why are these chip makers definitely going to fall?
The main reason is that these companies keep shouting every day,
"Oh~ capacity is not enough, inventory is insufficient."
Let's put it this way, if this thing could make money in the long run, these companies would have already expanded capacity and started competing on who sells more.
Anyone with money who doesn't make money is a fool, everyone understands that.
Now prices are skyrocketing, but when it comes to expansion, not a single word comes out. Obviously, these manufacturers or the entire industry have a consensus.
This thing could crash at any time. Do you really expect AI to become evergreen? The algorithms in the US are indeed powerful, but actual applications are basically zero, and the popularity and utilization are far lower than in China, unless there is another breakthrough in high-end algorithms.
Otherwise, 95% of users are using the free version, just asking a question and that's it, at most letting it help you edit something, produce a copy, or fix a few pictures. The very few who do AI videos, how much computing power can they really use?
The market is basically saturated now. Without any new technological breakthroughs, the market will head south at any time.
It's similar to the mining crash in 2022, but back then these manufacturers weren't so smart. They expanded mining capacity wildly, ending up with a huge pile of inventory. After the mining crash, they were stuck with it, and the market went straight down. The annual flash memory price drop exceeded 40%.
Now they play it smart. When the market is good, they raise prices aggressively but don't increase capacity. When the market is saturated or crashes, they shrink back like turtles, waiting for the next technological breakthrough to raise prices again. So essentially, a drop is guaranteed.#PalantirQ3Valuation #美国CPI超预期升至3.7%，今晚PPI再迎考验
Last night, as the US inflation data was released, Bitcoin immediately plunged from its high, causing tens of thousands of contracts to liquidate instantly! 😱
Right now, the market is focused on tonight's production-side data. CPI reflects whether everyday goods are getting more expensive, while PPI shows if factory production costs continue to rise.
If factory costs also rise tonight, US Treasury yields will likely struggle to fall, and Bitcoin at the 80,000 level will become the last key support.
Conversely, if the data cools down, the market will have reason to believe this was just caused by oil prices, and funds might boldly flow back in.
Currently, $BTC has dropped from its high to around 80,000, with the market repeatedly testing this critical support level. The price is stuck in the 80,000–81,000 range; if it breaks down, the next stop could be 78,000.
After tonight's data release, the direction will become clear. Whether it holds or breaks depends on how strong the PPI is.
$ETH $SOL
#台积电Q3净利润超预期，盈利创历史新高
#奈飞增长或放缓，广告业务成新看点 To be honest, in the short term, the market will most likely be a frustrating period of volatility rather than a sudden direct reversal and surge. $BTC $ETH $SNDK
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 does not rule out further rate hikes. As US Treasury yields continue to rise, all high-risk assets come under pressure, with US tech stocks, storage, and cryptocurrencies all dropping together.
Let's first talk about US stocks, especially the storage sector we've been watching.
It had risen too much before, and the stock price had already priced in the AI benefits in advance. Now, investors have become very picky; if earnings reports are slightly below expectations, the stock is immediately sold off.
Like SanDisk, stuck at the 1500 resistance level for a long time, with many profit-taking positions waiting to be cashed out. Plus, Musk’s tweet supporting Micron made the market worry that SanDisk’s price increase might be too aggressive and lose customers, so bulls dare not push it up hard.
In the short term, storage stocks are unlikely to rally straight back to new highs; they will most likely fluctuate repeatedly, bouncing on good news but being pushed back down by macroeconomic negatives.
Now, let's talk about the crypto space, like ETH and BTC.
They are most sensitive to interest rates. As long as the market expects rates to remain high, funds are unwilling to stay in crypto.
Once the market goes down, leveraged positions will cascade liquidate, 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 difficult to immediately start a new bull market.
Going forward, focus on two things:
US Treasury yields and statements from US officials.
As long as Treasury yields continue to rise, it will be very difficult for risk assets to have a big rally.Bitcoin miners are about to face a difficulty adjustment.
Around October 16, the difficulty is expected to increase by 5.68%.
Sounds like not much.
But it has a more direct consequence: with the same hash power, the amount of money mined will decrease by 5.37%.
Here’s how the 5.37% comes about.
There’s something called hashprice, which means how much you can earn per unit of hash power per day.
Right now, that number is 39.77 USD.
When difficulty rises, the same machine mining for the same day will get fewer coins.
Income drops first.
The key is this number hasn’t deducted electricity costs yet.
So whether mining is truly profitable depends on two things: how much you earn and how much electricity costs.
Old miners with cheap electricity can still hold on.
New miners with expensive electricity are in danger.
Here’s a calculation:
For machines with efficiency of 20 joules per terahash, at 8 cents per kWh electricity price, electricity costs alone wipe out profits.
In other words, it’s not that mining less causes losses, but that electricity costs exceed mining revenue.
So why is the difficulty suddenly adjusting?
Bitcoin automatically adjusts difficulty every 2016 blocks, aiming for an average of one block every ten minutes.
Because total hash power is increasing, blocks are being found faster, so difficulty must rise accordingly.
That’s also why it’s "expected" now, not "confirmed."
There are still over 300 blocks left before the adjustment; during this time, price, hash power, and costs may change.
The real watershed isn’t difficulty, it’s electricity price.
Miners with cheap electricity are expanding their machines.
Miners with expensive electricity are already shutting down#TSMC’s record-breaking revenue was already on the market’s radar before October 15, so presenting it as a fresh bullish surprise on earnings day may be missing the bigger picture.
On October 8, the company reported that September revenue jumped 54.6% year-over-year, while slipping 0.6% compared with the previous month. 📊
The takeaway? Strong annual growth doesn’t necessarily mean consistent month-to-month gains. Even in a booming AI chip market, short-term fluctuations.
#DailyOrbit Record numbers from TSMC, but I think the real story is what comes next. 🧐
Revenue was already disclosed month by month, so the surprise is limited. What matters now is the margin guidance (65%-67% vs 67.7% in Q2) and how much TSMC will spend on capex.
Still, with demand for AI chips this strong, I see the long-term picture as solid. Short term, the market may stay cautious. 📊
#TSMCQ3RecordProfit $TSM
Just my opinion, not financial advice.#奈飞增长或放缓，广告业务成新看点 The 81000 level is where others caught it, not where it held on its own.
$BTC dropped back near 81000 again early this morning.
This is the second time in two days testing this price.
What others think: holding here means a double bottom, looking back to 84000.
If it doesn't hold, then look at 80000, the anchor for the whole market is here.
How this number is calculated: 81000 was the entry point for funds during the last sharp drop.
The second test means the first batch of catchers are already at a floating loss.
The second time is more fragile than the first because fewer people dare to catch it.
$OKB rose 2.4%, money is moving into platform tokens, seeking stability.
$STRK dropped 10% again, those chasing highs are still stuck at the peak.
Before the direction emerges, the first movers get hit first.
#BTC spot ETF sees net outflow again, can institutional buying recover?
#$BTC $OKB # TSMC Q3 Net Profit Exceeds Expectations, Profit Hits Record High
TSMC's Q3 financial report is out, with net profit far exceeding market expectations and reaching a historic high. The core driving force is the explosive demand for AI computing power chips.
✅ Key Data: Net profit of 763.06 billion TWD, a year-on-year increase of 68.7%; revenue of 1.49 trillion TWD, a year-on-year growth of 50.9%, gross margin of 66.4%, net margin surpassing 51%, with profits rising for multiple consecutive quarters. Advanced processes (7nm and below) form the revenue base, with HPC high-performance computing business continuously contributing the largest income.
💡 My thoughts:
1. AI is the core of this growth cycle. AI large models drive HPC chip orders, with 3nm and 5nm high-end process capacity in short supply. High-margin advanced processes elevate overall profitability. However, growth is highly tied to the AI industry chain; if downstream customers slow capital expenditures, it will directly impact future orders.
2. High capital investment is a double-edged sword. To meet AI orders, capacity expansion continues with capital expenditures remaining high. The heavy asset model means significant future depreciation pressure.
3. The competitive landscape remains strong, but external variables are increasing. Advanced process barriers remain difficult to shake in the short term, but geopolitical issues, customer diversification, and local manufacturing costs are uncertainties to face in the medium to long term.
Currently, AI dividends are being realized with impressive performance, but high growth is not permanent. Going forward, focus will be on tracking AI capital expenditures and the pace of advanced process capacity release. TSMC's revenue record was already known to the market before October 15. Treating the same fact as a sudden positive surprise on the day of the earnings report might be a step behind the rhythm.
The company announced on October 8 that September revenue grew 54.6% year-over-year but declined 0.6% month-over-month. This contrast is quite interesting: strong long-term growth does not mean every month is higher than the previous one. Small monthly fluctuations.Let's talk about some thoughts and plans.
In the past few months, I've only been buying the dip and adding some long positions, without other operations. Mostly just enduring the market. The current market is mostly about endurance, and it will probably be like this in the future as well...
It's hard to make flashy moves. For ordinary people, this is the only way to actually make some money—above spot, below futures... Currently, I think this is just a small-level top.
If it pulls back to around 7.6 (if it doesn't give a chance, I'll just hold what I have), I might add some positions. Overall, I still see 9.5-10.
This is also roughly the position of the B-wave rebound. Overall, I still expect a large-level consolidation... Mainly, the index has been doing an internal wave 4 since April last year until now.
Personally, I think after the midterm elections, there will be a small rally of about 10-20 points. After that, there will be a mid-term major correction of 20-30 points (not ruling out that the correction might start directly here). So the risk is quite high... If the final expected wave comes out later, I will consider clearing my positions around 9.5-10... I think the future will be a long period of large consolidation. The big opportunity will have to wait until the Nasdaq pulls back 20-30 points... Only then will it be the chance to heavily invest in the index and various assets.Term Structure Radar
$BTC marked positive basis shows a significant shrinkage in the top bid-ask spread. The near-far basis is +2.27%, and the top bid-ask spread for near buy and far sell is +1.21%.
$ETH top positive basis scale is very small. The near buy and far sell spread is +0.6%, with the top pair scale about $198; at a scale of 10,000, this only covers about 2%.Now panic sentiment has completely spread across the Asia-Pacific market, with the KOSPI Korea Composite Index's decline further expanding, intraday drop reaching 9%, heavy selling pressure on the board, and market risk aversion sentiment sharply rising. Leading memory chip maker SK Hynix plunged nearly 12%, semiconductor sector becoming the hardest hit area, with a large amount of high-position long positions severely impacted—it's unclear how many bulls will be buried inside again 😇. The collective sell-off in Asia-Pacific stock markets is just the appetizer; tonight's U.S. stock market opening is highly uncertain, and the outlook is hardly optimistic. Additionally, two key data releases are scheduled for the evening: the U.S. weekly initial jobless claims and the PPI Producer Price Index. The continuous release of inflation chain data will directly amplify market-wide volatility. If the PPI rises again, it will further strengthen expectations for Federal Reserve rate hikes, and risk assets will continue to face pressure. Currently, global market logic is interconnected, with strong linkage among Korean and Japanese stock markets, U.S. tech stocks, and the crypto market. Korean retail investors hold heavy positions in both chip stocks and cryptocurrencies; after stock market liquidations, some funds will be forced to reduce crypto holdings to cover margin calls, indirectly bringing selling pressure to BTC and ETH. The crypto space, already fragile after the CPI negative impact, is vulnerable to further external negative factors, which can easily trigger contract liquidations again. Going forward, focus on two points: first, whether the tech and semiconductor sectors can stop the collapse after the U.S. market opens; second, the quality of the PPI data. At this stage, macro factors completely dominate the market, uncertainty is at its peak, so avoid bottom-fishing or gambling on rebounds. Leverage positions must be tightened, prepare in advance for sudden spikes, and avoid holding positions betting on direction. #美国CPIThe market is still paying back CPI, but RSR has been rallying for the third consecutive day. How many more bites can Upbit's bowl still offer?
Bitcoin is down 2.7% at 81,000, ETH down 2.7%, yet RSR (Reserve Rights) bucks the trend with a +10% surge to 0.00218, becoming the most eye-catching gainer on the leaderboard aside from ZK.
But to be clear, this isn’t a new story today. The real trigger was on October 12 when Upbit opened KRW and USDT trading pairs for it. Korean retail investors rushed in, pushing the price up by 50 to 60 points in a single candle that day; combined with the September community vote that changed the performance unlocking and burning mechanism, tightening supply expectations. Today’s +10% feels more like a rebound on the third day after listing. I didn’t see any new positive news after checking around.
My judgment: The first bite of meat on a new exchange listing is the most delicious; chasing on the third day usually just props up the first two batches. OKX’s 24-hour trading volume is only about 1.4 million USDT, with the main battlefield actually on Upbit’s KRW market. OKX is just a thin market following the rally. RSI is already at 67, price hugging the upper Bollinger Band; it’s easy to pump and dump, and slippage and spikes can bite.
Watch for two signals: whether it can volume-wise hold above the previous high of 0.0022 and if the Korean market volume doesn’t collapse; once Korean won buying power recedes and volume shrinks on the rally, it’s basically a short-term profit-taking window. DYOR, not investment advice.
#RSR #stablecoin #Upbit #Tesla delivery beats expectations, profit recovery becomes focus #Netflix growth may slow, advertising business becomes new highlight # Netflix Growth May Slow, Advertising Business Becomes New Focus
Three key variables truly determine the ceiling of the advertising business
1. Whether advertising ARPU can catch up with subscription ARPU
Currently, the per-user revenue of ad-supported plans is still lower than that of ad-free plans. The gap may only narrow if advertiser budgets, programmatic buying, and live CPM increase.
2. Whether live and highly engaging content can be continuously supplied
Live content like NFL, WWE, MLB has higher ad monetization efficiency and can boost user engagement, but it is costly and concentrated in scheduling, requiring proof of sustainability.
3. Whether AI advertising systems can improve delivery efficiency
Netflix is advancing AI-driven ad planning, creative adaptation, audience insights, and data clean rooms, which relates to whether advertisers are willing to increase budgets long-term.
Investment Judgment
Netflix's advertising business is worth tracking, but one should not focus solely on the slogan of "doubling ad revenue." More importantly, watch: ad plan ARPU, advertiser retention, live ad fill rate, programmatic share, and whether profit margins are diluted by content costs.
If these indicators improve, the ad business will shift from a "story" to "cash flow"; if not, it will only continue to support valuation rather than become a performance inflection point. #TSMCQ3RecordProfit TSMC's revenue record was already known to the market before October 15. Treating the same fact as a sudden positive surprise on the day of the earnings report might be a step behind the rhythm.
The company announced on October 8 that September revenue grew 54.6% year-over-year but declined 0.6% month-over-month. This contrast is quite interesting: strong long-term growth does not mean every month is higher than the previous one. Small monthly 💫20261015
1. Fed rate hike bets double
1. CPI September 3.7%, August 3.4%; Core September 2.6%, August 2.4%, both numbers slightly higher than expected; gasoline rose the most
2. Before the CPI data release, the market bet about 20% chance of a rate hike, after release it rose to 47%
3. BTC intraday low on October 14 was 81588.88, 30-day increase 4.27%; monthly still intact, intraday already broken
4. BTC pays no interest, 2-year US Treasury yield 4.87%, market bets split 50/50
2. ETH zero-knowledge proof speed race
1. Zksync has continuously processed 7200 mainnet blocks, 99% of proofs completed within 10s, one block every 12s; no proof is slower than block production
2. Selling point for banks: even if servers are hacked, no invalid ledger update can pass verification
3. Whose proof system first achieves formal verification
3. ETH down 4.45% this week, currently around 2500
1. Bought ETH 9 years ago at $50, spent $370k USD, now worth 19,260K; unstaked and deposited into Kraken.
2. Three signals:
(1) Coins moving to exchanges
(2) Whales going off-exchange: an OTC whale bought nearly 10,000 coins, cashed out 25,000k
(3) US buyers retreating: comparing the same coin price on exchange C$CRCL Don't rush to copy yet.
From a fundamental perspective, since October 4th, USDC supply has dropped from 75.4 billion to 72.67 billion, approaching 71.75 billion, putting Circle's fundamentals under pressure. The technical outlook is also bearish, with a high probability of continued decline.
There are two possible trends: support between 58-66, possibly stabilizing. If it breaks 58, then look at 50, or even 45-50. Although the second scenario is more likely, it's not guaranteed that 58 will be broken.
Operation is divided into two parts 3:7: buy low between 58-66, place a regular limit order near 50, and adjust later if USDC or the trend changes.#美国CPI超预期升至3.7%，今晚PPI再迎考验
US September CPI exceeded expectations, doubling the probability of a rate hike in October overnight: Tonight's PPI is the key variable!
US September CPI rose year-on-year to 3.7% (previous 3.4%, expected 3.6%), core CPI year-on-year 2.6% (expected 2.5%), month-on-month 0.6% (previous 0.4%). Inflation pressure is no longer limited to energy but has spread to entertainment, used cars, healthcare, education, clothing, and other sectors.
Market pricing for tightening has surged sharply: the bet on an October rate hike jumped from 21.6% to 43.7%, nearly doubling; under CME metrics, the probability of a rate hike before year-end has risen to over 96%. Risk assets are collectively under pressure, US stocks sold off, Nasdaq fell 1.88%, marking the largest single-day drop since July, Philadelphia Semiconductor Index plunged 4.3%, the US dollar strengthened, and gold plunged.
Tonight's focus: US September PPI. The market consensus expects a month-on-month increase of 0.5%, core up 0.3%. If PPI exceeds expectations, it will further confirm the stickiness of inflation and heighten rate hike concerns; if it moderately falls back, it could provide the market a breathing space.
In short: CPI has put the October rate hike back on the table, and tonight's PPI will determine whether to add fuel to the fire or temporarily cool down. #TSMC Q3 net profit exceeds expectations, profits hit a record high
I am Script Brother, and this TSMC financial report is like a strong shot of confidence for the AI industry.
Revenue grew 50.9%, net profit grew 68.7%.
This speed of making money even makes a money printer call it big brother.
But what I care about more is not the profit, but the underlying industry logic.
Is AI really a bubble?
My view is that AI valuations may have a bubble, but the real demand is definitely not a bubble.
Why?
Because tech giants can boost stock prices by storytelling, but TSMC must make money through solid orders.
If chips don’t sell, the financial report won’t play along with you.
What’s more interesting is that the AI industry chain is forming a huge wealth distribution structure.
Upstream chip sellers are making a fortune, while downstream AI application developers are still burning money desperately.
It’s like the gold rush back in the day.
The gold miners may not make money, but the shovel sellers have already gotten rich.
However, TSMC making money doesn’t mean all AI companies can make money.
The real future dividing line is not who has the most computing power, but who can turn computing power into profit.
It’s the same for the crypto world.
AI concept coins are loudly hyped, but very few projects actually generate cash flow.
So don’t blindly rush just because you see the two letters AI.
Stories can be hyped for a while, but profits can sustain a lifetime.
What do you think, will the biggest winner in this AI round be the chip sellers or the application makers?
Discuss in the comments. $BTC $ETH $ZEC There is a trader called Loracle who shorted almost the entire market.
He holds more than a dozen US stock contracts, all shorts:
NVIDIA, Oracle, SanDisk, Micron... all the names that have surged the most in this AI rally.
What he shorts is not a specific coin, but the "AI narrative." And so far, he is right.
But at the same time, he went long on Bitcoin, the only long position, which is $BTC .
There are also a bunch of altcoins, all short positions.
The result is:
Almost all the shorts are making money. And the only long position lost more than half of all the profits from his shorts.#Tesla's delivery exceeds expectations, making profit recovery the focus. Deliveries are good news in terms of "volume," but the financial report is the real "test" of profitability. On October 2, Q3 deliveries reached 486,532 vehicles, 5.3% higher than the company's compiled consensus (about 462,000), with the stock price rising 4.65% in a single day; however, production was 464,391 vehicles, and deliveries exceeded production by about 22,000 vehicles, indicating that part of the deliveries relied on inventory/in-transit vehicles rather than a pure surge in new orders.I think Binance's Ethereum deposit and withdrawal maintenance is just a precaution, trading continues as usual during the inspection. ETH spot is about 2365 USDT (Beijing 14:42, OKX).#Netflix growth may slow, advertising business becomes a new highlight The advertising business is currently just a growth highlight, not the main source of income — this is the easiest point to misjudge when looking at this issue.$SOL : Positive News Doesn’t Guarantee a Pump!
Solana’s Shanghai developer event and CZ’s public support are positive for ecosystem sentiment, but they’re not signals of a major price rally.
A short-term hype-driven bounce is possible, especially in Solana meme coins, but “buy the rumor, sell the news” could trigger profit-taking once the event begins.
The bigger picture still depends on macro conditions, US Treasury yields, and Fed rate-cut expectations.
#KOSPIDropsNearly8% 大盘在给CPI还债，资金却把一个"DAO都关了链"的老游戏币拉翻倍，这是闹哪样？
比特币趴8.1万、OKB跌5个点，热钱不去主流，反而扎堆MAGIC这种老牌游戏币——今天又+9.6%冲到0.1177，十月初还在7分出头，一周多翻倍还多，最高摸过0.12。
先泼盆冷水：这不是GameFi第二春。MAGIC背后的Treasure这一年基本在收缩，连自家链都交给DAO停了，代币九成都已流通，没有新合作也没有链上数据撑着。这波更像AI-agent游戏叙事点火后，薄盘里的杠杆逼空加高换手赌博，海外分析师干脆点破这就是个"杠杆故事"。
盘面也诚实：欧易24小时成交额才250万USDT上下，典型小盘薄口，价格正死压在0.1179压力线和MA5下面，今天一根针冲0.12又被按回来，易拉也易砸；头上还悬着一笔千万级代币解锁的抛压预期，逼空一旦没了增量，兑现常是一根直线。
盯两个信号：能不能放量站稳0.12前高，还是在压力位反复插针缩量，后者基本就是尾段。别用追大饼的逻辑接飞刀。DYOR，非投资建议。
#MAGIC #GameFi #TreasureKORU is currently in a contradictory state of "long-term fundamental logic intact + short-term price structure severely damaged." My view is that this position is more suitable for observation or waiting for a right-side signal rather than attempting to bottom-fish.
Fundamentals: The long-term narrative remains but is overshadowed by short-term panic
The underlying driver of the South Korean economy—AI-related semiconductor exports and equipment investment—remains solid. The Korea Development Institute (KDI) October report clearly points out that semiconductor and computer exports maintain high growth, and AI infrastructure-related categories (steel, non-ferrous metals, etc.) exports also maintain a good trend. Both the IMF and AMRO have raised South Korea's 2026 growth forecast to 3.3%.
Currently not touching it, but it is on the watchlist. There are three reasons:
First, triple leverage will continue to "decay" in a sideways and volatile market; even if the KOSPI remains flat, KORU's value will be eroded. The current panic sentiment in the Korean stock market has not been fully released, and the probability of sideways or slow decline is not low.
Second, the liquidity depth on OKX is insufficient to support high-leverage positions. A $6.5 million position size means large liquidations could trigger a chain reaction, and spread costs and slippage will significantly eat into profits.
Third, the technicals need to be rebuilt after the stock split. The 20-for-1 split changed the price coordinates, so previous support and resistance references have limited value; the market needs time to form a new trading range.
If you must participate, wait for a clear stabilization signal from the South Korean KOSPI The countdown will continue, but the market may not follow accordingly
$LTC's next halving is expected in July 2027, with the block reward dropping from 6.25 to 3.125 coins.
This event has clear rules, but there is no fixed timeline for when the price will react.
I think the easiest mistake is to interpret "future reduced supply" as "buy now and the price will rise."
People who buy before the halving might also sell before it actually happens. What really needs to be observed is whether, after repeated discussions of expectations, there is still new buying demand, not just that the countdown has fewer days.
$ETC's reduction method is different; every 5 million blocks, the block reward decreases by 20%, not a direct halving.
So you can't just see the word "reduction" and copy the historical price increases of other coins.
A decrease in new supply can be a reason for research, but if demand is also declining, the price may still face pressure.
I treat the reduction as one condition, and market performance afterward is needed for confirmation.
For $XLM, I pay more attention to how much coin needs to be retained for payment growth.
It is used for network fees, rent, and meeting minimum balance requirements.
But a large transfer amount does not mean an equivalent amount of XLM needs to be bought and held long-term.
I prefer to see how much balance users need to keep after continuous use. The growth of payment business is worth noting, but how much token demand increases still needs further calculation.On October 15, the KOSPI plunged 8.36% intraday, triggering the tenth circuit breaker this year and halting trading for 20 minutes. KOSDAQ was simultaneously circuit-broken.
SK Hynix fell more than 11%, Samsung Electronics dropped over 9%. KOSPI has declined for seven consecutive trading days, cumulatively down about 21% from its peak.
South Korea is the "canary in the coal mine" for global risk appetite. It fell first; who’s next?
The Nasdaq dropped 1.88%, down more than 510 points, marking the largest single-day decline since July 23. The Philadelphia Semiconductor Index closed down 4.3%, falling for the sixth consecutive trading day.
NVIDIA fell 3.16%, Meta dropped over 3%, Oracle plunged 8.8%. Dell Technologies fell more than 11%.
Tech stock valuations are being repriced due to rising interest rate expectations. No matter how strong NVIDIA is, it can’t withstand the surge in risk-free rates.
The US September CPI exceeded expectations across the board. Before the data release, the probability of a rate hike in October was about 22%; after the release, it surged directly to 43.7%, doubling in one day.
The market is starting to bet on three more rate hikes before the end of 2027. The September CPI annual rate rose to 3.4%, with inflation above the 2% target for 65 consecutive months.
This is the root cause of all the declines. It’s not war; it’s interest rates.
On October 9, 12, and 15, Bitcoin retreated three times near $81,000. Realized loss scale: October 12 was less than October 9, and October 15 was less than October 9.
October 9 looked more like a short-term sensitive chip concentration exit peak; the second dip showed weak selling willingness; the third was a macro negative catalyst causing some swing chips to stop losses. Selling pressure is decreasing but not over yet.
Short-term holders’ loss ratio is only about 31%, while the historical cumulative average is 42.7%. This means the "swing chips" have not been fully cleared.
Don’t mistake "repeatedly testing support" for "successful bottoming." Repeatedly stepping on the same floor will eventually make it creak.
Watch whether the realized loss scale continues to converge: if loss release weakens and the price holds $81,000, the phase correction may be nearing its end. If losses expand again, the market needs more thorough turnover.
Now is not the time to bet on direction; it’s time to wait for signals.
$BTC $ETH $SKHYNIX #韩股盘中跌近8%，全球科技股集体承压 $TSLA 观察：5年零利息确实狠，但救不了宏观大盘
5年0息这招确实绝，为了保销量，特斯拉把金融杠杆玩到了极致
Q3交付48.6万辆超了预期，但同比其实降了2.1%，说明价格战和贴息换来的是辛苦钱。
10月21号财报马上来，市场全盯着毛利率和现金流，还有Robotaxi和Optimus的新饼。
但看盘面，TSLA从388一路滑到362，跌了1.1%
台积电业绩爆表都照跌不误，特斯拉哪怕交付超预期，在宏观大抽水的背景下，照样硬气不起来。
车卖得再好，也抵不过美元走强、流动性枯竭的大趋势
贴息能刺激订单，但救不了估值。
大环境不回暖，个股再折腾也只能跟着泥沙俱下
#特斯拉交付超预期后盈利修复成焦点
看戏就好，别急着替资本操心。 Don't blindly jump in on the rebound! BTC, ETH, and SOL funds are contradictory 🚦
Looking at 24 hours, things seem stable, but in the last hour, they all turned: BTC, ETH, and SOL spot + futures show simultaneous net outflows, with short-term selling pressure dominating.
$BTC: Spot +95.65 million and futures +91.86 million for the whole day, the foundation is intact. But in the last hour, spot -48.64 million and futures -229 million, real selling pressure has arrived, watch for spot support.
$ETH: The most conflicted. 24h futures +415 million, but spot -89.43 million. Futures are active but direction unclear, spot doesn't follow, rebound lacks a solid base.
$SOL: Spot +30.07 million and futures +26.07 million for the whole day, still okay, but both sides outflow in the last hour, 4-hour trend is weak, endurance is questionable.
Summary: If there is support, don't chase the highs. Watch for 1-hour selling pressure to ease and 4-hour spot to strengthen. Warm colors on the daily chart ≠ safety on the hourly chart. Futures carry high risk, trade cautiously!
#US CPI unexpectedly rose to 3.7%, tonight PPI faces another test Once the September CPI was released, the market saw the least desired combination: both headline and core inflation exceeded expectations. The overall year-on-year rate was 3.7%, higher than the expected 3.6% and the previous 3.4%; month-on-month it was 0.6%, faster than the previous 0.4%. More troublesome was the core CPI year-on-year at 2.6%, also surpassing the expected 2.5% and previous 2.4%, with a month-on-month increase of 0.3%. This is not just a single energy disturbance, but price pressures spreading into broader areas. Gasoline price increases are the driver behind the rise in overall inflation, but the simultaneous rise in core inflation is the real signal. The Federal Reserve fears core inflation not falling the most, as this means a stronger case for rate hikes. After the data release, rate hike expectations were repriced; previously, the probability of a rate hike in October was about 23%, but now it is likely revised upward. The US dollar and US Treasury yields may strengthen simultaneously, putting pressure on risk assets. $BTC is currently only up 0.42%, which is clearly weak given the CPI surprise, indicating bulls lack the willingness to chase higher. Resistance is between 83,500 and 84,500, with support between 81,500 and 82,000. If the market starts pricing in a higher probability of an October rate hike, $BTC may retest the lower support, and if it breaks, look toward 80,000. $ETH and $ZEC are similarly unlikely to remain unaffected; when risk appetite contracts, high-volatility assets are usually the first to be hit. The core logic: the stickier the inflation, the longer interest rates stay high, and the tighter liquidity becomes. The combination of short-term pressure and a stronger dollar is a headwind for the crypto market. The confirming signal lies in whether the probability of an October rate hike continues to be revised upward, #特斯拉交付超预期后盈利修复成焦点
特斯拉的「交付悖论」
所有人都把交付超预期当成利好。但恰恰是这份超预期，藏着一个坏消息。
Q3交付486,532辆，高于市场预期，同比却下滑2.1%。超预期靠的是什么？大概率不是卖得更多，而是降价促销撑住了量。量的故事越成功，价的故事就越难讲。
更关键的是交付结构——高利润的美国车占比下滑，低利润车型占比上升。这意味着即便交付超预期，汽车毛利率大概率承压。你为销量欢呼的每一分，都在侵蚀利润的质量。
两头都不便宜：靠降价换量，交付越亮眼，盈利修复越难；砍量保价，交付故事又立刻崩塌。
所以10月21日的财报，别盯交付数字，盯毛利率。# Netflix Growth May Slow, Advertising Business Becomes New Focus
Netflix's "Growth Paradox"
Everyone says advertising is Netflix's new engine. But it is precisely this engine that exposes the old story can no longer continue.
The essence of the ad tier is Netflix's own admission: some users are only worth the ad revenue, not the subscription fee. The higher the proportion of ads, the more it indicates that the era of charging full price based on content premium is fading.
More crucial is the valuation—33% profit margin is given to the mature subscription business, but the premium is supported by a very small advertising business. It's like paying the price of a car already on the road for an unproven engine.
Neither option is cheap: if advertising succeeds, it proves the golden age is over; if it fails, the only path left is price hikes. #ZEC机构资金入场，高位杠杆开始出清 $TSM 观察：业绩爆表也难救主，宏观大锤下众生平等
台积电Q3净利润同比暴涨68.7%，2nm制程和AI芯片需求全拉满，创了历史新高。
再看看K线，从484一路泄到423，跌去2.4%。超预期财报在宏观大熊面前，连个水花都没溅起来。
这就是当下的残酷现实：微观基本面再硬，也扛不住宏观流动性的抽干
CPI超预期、降息落空、美债收益率高企，风险资产全在被无差别抛售。
台积电作为全球科技股的风向标都这待遇，BTC和ETH哪来的底气独立走强？AI叙事现在不仅撑不起盘面，还随时可能面临估值重估的连环反噬。
业绩好只能证明公司赚钱，但赚再多钱也抵不过资金撤退的踩踏
大盘大环境不转好，个股再优秀也只能跟着泥沙俱下。
别对财报抱有幻想。大雾还没散，看戏就行。
#台积电Q3净利润超预期，盈利创历史新高 #Meta巨额AI投入进入回报检验期 最近美股AI板块出现了一个挺有意思的现象。 Meta、微软、苹果这些科技巨头受到市场关注，但部分存储芯片和光通信企业的股价却承受压力。 说实话，之前大家只要听到哪个公司加大AI投资，第一反应就是利好。 现在好像不一样了。🤔 市场开始追问：AI投入这么多钱，到底什么时候能真正赚回来？ 📊 先看看Meta。 Meta这些年在AI基础设施上投入了大量资金，包括数据中心、算力和模型研发。 这些投入可不是小数目。 但问题是，花钱容易，想让AI持续创造利润就没那么简单。 Meta的优势在于，它本身就有庞大的广告业务和用户基础。 如果AI能提高广告推荐效率、改善用户体验，最终就有机会转化为实际收入。 这才是Meta和部分纯AI概念公司的区别：它有现成的商业场景。 💰 再看看AI硬件产业链。 以前大家觉得，AI需求增长，芯片、存储、光通信企业就能一直受益。 但市场逐渐发现，事情没那么简单。 需求增长不代表每家公司都能保持高利润。 如果产能扩张太快、产品价格下降，或者行业竞争加剧，企业利润同样可能受到挤压。 这里涉及一个很重要的投资概念：资本回报率（ROIC"The bonus from password sharing is over, will Netflix rely on ads to survive? Three numbers to watch on October 20"
Netflix will announce earnings after the market closes on October 20. Key points first: the company guides Q3 revenue at $12.86 billion (+12%), operating margin at 33.2% (28.2% same period last year)—the margin story is more attractive than revenue.
There are three real tests: ① Can ad revenue reach $800 million (Q2's $618 million missed expectations and serves as a warning); ② Will revenue growth fall into single digits (17.6% → 13.4% → guidance 11.7%, stepping down continuously); ③ Can price increases plus ads offset subscription slowdown. The bonus from password sharing is over, and the growth engine is shifting gears.
The stock price has dropped 24% in the past three months, forward P/E is 19.9x, about 60% of the 5-year average—expectations are already low, but don’t rush to bottom-fish; the 7% drop after Q2 earnings was due to soft guidance.
Mapping to crypto: US stocks have earnings reports to falsify expectations, crypto only has macro. BTC 80800 is holding the 80000 level; tonight’s triple PPI release is the first to watch. If macro doesn’t ease, all narratives are just distant hopes.
#Netflix growth may slow, ad business becomes a new focus