Orbit Post Sitemap
This time, there is a noteworthy signal: CME's hedge funds have unusually turned net long BTC futures, indicating that some traditional institutions are starting to lean bullish
But on the other hand, long positions have already piled up quite a bit. Currently, long positions hold about $23.4 billion, accounting for 57.6%. This structure doesn't mean a drop is imminent, but it means that as soon as the market moves a little down, liquidation can easily amplify volatility
It's not that no one is buying BTC right now, but rather that the buyers are a bit too eager
ETF inflows are a medium-term positive sign, but contract crowding is a short-term risk. Both can be held at the same time, so don't blindly chase rallies just because funds flow in
The biggest highlight for BTC right now is whether it can truly hold above the $65,400 to $65,800 range
There have been multiple false breakouts at this level before. If it can break out on increased volume this week and pulls back without falling back, then there will be opportunities later. If it surges and pulls back again, in the short term, be cautious of one lot pulling back to 64,000 or even 61,500 USD
I think the most painful trading right now is chasing long positions near resistance levels
A truly strong market will be confirmed after a breakout. A truly weak market won't miss all opportunities just because you hesitate for two hours. Rather than betting on direction around 65,000, it's better to wait for the market to give the answer first $BTC Complete practical briefing on the crypto market during the morning session of Tuesday, August 11
1. Macro Core Early Trading Trends (Determining the Overall Market Tone)
1. US Dollar + US Treasuries + Fed Expectations (Core CPI Countdown This Week)
1. The 10-year US Treasury yield rose to 4.699%, the US dollar index rose to 99.81, slightly strengthening, and oil prices surged 5.07% to $82.14. The market feared a rebound in inflation, prompting funds to shrink risk assets in advance and face pressure on cryptocurrencies in early trading.
2. Market consensus expectations: July CPI year-on-year 3.4%, core CPI 2.5%; Data exceeding expectations→ rate hike fears, causing deep corrections in the crypto and storage tokens; Low inflation directly triggered a rate cut rally, leading to a one-sided rally. Funds were mainly cautious and consolidating the market throughout the day, unlikely to see a major rally.
3. The yen depreciated again to 159.34. Previously, most of the gains from joint intervention by the US and Japan pulled back, and carry trading funds have not yet exited the leveraged market; if further depreciation triggers secondary intervention, contract funds will be quickly withdrawn, triggering concentrated selling pressure.
2. Japanese and Korean Stock & Storage Sector (SanDisk and SK Hynix tokens directly linked)
1. Korean KOSPI forecasts a lower opening and downward trend, while the memory chip sector continues its previous selling sentiment. Low long-term contract prices suppress industry prosperity, while native US stocks SanDisk and Western Digital continue their previous sharp declines, while XSNDK is naturally in a bearish market in early trading.
2. The Korean won appreciated slightly, and domestic storage companies' order expectations were weak. During the Korean stock market opening this morning, tokenized stocks experienced the largest volatility, marking a key high-leverage risk control period.
3. Geopolitics and precious metals
Negotiations on the Strait of Hormuz remain deadlocked, with rising oil prices fueling inflation concerns; Gold surged 2.49% to $4,448, with safe-haven funds flowing into gold, diverting some BTC buying pressure and putting Bitcoin under pressure.
2. Crypto Industry Morning News + Capital Liquidation Data
1. Market capital flows: The proportion of mainstream BTC funds has increased, and collective capital from altcoins is fleeing, showing a pattern of the strong getting stronger and counterfeit currencies accelerating differentiation.
2. 24-hour online liquidation: Larger volume of short liquidations, last night's decline swept away low-level long chips, short-term bearish sentiment dominates.
3. Spot ETFs: BTC and ETH saw small and continuous net inflows, while institutions were slowly positioning at low levels. The market had limited room for major declines, mainly consolidating momentum through volatile pullbacks.
4. Regulation: The CLARITY bill has been postponed to September voting, with no short-term regulatory headwinds, making it unlikely for the market to drop sharply before the negative vacuum period ends.
5. Warning: Multiple small Dogecoin tokens are experiencing contract insertion and dumping rally, avoiding hot, niche knockoff contracts.
3. Mainstream Coin Early Trading Levels + Market Analysis
BTC current price is $64,120, down 1.64% intraday
- Short-term support: 64,000, 63,700; strong trend support at 63,400
- Pressure: 64,700, 65,300
Trend analysis: Overnight funds fled for risk-averse and pulled back to test key support, holding the major bullish structure at 63,400 level; Today, the price remains oscillating in the 63,700~64,700 range, waiting for Wednesday's CPI to choose direction; heavy positions are not suitable for betting on one side.
ETH current price $1876, down -2.42%
- Support: Short-term level at 1870, strong support at 1850
- Pressure: 1908, 1935
Market Front: Volume increases and pullbacks downward, short-term bullish momentum is weakening; Holding above 1870 to maintain range-bound consolidation, effectively breaking below 1850 to open a deep pullback to the 1810 range; A pullback to 1850 is a stable low buying point.
4. Today's complete timeline + practical trading strategies
Today is a key milestone
1. Morning: Korean stocks opened around 8 a.m.→ SanDisk and SK Hynix experienced peak volatility
2. Evening: Volatility increased in pre-market trading of the U.S. stock market
3. Wednesday Evening: Core July CPI Data (the only major market flashpoint this week)
Overall trading strategy
1. Tuesday's overall market: shrinking volume and oscillating with shakeouts. The main force deliberately waits for inflation data. The probability of a one-sided trend is extremely low. Prioritize short-term trades within the range, avoid heavy positions betting on the direction.
2. Track Division:
- Mainstream BTC/ETH: Mainly buy on dips when pulling back key supports, leaning towards medium- to long-term bullish positions;
- Storage tokens (SanDisk, Hynix): The sector's negative factors are clear, and relying on the resistance above to position short positions offers better cost-effectiveness.
5. Overall Market Summary
Today is a pre-CPI safe-haven pullback, with the market base supported by spot ETF funds, leaving limited downside potential; The storage sector has independently emerged from a weak bearish market and is the most cost-effective track for trading today; All major price movements will wait for Wednesday night's inflation data before making any moves. #本周三CPI公布, will the September rate hike pricing be rewritten? #存储股抛压缓和, is the AI memory bull market stable? #现货ETF资金回流, can BTC and ETH take over? $BTC $ETH $SNDK Putting price aside, the BTC community data itself already shows two different clues. OKX Onchain OS recorded 66 BTC mentions in one hour at 02:00 on August 11, including 63 times on X and 3 on news; The total volume in 24 hours was 1,058. After conversion, the latest hour is 1.50 times the hourly average for the long window, which is about 50% higher than the 24-hour average. This ratio only answers whether discussions have heated up, not whether buying has increased. If you write it directly as a breakout signal, you take an extra step and make an inference that the data does not support. The structure of tone is another line. Within one hour, 48% are bullish, 8% bearish, and about 44% neutral, indicating a clear bullish advantage; Within the 24-hour period, the trend is 39% bullish and 19% bearish. The gap between the short and long windows is the part worth tracking going forward. In terms of origin, BTC is currently almost entirely driven by X. When a message is widely shared, mentions quickly increase, but independent information may not necessarily increase year-on-year. The trending list cannot tell us whether each piece of text comes from different participants, nor does it weigh by account influence or fund size. Long window sources can be used as background: BTC 24 hours has 946 times and 112 news events. If the proportion of sources in one hour suddenly deviates sharply, it could mean new news first broke out on a certain channel, or news updates just haven't caught up yet. Both explanations are reasonable, so we still have to wait for the original#本周三CPI公布，9月加息定价会改写吗？
I see the nonfarm payrolls as just an appetizer; CPI is the real anchor.
The nonfarm data knocked the probability of a September rate hike from nearly 60% down to just over 40%, but the real direction will be decided by this Wednesday's CPI. The market isn't focused on employment data now; it's waiting for the CPI to write the verdict for the September FOMC.
---
July nonfarm payrolls were -23,000, expected +80,000, missing expectations by over 100,000. May and June data were revised down by a total of 103,000. After the data release, the probability of a September rate hike dropped from nearly 60% to 44.4%, while the probability of holding rates steady rose to 55.6%. On Polymarket, the probability of no rate change rose to 63%.
The market has already moved ahead.
But the real variable is Wednesday. The market expects July's overall CPI annual rate to drop from 3.5% to 3.4%, and core CPI from 2.6% to 2.5%. On the surface, this looks like "inflation cooling further," but there's a detail that might be overlooked — core services inflation is expected to rise 0.3% month-over-month. In May and June, this figure was 0%. After two months of pause, service inflation might be starting to climb again.
This is where the divergence lies. Citi says if inflation softens for a second consecutive month, a September rate hike is basically off the table. Bank of America says the rebound in core services inflation might keep the September hike on the table. Both are major banks analyzing the same data but reaching completely opposite conclusions. BofA even says employment data won't change the Fed's judgment; the policy focus remains on inflation. Their analyst Kate Duguid puts it more bluntly: if inflation data is below expectations but concerns over rising core services inflation dominate, the hike could be delayed until December or later.
For the crypto market, there are basically three scenarios Wednesday night:
CPI below expectations, service inflation doesn't rise — probability of a September hike is further pushed down, risk appetite recovers, BTC has a chance to rally further.
CPI meets expectations but service inflation exceeds expectations — the market will reprice the likelihood of a September hike, and BTC will likely pull back.
CPI exceeds expectations, and service inflation also exceeds expectations — hike expectations are reconfirmed, risk assets come under pressure.
Nonfarm already moved the market once, but not thoroughly enough. CPI is the real catalyst for repricing. After the nonfarm data release, Bitcoin only rose for five minutes before falling back; the true catalyst is Wednesday.
Weakening employment data superficially weakens the basis for a rate hike, but the market is not yet convinced of a monetary policy pivot; all eyes are on Wednesday's CPI. Nonfarm moved the market once, but not thoroughly enough. Wednesday's CPI is the real factor that can make the market reprice.
Nonfarm just flipped the table; CPI decides how the next hand is played. 美股承压下资金正转向独立估值洼地，加密市场中具落地场景的板块或率先触发轮动。美股三大指数收跌之际，纳斯达克中国金龙指数逆势上涨1.65%，阿里巴巴上涨3%，显现AI落地对资金的吸聚效应。若资金持续在美股主线外寻找低估资产，加密市场将从全盘跟随转为围绕具体应用展开分化。后续观察加密山寨板块交投量能能否在美股风险偏好回落时保持独立。
#闪迪8月13日投资者日临近，财报分歧待解 #交易之声：你的经验值得被听到Here's a small emotional signal: tonight, while the three major US stock indices all closed lower, the Nasdaq Golden Dragon China Index bucked the trend and rose 1.65%, and Alibaba rose 3%. This wave of independent strength in Chinese concept companies is backed by Apple's connection to Alibaba's Qianwen and the launch of domestic AI applications, which are gaining momentum. Looking at this from the crypto perspective is somewhat insightful—when funds start searching for "cheap, independent narrative" lowlands outside the main US stock market, those fake sectors with real implementation scenarios may react before the miscellaneous coins lying down with BTC. Rotation always starts with the hardest part of the narrative. Those who understand, understand.Here's an unconventional reading: $BTC tonight just over 64,000, down 1.6% intraday, with 15-minute and 1-hour RSI both plunging into deep oversold levels—but that's not the point. The key is divergence: gold hits new highs, oil prices jump 5%, both safe-haven and inflation lines are moving, but crypto doesn't follow risk-on or buy safe-haven risks—both sides fail. At times like this, the most honest signal in the market is volume: OI 24-hour de-leveraged, tariffs moderate but not at extremes. Without extreme sentiment, there is no one-sided rally. Don't rush to pick directions for it before the box is broken. Look at your positions.Here's something I always want to laugh 😅 about. Whenever war breaks out or tensions rise, someone immediately jumps out to shout, "Buy Bitcoin to hedge!" 🤣 Tonight, oil prices soared 5%, and this joke is back on time again. But reality is especially harsh—gold is skyrocketing to new highs, while Bitcoin is just lying on the sidelines, playing dead 😭. I've gradually come to understand: at times like this, the market doesn't treat coins as safe havens, but rather as victims of "inflation is coming, rate hikes won't go away." So next time someone shouts "war bullish coins," you can smile and take a look at the 2-year US Treasury bond 🫠. Many people tell stories, but few are right.霍尔木兹协议又卡住了，油价直接往上飞了。
WTI原油$CL 涨到80.98美元，布伦特原油$BZ 也同步走高。
伊朗的条件摆出来了，解除美国海上封锁、美军撤出伊朗周边、赔偿冲突损失、解冻伊朗资产
这些条件一个比一个硬，短期内达成协议的可能性不大。阿曼在中间撮合安全航道谈判，但离全面重新开放还有距离。
地缘风险今天一直在持续升温。
胡塞武装又炸了沙特阿美的炼油厂，整个中东的紧张局势没有缓解的迹象，油价继续在定价这个不确定性。
那BTC和ETH为什么在跌？
市场在重新算账，油价持续走高意味着通胀预期升温，通胀预期升温意味着美联储更难降息，甚至可能再加息
BTC跌到64300附近，ETH跌到1878，回吐了之前的涨幅
主流币现在这个位置挺尴尬的
地缘风险推高油价→油价推高通胀预期→通胀预期推迟降息→风险资产承压
这条传导链目前是通的，BTC和ETH短期被宏观逻辑压着走
现在不要急着抄底，地缘风险和油价走势还没明朗，CPI数据也马上要出，不确定性太大了
等这两个变量落地了再说，要管住手，方向明确了再动
$BTC $ETH #霍尔木兹协议未落地，油价风险再升温？ Here's a geopolitical narrative repeatedly misinterpreted by the crypto community. Let's see: tonight Trump declared "only the US can control Hormuz," and claimed he'd cleared the strait; The German Foreign Minister is calling for the strait to be opened unconditionally. On the surface, tensions are at stake, but in reality, all sides are looking for a way to "reopen." Why does it have to do with you? Because oil prices have jumped 5% in response to this game. Don't react to the "positive for Bitcoin hedge" just because of Middle East tensions—this round of oil prices is driving inflation and rate hikes, weighing on all risk assets, and crypto can't escape either. Geopolitical narratives only work when it comes down to pricing logic. Let's see.英伟达拉着六家华尔街巨头搞了个 5000 亿美元的 AI 基建融资平台，把算力包装成「可投资资产类别」。听着很性感，我提个扫兴的问题：当一门生意开始需要华尔街帮它把未来收入证券化、提前变现的时候，通常说明它靠自己的现金流已经喂不饱这个扩张速度了。这不是看空英伟达，是提醒一句——AI 叙事正在从「卖铲子赚真钱」滑向「用金融工程续命」。我见过太多周期，故事讲得最圆的时候，往往就是最热的时候。Raindrops dripped down the scope hood, and my breathing rate grew to four beats per minute. In the muzzle field of vision, the surge above 7757.64 wasn't a celebration—it was just the target slowly entering the uncovered, extreme kill zone.
The weekly 3.6% advance trajectory is merely a parabolic lift caused by changes in air pressure. The weakening nonfarm payroll data temporarily blew away the sidewinds of frequent rate hikes, giving the main battlefield a brief ballistic correction period. Those drummers shouting "straight to 8000 points" are merely observers shouting from the safe zone—they have never tried pressing their fingers on the cold trigger and have no idea how much recoil recoil is required to pull the trigger high on the ridge.
The deputy gunner urges fire in his ear, but the hidden threat in enemy positions has never been eliminated: the overheated valuation of the hash rate is like barrel unfired after consecutive bursts, the rise in bond yields is a storm that strikes at any moment, and the midterm elections are like the thick fog of the battlefield looming.
At this moment, the linked target's $XIWM is emitting a faint vibration at the edge of the crosshair. As the small-caliber pathfinder most sensitive to wind direction, the $XIWM's trajectory directly reflects the airflow and fluidity at the edge position. If the main position's performance cannot provide sufficient muzzle velocity, blindly raising the target will only cause the bullet to bounce off hard cover and backfire.
My principle is: without an absolute overwhelming profit-loss ratio, never pull the safe; without locking onto the target's throat, never waste a bullet.
When heatwaves are at their worst, lurking is the only survival rule.原油今晚收涨 5%，WTI 站上 82、布伦特破 87，霍尔木兹的博弈还没解。把它放进本周的宏观日历读：周三 CPI、周四 PPI，而这一跳油价大概率还没完整反映进上一份通胀数据。市场现在的逻辑链是——油涨→通胀黏性→加息预期不退→美债 2Y 撑住→风险资产承压。这也是为什么黄金创新高、加密却跟不动：这轮地缘不是被定价成避险，是被定价成又一次加息。方向别猜，先看两年期美债往哪走。数据不会陪你演戏。Here's a quick news item that's easy to skip over. Let's take a look: the SEC has sued Adit Ventures, accusing it of defrauding investors and misappropriating client funds in pre-IPO stock investments like SpaceX and Klarna. This signal is more significant than an isolated case—the hotter the star stocks in the primary market, the more gray tactics are trying to "help you get SpaceX's original shares." Retail investors get excited when they hear about SpaceX or the "last round before listing," which is precisely when they're most likely to be harvested. If you want to touch these kinds of assets, first figure out whether you're buying equity or a story. Those who know, understand.大型算力基地的机柜正在重组布局，高密度的液冷管线与高性能计算设备挤入电厂旁边的园区，电力储备的竞争张力陡然上升。
设备运维的焦点从单纯增加机器台数，转向对极值功耗下的散热效率和自动化响应速度进行精确微调。
长期电力协议的谈判门槛随之抬高，算力园区复用为AI训练中心的商业尝试正在锁定更多低成本能源额度。
当长期电力履约能力与智能调度效率相匹配时，基础设施的租用价值与现金流稳定性得到同步提升；若二者脱节，重型资产转型升级的沉没成本仍待确认。
如果AI计算需求保持强劲增长且园区电价调配顺利完成，多元化算力的产能释放将推动基础设施价值重新定价；倘若长期电价协议谈判破裂或交付延迟，这一估值修复通道将迅速关闭。
一旦AI计算基础设施的实际需求释放不及预期，高昂的液冷改造与自动化运维成本可能侵蚀既有现金流；只有当传统加密算力效率出现突发突破时，设施下行压力才可能得到缓解。
算力基础设施从单一用途向综合数据中心演进的逻辑是否立得住，关键看能效治理能否持续转化为运行时间上的确定性优势。
未来七天最值得观察的变量，是具备弹性用电调配能力的算力园区在能耗监测数据上的波动轨迹。
#本周三CPI公布，9月加息定价会改写吗？ #CLARITY表决推迟至9月，监管窗口后移 #财报观察员：空头回补成焦点，SpaceX后续怎么看？Oil prices soared 5% tonight, and the group chat about "fighting a war, hedge safe, good for Bitcoin" jokes were back on time again. I said bluntly: completely wrong. This round of geopolitical markets wasn't even safe-haven pricing but a rate hike pricing — oil prices → inflation → US Treasury yields supporting → risk assets get beaten together. Gold hitting new highs is true safe-haven risk; Bitcoin not keeping up is the best proof. If you truly believe in "war bullish coins," first look at where the 2-year US Treasury yield is headed. The most expensive thing in trading is never misdirection, but a story that sounds good but is actually the opposite to bet.NVIDIA has teamed up with Apollo, BlackRock, BlackRock, Goldman Sachs, and KKR to build an AI infrastructure financing platform mobilizing over $500 billion. The focus isn't on the big numbers, but on one thing: turning 'computing power' from one-time hardware sales into asset classes that global capital can invest in long-term. To translate—the AI business model is shifting from 'selling shovels' to 'collecting rent.' What does this mean for valuation? Cash flow is stretched and securitized, and volatility theoretically decreases, but this narrative is now tightly tied to interest rates and financing costs. The sexier the story, the more you need to see what cash flow it relies on to support it. Structure speaks for itself.#现货ETF资金回流, can BTC and ETH take over?
BTC 64,119 (-0.25%) | ETH 1,879 (+0.17%)
[Derivatives Perspective] Last week, Strategy sold 1,690 BTC at an average price of $64,262, cashing out $108.6 million to buy back its own shares. This year, it has sold a total of 6,916 BTC, with institutional reductions causing short-term pressure
However, the derivatives side did not turn bearish: BTC open interest across the market rose to $48.25 billion, up 3.8% over 30 days, with funding rates neutral to positive and bears not united
Reasons for price fluctuations:
(1) Strategy reduced holdings by 1,690 BTC, selling pressure suppressed the market (2) Open interest rebounded, widening the divergence between long and short positions (3) Before CPI was released, funds waited and saw insufficient volume (4) Marex planned to buy BTC/ETH as margin funds and institutions accelerated entry
Short term (1-3 days): fluctuating between 63,000 and 65,000 yuan, holding 64,000 with a slightly stronger edge
Medium- to long-term: institutional long-term allocation remains unchanged, holding 65,000 is expected to challenge 68,000 SPCX's counterattack was wild! The bears were cornered, but August 20th is the real test
$SPCX Today it hit 134.9, just one step away from its IPO price of 135.
Who would believe this script? In the days before the lock-up, the whole internet was shouting, "The 100 billion selling pressure is coming, run quickly," but what happened? On the day the ban was lifted, it jumped 6%, then rallied the next day, surging 23% in two days, forcibly pushing up the market value by over 327 billion. This isn't just dumping; it's digging a grave for the bears.
The bears really hurt this time
Now, the bears hold $24.6 billion in positions, larger than Tesla's short positions. This rebound in stock prices has pushed these people into a dead end—they have no choice but to close their positions and cover them. And this buying move, in turn, has become fuel for the stock price to rise—a classic 'long sell-short stampede' scene.
Why wasn't he crushed to death?
The reason is simple: before the lock-up, SPCX had already fallen to a historic low of 104.85 along with the earnings report. By then, all the panic positions had already left, leaving only "dead bulls" or those out of bullets. The moment the lock-up was lifted, it actually became a signal that "all the negative news has been gone." Although there are still 250 million shares shorted, accounting for 16% of the float, it seems the bear forces still exist, but the current chip structure is clearly more favorable to the bulls.
The next two hurdles
There are still 10 days left until the second batch of restrictions is lifted on August 20. Here comes the main event:
• If it can firmly hold above 135, there is still room for it to rise in the short term;
• If it fails to break through, it is highly likely to pull back to confirm support.
The money is burning too fast
Looking at the fundamentals, revenue of 7.8 billion looks decent, and losses in AI are narrowing, but the capital expenditure is shocking—18.37 billion! A year-on-year surge of 550%. Although Citi set a target price of 220, to sustain this valuation, the pace of burning money must slow down, or it will inevitably become a hidden danger.
The harsh contrast of the market
Look at SPCX's surge, then look next door: SanDisk$SNDK is still playing dead around 1200, $BICO worse, dropping half (49%) in a single day. SPCX's strength is entirely due to the combination of "unlocking negative news sold out + short covering."
To sum up
It's not time to celebrate just yet. The real outcome will have to wait until the August 20th batch of restrictions is lifted, plus the CPI data, and see if the macro environment cooperates. Once both of these boots are in place, this "new story" will finally come to an end.
#财报观察员: Bearish buying becomes the focus—what is SpaceX's outlook going forward? Just opened a 10x short on $PUMP.
$BTC
The token is up roughly 85% over the past 30 days, with a major unlock for the team and early investors coming in around 36 hours.
$ETH
There’s also been a wave of disclosed and undisclosed bullish promotion ahead of the unlock, which makes me think the market could see heavy selling the moment those tokens become available — potentially even hours before.
$SOL
This setup looks extremely attractive, but the real test is whether the unlock triggers the sell-off I’m expecting.
#CPIToResetFedBets #AIMemorySelloffEases #BTCETHETFInflowsReturn Selling another 1,690 coins, Strategy reduced holdings by nearly 7,000 $BTC in just over a month!
According to documents submitted by Strategy to the U.S. SEC, Strategy sold another 1,690 BTC from August 3 to 9, cashing out about 108.6 million yuan, with funds to be used to repurchase $STRC preferred shares.
Since the end of June, Strategy has sold approximately 6,948 BTC, recovered approximately $432 million, with an average transaction price of about $62,100.
Based on the company's overall holding cost of about $75,000, the difference corresponds to approximately $93.12 million.
This is a market valuation calculated based on average cost and is not the formal accounting loss disclosed by Strategy.
Looking at the 1,690 BTC sold last week, it can be interpreted as a capital structure adjustment, but combined with more than a month of continuous reductions, the signal is different.
Strategy has not abandoned Bitcoin, still holding 840447 BTC as of August 9, with cumulative sales accounting for less than 1% of its current holdings.
However, current operations indicate that when preferred share buybacks, cash reserves, and financing arrangements require funds, Bitcoin will also be sold in priority for cash.
This sell-off has limited direct impact on BTC's overall liquidity, but has a broader impact on market sentiment and the valuation logic of $MSTR.
If selling BTC becomes the norm, then Strategy's Bitcoin narrative valuation may need to be repriced.Bottom signals are stacking up:
$BTC
• Dats have been unwound
• The CME basis trade is basically gone
• Miner capitulation + an 18.5% difficulty drop — the largest since the 2021 China ban
• Saylor was reportedly forced to sell BTC, while STRC suffered a major depeg
• BTC spent meaningful time below the 200-week MA
• Price fell 17% below the previous cycle ATH
• Weekly RSI reached oversold territory and has now printed a higher low
What could still be missing:
• One final high-volume capitulation event
• A retest of realized price around ~$54K
A lot of the classic bottoming signals are already here.
We’re getting close. 👀#CPIToResetFedBets #AIMemorySelloffEases #BTCETHETFInflowsReturn Is this round of AI memory crashes actually making room for cryptocurrencies?
Recently, the AI memory stock has plummeted dramatically.
$SNDK and $SKHYNIX are bouncing back and forth, causing many believers to question their lives. Everyone is asking: Is the AI story just too late? Is the valuation too high and can't hold on?
Don't rush, this matter is actually much more interesting than it appears on the surface.
1. Not the end, but the "halftime"
To start with the conclusion: the big logic of AI infrastructure hasn't changed. SK Hynix and SanDisk are pulling out a big move—HBF (High Bandwidth Flash). This is designed to solve the current pain point of HBM capacity shortages, specifically catering to next-generation AI systems.
The current adjustment feels more like the market itself "washing the market." Previously, the rally was too aggressive, with too many profit-takers, so institutions used this as an excuse to sell first. Morgan Stanley also stepped in to smooth things over, saying the worst might be over. Although the second half still depends on supply and demand, at least there is no need for panic stampede.
2. The Real Chain Reaction: Where Does the Money Go?
This is not just a matter for the semiconductor industry. What we need to consider is: when tech stocks stabilize, where will the money go?
Think about it: if a high-growth track like AI memory can stabilize at this level, it means market risk appetite isn't dead but is just temporarily shaken. Once this sentiment recovers, plus US Treasury yields don't fluctuate, the dollar doesn't get too strong, and institutions still hold their wallets, then the liquidity overflowing from tech stocks will have to find a place to go, right?
3. The "Succession" Opportunity for Cryptocurrency
At this point, the crypto market should come into view 👀
If the script follows this path:
📈 Tech stocks stabilized (memory stocks stopped falling→ 💵 macro environment supported (ample liquidity) → 🏦 capital overflow
Therefore, venture capital is likely to spill over from traditional AI hardware into digital assets.
What does this mean for the crypto world?
• ₿ $BTC: First and foremost, it is the preferred safe haven and liquidity gateway for large institutional capital.
• ♦️ $ETH: It combines the fundamentals of the "world computer" with the new narrative of RWA (tokenization), which suits institutional tastes very well.
• ⚡ $SOL: High elasticity; once risk appetite picks up, this high-beta product often runs fastest.
• 🚀 Altcoins: Only after the overall environment is confirmed to warm will there be a broad rally.
4. Don't rush to "switch with one click"
But I need to pour some cold water on it: don't rush into the crypto world just because memory stocks drop.
There is a strict transmission sequence for this matter:
Tech stocks stabilize →, macro liquidity improves→ Bitcoin leads → Ethereum follows→ and altcoin rotation finally follows.
Now, we are in the stage of observing the "first step." If SanDisk and SK Hynix can hold their ground at this level or even slowly rise, then cryptocurrencies are very likely to be the next stop.
👀 Focusing on memory stocks, if they hold steady, the next round of the crypto market may not be far off. TSMC's monthly revenue record diverged from a 5% stock price correction, reflecting capital's recalculation of AI data center depreciation pressure and ROI.
In July, consolidated revenue reached NT$467.58 billion, a year-on-year increase of 44.7%, directly confirming that advanced 2nm to 5nm process capacity remains fully loaded. The nearly $2.4 trillion investment plan by the four major cloud providers has become the core driving force behind current hardware delivery growth.
The trigger for the upward scenario is that cloud giant Capex continues to expand and supports the high prosperity of AI chips into 2027. If the annual revenue growth target of over 40% continues to be realized, strong revenue growth will fully absorb depreciation cost pressures.
Once Q2 profit margins are confirmed to have peaked temporarily, the upward logic of continued valuation expansion will be broken.
The downward scenario is based on the logic of slowing data center construction or depreciation in the early stages of 2nm mass production exceeding expectations, eroding gross margins. If cloud providers lower their capital expenditure guidance, sector valuation premiums will be quickly recovered; If there is a new round of unexpectedly increased orders, the downward path will be interrupted.
The most noteworthy variable in the next seven days is the latest regulatory stance on capital expenditure guidance in the financial reports of the four major cloud providers.
#三星钱包将接入稳定币, payment scenarios continue to expand. #财报观察员: Bear buying becomes the focus—what is SpaceX's outlook going forward?Here's a thread from the industry narrative I've been following: the real demand for AI isn't graphics cards, it's memory. Large model training and inference rely on high-bandwidth memory, HBM and DDR5 are out of stock, and even with hundreds of billions spent on capacity, several big companies can't keep up with demand. So tonight, storage stocks pulled back along with the broader market, and SK Hynix's US stock dropped 1.9%. Instead, I'm paying more attention—the narrative isn't broken, orders haven't crashed, it's just sentiment. Those who understand this chain know: AI's accounts ultimately have a big chunk on memory. Protect your bullets, don't throw away your good cards in emotions.Samsung's move is ruthless! 800 million phones directly become "banks"—will USDC become the biggest winner?
Many people have yet to realize that Samsung quietly dropped a nuclear bomb at its launch event in early August.
It's not a new phone, nor a new foldable screen, but the feature showcased at Galaxy Unpacked 2026: Samsung Wallet natively supports stablecoins.
Don't swipe away—this is far more important than camera upgrades or processor speeds. Simply put, Samsung plans to turn the world's 800 million Galaxy phones into a walking digital dollar account.
1. What exactly is Samsung up to?
In the past, if you wanted to use stablecoins, you had to download an exchange app and complete KYC verification, which was a long hassle. You won't need it anymore.
Samsung's plan is: directly open a "stablecoin account" in your wallet. When you open Samsung Pay, your balance isn't just in RMB or Korean won, but also in USDC. When you go to the supermarket, the POS machine is tapped and money is directly deducted from your stablecoin balance, just like swiping a bank card.
This app already has 19 million users in South Korea and is now being rolled out to 61 countries worldwide. The demo clearly shows Circle's USDC. Moreover, Samsung isn't messing around on its own; the underlying compliance, custody, and clearing are handled by licensed giants like Circle and Coinbase, while Samsung focuses only on its strengths—hardware entry points and front-end experience.
Even more aggressively, Samsung wants not only a US dollar stablecoin but also a Korean won stablecoin. The technical support behind this comes from Samsung SDS and its stake in Dunamu (the parent company of Upbit, South Korea's largest exchange). This move is a big move.
2. Why is this called a "dimensionality reduction strike"?
Joseph Goh of crypto investment bank Areata spoke the truth: "Distribution channels are the scarce assets." ”
What was the biggest pain point of stablecoins before? It wasn't technology or liquidity, but ordinary people didn't know how to use them. Now, Samsung has completely removed the threshold. You don't need to know private keys, mnemonic phrases, or even know you're using blockchain. As long as you have a Samsung phone, you can use it by opening your wallet.
It's like WeChat Red Envelopes back in the day, which overnight gave hundreds of millions of people WeChat Pay accounts. Samsung's 800 million pre-installed devices equals 800 million potential bank branches. Even a 1% conversion rate would still represent 8 million real users. This is far more terrifying than any DeFi protocol or exchange for new user acquisition.
3. This is the real "financial ecosystem"
Samsung Wallet product manager Lee Dinham said they are aiming for a "connected financial ecosystem." How should this be understood?
Previously, your wallet only had money; in the future, your wallet will have assets, points, and identity. Samsung began deploying the Knox security module in 2019, opened up a purchase channel with Coinbase last year, and now directly supports stablecoin payments. This is a complete chain: secure storage → convenient purchases → daily payments.
Especially this time, the choice of the "custody model" shows Samsung's determination to comply with regulations. It's not just playing around; it's using its brand reputation as a guarantee, telling regulators and users: this is reliable, regulated, and safe.
4. Who benefits the most?
Anyone with eyes can see that Circle (the issuer of USDC) is winning effortlessly.
Samsung is currently the only mainstream phone manufacturer to publicly showcase USDC at the event. If USDC is the default stablecoin pre-installed in these 800 million phones, then Circle would directly have the world's largest retail distribution network.
On the other side, Circle is busy building the Arc mainnet, building the "highway" (institutional side); Samsung is working on the "community entrance" (retail side). Once the two sides connect, the last mile of stablecoins turning from "trading tools" to "payment tools" is thus connected.
To sum up:
Samsung's move is not just about simple feature iteration, but a battle for the financial gateway. When phone manufacturers start handing out "money," the traditional banking system and payment giants really can't sleep.
$SAMSUNG $USDC $BTC
#三星钱包将接入稳定币, payment scenarios continue to expand A cold wallet has a problem, and this time it's the least likely type to explode.
Starting at the end of July, Coldcard's hardware wallet was hollowed out from a firmware vulnerability buried in March 2021. To put it bluntly, when generating mnemonic phrases, it didn't use hardware random numbers, but rather a crippled software random number. The key strength of older models dropped directly from 128 bits to 40 bits—40 bits meant hackers didn't need to touch your device, your PIN code, or any trojans, sitting on the other side of the world and using computing power to extract the private key. So far, over 1,800 BTC stolen have been confirmed, which, at current prices, amounts to $120 million and involves over 5,200 addresses. Galaxy gave a quote of $130 million and 7,300 wallets. This is known as the most serious self-custody security incident in Bitcoin's history, bar none.
The most shocking detail is Coinkite's official statement: upgrading firmware is useless. Seeds you generated after 2021 have basically been running naked for five years; the only way to survive is to recreate a wallet and move all the coins. The on-chain reaction after the news came out was very real: on August 3rd, 935 Bitcoins dormant for over ten years moved in one day; on July 31st, 6,388 coins dormant for five to seven years moved away; and an old wallet from 2013 suddenly transferred 500 BTC worth over $31 million. These veterans who hadn't touched for over a decade were collectively frightened awake.
Now, back to that question: BTC holders or ETH holders, who panics more?
The answer is BTC holders, and their panic is justified. Coldcard is a Bitcoin-specific wallet, and its buyers are the hardest group of hoarders—no DeFi, no staking, no interaction. The coin sits in a drawer for five years, all for the word "absolute security." This incident exposed exactly the belief of this group: you think offline is the end, but the vulnerability is written in the firmware, and your sense of security is an illusion. Over the weekend, on-chain data showed a significant increase in $BTC flows into exchanges, a typical sign of holders losing confidence in self-custody. On August 10, BTC was priced at $65,003. After two weeks of the event, the price barely dropped but only rose slightly, indicating there was no panic selling, but underlying trust shifts were already underway.
$ETH The situation on the holder side is completely different. Ethereum ecosystem money rarely lies in hardware wallets—staked on validator nodes, locked in DeFi pools, and suspended on L2s; risk exposure is decentralized, online, and at the smart contract level. For them, incidents like Coldcard's "seed being calculated" are far from them; they constantly guard against phishing signatures and fake authorizations. Ironically, on August 10, ETH was priced at $1916, climbing from $1695 all the way up. The ETH/BTC exchange rate is still recovering, but hardware wallet issues have instead highlighted the resilience of the "active custody" approach.
But the real impact of this isn't about price, it's about the capital structure. BTC's core narrative is digital gold, and the foundation of digital gold is self-custody, and the self-custody base is 'hardware wallets can't afford to fail.' Now this foundation has cracks. In the short term, some long-term capital will reassess: should they continue trusting single hardware devices, or switch to multisig and institutional custody like ETFs? Think about it, this is actually a hidden boon for spot ETFs—BlackRock and others hold your coins, and if something goes wrong, someone pays for it. Coldcard manages your coins, and if something goes wrong, you cry yourself.
$120 million is a drop in the bucket compared to Bitcoin's $1.33 trillion market cap, but it damages a psychological account. Next, watch for two signals: first, dormant old coins will continue to move, and once moved, all the negative news will be gone; second, if BTC balances on exchanges keep rising, it means self-custody faith isn't recovering as quickly.
Cold wallets are not dead, but the era of "buying a hardware wallet means worry-free" has come to an end after August.82.5亿枚$PUMP 解锁，价格翻了2倍
你敢信？7月12日，82.5亿枚PUMP代币一次性解锁，占当时流通量的20%。搁别的项目，这种量级的解锁早把价格砸穿了
结果呢？PUMP从$0.0013干到$0.0027，翻了整整一倍。市场硬生生把82.5亿枚新供应全吃了
凭啥这么硬？
两个原因。第一，Pump.fun每天用50%的平台收入回购并销毁PUMP。过去一周就烧了$450万。第二，过去30天收入干到$3373万，直接超过了Hyperliquid。平台在真金白银地赚钱，不是空气
但别高兴太早
交易所余额快接近历史新低了——但8月12日还有68.75亿枚PUMP要解锁，价值大概$1000万到$1600万。好消息是，后面每月解锁只剩71亿枚，比7月那次少太多了
说白了，82.5亿都扛过来了，后面的压力只会越来越小。但短期波动躲不掉
操作上我不追高。等8月12日解锁砸完再说Title: SanDisk's Life-and-Death Gamble: On August 13, Can HBF Pull SNDK Out of the "Ruins"?
$SNDK Holders, attention, the countdown has begun.
On Thursday, August 13th, at 9 a.m. Eastern Time, SanDisk Investor Day. This was not just a press conference, but also an "ultimate judgment."
The stock price halved from its peak, now lying in the "ruins" of $121. Is this a gold pit or a value trap? It all depends on whether the story can be told this time. I suggest bookmarking it first; the information density is extremely high.
🔥 There was only one focus in the entire event: HBF
There were many highlights at this conference, but they were all just side dishes. The main dish was just one dish: the HBF (High Bandwidth Flash) product roadmap and commercialization timeline.
HBF is a "freak" created by SanDisk and SK Hynix, with a very tricky positioning—caught between HBM and SSD:
• HBM: Like a supercar, fast but expensive, with a small trunk capacity.
• SSD: Like a cargo ship, pulls more but slowly.
• HBF: Aiming to make an "armed pickup" requires both speed and flexibility.
It uses 8 or 16 layers of NAND stacking, with a single module packed up to 512GB, bandwidth down to 0.4-3.0TB/s, and it also uses a UCIe interface to connect directly to the GPU/CPU. Google and Tenstorrent have already tested it.
Its mission is simple: to smash the "storage wall" of AI inference. No matter how powerful the GPU computing power is, if data can't be fed in, it's just a pile of scrap metal.
🃏 Another trump card in hand: BiCS10
Besides HBF, SanDisk also has a technological card: BiCS10 (10th generation 3D NAND), jointly developed with Kioxia.
• 332-layer stacking, QLC version.
• Bit-density surges 59%, samples to be delivered in the second half of the year.
• CBA wafer bonding is used (logic circuits and memory arrays are done separately without interference).
• Designed specifically for AI data lakes and RAG knowledge bases, which are "data devouring beasts."
🤔 Why is the market in such a heated debate?
SNDK now is full of contradictions:
The stock price has been cut 48% from its peak, with a price-to-earnings ratio of only 6 times.
The earnings report was actually decent, but the guidance was $250 million less, and the market immediately turned hostile.
Citi is still holding on, setting a target price of $222.
This leads to the core disagreement:
AI storage leaders with 6x PE: are they picking up bargains or stepping on a pitfall?
⚖️ Two endings on August 13:
✅ Optimistic script: HBF roadmap is clear, mass production milestones are specified, and major clients are locked in orders. Market repricing, valuation recovery + earnings expectations = Davis double click.
❌ Pessimistic script: Keep making empty promises, full of nonsense about 'ecosystem construction,' with no hard timing. So keep grinding for a bottom around $120 to wear down patience.
As for executives selling shares? It really is annoying. But if you put yourself in their shoes, if you reduce your position at $200+, why not sell?
The key isn't whether they sell or not, but whether HBF can actually be sold! This wave of storage correction is worth looking at: SK hynix fell 1.9% in the US stock market tonight, Intel dropped 4%, ostensibly showing a collective semiconductor weakening. But zooming in on the lens—DDR5 spot prices, major company capex, and AI servers' rigid demand for high-bandwidth memory—these three fundamental curves did not fall along with the stock price. Stock prices are sentiment; capacity and orders are the ledger. So the real question in a pullback is not "how much has fallen," but "which fundamental data has been disproven"—so far, not a single one. $MU data won't accompany you.$ETH community is fighting a civil war
Justin Drake led his team to propose EIP-8363, with the core saying: the higher the staking rate, the more returns burn. Currently, a 33% staking rate yields 2.6%, but according to the proposal model, it was cut directly to 1.2%, which is more than halved
Core members of Aave, Lido, and ether.fi all got blown up. SharpLink CEO bluntly called this "value destruction"—institutions buy ETH just to get that little interest-bearing income, who would bother cutting it to zero?
While cutting yields, ETFs are buying like crazy. Last week, ETH ETFs saw a net inflow of $245 million, while BlackRock's ETHA alone made $203 million. But the price was still hovering around 1900—someone was quietly selling off
Wednesday's CPI will be the real test. Beyond expectations, rate hike expectations are making a comeback; Below expectations, risk assets take off
To put it bluntly, ETH is now under pressure from both inside and outside. Staking yields are about to be cut, ETFs are being bought but prices aren't rising, and CPI is hanging overhead
I'll keep waiting for the move. Don't act before CPI is implemented; wait until the direction is clear$SNDK SanDisk | August 13 Investor Day = Final review date, HBF decides whether it can save half its life
It fell from 2354 to 1238, a 48% slash, with the PE ratio returning to just above 16 times. After Citi cut the target price, there was still 2100.
On Thursday the 13th at 9 PM in the US East Coast, SanDisk revealed its cards, focusing on one thing at the event: HBF roadmap + commercialization timeline.
HBF = SanDisk × SK Hynix, sandwiched between HBM and SSD:
NAND stack 8/16 layers, single-mode 512GB, bandwidth 0.4-3.0TB/s, UCIe direct connection to GPU/CPU, Google+Ttenstorrent has joined the OCP alliance.
What they do is build an AI inference "storage wall"—no matter how powerful the GPU is, if it can't feed data, it's pointless.
The companion BiCS10: 332-layer QLC, density +59%, samples to be sent in the second half of the year, CBA bonding, dedicated to feeding AI data lake/RAG.
The financial report itself is solid (Q4 revenue was 8.97 billion, exceeding expectations), but next quarter's guidance is 250 million less, causing the market to crash.
Long position: 6x PE (old caliber) AI storage leader + HBF revenue = Davis double click.
Bears: Cyclical stocks are just empty promises; HBF will only be sampled in 2027 and fully commercial in 2030. Distant waters can't put out immediate fires.
Number 13 has two scenarios:
✅ Roadmap nails sample/mass production nodes + major customer orders → repricing, establishing a golden pit
❌ Still the nonsense of "ecosystem construction" → continues to stall around 1200, and the shadow of executive share reductions cannot be hidden
Don't focus on whether executives have sold, or whether HBF will sell.
$SNDK #闪迪8月13日投资者日临近, divergences in the financial report remain to be resolved #闪迪8月13日投资者日临近, divergences in the earnings report remain to be resolved
SanDisk Life-or-Death Match: The Ultimate Judgment on August 13! Can HBF pull it out of the 121 pit?
$SNDK The countdown begins, August 13th is Investor Day, which is definitely the toughest battle in recent times.
The information is dry and the logic is twisted; it's recommended to read quickly and read later.
Key point: What exactly should you watch on the 13th?
At 9 a.m. Eastern Time on Thursday, SanDisk was about to reveal its cards. The only thing that was most explosive was the product roadmap and launch date for HBF (High Bandwidth Flash).
The remaining highlights are considered "side dishes":
• Latest progress on BiCS10 (10th generation 3D NAND).
• How to stack SSD capacity
• Are there any major companies signing "long-term contract orders" (locked in contracts is confidence)
What is HBF? Why is the whole village's hope resting on it?
Simply put, HBF is a "middle-tier" storage developed by SanDisk with SK Hynix.
Let's give an example:
• HBM: Like a sports car, super fast, but the trunk is too small and extremely expensive.
• SSD: Like a large truck, it hauls a lot and is cheap, but not fast.
• HBF: Want to make a "pickup"—powerful (bandwidth) and able to drive (capacity).
It uses NAND stacking, with 8 or 16 layers, and a single module can fit up to 512GB, with bandwidth between 0.4 and 3.0TB/s. It also uses a UCIe interface, allowing direct connection to the GPU and CPU.
Currently, Google and Tenstorrent have already tried it out. Its mission is to eliminate the "storage wall" of AI inference—no matter how powerful the GPU computing power is, if data can't be fed in, it's just a decoration.
BiCS10: The trump card in hand with 332 layers
This is the tenth generation 3D NAND developed jointly by SanDisk and Kioxia, with top-notch specs:
• 332-layer stacking, QLC version
• Bitdensity surged by about 59%
• Sample deliveries will begin in the second half of the year
• Uses CBA wafer bonding technology (logic circuits and memory arrays are handled separately without interference)
• Specifically catering to AI data lakes and RAG knowledge bases, which are major data swallowers.
The current awkwardness: gold pit or value trap?
SanDisk's stock price has dropped from 121 to now, being halved to just 48%, with a PE ratio of just 6 times.
The earnings report was actually quite good, but the guidance was $250 million less, and the market immediately turned its back on the market.
Citibank was stubborn and directly set a target price of 222.
The core dilemma now is: with a leading AI storage company with a 6x PE, do you dare to buy the dip?
Final conclusion: The key is to look at number 13
This trend depends entirely on whether HBF can turn "PPT technology" into a "money printer."
✅ If the roadmap is clear and the timeline is reliable: the market immediately repriced, Davis double-clicks, and takeoff.
❌ If you're still making empty promises and giving vague answers: then just keep lying flat and grinding down near $120.
As for executives selling shares? It is indeed annoying, but if you put yourself in their shoes, you reduce your position at over 200 yuan, and you don't sell? The key isn't whether they sell or not, but whether HBF can actually be sold! Gold, BTC, US stocks—the hidden logic of the market after non-farm payrolls
🌐 Cross-market observation: after the cooling of nonfarm payrolls, don't simply interpret it as broad positive news
After nonfarm payroll data fell short of expectations, the market collectively traded expectations for rate cuts, gold surged sharply, US tech experienced volatility, BTC and ETH surged slightly but then fell back again. Many people simply interpret this as negative for the dollar, with all risk assets bullish together, but the reality is not that simple.
Gold has already stabilized above 4370, with global gold ETF funds flowing back again. Coupled with continued gold purchases by central banks in multiple countries, the medium- to long-term allocation logic is solid, but in the short term, many long positions have already accumulated profit-taking. There is strong selling pressure between 4420-4450, and the risk of chasing the high is considerable.
$BTC and $ETH Although benefiting from rate cut expectations, they are high-risk assets and β not entirely the same logic as gold. Rate cut expectations indicate economic weakness. If CPI remains stubborn, the market will adjust its rate cut pace again, putting pressure on risk assets immediately.
In other words: rate cut expectations are good for gold, but for crypto and US tech stocks, it's a double-edged sword. A soft economic landing will benefit both the stock market and crypto; If the economy cools rapidly and recession trading arrives, risk assets will still be sold off, with only gold's safe-haven properties prevailing.
The biggest turning point in the market now is the upcoming CPI. This inflation data will rewrite the Federal Reserve's pricing.
If inflation eases and rate cut expectations are confirmed, then U.S. tech stocks, BTC, ETH, and gold will collectively recover;
If inflation remains elevated and the market withdraws from rate cut ideas, gold will also pull back in the short term, putting greater pressure on risk assets.
The market is already betting on the non-farm payroll results in advance; stop using them as trading basis, and shift all focus to CPI data. In a volatile environment, don't bet one-sidedly; responding well to different scenarios is far more important than predicting price movements.大多数目光都盯着BTC的ETF资金流入，但近期一份持仓数据透露出不一样的信号：美股上市企业BitMine上周再度增持7391枚ETH，目前总持仓已经来到580万枚，占到以太坊总供应量4.8% 不同于散户追涨杀跌，这家企业是持续分批累积，并不会因为短期价格涨跌随意抛售。但现实盘面我们看到，ETH走势持续弱于$BTC，价格跌破多条短期均线，并没有因为大额机构囤币走出反弹行情。 这里就出现一个很值得思考的背离：链上实体在持续收筹码，二级市场价格却持续承压。核心原因在于，现货ETH‑ETF的日内资金流入并不稳定，机构现货买盘，更多来自上市公司直接链上囤积，而不是ETF通道进场。 BTC的机构资金绝大多数走ETF，资金进出直接反应在盘面；$ETH 有很大一部分买盘沉淀在链上冷钱包，不在二级市场交易，短期不会拉动币价。 放到当前宏观环境，非农数据降温，市场已经开始提前交易降息预期，但是CPI还没有落地，市场不敢直接大举进攻。 $BTC 依旧在63300‑65500箱体来回震荡，属于整个风险资产的锚；ETH 1840支撑是当下的关键考验，如果支撑守住，后期弹性会被充分释放；一旦跌破，会迎来一轮杠杆清Why has $BTC been trading sideways around $65,000 for several days? Is this a warning of a decline or a sign of an upward trend? I will simulate three scenarios for reference only to make predictions.
My view is:
1. This is not weakness, but a typical "chip exchange period"
The most obvious current market features are:
* ETF funds are flowing back in
* Whales continue to accumulate shares
* Retail investor sentiment remains cautious
* Low level of leverage
Such combinations rarely appear at the top of bull markets, but rather resemble a consolidation phase. Recent data shows that US spot BTC ETFs have seen large-scale net inflows, while market sentiment remains in the fear zone, indicating institutions are buying while retail investors are not chasing the rally.
2. 65,000 is the dividing line between bulls and bears
From the perspective of capital and technology:
* Around 63,000 = Strong support
* Around 65,000 = Current Equilibrium Zone
* 66,800-68,000 = Key breakout zone
* Above 72,000 = Medium-term trend reversal confirmed
Currently, BTC most resembles the kind of volatile buildup seen after ETF approval in 2024, rather than the bear market decline of 2022.
3. What is the market waiting for?
Essentially, it is waiting for macro catalysts:
* CPI data
* Expectations of Fed rate cuts
* Changes in U.S. Treasury yields
* Continued ETF inflows
Many analysts believe that BTC has now shifted from being "crypto-driven" to "macro liquidity-driven," with oil prices, interest rates, and dollar liquidity increasingly influencing prices.
4. Three types of future scripts
Scenario 1 (50% chance)
The 63,000-68,000 range will continue to fluctuate for 1-4 weeks
Features:
* Every time it drops to 63,000, someone buys it
* Every time it hits 67,000, it gets sold; this is the main player's favorite shakeout method.
Scenario 2 (35% chance)
After breaking through 68,000, it quickly surged to 72,000-75,000
Trigger conditions:
* CPI came in below expectations
* ETFs continue to see significant net inflows, and current ETF capital flows are a key factor supporting this scenario.
Scenario 3 (15% chance)
If it falls below 63,000, it will pull back to 58,000-60,000
Trigger conditions:
* CPI exceeded expectations
* The Fed is hawkish
* The ETF has turned into outflows again
But institutional demand still exists, so this situation is not the main scenario for now.
My conclusion
If you broaden your perspective to the next three months:
BTC is trading sideways near 65,000, which I believe is more of a "build-up" rather than a "top."
The probabilities I currently give are approximately:
* Breakout after sideways movement: 60%-65%
* Continued long-term volatility: 20%-25%
* Entering a new bear decline below 60,000: 10%-15%
The real signal to watch out for isn't sideways movement, but rather:
1. Consecutive large-scale outflows from ETFs;
2. Whale addresses continue to reduce their positions;
3. BTC cannot recover after falling below 60,000.
If all three occur simultaneously, I will start considering deeper levels of bear market risk.ETH/BTC fell to around 0.03, the lowest point since the DeFi summer of 2020. BTC's market share has surged to around 60%, ETH is stuck in the middle, struggling financially.
Bears have a straightforward reason: Layer 2 has drained mainnet traffic and fees, gas fees have plummeted, ETH's burning mechanism is virtually useless, and the deflationary narrative has completely collapsed. DeFi regulation still lacks reliable trust, the SEC has changed chairs several times, and whether ETH counts as a security or is just a confusing account. Moreover, ETF funds clearly favor BTC, and the ETH narrative increasingly resembles "the old story of the previous cycle."
But Tom Lee has been shouting lately that ETH/BTC will rebound in the second half of the year. His logic isn't to speculate on technology, but to bet on stablecoins and RWA tokenization. Most stablecoins like USDT and USDC run on Ethereum, RWA (Real-World Asset Tokenization) is being developed by BlackRock and JPMorgan, with ETH as the underlying infrastructure. If global capital really starts moving on-chain on a large scale, ETH's value as a settlement layer will be repriced. He even declared that if BTC reaches 250,000, ETH could reach 12,000–22,000 at the historical average, and in extreme cases, 60,000.
The core of this controversy is: $ETH Is the current downturn a "value trough" or a "structural lagging behind"?
If it's a lowland, then 0.03 is the right time to pick up bargains. After all, ETH's developer ecosystem, TVL, and stablecoin circulation are still the best in the industry. Although Layer 2 has diverted gas resources, it has also expanded Ethereum's overall layout. Once regulations become clearer and capital rotates, ETH's elasticity may be greater than $BTC.
But if it's structurally lagging behind, the problem becomes serious. BTC's "digital gold" narrative is becoming increasingly rigid, with clear institutional allocation logic. ETH's narrative, however, keeps shifting—from "world computer" to "ultrasonic currency" to "settlement layer," with none of these stories fully realized. The more layer2s there are, the more the mainnet feels like an empty shell; competitors (Solana, Sui) are catching up quickly in performance and user experience. ETH may be shifting from the "king of altcoins" to "large-cap altcoins."
The market is very fragmented right now. On-chain data shows ETH staking volume is still growing, indicating long-term holders haven't left; But ETF fund flows and exchange rate trends show short-term funds are voting with their feet. Tom Lee's bullish view is more like a "macro belief"—believing that the major trend of financial digitalization will boost ETH. But bearish investors see immediate bleeding: no fees, less narrative, and funds fleeing.
I think the key to this game isn't technical, but liquidity. If the Fed cuts rates and the dollar weakens in the second half of the year, risk assets will collectively rebound, and ETH, as a high-beta asset, may outperform BTC. But if macro conditions remain tight, BTC's safe-haven nature will further concentrate funds, and ETH/BTC may hit new lows. Holding the 0.03 level means a double bottom; failing it opens up a new round of downward potential.Nến xanh không có nghĩa toàn bộ thị trường đang cải thiện 🚨
Đợt tăng này trông mạnh mẽ, nhưng bên dưới bề mặt, thanh khoản đang ngày càng thận trọng. Dòng tiền không lan tỏa đều mà chỉ xoay vòng trong một nhóm nhỏ các đồng thắng cuộc, trong khi phần lớn dự án âm thầm mất đi sức mạnh tương đối.
Dữ liệu cho thấy rõ điều này: lợi ích mở đang hạ nhiệt, khối lượngTSMC's record-breaking monthly revenue and the partial correction in its stock price intersected at the same time, shifting the trading focus of the AI industry chain from blindly chasing highs to financial actuarial analysis.
July consolidated revenue reached NT$467.58 billion, a year-on-year increase of 44.7%, with the full-year revenue growth guidance further raised to over 40%.
The nearly $2.4 trillion investment committed by the four major cloud vendors is being converted into orders, driving advanced process capacity from 2nm to 5nm to continue reaching full capacity.
This strong hardware delivery capability is linked to the current 5% correction from the stock price peak, highlighting capital concerns about the actual return rate of data centers and the cash flow consumption of large capital expenditures.
If capital expenditure by giants continues to expand and the chip boom continues as expected into 2027, strong revenue growth will offset depreciation pressure; If Q2 profit margins are confirmed to have peaked temporarily, this upward logic will be broken.
If the pace of data center construction slows down, or if depreciation costs in the early stages of 2nm mass production exceed expectations and erode gross margins, valuation premiums will be quickly recovered; If hardware orders receive a new round of unexpected additions, the downward trend will be interrupted.
Earnings support for the US tech sector remains stable, and the divergence essentially stems from the market's recalculation of the pace of AI dividend realization.
The most noteworthy variable to watch over the next seven days is the latest regulatory stance on capital expenditure guidance in the subsequent earnings reports of the four major cloud providers.
#现货ETF资金回流, can BTC and ETH take over? #比特币BIP-110 fork stalled, miner support insufficient最近和几位穿越过好几轮牛熊的老朋友深聊了几次，几个观点反复碰撞后，慢慢拼出了一张当下周期的生存地图。这张图的核心转变在于：市场已经彻底告别“听故事、炒预期”的浪漫阶段，全面进入“看现金流、验证落地”的硬核模式。换句话说，行情不再奖励想象力，而是只认那些能真金白银把价值装进持有者口袋的项目。以下这些心得，来自这几场深夜长谈，整理出来作为观察周期的参考框架，不构成任何买卖建议。 第一个生存法则：优先拥抱有真实价值捕获能力的资产。牛市里，市场愿意为一个宏伟叙事提前买单，但在熊市和震荡市里，唯一被认可的只有实打实的手续费收入和回购销毁记录。本周期真正的免死金牌，属于那些能持续产生费用，并且把这笔费用通过回购、销毁或分红返还给代币持有者的协议。你会发现最近走得比较稳的叙事，比如$UNI、$PUMP、$PONS这些发射台概念，以及被市场戏称为本周期回购之王的$HYPE，本质上都是这个逻辑的受益者。它们不靠情绪硬撑，靠的是每笔交易背后真实燃烧的gas和费用支撑。 第二个法则：只碰已经验证了产品和市场契合度的项目。接下来的周期里，如果不出意外，值得关注的叙事只会剩下两条主线——真实资产代币化和AI代理TSMC's July revenue surged 44.7% year-on-year, confirming the AI hardware boom, but depreciation pressure is reshaping risk appetite. The core conflict now lies in the position game between strong wafer shipments and the absorption of high valuations in tech stocks.
The nearly $2.4 trillion AI capital spending commitments from the four major cloud providers and Nvidia's $500 billion financing provide reliable support for fully loaded advanced process capacity. However, the indication that TSMC's stock price has pulled back 5% from its peak and that Q2 earnings may hit a temporary peak indicate that the market's pricing mechanism is shifting toward a strict calculation of capital expenditure returns.
From the perspective of event risk transmission, strong hardware shipments maintain the fundamental bottom line of tech stocks, thereby consolidating the defensive floor of overall risk assets. However, the depreciation pressure from the $60-64 billion capital expenditure guidance will short-term erode profit margins and suppress liquidity premiums, prompting high-leveraged positions to shift toward defensive assets.
Under a bullish scenario, if concentrated selling pressure is released and cloud vendors deliver returns on schedule, risk appetite will rebound. The trigger condition for this scenario is that TSMC's annual revenue growth rate remains above the 40% guidance limit, and after the pullback, the US tech sector completes full turnover. The failure signal is delays or cuts in supply chain orders.
Under a bearish scenario, depreciation in the early stages of 2nm mass production erodes profit margins more than expected, combined with concerns over overcapacity, suppressing cross-market risk appetite. The trigger conditions are a deep valuation adjustment in the U.S. tech sector and triggering deleveraging selling pressure, with core customers placing more orders than expected.
The overall invalidity premise of the above scenario is a significant reduction in the $2.4 trillion long-term capital expenditure commitments of global cloud providers, or a sudden drastic change in the macro liquidity environment.
In the next 7 days, focus should be paid to the position turnover rate and leveraged capital flows of the U.S. tech sector after the revenue data is released.
#CLARITY表决推迟至9月, the regulatory window has moved back to #标普收盘再创新高, and expectations for 8,000 points are heating up#本周三CPI公布, will the pricing for a rate hike in September be rewritten?
Nonfarm payroll data has sounded an alarm for the market, but it may also be opening the door to the next rally.
US nonfarm payrolls unexpectedly fell by 23,000 in July, and the combined May and June figures were revised down by 103,000, indicating that the US labor market is clearly cooling. The market quickly adjusted its expectations, the probability of a rate hike in September decreased, and funds began trading again as the Fed shifted its policy.
If Wednesday's CPI continues to prove inflation is cooling, pressure on the Fed to maintain high interest rates will increase, and expectations for rate cuts may heat up further.
Regarding Bitcoin $BTC, high interest rates have always been a major factor suppressing BTC. Once the market confirms the start of a rate-cutting cycle, dollar liquidity improves, institutional investors' risk appetite increases, and Bitcoin may become the first asset to benefit. But in the short term, don't blindly chase gains; the market often reshapes before expectations are realized.
For gold $XAU, weak employment combined with rate cut expectations provide support for gold. A weaker US dollar and falling US Treasury yields will both encourage continued allocation to gold. However, since gold has risen significantly earlier, it is also important to guard against short-term corrections after favorable momentum is realized.
For US $QQQ, if the economy achieves a soft landing, with rate cut expectations combined with AI industry earnings growth, tech stocks still have room to strengthen. However, if economic data deteriorates rapidly, market logic may shift from rate cut positive to recession trading, increasing volatility.
However, before the actual data is confirmed, I believe the most important thing is not to predict direction, but to control positions.
Short-term liquidity positions are kept to cope with market fluctuations, while medium- and long-term positions await confirmation of liquidity turning points.
Next, focus on CPI. If inflation continues to cool, the logic of rising risk assets may officially kick in → interest rate cut expectations improved liquidity → risk assets.
The above is just my personal opinion and does not constitute any investment advice!Miners Transform into AI Computing Power: Bitcoin's most robust narrative is being demolished by its own "miners."
Let's talk about the most ironic phenomenon this year: Bitcoin talks every day about the "world's strongest computing moat," but those who contribute to this moat are now working for AI in waves.
Let's first look at what happened. Listed mining companies have signed AI/HPC contracts totaling over $70 billion in the past two years. TeraWulf just confirmed in its Q2 financial report that it signed a long-term lease worth about $19 billion with Anthropic, with HPC revenue accounting for over 60% in Q1; IREN secured a $9.7 billion contract with Microsoft, and the treasury's Bitcoin holdings are zero—note, it's zero, which is a proactive choice; Even MARA, known as a "HODL" maniac, sold $1 billion worth of coins to buy GPUs. In the first quarter of this year, listed mining companies sold a total of 32,000 BTC, more than they sold for all of 2025.
Why? Because pure mining business can no longer sustain. After the halving, the block reward was cut to 3.125 coins, and the token price fell from a peak of $126,000 in October 2025. The hash price once dropped to $29 per PH/day, and the cash cost for a listed mining company to mine one coin soared to nearly $80,000. CoinShares estimates that 15% to 20% of mining machines across the entire network are operating at a loss. And what about AI? CoreWeave's data is straightforward: 10 megawatts of GPU hosting revenue equals 100 megawatts of mining. Even an elementary school student can calculate this amount.
As a result, the network's hash rate has slipped from a peak of about 1,160 EH/s in October 2025 to below 1,000 EH/s, with three consecutive negative difficulty adjustments at the end of last year, the first since July 2022. Previously, this was the classic bottom-fishing signal of "miners surrendering, near bottom." This time, the atmosphere is different—the portion of hash power that leaves isn't shutting down, it's moving, serving AI, and it's unlikely to return.
This brings us to the real comparison we want to discuss today: the divergence between $BTC and $ETH in the "security at the bottom" narrative.
Bitcoin's security is bought with real money, relying on miners continuously burning money. Its security budget = block reward × token price + fees. Now, block rewards are halved, price drawdowns, miners are shifting—three variables are putting pressure on each other at once. Charles Edwards warns that within two to three years, mining revenue will fall from about 90% to 30% of total mining company revenue. Of course, people like Adam Back will argue: if hash power decreases, difficulty will decrease, profit margins will stabilize, and eventually balance will be reached. That's true—the network won't die. But "won't die" and "security narratives remain unharmed" are two different things—a security model that survives on miners' profits loosens the premium logic once miners find better customers.
Ethereum was heavily criticized when it transitioned to PoS, with people saying it lost the "physical anchor" of PoW. Looking back now, that move actually seemed like defusing the bomb early. ETH's security relies on staked coins that don't rely on electricity, mining machines, or AI for data centers. Amid the wave of collective transformation by mining companies, ETH's security budget remains unchanged. The once mocked "moving from physical to virtual" approach has now become an immunity.
Ultimately, this round of differentiation teaches a simple truth: security models are also a business model and must withstand the test of cost-effectiveness. PoW security is built on real money spent on electricity and machines; its enemy has never been the 51% attack, but the tenfold rent offered by the neighboring AI data center.
Of course, don't turn this into a Bitcoin obituary. The difficulty adjustment mechanism will self-repair; national miners and private miners with extremely cheap electricity are filling the pit, and the survival pure mining companies will face less competitive pressure. But there's an old script that really needs to be torn up: next time you see "hash rate crash = miners surrender = bottom-fishing signal," don't rush to pull the trigger. First, see if those machines lost money to shutting down or were rented out by Anthropic and Microsoft.
To understand Bitcoin's cybersecurity in 2026, you first need to understand the supply and demand table of AI data centers. The two worlds have already been welded together.Watching BTC fluctuate at high levels, thinking about niche coins fluctuating within ranges to do contract grid arbitrage, with 5x low leverage for a steady and easy profit, but then a single coin broke out and plunged, swallowing all grid arbitrage profits and even causing losses. Here's a good lesson for everyone to review. SKHYNIXUSDT perpetual 5x long grid running for over 27 days, cumulative investment of 300U, total return -117.87U, loss 39.29%; Grid arbitrage earned 103.69 USD, but the coin's one-sided decline led to an unpaired floating loss of 221.57 USD, directly wiping out the gains and causing a significant drawdown. The strategy range was 850-1650, starting price 1254.47, current price dropped to 1002.92, deeply breaking below the opening benchmark price. In a one-sided downward trend, the grid declined passively and losses continued to expand. Currently, BTC is fluctuating in a range, and many small-coin counterfeit coins lack capital support. Once the market pulls back, it is easy for the market to break out of a standalone one-sided decline. Many people mistakenly believe that grid trading is a blind win and blindly opening, ignoring the liquidity, fundamentals, and overall trend of the target. In a volatile market, repeated grid trading and price differences are very attractive. Once the price breaks and the side weakens, passive grid positions will only get deeper and deeper. When mainstream synergies are weak, don't blindly adopt grid strategies for niche knockoffs. Currently, BTC is under pressure and volatile, with intensified altcoin market differentiation. Will you prioritize grid allocation only for mainstream coins, or simply avoid small-cap coins and wait and see? Share the pitfalls 👇 your grid has fallen into in the comments. #本周三CPI公布, will the September rate hike pricing be rewritten? #存储股抛压缓和, is the AI memory bull market still stable? #现Major news has been released! Positive?
At 20:30 Beijing time tomorrow night, the non-farm payroll will be implemented, and US stocks are set to face a key decision
At 20:30 this Friday evening, the July nonfarm payroll report will be released. This is the most important employment data since the Federal Reserve's July meeting, and it will directly rewrite September rate expectations, affecting all assets in US stocks, Treasuries, and cryptocurrencies.
Previously, ADP's small nonfarm payroll data was clearly below expectations, giving the market an early warning as employment gradually cooled.
The three data scenarios correspond to the U.S. stock market trends
Scenario 1: Nonfarm payrolls are significantly stronger than expected, and wages rise in tandem
Strong employment will delay rate cut expectations, pushing U.S. Treasury yields higher. High-valuation AI technology and storage sectors are under the heaviest pressure, while growth stocks like MU and SNDK are prone to selling pressure; Dow blue chips are relatively resilient to declines, and the overall index shows divergence.
Scenario 2: Nonfarm payrolls weaken significantly, unemployment rises
The market will strengthen expectations for rate cuts, U.S. Treasury yields will fall, which is positive for tech growth stocks. Storage and AI hardware have the opportunity to see a recovery and rebound. But one risk must be watched for: if the data is too poor, it could trigger market concerns about an economic recession, leading to a short-term broad drop.
Scenario 3: Data and expectations basically match
Employment cooled mildly, neither hot nor lukewarm. U.S. stocks continue the current tear-off pattern, with the Dow slightly strong, the Nasdaq oscillating at high levels, the market returning to earnings report logic, and sector rotation continues.
Putting aside nonfarm payrolls, the U.S. stock market will be the next outlook
1. The storage sector is currently in a phase of intense volatility following the financial report falsification. SNDK has made a deep V reversal, but the issue of downward expectations brought by the earnings report has not completely disappeared. Looking ahead, focus on whether the key support in MU can be held; holding it will mean sector differentiation and recovery; Once it effectively breaks below the threshold, the current round of storage will enter a mid-term valuation digestion phase. Don't treat the oversold rebound as a new main rally.
2. Structural market differentiation will continue to unfold. Stocks whose guidance exceeds expectations will continue to enjoy premiums; Even if profits are high, companies with conservative shareholder returns and future guidance will continue to be abandoned by capital. The broad rally has ended, making stock selection more difficult.
3. Risk points cannot be ignored. $SPCX massive unlocking pressure remains, which will occasionally disturb the market and amplify the spike volatility.
Key Targets to Watch:
$MU • $SPCX • $SNDK • $SKHY • $CL • $XAU • $NITC • $AMD
Stocks with Weakening Momentum and Capital Exits:
$BEAT • $EDGE • $COAI • $TRUMP • $RAVE • $SPACE • $SOPH • $IP • $AVNT • $ZAMA • $OFC • $PIEVERSE • $VIRTUAL • $ACU • $H • $MEGA
Waiting for signal confirmation in the observation pool:
$MEME • $EDEN • $HUMA • $ZKP • $METIS
Strong stocks favored by capital:
$JTO • $JELLY • $BTC • $OPG • $BTCSLX • $LAB • $BSB • $ALLO • $CHIP
Current market logic summary:
$BTC — The liquidity center in the crypto market, which determines the overall temperature of the market
$ETH — Institutional funds continue to build positions, gradually accumulating shares through volatility
$SOL — The elastic role of the Layer 1 sector, with considerable upside potential at market launch
$TAO & $WLD — AI remains hot and repeatedly favored by capital
$HYPE — A market speculative sentiment gauge used to assess current risk appetite
$DOGE & $ZEC — Retail investor sentiment window, intuitively reflecting short-term speculative heat🔥Robinhood has finally brought crypto trading to the UK, a process that took a full two years
Robinhood officially launched cryptocurrency trading services for UK users this week, offering trading of over 50 crypto assets including BTC, ETH, XRP, and Hyperliquid through Bitstamp UK. Zero trading fees, no account management or custody fees.
This calculation is quite interesting—Robinhood spent $200 million to acquire Bitstamp and now has finally opened the door to the UK market with it.
Why now?
On July 31, Robinhood UK Ltd officially entered the UK FCA's crypto asset registration list. This move has been in place for a long time—in 2024, Robinhood's CEO publicly criticized the UK's "regressive" stance on cryptocurrencies. Now, having obtained compliance qualifications is equivalent to having a formal entry ticket.
Additionally, the FCA has just released the final draft of its crypto regulatory framework, and the licensing system will officially take effect in October 2027. Robinhood is rushing to launch before the window period, seizing the time window.
Compared to competitors, their approach is different
The UK market is already packed—Coinbase, Kraken, Revolut, eToro, none of which are easy to deal with. As of June, Coinbase held $246 billion in assets, accounting for 10.3% of the trading volume share.
Robinhood's differentiation strategy is clear: one-stop solution. Users can trade stocks, ISAs, options, futures, and also buy and sell cryptocurrencies in one app. For UK retail investors who trade both stocks and cryptocurrencies, the appeal of adding one less app is real.
It also comes with an AI tool called Cortex Digests, which analyzes news and market data in real time, explaining the reasons behind price fluctuations in plain language. This feature is quite beginner-friendly, but whether it becomes a core competitive advantage for retaining users depends on actual experience.
Robinhood Chain's data is actually quite interesting
Since its global launch on July 1, Robinhood Chain's DEX trading volume has exceeded $18 billion, with TVL surpassing $840 million. A traditional brokerage transforming into a Layer 2 has achieved this figure in just two months, indicating that user acceptance of "on-chain Robinhood" has exceeded expectations.
But there is a practical issue that needs to be clarified
Crypto trading is provided through Bitstamp UK and is not protected by FSCS or FOS. If UK retail investors' money goes into trouble, there is no government back. Although Robinhood's brand endorsement can alleviate some concerns, it still lacks the level of protection compared to traditional stock accounts.
Additionally, although trading is fee-free, UK users have to pay a 0.1% foreign exchange fee, with weekend transactions rising to 0.3%. Though free on the surface, there are costs upon closer inspection—whether this pricing strategy will be accepted by UK users remains to be seen.
What does this mean for the crypto market?
Robinhood's entry into the UK has most directly lowered the barrier for retail investors to access crypto assets. Zero commissions, one-stop operations, and familiar brands—these three factors combined will bring a wave of new user growth in the short term.
But Robinhood itself has its own problems—crypto trading revenue in the first half of the year fell 43% year-on-year to $234 million. Although the UK market offers new growth points, whether it can reverse the downward trend remains uncertain.
From a broader perspective, Robinhood is seizing the window before the FCA's official licensing system is implemented. Once the new regulations fully take effect in October 2027, compliance costs will rise sharply. If you don't enter now, the threshold will only rise in the future.
After two years of moving the chess game, it finally made its move. For British retail investors, it gained another option. For Robinhood, this was a key step in its transformation from a domestic US brokerage to a global financial platform.
👇 Friends in the UK, will you use Robinhood for cryptocurrency trading? Let's talk in the comments.
$BTC $ETH 🔥 TSMC's July revenue soared again—how much longer can AI survive?
TSMC has just released a report card that silences the bears.
In July, consolidated revenue reached NT$467.58 billion, equivalent to USD 14.5 billion, a year-on-year surge of 44.7% and a month-on-month increase of 5.6%, marking the third consecutive month of record-breaking single-month records.
Cumulative revenue for the first seven months of this year reached NT$2.87 trillion, a year-on-year increase of 37%. Last month, TSMC raised its full-year revenue growth forecast from 30% to over 40%, and its capital expenditure guidance reached a historic high of $60-64 billion.
AI is a livelihood we can't finish for now.
NVIDIA and Apple are TSMC's two largest customers, and their orders have pushed advanced process capacity to the limit. 2nm, 3nm, and 5nm are fully loaded, with demand for AI accelerators, custom chips, and flagship smartphone processors heating up simultaneously.
The world's four major cloud providers—Google, Meta, Microsoft, and Amazon—have already pledged to invest nearly $2.4 trillion in AI over the next few years. Most of this money will ultimately flow into TSMC's wafer fabs.
TSMC believes that the strong growth momentum of AI chips will continue into 2027 and beyond.
But the market is not without disagreements.
TSMC's stock price has pulled back about 5% from its peak at the end of June, but its cumulative gain this year still exceeds 50%. It's normal for some to take profits after a large rise; what really matters is what the market is hesitating about.
First, concerns about data center overcapacity. Nvidia just announced a 500 billion yuan infrastructure financing round, and the four major cloud companies have pledged 2.4 trillion yuan, but whether investing in these funds will yield real returns remains uncertain.
Second, TSMC itself warned that Q2 earnings may have reached a temporary peak. Large-scale capital expenditures consume cash flow, along with depreciation pressure from the early stages of 2nm mass production, will erode profit margins in the coming quarters.
What does this mean for the crypto world?
First, demand for AI hardware hasn't collapsed, so this is at least not a bad news for AI concept coins. Nvidia's GPUs are not a problem of selling, TSMC's wafer orders are not a problem, and the overall AI industry chain remains highly prosperous. Projects with real technology implementation and the benefits of AI infrastructure remain largely unbroken.
Second, don't expect a single monthly report from TSMC to take AI concept coins to the skies. Market enthusiasm for AI has shifted from "storytelling" to "settling accounts" mode. The fact that TSMC's stock price pulls back 5% from its peak itself shows that good earnings are expected; what truly drives the market is whether earnings can consistently exceed expectations.
Third, for BTC, TSMC's monthly report is more of sentiment support. As long as the earnings fundamentals of US tech stocks haven't collapsed, the Nasdaq won't experience a systemic collapse, so BTC's lower bound as a "tech stock Pro Max" is relatively manageable.
How much longer can AI survive? TSMC's answer is: at least until 2027.
But food is food, and pots are pots. No matter how fragrant the rice in the pot is, it depends on whether you can handle it yourself.
👇 How much longer do you think the AI chip boom cycle can last? Does the AI concept in the crypto world still have a chance? Let's talk in the comments.🔥革命卫队前总司令掌舵国安会，伊朗这步棋传递了什么信号？
8月9日，伊朗总统佩泽希齐扬签发总统令，正式任命71岁的穆赫辛·雷扎伊为最高国家安全委员会秘书。同一天，最高领袖穆杰塔巴·哈梅内伊也任命雷扎伊为自己在该委员会的代表。
雷扎伊的简历，是这份任命最值得琢磨的地方。1981年至1997年，他担任伊朗伊斯兰革命卫队总司令长达16年。1997年至2021年，担任确定国家利益委员会秘书长。2021年至2023年，又担任分管经济事务的副总统。
从执掌军队，到协调顶层政治，再到分管经济，雷扎伊的履历覆盖了伊朗决策层最核心的几个维度。
但真正让市场需要留意的，是他7月说过的一句话。
据伊朗伊斯兰共和国通讯社报道，雷扎伊当时明确宣称：“控制霍尔木兹海峡比数十枚原子弹还要重要”。
这句话放在当前霍尔木兹海峡局势持续紧张的背景下，分量很重。这个占据全球约五分之一石油贸易量的咽喉要道，至今仍处于实质性封锁状态。而说出这句话的人，如今成了伊朗最高安全机构的掌舵者。
这次人事变动发生在三个背景之下：
首先，伊朗正面临内外压力。特朗普刚表态对伊转入“半谈判”的经济消耗战模式，美军海上封锁持续，国内通货膨胀高企。
其次，前任秘书佐勒加德尔今年3月才接替在美军空袭中身亡的拉里贾尼，任职不到半年即转任最高领袖政治顾问。
再者，雷扎伊同时身兼“秘书”和“最高领袖代表”两个职位——既由总统任命、又受最高领袖直接委派，身份的特殊性意味着他在国安会的决策话语权可能比前任更大。
对市场的影响，可能分三个层面传导：
第一，油价的地缘溢价短期难以消退。 一位公开宣称霍尔木兹海峡“比原子弹重要”的强硬派人物掌管国安会，意味着伊朗在航道问题上的立场出现实质性松动的概率不大。
第二，美伊谈判的复杂性在上升。 特朗普自称“半谈判”状态，而伊朗这边换上一位革命卫队出身、经历过两伊战争的老将主导安全决策，双方的信任成本和博弈烈度都可能增加。
第三，对加密货币市场的间接影响。 油价若因地缘因素维持高位，将直接推升通胀预期，压缩美联储政策空间，进而对BTC等风险资产形成宏观层面的压制。
伊朗这次人事调整，传递的信号很清晰——国安会正在由一位经历过战争、掌管过经济、且对霍尔木兹海峡有明确强硬立场的元老级人物主导。在霍尔木兹海峡仍处封锁状态的当下，这步棋对油价和全球风险资产的潜在影响，值得持续跟踪。
👇 你觉得雷扎伊上任后，霍尔木兹海峡的局势会更紧张还是可能出现转机？评论区聊聊。#闪迪8月13日投资者日临近，财报分歧待解 SanDisk $SNDK tonight surged straight up to 1277, nearly a 100-point increase. The volume-backed rebound at this level indicates that around 1200, there are buyers willing to take positions, and market sentiment is gradually recovering from pessimism. The core catalyst is the Investor Day on August 13. The biggest issue with the earnings report is the weak guidance; the market has been worried whether this is due to conservative management or if AI storage demand is truly $XSNDK $BEAT SNDKUSDT: After dropping 47%, I actually started to take it seriously.
SNDK's recent trend has really been quite interesting.
In June, the highest was over $2,300,
Then they slaughtered along the way.
At its peak, the maximum drawdown was close to 47%.
Many people's first reaction is:
The AI hype is over.
Storage cycles have peaked.
SNDK's wave is completely over.
But after reviewing the recent financial report again, I actually felt things might not be so simple.
Because the strangest part is right here:
The stock price fell sharply, but the company's fundamentals did not collapse along with it.
In fact, quite the opposite.
⸻
SNDK's latest quarter:
Revenue was $8.97 billion, a year-on-year surge of 372%.
Adjusted EPS is $39.25.
The data center business alone reached nearly $3 billion.
Moreover, the median revenue guidance for the next quarter is about $10.55 billion.
What does this mean?
The market ahead is trading:
"AI → Storage Demand Explosion → NAND Supercycle → SNDK Valuation Reassessment."
So the stock price once surged to over 2300.
Later, the market suddenly began to worry:
"Is it going up too fast?"
Thus, the price was appraised and the price was slashed
One kill is nearly halfway through.
⸻
But what really stands out:
Demand for AI data centers has not disappeared despite the stock price decline.
Sandisk disclosed that it has signed multiple long-term supply agreements, eight of which involve six customers, with a total contract value of about $93.9 billion and an average term of about four years.
According to the company, by the end of fiscal year 2027, about half of capacity will have been covered by these long-term agreements, with further increases expected in fiscal year 2028.
So now comes the most interesting part of SNDK.
The market is pricing it as a "cyclical stock."
But the company is working hard to transform itself into:
"Long-term AI infrastructure supplier."
The two valuation logics are completely different.
⸻
So now, I actually don't care much:
"Can 2300 come back?"
I'm more concerned about one question:
Around 1200, is this a rebound relay, or the starting point for a new round of decline?
Currently, SNDKUSDT is oscillating around 1200–1230.
This position is very sensitive.
If it can regain above 1200 and break through previous resistance with increased volume,
Then the market is likely to resume trading:
AI storage supercycle.
At that point, the ones suffering most won't be the bears.
Instead, it is those:
"Wait until it drops to 800 before buying."
⸻
Of course, the reverse is also true.
The biggest risks of SNDK are very clear:
It has already risen too much.
Since 2026, the gains have remained extremely exaggerated, and even after a recent sharp drawdown, valuations remain high. The latest market reports also point out that one of the biggest concerns right now is whether high valuations can match future profit growth.
So I won't blindly shout here:
"SNDK will double immediately."
Such words are meaningless.
What I really want to say is:
Now, SNDK has shifted from "chasing rallies" to "studying pullback value."
These two stages are completely different.
⸻
My own observation points:
1200: The first crucial defense.
Around 1300: Short-term sentiment watershed.
If volume surges again and breaks through the aboveward resistance,
So the previous round of valuation cuts was likely just a massive shakeout.
But if 1200 completely falls,
Then caution is needed as the market continues to restructure the valuation of the AI storage sector.
⸻
So at this position, I have only one question:
SNDK fell from 2300 to 1200—do you think it's a "bubble bursting" or a "gold pit"?
I'm more curious to see how many people in the comments dare to go long at this position.
If you're bullish, answer 1.
If bearish, play 2.
Let's see whose faith SNDK really is.
These are personal market views only and do not constitute investment advice. SNDKUSDT is a high-volatility stock perpetual contract, and leveraged trading carries extremely high risk.