Orbit Post Sitemap
Target locked, Huang Renxun's capital trajectory has shifted.
The readings in the rangefinder have changed. Holding $21 billion worth of SpaceX equity, yet compressing the first-round guarantee for OpenAI's Ohio data center from 250 billion to 120 billion — this is not a retreat, but an adjustment of the bunker. Equity is long-range ammunition, credit is the exposed flank. When a 400 billion GPU order needs to be backed by the company's own balance sheet, this battlefield is no longer just about buying computing power, but a crossfire network exchanging equity for demand and guarantees for exclusivity.
Wall Street watches the debt ceiling, I watch the trajectory deviation angle. Nvidia's move is essentially a sniper's approach to positional warfare: delivering equity ammunition precisely from a distance, while shrinking credit exposure up close. GPUs are bullets, data centers are magazines, customer financing is the rifling. When the rifling wears out, no matter how fast the bullet flies, it won't hit the target accurately.
But from the scope, the term "revolving financing" indeed looks like a sign of trajectory deviation. If a significant portion of the money customers use to buy chips comes from Nvidia's own guarantees or investments — then this impressive revenue data is like shooting through fog; the bullseye looks close, but the actual impact points are all on the outer ring. Market momentum indicators like XMeta, HBM inventory, and even the so-called monthly peak are just paper targets on the shooting range, trembling with the wind.
I've observed enough cycles through the scope to see too many companies quietly deepen leverage under the halo of growth. Equity is durable goods, credit is consumables; using consumables to exchange for durable goods means short-term cash flow losses, betting long-term on whether the customer can hold their ground on the battlefield. The shrinkage of OpenAI's data center guarantee may mean the shooter’s judgment of the wind direction has changed — the wind is coming from the left front, humidity increases, and the trajectory correction must be greater than expected.
As for whether Nvidia's equity, guarantees, and financing chains can convert computing power demand into lasting returns? For now, only two things are certain: one, the GPU trajectory remains straight; two, the mud spots on the bumper are clearer than anyone's eyes. When pulling the trigger, I only need to know one thing — whether I am controlling the trajectory, or the trajectory is controlling me.
Retract the ruler, the muzzle dips slightly. I see those dense ribbon cables sliding in the dark compartments, golden handcuffs locking the chips, while the audience's eyes are all fixed on the glowing screen at the center of the stage—they think that's the truth.
SK Hynix pulled out the 180 trillion won it invested in factories in the first half of the year from backstage, 70% more than the same period last year. Thunderous applause erupted below, but the real hustlers all know this is just the climax of a sleight of hand. HBM, advanced packaging, NAND capacity—these intricate mechanisms are indeed setting the stage for the AI server show, but remember, any magician showing off expensive props is secretly eyeing your wallet.
Capital expenditure is never a promise; it's hypnosis. It uses the narrative of technological leadership to pin the spotlight on the stage, making everyone forget that the real game is played with the three hole cards of orders, utilization rates, and memory prices. Capacity ramps up stepwise, like the fan slowly unfolding in my hand, each rib creating the illusion of "continuous flow." But when the fan closes, whether the applause can be exchanged for cash flow depends on whether AI demand is willing to be a long-term babysitter.
Retail investors stare at that 70-degree year-over-year rising curve like they're checking for a safety rope on a tightrope. Meanwhile, I focus on the light leaking through the hand-sewn seams—when revenue cycles become fuel for reinvestment, when every quarter's profit is swallowed back into the furnace, is this compound interest magic or just stacking chips into a bottomless pit? If demand and pricing can't fill utilization, these shiny machines are the world's most sophisticated nooses.
Nvidia's HBM architecture remains solid; that's the only trump card worth revealing today. But it protects the star on stage, not you.
Magicians don't care how expensive the props are; they only care when the next audience member takes their seat. And cash flow is just the hand speed that lets me perform the next act.当K线在亚盘午盘逐渐收缩，波幅的收窄往往是变盘前的蓄力。在消息面处于真空期的当下，我们剥离外部叙事，纯粹从量价结构与筹码分布来拆解 $BTC 的技术形态。 ══════════════ 📌 【$BTC 价格与趋势】$63,051 | 24h -0.38% | 7d -2.9% 大饼目前运行在周线级别的回调结构中。7日累计近3%的跌幅，表明中期均线系统正在向下发散。当前价格紧贴63K整数关口，日线级别呈现典型的收敛三角形态，亚盘时段的横盘并未改变短期的弱势基调。 📌 【量价关系与成交量】24h成交量 $640.5亿 在全市场总成交量大幅放量58.67%的背景下，$BTC 自身的成交量并未出现等比例放大。这种大盘放量而大饼缩量的背离，说明当前63K附近的盘整属于缩量抵抗，而非主力资金的大举建仓。没有成交量的配合，向上突破的动能严重不足。 📌 【关键支撑与阻力位】 短期核心支撑位：$62,000 - $62,500。这是前期密集成交区，若跌破此区间，将打开向下寻找$60,000心理关口的空间。 短期核心阻力位：$64,500 - $65,000。上方均线压制明显，需等待放量长阳才能确认SOL 的熱門數字不難讀，難的是別把語氣和資金方向混在一起。 OKX Onchain OS 於 08 月 15 日 11:00 統計到 SOL 一小時 26 次提及，其中 X 26 次、新聞 0 次；二十四小時總量為 465 次。 最新一小時相當於長窗每小時平均的 1.34 倍，也就是比二十四小時的每小時平均高約 34%，可歸為「略有加快」。這個速度描述的是新增討論，和行情漲跌沒有必然關係。 文本語氣則是偏多 62%、偏空 4%、中性約 34%，目前屬於「偏多明顯佔優」。二十四小時偏多 52%、偏空 8%；兩個窗口若出現差距，應先理解為討論結構在變，而不是直接推導價格目標。 我會把這兩條線分開畫。語氣偏多、提及速度卻放慢，代表現有討論比較正向，但新注意力沒有加速；提及速度上升、偏空又佔優，則可能是風險或故障消息把人吸引過來。就算熱度和語氣同向，也還不能直接等同真實買盤。 來源是另一項限制。目前 SOL「幾乎全由 X 驅動」。社群渠道反應最快，同一個話題也可能被重複轉發；來源越集中，越需要下一個窗口確認。新聞提及增加也不自動等於事件屬實，原始公告仍是最後的查證基準。 二十四小時內，SOLBTC is highly likely to pull back to the two major supports at 62500 and 62150 in the short term, with today's resistance at 63200.
👉 Enter long positions around 62555, stop loss at 62000, target 63000‑63300, mainly for short-term profit-taking.
63450 is an important dividing line between bulls and bears:
Once effectively broken, the short-term trend turns bullish, and long positions can be added accordingly; this level also serves as the stop-loss line for short positions.
Before a breakout signal appears, the market is defined as ranging, so avoid chasing a one-sided rally with the trend. $BTC #加密估值转向收入，BTC如何定价？ Actually, I've always thought Musk was a big talker because many of the things he said before were delayed.
After seeing chopsticks holding a rocket, I felt he might actually have something.
Let's talk about SpaceX. Actually, when SpaceX went public, many TSLA bloggers I follow clearly said they would buy a little and keep an eye on it, which basically means they are not optimistic and won't hold heavily.
Because it's simple: space exploration is not profitable and can't support a $2 trillion valuation.
But recently, I seem to have gotten Musk's thinking.
If there are ten gigawatt data centers, $2 trillion doesn't seem overvalued. It might even be reasonable.
Can he build it? The first two data centers have already proven it.
Don't forget, Jensen Huang once praised Musk's efficiency.
So I think SpaceX as a company needs to be revalued. It's not just about space exploration or interstellar migration. Putting aside these grand narratives, what they're really doing is a super data center, building data centers with the efficiency of a rocket team.
Based on this, let's see if it can be realized.
#马斯克称AI将占SpaceX价值99% Recent data shows that the U.S. consumer side is starting to falter a bit, with retail sales and consumer momentum clearly weakening. Normally, with the economy cooling and demand softening, the Federal Reserve should be easing to support the market, right?
But inflation data hasn't fully dropped to a level that reassures the Fed, and the internal hawks are still closely watching prices.
Consumption is weakening, but policy is tied up by inflation. This awkward situation, where the economy faces downward pressure but policy dares not easily shift, directly leaves the entire risk asset market expectations hanging in midair.
So what does this mean for $BTC and $ETH?
First, liquidity expectations have no "surprises" in the short term.
Everyone was originally hoping that with weaker consumer data, the Fed would directly ease and offer support at the September meeting. But the shadow of inflation lingers, and Fed officials' back-and-forth continues endlessly.
Without the expected "big liquidity package," incremental off-exchange funds naturally dare not enter the market on a large scale. The current market is entirely reliant on existing funds battling it out.
Second, BTC resists pressure, while Ethereum lacks resilience.
In this uncertain macro squeeze, BTC, supported by spot ETFs and its own safe-haven attributes, can barely maintain a sideways consolidation, with overhead resistance from trapped positions and long-term funds supporting the bottom.
Ethereum's situation is not so good. Cooling consumption means a decline in on-chain activity and speculative enthusiasm. Without a macro liquidity surge or strong on-chain narratives, Ethereum struggles to break out independently amid macro data fluctuations. Everyone is waiting for the wind, but the wind is scattered by the divergence between inflation and consumption.
In terms of allocation, focus more on BTC, keep Ethereum and altcoins light and watchful before liquidity truly loosens, and control drawdowns well.
#消费动能转弱，9月政策仍受通胀制约 The US ban is expected to compress Apple's storage chip supply chain, passively increasing Micron's bargaining power, but macro inflation pressure and the risk of slowing AI capital expenditure growth are limiting the risk appetite of long positions.
Washington's discouragement of Apple purchasing Chinese storage chips directly narrows the buyer's alternative pool, thereby raising the order certainty for suppliers like Micron. The primary factors driving market pricing currently are policy access restrictions, followed by AI's competition for capacity, and finally the expectation of contract price increases.
Policy intervention transmits through supply contraction to inflation expectations. If procurement costs are forced to rise, the profit margin of end consumer electronics will be squeezed. This causes market risk appetite to diverge within the tech sector, with capital positions beginning to concentrate on targets with monopoly premiums.
The trigger for the upward path is Apple placing additional orders with Micron. If $MU price holds above $953, it indicates the market has priced in the policy premium. At this point, long positions may push toward the $1012 resistance level; if it falls below $935, this upward scenario is invalidated.
The trigger for the downward path is AI capital expenditure signaling a slowdown in Q4, leading to a contraction in overall market risk appetite. If the price falls below $953, long position liquidations will accelerate the price decline to the $900–$920 range; if the price breaks above $1012, the downward scenario is invalidated.
The most critical observation variable in the next 7 days is whether the US will escalate its discouragement into formal regulation, and the turnover of Micron's chips at the $953 support level.
#特朗普因TruthSocial付费数据流遭起诉 #AMD完成历史最大美元债发行：融资47.5亿美元 #财报观察员：AI基建财报接力登场The reason why Btc hasn't been able to rise these days has been found.
This morning it was 63,016, the lowest point at midnight dropped to 62,667, the rebound is negligible. ETH 1,882, SOL 75.6, among the three brothers SOL fell the most today.
ETF has had net outflows for three consecutive days, and the price has been suppressed like this.
Yesterday's total net outflow was $57.63 million, with BlackRock IBIT alone running off $55.5 million. Last week when $853 million flowed in, BTC was pulled from 62,000 to 65,000; this week, with four trading days and $332 million outflow, the price directly returned to the starting point. The market is extremely sensitive to ETF capital flows, more effective than any macro data.
The liquidation data also looks bad. In the past 24 hours, $166 million was liquidated, with long positions accounting for $96.42 million. In BTC liquidations, longs are nearly 7 times the shorts, and ETH is 3 times. Every rebound sees someone bottom-fishing, but every time they get pushed back—buy orders are invisible, sell orders are not urgent, the market weakness is obvious at a glance.
Liquidity is thin on weekends, most likely it will just move sideways like this. After breaking 63,000, there has been no decent rebound; 62,700-62,850 is short-term support, breaking it means 62,000. Above, 63,300-63,500 has already become a resistance wall.
Not watching during the day, will wait until Monday to talk again. For platforms, no matter which one, they all rely on trading volume. Whether or not there are long-term projects doesn't significantly impact the platform. What platforms want is a continuous influx of new attention-driven projects, not a few scattered projects lasting longer. Is it easier to expect a project to maintain high daily trading volume for 2 years or to have new projects popping up every day for 2 years? So now platforms don't care if a project is long-term; as long as there are constantly new projects launching, whether they last three days, five days, or 3.5 hours, it doesn't matter. I remember some platform even encouraged devs to launch more projects recently. This is the underlying driver. So retail investors should stop expecting long-term projects. No one has the patience anymore. Just accept that the survival time of meme markets will inevitably get shorter, and slowly adapt within this rule.$BTC is in a painful transition.
US stocks are rallying while Bitcoin falls below $63K, with ETF outflows adding pressure. But this may be more than a simple fund rotation—BTC is being repriced as macro conditions change.
With yields high and Fed cuts delayed, patience matters.
Volatility is compressed, options still show interest around $70K, and the next major catalysts are Fed policy and the Clarity Act.
For now: less emotion, fewer moves, more patience.
$BTC $ETH
#WeakConsumptionFedSplit Volatility and Sharpe Ratio Comparison
$ETH has an annualized volatility of 26.1%, more than twice that of $BTC's 11.5%—high volatility can be both bad and good. Bad because it can be hard on the nerves, but good because this elasticity means it can move faster than $BTC when going in the right direction. Sharpe ratios are -5.21 vs -17.63, meaning for the same downside risk, $ETH loses less money per unit of risk than $BTC, making $ETH the more resilient one after risk adjustment.
Capital Attraction Comparison
Funding rates are expensive on both sides: $BTC at 0.0068%, $ETH at 0.0060%, with longs effectively paying shorts. But note that this week $ETH's rate climbed from 0.0019% to 0.0093%, growing faster than $BTC's—money is moving into $ETH. Coupled with a record high staking amount of 41.7 million tokens, $ETH's chip lock-up effect is strengthening. Meanwhile, $BTC's ETF inflows of $930 million haven't been able to push the price, indicating smart money is switching sides.Good afternoon, BTC is around 63,150, up slightly by 0.26% in 24 hours. The highest point today was 63,406, the lowest 62,521, another day oscillating within a 1,000-dollar range. The price is still just above 63,000, with MA5 pressing down at 63,321, MA10 and MA20 at 64,075 and 63,949 respectively, all lines breached.
CPI and PPI are cooling down, the probability of a rate hike in September has dropped to 32.4%. Macroeconomic data looks good, but the market just isn't buying it. Over the past week, BTC fell from 65,000 to 62,500, a drop of nearly 4%. Whether the good news is fully priced in or there are no new stories, the result is capital is withdrawing.
ETF net outflows have continued for four consecutive days, totaling $332 million. Yesterday, BlackRock's IBIT saw a single-day outflow of $55.5 million. Although there is still a monthly net inflow of $521 million, buying momentum is indeed weakening. Geopolitically, the Strait of Hormuz is still tense, oil prices remain above $87, and inflation expectations cannot be suppressed. BTC options market skew is near the lowest level of the year, traders are extremely pessimistic about an upside. On the US stock side, SPX, NDX, and MSCI global indices are close to previous highs, but BTC has not kept up.
The 63,000 level has been defended several times. The recent low near 62,500, if broken, could see 60,000 or even 57,800. The 64,000 level has become resistance. Bitwise says the market has become desensitized to bad news, which could be a bottoming signal—but bottoming is usually a long process, not an overnight event.
To be honest, this position is quite awkward. Neither up nor down, I choose to wait and act once the direction is clear.
Personal opinion, not investment advice.
$BTC $ETH $BTC vs $ETH One-Week Earning Ability Comparison
In this round, $ETH has thoroughly outperformed $BTC, no denying it. Over the past 7 days, $BTC dropped 3.31%, while $ETH only fell 2.26%. Both lost money, but $ETH lost with more dignity—after bottoming out on 8/10, it climbed steadily for three consecutive days, whereas $BTC was still struggling in the mud. The Sharpe ratio crushes it: $BTC at -17.63 means "money spent, pain endured," while $ETH at -5.21 is at least "a loss with value." In short: this round is about who can endure better, and $ETH wins.
Comparing Returns and Drawdowns
Looking at the curves: on 8/9, both took a dive, with $ETH falling harder than $BTC (97.45 vs 98.58), but after hitting a solid bottom at 1852 on 8/10, $ETH immediately bounced back, recovering 1.5% in one day on 8/11; $BTC? It kept sliding down, hitting a 7-day low on 8/13. Drawdowns are roughly the same (3.4% vs 3.3%), but "falling and bouncing back" versus "falling and staying down" are two different species. $AMD JUST RAISED $4.75B — ITS LARGEST USD BOND DEAL EVER.
AMD has completed the biggest dollar-denominated bond issuance in its history, raising $4.75B across four tranches with maturities ranging from 3 to 10 years.
The 10-year notes came with a 5.5% coupon, reportedly 25 bps below the company’s initial guidance after strong demand from 16 Wall Street institutions pushed borrowing costs lower.
What’s interesting is that AMD doesn’t appear to need the cash urgently.
The company reportedly has around $13.1B in cash against just $3.2B of debt. But with $875M of debt coming due next month and a planned $5B investment in Anthropic, AMD is clearly building liquidity ahead of major commitments.
And AMD isn’t alone.
NVIDIA raised $25B in June, while Google issued another $25B in early August.
Big tech is increasingly tapping the bond market to secure capital for AI infrastructure and expansion.
AMD’s choice to issue debt rather than shares also avoids immediate shareholder dilution — effectively using leverage to fund its AI ambitions and compete for more market share.
The AI arms race isn’t slowing down. 💰🤖
#OpenAIAnthropicRace
#WeakConsumptionFedSplit
#NvidiaAICapitalChain July retail sales month-over-month, I was stunned when I saw the numbers.
The expectation was at least a 0.1% increase, but it actually dropped by 0.6%, a 0.7 percentage point difference, a cliff-like plunge. Consumption willingness has weakened to this extent, how can prices still rise? CPI and PPI just fell, retail is collapsing next, I estimate the probability of a rate hike in September has dropped below 30%.
But the crypto market didn’t follow. Bitcoin dropped to 62,790, Ethereum hovered around 1,875, no rally despite the good news, liquidity is too thin.
SanDisk actually rose nearly 4% pre-market, JPMorgan just upgraded to overweight with a target price of 2250. The logic is that the worse retail performs, the closer the rate cut, and long-duration assets like AI hardware benefit more. But honestly, I have doubts—consumption is collapsing like this, how long can AI capital expenditure remain unaffected?
My strategy: reduce gold positions near 4400 and wait for a pullback, buy again below 4350. Bitcoin must hold above 62,000 before considering entry; entering now is no different from catching a falling knife. SanDisk’s logic is sound but the stock price has already priced in expectations, wait for a correction. The real good opportunity is next week, no rush. $BTC $ETH $SNDK
#Consumption momentum weakens, September policy still constrained by inflation
#SanDisk investors see future stock price surge, long-term target to be verified
#CLARITY vote pending, SEC rules not finalized 【ADA, SOL, XRP: Which Major Holder Is Most Deeply Underwater?】
Based on the current estimated average cost for major holders:
$SOL: approximately $105, unrealized loss about 26%
$XRP: approximately $1.82, unrealized loss about 42%
$ADA: approximately $0.57, unrealized loss about 68%
Among the three, ADA holders are currently bearing the deepest paper losses, while SOL's position is relatively stable. However, a deeper unrealized loss does not necessarily mean greater potential for future price gains. What really needs attention is that these major holder costs could become significant resistance zones during subsequent rebounds.
When prices return near ADA $0.57, SOL $105, and XRP $1.82, previously underwater funds might choose to exit, creating concentrated selling pressure. If the market is only experiencing a typical rebound, prices may not break through these levels in one go; only when the altcoin season returns with noticeable increases in capital and trading volume will the market have a chance to fully absorb these positions.
Therefore, major holder costs are not just about observing who is losing the most but can also be used to identify potential future resistance points.
Which one do you think will break through the major holder cost first? Today, $CAP overall is in a wide oscillation. If you do both $APR and $CAP, you'll find that the performance of these two coins after rising is very similar. They all oscillate at high widths. If I only compare the trends of the two, I might still be uncertain about the direction of $CAP. However, after analyzing $CAP's data, I can draw one conclusion—$APR today is the tomorrow of $CAP. —————————————————— Let's take a look at the contract data for $CAP. It can be seen that its contract long-short ratio experienced a sharp drop at 7:45 a.m. today, with a slight decrease in contract open interest. Looking at the candlestick at that time, we can see it was already in a consolidation phase. So, I infer that $CAP's short-term long funds have already moved out during this morning's wide oscillation. Looking at this set of data, we can see that at midday today, both contract open interest and the long-short ratio have declined simultaneously. This also confirms what I just said: $CAP short-term long funds withdrew this morning. This situation is very similar to before $APR, so I believe $CAP is very likely to drop sharply like $APR. —————————————————— I know many people are currently going long on $CAP because they want to take on the fees. Currently, its fees are heavily undermined, and going long can incur a lot of fees. But, ITokenized assets on Wall Street are accelerating their integration into the underlying layer, yet $ETH continues to trade sideways around $1,880, creating a dull contrast with the asset migration pace in the US stock and Treasury markets.
Bitcoin maintains over half of the market dominance above $63,000, while Ethereum has been suppressed below $2,300 throughout the year.
As institutions like Goldman Sachs and BlackRock advance pilot projects for real Treasury bonds and fund tokenization, macro funds prioritize Bitcoin as an interest-free reserve allocation and view Ethereum as a financial facility bearing execution costs.
Layer 2 networks have significantly reduced transaction fees, causing the expansion of settlement asset scale not to directly translate into burn consumption on the Ethereum mainnet, resulting in a tiered diversion of on-chain asset benefits.
If, after institutional commercial launch in October, large-scale securities clearing and high-value collateral settlement concentrate back on the Ethereum mainnet, a rebound in spot fees will trigger valuation repair; but if migration remains confined to private environments, this hypothesis automatically fails.
If tokenized flows of US stocks and interest rate assets are fully diluted across various branch networks, the mainnet's capture capability will continue to be insufficient, and the price range may further slide toward the $1,800 bottom.
Once macro liquidity shifts and drives down the risk-free rate, reconstructing institutions' return expectations for the base chain, the judgment of decoupling between the two will be disproven by reality.
The most important variable to observe in the next seven days is the actual change in mainnet base settlement layer fee consumption when traditional institutions deposit Treasury assets on-chain.
#霍尔木兹通航谈判未果，美伊施压升级 #财报观察员：AI基建财报接力登场 #韩股十日反弹逾22%，芯片股领涨As of August 15, 2026, BTC is priced at $63,010, up slightly by 0.06% in 24 hours; ETH is at $1,880, up 0.12%. The Fear and Greed Index is 30, with the market still lingering in the fear zone. Prices are relatively stable, but another curve is more worth watching—the total stablecoin supply is $308 billion. Although this is a 4.5% drop from the May peak of $322.4 billion, it still represents a net increase of 14.3% over the past 12 months. During this correction, coin prices have halved, but stablecoin supply has not fled. The money hasn’t left the market; it has just shifted its position.
This is the most easily misunderstood point right now: stablecoins have long ceased to be just a transit station for trading cryptocurrencies. The on-chain transfer volume for 2025 ranged from $33 trillion to $62 trillion, and in February 2026, the monthly settlement volume reached $7.2 trillion, surpassing the US ACH clearing network for the first time. With the implementation of the GENIUS Act and licensing in the EU and Hong Kong, stablecoins have officially transformed from crypto tools into payment infrastructure. This is the most solid and verifiable real demand in the crypto market.
This demand curve affects $BTC and $ETH in completely different ways.
BTC benefits from reserve premium. Stablecoin issuers are now major buyers of US Treasuries; Tether alone has about $141 billion exposure to US Treasuries, ranking among the top twenty globally. The larger the on-chain dollar system, the more funds are locked in US Treasuries and dollar assets, and these funds require a final settlement collateral that does not rely on any sovereign credit—BTC plays this role. It generates no cash flow; it sells "trust itself" and serves as the foundation of the entire on-chain financial edifice. The logic behind institutional BTC allocation is essentially buying systemic insurance for the on-chain dollar system.
ETH benefits from usage. Of the $308 billion stablecoin supply, 48.7% runs on Ethereum, with nearly $150 billion of "on-chain dollars" carried by this single chain. Every stablecoin transfer, every DeFi collateralization, every RWA liquidation, every batch settlement on L2 ultimately pays rent to the Ethereum mainnet. USDC’s annual transaction volume of $18.3 trillion is mostly settled within the Ethereum ecosystem. Every expansion of stablecoin supply re-prices ETH’s gas economy, staking collateral demand, and block space value. ETH is not digital gold; it is the settlement layer for on-chain dollars, charging fees based on usage.
So the core contradiction in the market is clear: fundamentals are expanding, but prices are playing dead. Stablecoin supply is up 14% year-over-year, yet BTC has dropped by half from last October’s high of $126,000, and ETH has fallen from above $4,000 to below $1,900. The SSR (BTC market cap/stablecoin supply) has compressed to 4.16, placing on-chain purchasing power at a significant level relative to BTC market cap. This divergence won’t last forever—either stablecoins continue to shrink, disproving demand, or prices catch up with liquidity. Historically, the answer is usually the latter, though no one can give you a timetable.
The division of labor is already very clear: BTC is the treasury of this system, ETH is the cash register. As long as stablecoin expansion continues, the repricing of these two assets is just a matter of timing—BTC first captures the reserve premium from institutional allocation, ETH later captures the usage premium from real settlement volume. The current fear zone is precisely the cheapest entry window for this long-term logic.WHEN THE MARKET WEAKENS, CAPITAL SHOWS ITS HANDS
$BTC is hovering near $63K, while $ETH remains below $1,900, keeping overall sentiment cautious.
But weakness in the majors can also reveal where capital is rotating. $SOL, $XRP, $HYPE, and especially $OKB are worth watching for relative strength.
$OKB is trading around $107, up roughly 5% over the past 24 hours, with a fixed total supply of 21 million tokens.
The important point isn’t simply which tokens are falling.
It’s which assets are holding up — or even gaining — while $BTC struggles.
Relative strength often reveals where the market’s attention is moving. 👀
#WeakConsumptionFedSplit
$BTC
#OpenAIAnthropicRace
$ETH
#NvidiaAICapitalChain
$OKB SNDK has been chased by funds like this, but no one wants BTC? The money actually hasn't disappeared at all. These days when I look at SNDK, I have a very strong feeling: the market is not lacking money willing to take risks. In areas like AI storage and semiconductors, funds still dare to chase. Even after rising this much, there are still people willing to rush in. But turn to look at Crypto. BTC is grinding around $63,000. ETH also shows no particularly strong performance. The whole market is so quiet it doesn't feel like a risk-on environment. So now I actually don't quite accept the phrase: "The market has no liquidity." Not necessarily. The money might still be there, just not choosing Crypto. This is the most worth pondering. Because AI at least has a very clear story: data centers, computing power, storage demand, orders, performance. But BTC now, no matter how you talk about it, is still ETF, interest rate cuts, macro. These stories are of course important, but the market has heard them too many times. So funds start to be picky, which is very normal. I even think this is more worth worrying about than BTC dropping a few percentage points. Because the real trouble is not that everyone is fleeing. But: everyone still dares to take risks, just unwilling to buy you. Of course, this logic could also be reversed. Once the AI boom cools down, funds will look for high-volatility assets again, and BTC might suddenly become the most popular direction again. Do you think Crypto has no money now, or is it just that funds temporarily don't favor BTC? $BTC $ETH $SND#S&P closes at a new high again, expectations for 8000 points heat up
The S&P 500 is just 2.7% away from 8000 points, but chasing at this level no longer offers good value.
On Friday, the S&P closed at 7785.76, down 0.17%, but it marked the third consecutive weekly gain. This week it broke through 7800 for the first time, reaching an intraday record high of 7816.7.
Why the rise? Inflation is cooling, the probability of a rate hike in September dropped from 44% to 28%. Q2 earnings season exceeded expectations, with profits up 50% year-over-year. Citi raised its target price from 7700 to 8100, JPMorgan from 7800 to 8000. Fundstrat's Tom Lee is more aggressive, believing it could reach 7900-8000 by the end of August. Kalshi bets the probability of hitting 8000 by year-end is 66%, and 8200 is 33%.
The trend is still up, but short-term sentiment is a bit overheated. At the end of July, everyone was calling a bear market; now everyone is shouting 8000, the switch has been too smooth. Consumer data is weakening, July retail sales unexpectedly declined month-over-month, the US-Iran agreement has not been finalized, and oil prices could fluctuate at any time.
The direction is most likely upward, but missing a day or two won’t matter. I won’t add positions at this level; I’ll hold what I have and keep cash ready for a pullback. From 7798 to 8000 is only 2.7%, one more push and it’s there, but what then? August to October is seasonally weak, and chasing highs no longer offers good value. $SPY The strong narrative of the dollar is unraveling. U.S. July retail sales data was unexpectedly weak, and the US dollar index (DXY) fell to its lowest level since May, prompting traders to significantly reduce bets on rate hikes within the year. What happened? Retail sales data fell short of expectations, becoming the final straw that broke the narrative of US dollar rate hikes. After the data release: Bond market traders pulled back bets that the Fed would raise borrowing costs in 2026. The US dollar index fell to its lowest point since May. The dollar is expected to record weekly declines for the sixth week in the past seven weeks. The core logic that previously supported the dollar's strength—"strong economy → stubborn inflation→ the Fed has to raise rates"—is being gradually eroded by consecutive data signals. Triple data stacking: Three key recent data points have collectively contributed to the dollar's decline: last Friday's labor market report was unexpectedly weak; this week's CPI and PPI data were moderate below expectations; retail sales data unexpectedly fell short of expectations. These three data points point in the same direction: the U.S. economy is cooling, inflation is falling, and the Fed lacks urgency for further rate hikes. Implications for the crypto market: The weakening dollar and the fading of rate hike expectations are marginally positive for risk assets: a weaker dollar → dollar-denominated assets appreciate relatively, supporting the narrative of crypto assets as alternatives to the dollar. As rate hike expectations fade → concerns about liquidity tightening ease, risk appetite may rebound. The "tightening narrative" is breaking down→ a core macro variable that has weighed on the crypto market in recent months is being unwinded. Of course, the flip side of weak retail sales data is the "economy."“This market cures all kinds of itchy hands. The more eager you are to make quick money, the more likely you are to repeatedly pay tuition. The premise of waiting for the wind to come is that the account is still there and you still hold chips. In a volatile market, less trading and more observation is itself profitable.” Concise summary: 1. The only strategy where chips are king—hold your bullets, refuse to heavily bet on one side in a choppy market. Don’t act rashly without a clear trend; patiently wait for signals, which is more practical than frequently bottom-fishing and topping out. 2. The bull market has already forked—BTC and ETH have stronger institutional support and resilience; altcoins are driven by sentiment, with violent rises followed by equally fierce drops, rotating quickly and trapping people fast. 3. Contract trading requires extra restraint—frequent spikes in volatility and high leverage easily cause stop-loss sweeps back and forth; don’t be fooled by short-term fluctuations, leverage risk is amplified in a choppy market. 4. Crypto and US stocks are worlds apart—while the S&P 500 pushes toward 8000 points, BTC grinds sideways at $63,000 with low volume, spot trading declines, crypto lacks incremental funds, and the two markets clearly diverge. 5. Value of position hedging—don’t fully load risky coins; allocate some hard assets as a base to hedge, smooth account drawdowns, and enhance risk resistance. 6. Reality of fading bullish factors—CPI/PPI cooling, ETFs continue net inflows, yet the market can’t rise. The root cause is lack of incremental funds; stock market games rarely produce big moves. 7. Altcoin risks amplified—stock market themes rotate very fast, hotspots come and go quickly, chasing highs easily leads to idling; avoid heavy positions in small coins without fundamentals. 8. Market support situation—BTC 628 Top 5 Factors Affecting Cryptocurrency ($BTC) or US Stocks
① Federal Reserve Rate Cut Expectations (Most Important)
The biggest driving force for risk assets ($BTC, US stocks) remains liquidity.
Market logic:
* Increased rate cut expectations → US stocks rise → BTC rises
* Rate cut delays → US stocks pull back → BTC under pressure
Currently:
* CPI and PPI have been relatively mild recently
* Latest retail sales data significantly weaker than expected
* Market begins to bet that the Fed’s future policy will be more dovish (easing)
For BTC:
Rising rate cut expectations = biggest positive factor
② Whether the US Economy Soft-Lands
Last night’s retail sales data was far below market expectations:
* Expected +0.1%
* Actual -0.6%
Indicates consumption is cooling.
There are two scenarios here:
Mild cooling (positive)
Economic slowdown but no recession
Result:
* Rate cuts
* Corporate earnings remain stable
* Both US stocks and BTC rise
This is what the market hopes to see most.
Rapid recession (negative)
If in the future:
* Unemployment rises
* PMI falls below 50
* Consumption continues to deteriorate
The market will start pricing in a recession.
Result:
* US stocks fall
* BTC falls in sync
③ ETF Fund Flows
BTC spot ETFs have become the core of pricing.
Key observations:
* Continuous net inflows into ETFs
* Continuous net outflows from ETFs
Experience shows:
* More than 5 consecutive days of net inflows → BTC likely to start an upward wave
* More than 5 consecutive days of net outflows → BTC likely to enter a correction
Same applies to ETH.
④ Whether the AI Rally Continues
The biggest engine for US stocks currently is AI.
Focus on:
* Nvidia earnings
* Microsoft AI business
* Meta AI business
* Data center construction
If AI continues to exceed expectations:
* Nasdaq rises
* BTC usually follows the rise
The market is still betting on AI profit growth recently.
⑤ Middle East and Geopolitics
Currently one of the biggest black swans.
If:
* Iran situation escalates
* Oil breaks through $90-100
Then:
* Inflation heats up again
* Rate cut expectations decline
* US stocks and cryptocurrencies come under pressure
The market is still watching energy prices and inflation risks from Middle East conflicts.
Most Favorable Combination Currently
If in the next two months:
✅ CPI continues to decline
✅ PPI continues to decline
✅ Moderate cooling of non-farm payrolls
✅ Continuous ETF inflows
✅ Fed signals rate cuts
Then:
* High probability of BTC breaking previous highs
* ETH gains may exceed BTC
* AI stocks continue to hit new highs
* Altcoins enter a catch-up phase
This is a typical "liquidity bull market."
Most Dangerous Combination Currently
If in the future:
❌ Oil prices surge
❌ CPI rebounds
❌ Fed signals hawkish stance
❌ Continuous ETF outflows
❌ US economy enters recession
Then:
* BTC may enter a 20%-30% correction
* ETH usually falls more
* Altcoins generally get halved
Importance Ranking (Currently)
1. Fed rate cut expectations ⭐⭐⭐⭐⭐
2. BTC ETF fund flows ⭐⭐⭐⭐⭐
3. US employment data (non-farm) ⭐⭐⭐⭐⭐
4. CPI/PPI inflation data ⭐⭐⭐⭐
5. AI company earnings ⭐⭐⭐⭐
6. Middle East situation and oil prices ⭐⭐⭐⭐
7. US Dollar Index DXY ⭐⭐⭐
For the coming week, I think the most worth watching is not altcoins, but: Fed September meeting expectations + ETF fund flows + US economic data (retail sales, PMI, employment). 𝗝𝗨𝗦𝗧 𝗜𝗡: Bitcoin spot ETFs recorded a net outflow of $56.2M on August 14.
BlackRock clients sold $55,500,000 worth of $BTC . Bybit launched 6 traditional Korean financial perpetual contracts today → Samsung Electronics, LG Electronics, KODEX200 ETF, with up to 25x leverage.
Use a crypto account to trade Korean stocks, 24/7 nonstop. This channel didn’t exist three months ago.
Binance listed Apple and Tesla → OKX listed Pop Mart and Xiaomi → now Bybit is connecting to the Korean market. Exchanges are no longer just gateways for crypto, but gateways for global assets.
The logic for buying ETH: the more TradFi assets enter the chain, the greater the demand for USDT settlement, and Ethereum’s throughput follows. Bybit added 6 at once today.
$ETHUSDCThe most eye-catching data across all platforms today wasn't $BTC's sideways movement, but $SNDK surging to second place with a turnover of 2.77 billion, pushing $ETH to third. Single-day +6.5%. Against the backdrop of shrinking market volume and chip stock differentiation, this money isn't just following the crowd but targeting the cyclical turning point. Outline - 🔍 Who is $SNDK? Why is it being speculated today? - 📈 Fundamental highlights: Cycles, AI, and storage price increases - ⚔️ Long-short battles: Disagreements behind 2.77 billion in trading volume - 🎯 How to participate: signals, positioning, and risk - 🧠 Conclusion: Where does the storage pendulum swing? Today's snapshot $BTC 63,063, -0.39% $ETH 1,883, +0.00% $QQQ -0.14%, $SPY -0.20% $DXY -0.31%, $GLD +0.63% $IBIT -0.70% $SNDK Turnover 2.77 billion, +6.5% $SPCX Turnover 420 million, -1.9% VIX 14.26, -2.60% US crude oil +1.26% 1. Who is $SNDK? Why is it trending 🔍 today$SNDK is the tokenized asset of storage chip giant SanDisk, specializing in NAND flash memory, with end-users covering data centers, consumer electronics, and automobiles. Today, its turnover has reached 2.77 billion, second only to $BTCHarmony public chain was attacked again, with over 30 trillion ONE tokens minted, directly breaking the protocol token issuance's 'verifiability' expectation. This round of attack is not the first, but the scale and token minting amount this time have caused cracks in the market's confidence in the 'self-healing ability of decentralized protocols'.
2) The weight of three pieces of news: The Manus acquisition was reversed due to a Ministry of Commerce ban, showing that regulatory intervention in cross-border tech ecosystems still has tangible strength; Harmony was attacked and minted tokens, exposing security vulnerabilities at the protocol level, which may trigger users' doubts about the 'effectiveness of on-chain governance'; Kalshi's fundraising plan proposed a $40 billion valuation, reinforcing the AI and crypto crossover narrative but without providing actual security capability verification.
Supporting factors include that over 40 crypto companies have applied to AI labs to use their models for security testing, indicating the industry is trying to use technical means to address attack risks. The pressure lies in that if such attacks occur frequently and lack transparent response mechanisms, the market will reassess whether 'decentralization' can still guarantee asset security.
Next steps to watch: Whether Harmony's official team will release details of the attack and compliance explanations for the token minting, and whether AI labs will disclose test results and actual protection capabilities. Without public verification, risk appetite still needs validation.
For information and market scenario analysis only, not investment advice. Crypto assets are highly volatile; please conduct independent research and manage risks. Shorted it, shorted it! A wise person knows the times!
Previously, the judgment was based on macroeconomic bullishness, combined with a 4-hour RSI divergence.
BTC's RSI6 has diverged 3 times, and RSI14 has diverged twice. But yesterday, RSI clearly dropped, so it's not advisable to be bullish anymore.
Moreover, BTC obviously does not buy into the macro bullishness.
The FOMC on September 16 is likely to be bullish, or at least not bearish. It can basically be asserted that there will be no rate hike in September, though the dot plot might lean a bit hawkish. However, the judgment is that the rate hike is more likely in December rather than October. This is due to declining employment data, the midterm elections, and cooling inflation.
From a macro perspective, the next key dates to watch are:
◆ August 26 PCE data, the Fed's core inflation indicator is actually PCE, not CPI.
◆ September 4 non-farm payroll data,
◆ September 11 CPI data. Breaking news! $AAPL wants to seek emergency storage from China, but Washington says no! #消费动能转弱，9月政策仍受通胀制约
According to The Wall Street Journal, U.S. Commerce Secretary Lutnick clearly stated that he does not want Apple to purchase storage chips from China.
The problem is, Apple has just started testing CXMT and YMTC, hoping to find an additional supply source and also to put some pressure on the current suppliers' pricing.
• If Chinese suppliers are blocked out, Apple will have to rely more on MU, SKhy, Samsung, and $SNDK, giving the old suppliers stronger bargaining power.
• Chinese storage is no longer about "low-price market penetration." CXMT and YMTC have also raised prices on some products, indicating the real issue is a shortage, not just politics.
• For MU and $SNDK, the market is not just buying a piece of news but anticipating firmer prices and steadier orders for the coming quarters.
U.S. stock investment sites believe this news truly changes Apple's bargaining power. In the past, Apple used its scale to pressure suppliers; now with AI major clients competing for capacity and Washington restricting alternative sources, Apple's leverage is clearly reduced.
Next, we will watch three things:
Whether the U.S. will escalate "discouragement" into formal restrictions;
Whether Apple will ultimately sign with CXMT or YMTC;
Whether DRAM and NAND contract prices can continue to rise. $MU $WDC $SNDK $SKHYYesterday the market dropped sharply, with a bunch of coins closing below the day-before-yesterday's lowest point, and the whole market was filled with panic. At times like this, my usual rule is: don't rush to bottom-fish, first set up breakout orders and wait for the price to move on its own.
$ICP hit a low of 2.13 yesterday and closed around 2.19. I followed my plan and only entered a long position after it reclaimed 2.22, with an average price of 2.225. Now it's around 2.27, with a small floating profit 😄. This "first break down, then recover" pattern means I wait for confirmation before entering, which gives me peace of mind.
$DOT was a lesson. I chased a short position around 0.763 yesterday afternoon, but it still closed weak, and the short got stuck, now with a floating loss of 0.036U. Reviewing the chart, DOT actually closed below the day-before-yesterday's low yesterday, so by my own rules, I shouldn't have rushed in at that level; I should have waited for a clear signal. Impatience comes with a price 😮💨
$XPL was entered late at night; it closed strong against the trend yesterday, reclaiming the day-before-yesterday's high. I'm still holding the long position around 0.076, with a floating profit. But today it gave a reverse signal, and I'm watching the 0.0735 line closely—if it breaks below, I'll exit and not fight it.
Today I set up a bunch of breakout orders waiting for the price; if they don't hit the price, I'll cancel them tomorrow and won't chase hard. Controlling your impulses is better than anything else.SOL is now around 75.5u, still grinding within that 73 to 77 range. Today, no other topics, just one thing: sentiment is sky-high, but money keeps flowing out. I’m not chasing longs at this level.
First, look at the money. On the spot side, in the 3-hour window with 12 bars, not a single net inflow has appeared; large orders are still net outflows; on the contract side, the active buyer transaction ratio is down to 40%, sellers are pressing down hard. To put it bluntly, funds are withdrawing, not that I’m not seeing it.
Now look at sentiment, it’s very lively. KOLs bullish ratio is as high as 8.3, Grayscale is talking about SOL’s tokenomics to argue for inflation reduction, ETF infrastructure, institutional entry, treasury accumulation, the whole square is full of hype, everyone acting like they found a money printing machine.
Here’s the problem — a basket full of positives, yet the price is still below the 50-day and 200-day moving averages, far from the historical highs, leverage is shrinking, and the 12-hour borrowed money betting on direction has been cut by nearly 30%. Sentiment is sentiment, money is money, with such divergence, those hyping won’t tell you.
So chasing longs at this level has mediocre cost-effectiveness. Wait until large spot orders turn positive and funds truly return; if you want to buy near the lower end of the range, wait for a pullback confirmation, much more comfortable than chasing now. Avoid short-term trades, wait for the money to decide.
#sol $SOL$OKB 这波单边行情，从进场到持有，整整拿了20天，价格直接走出6倍空间。现在的悬念只有一个：目标位放在124，但到了这个位置，手里的仓位能不能顺利出掉？没人能提前知道答案。📈 很多时候，交易的难点不是“看不看得准”，而是“拿不拿得住”。短线和长线，表面上是两种玩法，背后其实是两种完全不同的心态体系。短线赚的是市场情绪和节奏的钱，拼的是反应速度和纪律；长线赚的是趋势和逻辑的钱，拼的是信念和耐心。你说短线是赌博，其实也有一点道理，因为短线确实更依赖临场判断，而长线更像是在用时间验证逻辑。🎯 真正能让长线变成现实的，不是判断有多准，而是仓位有多小。小仓位才拿得住，拿得住才能等到趋势走完。很多人之所以拿不住，不是因为不知道方向，而是因为仓位太重，一个回调就慌到睡不着。进场之后最好的状态，就是把单子“忘掉”——不是完全不管，而是不再被每一根K线反复折磨。该止盈止盈，该止损止损，中间的过程越少干预越好。这种“忘记”，不是逃避，而是对逻辑和风控最大的尊重。💎 短线最大的危险，从来不是方向看错，而是人性里的重仓冲动。很多人做短线，赚了想乘胜追击，亏了想立刻翻本，结果就是不断加大仓位，把交易活#Consumption momentum weakens, September policy still constrained by inflation
Let's casually talk about the current macro-driven market changes. On one hand, consumption data is weakening and the economy shows signs of cooling, but inflation hasn't been fully suppressed, which directly limits the policy easing space in September, creating a rather tangled situation.
$BTC is affected by this dilemma, making it difficult to have a smooth, one-sided market trend. Economic weakening brings some demand for safe havens, but inflation stickiness suppresses rate cut expectations. The expectation of high interest rates lasting longer lingers, so Bitcoin mostly oscillates within a range. It's hard to see large capital inflows driving a rally; most of the time, sentiment fluctuates back and forth, and we can only wait for clearer policy signals.
$ETH has greater elasticity and is more obviously disturbed by liquidity expectations. People originally expected economic weakening to accelerate easing, but inflation drags it down, pushing rate cut expectations further back, which is not favorable for high-risk ETH. The market tends to show pulse rebounds that fail to sustain, with more frequent rallies followed by pullbacks. In trading, one should not blindly bet on easing landing; be cautious of drawdown risks caused by unmet expectations.
This is currently a rather conflicted environment, with the economy and inflation pulling against each other. News appears mixed, making it hard for a one-sided market to emerge. Trading should not be based on a single data point alone; don't subjectively fantasize that easing will come immediately. Leave some room, control position size and leverage, and wait for the situation to become clearer before increasing exposure. Transaction fees reveal two completely different "temperatures" hidden in two blockchains.
To judge whether a blockchain is thriving, some look at the coin price, some at TVL, but I prefer to focus on a less conspicuous indicator: transaction fees. It is the only on-chain vote that cannot be faked—real demand paid with real money. Recently, the fee structures of BTC and ETH have told two completely different stories.
Let's start with BTC. The cumulative number of Ordinals inscriptions has surged to 80.99 million, with a total minting fee of 7,016 BTC, equivalent to about $628 million. What does this mean? Bitcoin's network originally only did one thing—transfers. Now, a group of people are willing to spend hundreds of millions of dollars to "engrave words" on-chain. You could say inscriptions are hype, but hype itself is demand, proving one thing: BTC's block space is expensive for a reason. Its fees are essentially a scarcity premium; blocks are limited in size, and whoever bids higher gets in. The higher the price congestion, the more valuable the space.
ETH, on the other hand, follows a completely different logic. Ethereum's fees now reflect two things: DeFi and stablecoin settlements on the mainnet, and the cost of L2s posting blob data to the mainnet. Blob fees once surged to 42,000 Gwei driven by L2 demand—translated, this means second-layer chains like Arbitrum, Base, and Optimism are doing so well that they compete to "pay rent" on Ethereum. So ETH's fees are not a scarcity premium but a thermometer of application demand, measuring the ecosystem's overall activity.
One is "valuable territory," the other is "thriving business." These two fee economics are worlds apart, and so are their investment logics.
Looking at the market status of these two chains as of 10 PM on August 14: BTC is around $63,500, basically flat in 24 hours, down 1.16% over the week, consolidating between $62,000 and $66,000 for five weeks. Support lies between $62,000 and $62,800, resistance between $64,000 and $65,500. ETH is around $1,885, with little 24-hour movement, basically dead in the water along with the broader market. SOL is at $76.08, up 0.7% in 24 hours and 4.6% over the week, showing rare strength among major coins. DOGE is at $0.0694, down about 1%. The Fear & Greed Index is 30, indicating a generally cold market.
When the market is cold, on-chain data is especially meaningful—prices can be distorted by sentiment, but fees cannot. When the inscription market is quiet, BTC's on-chain revenue bottoms out; as long as L2 transaction volume on ETH keeps rising, blob fees are its hidden performance. Looking long-term, both chains are answering the same question: can they sustain themselves through fees? BTC faces halving pressure, so miner income ultimately depends on block space demand; ETH faces more direct pressure—the more prosperous L2 is, the more ETH is burned on the mainnet, supporting the deflation narrative.
So my conclusion is straightforward: don't apply BTC's fee logic to ETH, nor ETH's to $BTC. One bets on scarcity, the other on ecosystem vitality. Fee divergence is not a bad thing—it shows this market has finally grown two distinct business models.BTC and ETH, despite favorable inflation indicators, why are prices staying flat? If the market buys expectations first and checks the news later, what does the current price say? The U.S. CPI (CPI) for May was tallied at 3.4% year-on-year, and the PPI also showed a slowing trend, reigniting expectations for rate cuts. Looking at the macro headlines alone, the environment is favorable for risk assets. However, the core pillars of the market, Bitcoin and Ethereum, are struggling to find direction. Bitcoin remains around $63,552, with intraday volatility narrowing to less than 500 points, and $64,000 remains a strong resistance level. Ethereum has repeatedly tested the $1,900 level near $1,886, but no clear breakout has emerged. The reason prices don't react is simple. This is because the market has already priced in a significant portion of inflation easing and the possibility of interest rate cuts within the year. In other words, improving the indicators itself is not new information but rather a confirmation of existing expectations. The problem is that confirmed expectations will provide further upward momentum.一年市值暴涨5倍、黄仁勳亲笔写信催货、全球科技巨头高管挤满韩国酒店——这是SK海力士的高光时刻，也是它正在面对的一场前所未有的投资考验。 这轮扩产有多大？ 2026年8月，SK海力士正式宣布投资54.3万亿韩元（约382亿美元），在韩国建设两座新工厂： 龙仁Y2工厂：投资35.2万亿韩元，生产HBM等新一代DRAM产品，预计2027年7月开工，2029年6月启用首个洁净室 清州M17工厂：投资19.1万亿韩元，建设NAND闪存生产基地，预计2028年12月启用首个洁净室 加上此前已宣布的600万亿韩元龙仁半导体集群长期投资计划，SK海力士的扩产蓝图正以前所未有的速度铺开。2026年全年资本开支预计将达到40万亿-50万亿韩元区间的高端。 钱从哪里来？用在什么时候？ 2026年7月，SK海力士在纳斯达克上市ADR，募资265亿美元，创下外国企业在美国上市的最高募资纪录。 但巨额投资能否兑现回报，关键取决于两个时间维度： 短期（2026-2028年）：HBM4已量产发货，HBM4E已送样，10份长期供货协议（LTA）锁定了中长期的订单可见度。公司预计下半年进一步扩大HBM产能。 中长期（2The market situation on the afternoon of August 15, 2026, is clear: BTC is priced at $63,103, down 0.7% in 24 hours, having halved from the all-time high of $126,080 set in October 2025; ETH is at $1,880, fluctuating mostly between $1,800 and $2,300 throughout the year. Market sentiment is cold, with Bitcoin dominance holding steady at 56.6%. At this moment, the story of RWA (Real World Assets) is becoming increasingly tangible—on July 15, DTCC completed a tokenization production pilot, with over 30 institutions including BlackRock, Goldman Sachs, JPMorgan Chase, and Vanguard running real transactions on the private Besu network (Ethereum Enterprise Edition) and the Canton public network. Official commercial service is set to launch in October; Standard Chartered Bank covered Chainlink for the first time this Monday, setting a $200 target price for 2030, betting that on-chain tokenized assets will expand from the current $340 billion to $4 trillion by the end of 2028. RWA is no longer just a PPT slide; it is becoming a pipeline connected to Wall Street's main artery.
But which chain the pipeline is laid on determines who gets paid. Here, the divergence between $BTC and $ETH is fundamental. The collateral management, securities settlement, and margin processing used in the DTCC pilot run on an Ethereum-like architecture; BlackRock's BUIDL fund and Franklin's Treasury tokenization also settle within the ETH ecosystem. ETH is becoming the settlement layer for financial assets, and its price logic should be linked to on-chain economic activity—transaction volume, Gas consumption, and tokenized asset scale. BTC, however, does almost nothing in this process; it is merely borrowed as collateral and held by institutions on their balance sheets as reserves. Its price drivers remain the same old factors: macro liquidity and allocation-driven capital flows.
The problem lies in pricing misalignment. The market's "reserve premium" for BTC is actually quite honest—holding above $60,000 and maintaining dominance after halving shows that during macro tightening and capital withdrawal, "productive assets" are cut first, leaving "value storage" behind. In contrast, Standard Chartered's valuation method for LINK uses a 25x growth in fee revenue to back-calculate the token price, but the same logic does not apply to ETH: L2s have siphoned off economic activity, with transaction costs on Base and Arbitrum being just a fraction of the mainnet's, diluting Gas fee revenue. The EIP-1559 burn mechanism is nearly dormant under low fee conditions. The busier the network, the less money the toll booths collect—this is ETH's biggest structural flaw in "track value." The dividends from settling $4 trillion in on-chain assets may largely flow to L2 tokens and infrastructure providers, not ETH itself.
Therefore, my judgment runs counter to the mainstream narrative: ETH's track value is not underestimated; rather, its value capture mechanism is malfunctioning. Unless the mainnet reclaims its role as the default high-value settlement layer (for example, for Treasuries and institutional collateral), the more prosperous tokenization becomes, the weaker the price transmission to ETH. The real beneficiaries may be middleware like LINK that collects "tolls"—Standard Chartered's price target of $200 from the current $8 bets exactly on this. BTC will continue to enjoy its reserve premium, diverging from ETH, with correlation decreasing. The so-called divergence is not competition between two paths but a complete disconnection from each other going forward.This week's macro data has truly given the market a shot of confidence. CPI fell to 3.4%, PPI surged back to 4.7%. The market's logic is simple: inflation goes down, rate hikes stop. Since prices are no longer soaring, it would be rude for the Fed to wield the big stick of rate hikes again.
* The market has already started lowering rate hike bets, with some even betting on when the bell for rate cuts will ring. For BTC, this "digital gold," as long as the dollar doesn't rise, it wins halfway.
*Wall Street funding is very realistic. When Treasury yields are no longer as attractive, money rushes like sharks smelling blood, rushing toward BTC, ETH, and growth public blockchains like Solana.
Looking back at our analysis of capital flows, you can match the numbers:
* $BTC (about $63,075): Despite Jump Crypto's billion-yuan sell-off, BTC still maintained net inflows. Why? Because the macro environment has changed, institutions know that selling pressure is temporary now, while the trend of "declining inflation" is long-term.
* $ETH (about $1,884): The intensity of that $1.951 billion inflow now makes sense. Large funds are betting on a rebound after falling inflation, and ETH, as the king of public chains, is the first stop for liquidity recovery.
* $LINK (about $9.68) and $SOL (about $75.60): The massive inflow of these high-beta (high-volatility) varieties is exactly the hot money pair#消费动能转弱，9月政策仍受通胀制约
Last night, multiple major U.S. economic data releases came out simultaneously, putting the long-anticipated narrative of "consumer resilience" to a significant test.
July U.S. retail sales unexpectedly fell by 0.6% month-over-month, sharply deviating from the market expectation of 0.1%, marking the largest single-month drop since May 2025; meanwhile, the University of Michigan Consumer Sentiment Index for August declined to 51.0, also below market expectations. A series of data clearly signals that U.S. consumer spending momentum is continuously cooling.
Weakened consumer demand, combined with the gradual easing of prior CPI and PPI figures, theoretically reduces the Federal Reserve's motivation to raise interest rates again in September. However, the market cannot ignore another set of warning signs: the University of Michigan's one-year inflation expectations rose from 4.2% to 4.3%, indicating renewed public concern over rising prices.
These two forces are pulling against each other, directly placing the Federal Reserve in a dilemma.
Two main scenarios will determine the future direction of global assets:
Scenario one: Continued consumer slowdown. The U.S. dollar and short-term Treasury yields face downward pressure, while assets like gold and Bitcoin will receive favorable support;
Scenario two: Persistent rise in inflation expectations. The Federal Reserve is forced to extend the high interest rate maintenance period, liquidity tightening continues, and global risk asset valuations remain under pressure.
The biggest challenge in the current market is that both logics hold true simultaneously, eliminating the foundation for a one-sided market.
The Federal Reserve's policy path in September no longer has a clear answer; every upcoming inflation and employment data release will trigger intense capital market battles. Amid macroeconomic uncertainty, whether in stocks, gold, or crypto markets, volatile fluctuations will become the norm, significantly increasing trading difficulty. #消费动能转弱，9月政策仍受通胀制约
Looking at the recent data from the US, I think the Federal Reserve is really caught in a dilemma right now.
Retail sales in July unexpectedly dropped by 0.6% month-over-month, while the market had previously expected a 0.1% increase. This decline is the largest since May 2025, and the previously touted consumer resilience was directly shattered by this data. Not only retail sales, but the University of Michigan Consumer Sentiment Index for August also fell to 51.0, below the market expectation of 54.5, clearly showing that US consumption has started to cool down.
Logically, weakening consumption combined with falling CPI and PPI would reduce the necessity for further rate hikes in September. But one point that cannot be ignored is that the one-year inflation expectations of consumers actually rose from 4.2% to 4.3%, indicating that concerns about prices have not disappeared.
The market will likely follow two divergent paths next.
If consumption slows further, the US dollar and short-term Treasury yields will come under pressure, which would provide supportive benefits for gold and BTC.
However, if inflation expectations continue to rise, the duration of high interest rates will be extended, and valuations of various risk assets will continue to be suppressed.
There is no clear one-sided certainty now; on one hand, weakening consumption data, and on the other, rising inflation expectations. The Federal Reserve’s September decision will be pulled between these two forces. Going forward, whether it’s the US stock market, gold, or the crypto market, close attention must be paid to changes in these two sets of indicators. #闪迪投资者日后股价大涨，长期目标待验证
After watching SanDisk's recent Investor Day, my first reaction was: the stock price will have a nice short-term boost, but whether the long-term goals can be fulfilled is a big question mark.
After the Investor Day on August 13, the market immediately repriced its growth story. The company set targets for FY2028-2030, aiming for mid-to-high double-digit revenue growth, an adjusted gross margin close to 80%, an operating margin reaching 75%, and stated that after completing business investments, it will return all excess cash to shareholders.
The news caused the stock price to surge sharply, rising about 13.7% in a single day, and subsequently holding steady above $1600.
But I am pondering a very realistic question: what exactly is this rally betting on?
Is it purely speculation on the high demand for AI storage combined with expectations of cash dividends returning to shareholders, or has the market already priced in the full realization of all future targets for the next few years?
The targets are very attractive: high gross margin, high profitability, large cash returns—each point appeals strongly to capital. However, visions are visions; from paper targets to actual financial reports, there will be many uncertainties. AI storage demand might fall short of expectations, and intensified industry competition could squeeze profits, both of which could disrupt this blueprint.
Short-term sentiment is already reflected in the price; going forward, it depends on whether performance can gradually meet expectations. Before chasing the rally, it is crucial to distinguish: which parts are already realized realities, and which are still just promising long-term stories. #闪迪投资者日后股价大涨，长期目标待验证
Amazon's earnings report is really interesting; despite guidance falling short of expectations, the stock surged 9%.
After reading Amazon's Q2 earnings, I was honestly a bit confused by the market's reaction.
Looking at the data alone, it's actually quite strong: Q2 revenue was $200.6 billion, up 20% year-over-year; AWS jumped to $42.2 billion, growing 37%, hitting a new high since the end of 2021. Operating profit was $16.6 billion, a 64% increase, with a profit margin raised to 39.4%, showing solid fundamentals.
But here's the contradiction: the company raised its full-year capital expenditure to $220 billion, while the Q3 revenue guidance was below market expectations. Based on Meta's previous trend, such guidance below expectations would normally cause a drop. Instead, after-hours trading saw a surge of over 9%.
It's clear the market logic has changed; as long as cloud business AI-related growth keeps accelerating, investors are willing to accept high capital expenditures.
The three giants are taking three completely different paths with the AI story: Microsoft is delivering on implementation, Meta leans towards painting a future vision, and Amazon is both burning cash and making money.
I also have a question for everyone: Is AWS's impressive growth a real proof of solid AI demand? Or is the huge $220 billion capital expenditure being spent now a bill that will have to be paid back sooner or later? Which side do you lean towards? Putting BTC's short-window numbers together with the full-day average makes the picture much more complete than just looking at the popular rankings. OKX Onchain OS recorded 42 BTC mentions in one hour at 11:00 on August 15, including 40 times in X and 2 in the news; The total 24-hour volume was 1,374 times. After conversion, the latest hour is 0.73 times the hourly average for Long Windows, which is about 27% lower 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, 38% are bullish, 31% bearish, and about 31% neutral, indicating a 'slightly bullish advantage'; Within the 24-hour period, the trend is 26% bullish and 30% 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 has 1,184 times in 24 hours, 190 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 fitIt's not that they don't want to spend it, it's that they really can't afford it. Salaries haven't increased, but consumption has increased
📉 Consumption data
Retail sales month-on-month: -0.6% (expected +0.1%), marking the largest decline since May 2025
Michigan Consumer Sentiment Index: 51.0 (previous 55.2, expected 54.5), a sharp decline
📈 Inflation expectations
One-year inflation expectation: 4.3% (previous value 4.2%), rising instead of falling
Logic breakdown: Last year you could carry a whole truck for 100 yuan, this year you carry two bags for 100 yuan—after shopping, you quietly put things back, not because you don't want to spend them, but because it's too expensive to afford. Inflation PTSD has become a reality; people still fear price hikes but don't dare to spend money on their hands.
Final summary: A rate cut in September is not possible; it is necessary. Only when prices really come down will consumption return. 💸
#消费动能转弱, September policy remains constrained by inflation Trump signed a drone tariff this week, let's talk a bit
High tariffs will be imposed on imported drones and parts
Large drones with thermal imaging get a direct 100% increase, small ones also 25%
But the whole drone is not the main focus
Parts and components are also covered
This is the real tough part
Previously, many US drone companies' tactic was to buy motors, ESCs, flight controllers, batteries from China, assemble them back in the US, and label them as Made in USA
Now parts themselves are subject to high tariffs, this route is blocked
It's not that you can't buy foreign drones, it's forcing you to move the entire supply chain back to the US.
//
The market's first reaction was to speculate on whole drone manufacturers. Red Cat, Unusual Machines surged that day, UMAC jumped over 20%
But that's not what I'm focusing on
Whole drone manufacturers assemble, parts suppliers make screens, chips, batteries
The policy says you can't use imported parts anymore, so whoever can make these things domestically in the US will reap the biggest benefits
ESCs, flight controllers, lithium batteries, carbon fiber frames
China has had a long-term cost advantage, the US has almost no domestic capacity
Now forced to cut suppliers, whoever can prove US-made compliance first will get both military and civilian orders
These companies have small market caps, from hundreds of millions to tens of billions of dollars, very low attention
But their growth rate might be much faster than the already surging whole drone leaders
//
Everyone only sees the whole drone manufacturers rising, but hasn't seen the supply chain restructuring underneath. After tariffs officially take effect, these small manufacturers might become the focus
Small market caps also mean big volatility, don't get too excited #英伟达深入AI资本链，协同与风险如何平衡
Recently, I carefully reviewed NVIDIA's latest series of moves. Honestly, it feels like this company has stepped beyond the role of a simple chip manufacturer and is fully embedding itself into the entire AI capital chain.
Previously, we understood NVIDIA as just selling GPUs, with downstream customers paying for chips—a straightforward business transaction. But now the approach is completely different, extending from hardware supply to equity investment, and even providing financing guarantees.
In the latest disclosures, NVIDIA's holdings in SpaceX have reached a market value of $21 billion, most of which comes from previous investments converted from xAI. This means it is no longer just a supplier but has become a deeply tied shareholder, with the success or failure of its customers directly impacting its own asset returns.
Interestingly, on the other hand, NVIDIA has taken a much more cautious stance toward the OpenAI project. The originally discussed guarantee for the Ohio data center, nearly $250 billion, has now been reduced to less than $120 billion, significantly shrinking its credit exposure.
These two contrasting moves together reveal a very clear strategy: locking in long-term downstream computing power demand through equity investments to stabilize its chip business foundation; while tightening financial risk on some operations to avoid excessive exposure to downstream corporate debt. It's like advancing on one front while building a firewall for itself on the other.I haven't shared much about $BTC recently because, as I mentioned before, I don't want to guess the bottom or try to predict exactly when the decline will end.
Given the current volatility, I see only two possibilities: either there's one last drop, or the bottom is around 57.
If there really is one last drop, how much further can it fall?
In the last cycle, around 17000, the final drop coincided with the FTX crash, and it ultimately fell to about 15000, a decline of just over 10%. Moreover, what was Bitcoin's market cap back then compared to now?
Currently, Bitcoin still has a market cap of about 1 trillion USD. Even if another FTX-level crash happens now, the actual impact might not be as severe as before. Even if it falls another 10% from the current lowest price, it would be hard to drop to the low 40,000s.
Besides, the 60,000 level has been tested repeatedly with huge trading volume.
If you still mechanically apply the Realized Price from previous cycles to this one to determine where Bitcoin's true bottom must be, I think that's somewhat misapplied.
The cycle measures time, not price.