Orbit Post Sitemap
$CHZ, the leader in sports fan tokens, is backed by various football giants and top-tier event IPs, claiming to bring the fan economy onto the blockchain. During the World Cup, fans rushed into the stadium, shouting "champion" while pushing token prices upward—the scene was incredibly lively. But today, the market is heading south, with obvious capital flight. No competitions, no passion, no money. Fan tokens have a very obvious "cycle": major sporting events are their adrenaline. During the World Cup, Champions League, and Premier League sprint phases, fans' emotions are on point—buying tokens to support or participate in team voting makes the logic smooth, emotions are maxed out, and funds naturally flood in. But once the competition gap came, without major showdowns and championship celebrations, fans' enthusiasm faded. With tokens in hand, all you can do is wait for the next season—there's nothing else you can do. Without emotional catalysts and speculative grounds, why should short-term funds endure with you? Of course, they all scattered and rushed into hot tracks like AI and RWA. This drop in CHZ is, frankly, "no story left." The entire fan token sector is cooling down—not just because it is underperforming, but because the entire sector has entered a dormant state. Its market performance basically relies on "waiting for notices." The weakness of these stocks is extremely obvious: the market is highly tied to sports events and is entirely event-driven. With major collaborations announced, top clubs issuing new currency, and major tournaments kicking off, it can take off on the spot; Without these, it would just be a long-term bearish decline, with trading volume becoming sparse and prices falling lower, making the holding experience extremely agonizing.🚀SPCX (SpaceX)｜Q2 revenue beat + Capex explosion + 8/6 trillion unlock, all good news priced in is just the starting point
Current price $125.33 (8/4 close +9.43%, after-hours -8.37% to $114.84)｜Market cap $1.65 trillion｜PE TTM negative
📅 First 10-Q after market close on 8/4
Revenue $7.814 billion (+92% YoY, beat $6.9 billion)
Adj EBITDA $3.54 billion (+191%)
Net loss $541 million (narrowed 46%)
Capex $18.37 billion (+550%, including $15.83 billion invested in AI infrastructure) → main reason for after-hours -8%
Cash $100 billion, backlog $47.5 billion
🟢 Three business segments (Starlink supports all)
Starlink (connectivity) $4.29 billion +66%, operating profit $1.66 billion, 12 million users / 10,200 satellites / Starshield government contracts >$6 billion
AI computing power $2.56 billion +247%, operating loss $1.26 billion, Q2 new cloud contracts $14.1 billion, year-end computing power target 2GW → 10GW by 2027
Aerospace $962 million +29%, operating loss $542 million, 78 launches / 1041 tons payload in H1, Starship V3 flew twice
🔗 Partner order amounts (signed)
Anthropic computing power lease $1.25 billion/month (until May 2029, total ~ $45 billion)
Google computing power lease $920 million/month (from Oct 2026 to mid-2029, total ~ $30 billion)
Reflection AI $150 million/month｜Cursor acquisition progressing｜Q2 cloud contract sales $14.1 billion
NVIDIA exclusive Rubin architecture + Starmind orbital AI satellite (launch in 2027)
⚠️ Biggest gray rhino: phased unlock starting 8/6
8/6 first batch 911.5 million shares (20% of lockup pool, ~ $104 billion) unlocked, circulating shares jump from <5% to ~12%
Late August–October every 2–3 weeks release 7%, full unlock on 12/8 after 180 days
Short positions account for 32-34% of float, average cost far above current price, insiders’ cost extremely low → selling cost basis far above $125
🎯 Four scenarios (expectation ~ $120)
🐂BULL 20% → $150-180 (all AI contracts confirmed + unlock volume shrinks and stabilizes + daily Starship launches)
⚖️BASE 35% → $115-140 (Starlink stable + AI cash burn + unlock sideways) Current price here
🐻BEAR 25% → $90-110 (unlock crushes price + AI losses widen)
💥CRASH 15% → $60-85 (contract termination + Starship delay + panic sell-off)
✋ Practical advice
Revenue beat + Capex explosion = all good news priced in; 8/6 unlock more important than earnings. Break above $140 requires unlock weekly volume shrink and stabilization, safer to retest $104.83 previous low (8/3 low), do not chase highs.
⚠️ SPCX has dual attributes of equity tokens + crypto, triple risks of liquidity/discount/unlock, not investment advice, refer to SEC 10-Q.
SPCX #SpaceX #UnlockSellingPressure #Q2Earnings #AIComputingPowerLease
$SPCX The SNDK earnings report hasn't been released yet, but smart money bulls have already started to flee.
Four major bull whales today directly withdrew $11 million. Whale 0x364a liquidated overnight to break even. Big players are cashing out on rallies, leaving only retail investors foolishly taking the risk.
Backend data shows 191 short whales holding $156 million in short positions, nearly double the capital of the bulls. Bull whales are retreating while short whales are heavily defending their positions. The current rebound is all an illusion.
With this capital structure before the earnings report, either smart money is hedging early or they know something.
In terms of strategy, just short the rebound and wait for the earnings report to land before deciding the direction. Don't bet on direction before the earnings, especially when whales are fleeing.
$SNDK $BTC $ETH #财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？ #SpaceX首份财报超预期，解禁仍是关键变量 #AMD财报超预期，增长已被透支？ Base is the strongest "application" for ETH in this cycle, not Arbitrum, not the L2 war winner.
If the mainnet is the "$ETH" "central bank clearing layer," then Base is its ATM interface reaching 100 million Coinbase users.
On August 24, Base's daily active users surged to 1.05 million, up 60% from the beginning of the month; during the same period, Arbitrum had 394,000, Linea 198,000, Mantle 69,000—Base alone has more than these several L2s combined.
In July, Base's DEX trading volume was $98 billion, accounting for 28% globally, second only to Solana; over the past 12 months, Base's share of DEX volume within the ETH ecosystem rose from 2.8% to 9.25%.
Basenames minted over 200,000 in one week after launch, while ENS took nearly two years to reach that number—this is putting the ".eth identity" into the wallets of Coinbase's novice users.
More importantly at the base layer: Coinbase Smart Wallet hides seed phrases, gas, and bridges behind FaceID + gasless transactions; Aerodrome / Uniswap / Morpho on Base all connect through the same compliant gateway. Previously, accessing ETH required learning MetaMask; now opening Coinbase is equivalent to accessing ETH.
This is completely different from "another OP Stack chain":
Arbitrum competes for DeFi depth (GMX, Pendle, Aave, BUIDL II are all there);
Base competes for the moment when the next billion users first encounter blockchain—Farcaster, Zora, AI Agent payments, JPMD deposit token pilot all run on Base.
In short:
Arbitrum is Ethereum's "NYSE trading floor," Base is Ethereum's "Robinhood App." The former serves institutional market makers, the latter puts ETH into the icon ordinary people open every day.
Whether $ETH can surpass 4000 this August depends not on L1 gas fees but on two things:
① Whether Base's daily active users can stay above 1 million;
② Whether traditional brokerage accounts will really start increasing ETH exposure after the ETHE cash dividend on 8/7.
If both hit, ETH will no longer be "just another public chain," but the global financial system's backup clearing mainnet.
$ETH The recent rebound in the Asian market essentially involves trading two expectations.
One is the easing of geopolitical risks, leading to funds withdrawing from safe-haven positions.
The other is the AI cycle regaining capital recognition, with semiconductor leaders becoming buy targets again.
But the real trend reversal depends on trading volume and the sustainability of capital.
Panic is responsible for creating low-level chips.
The AI industry delivering results will be responsible for driving the next round of gains. $BTC $ETH #闪迪财报前夕，HBF与存储紧缺引发热议
SanDisk will release its fiscal 2026 Q4 results after the market closes tonight. The options market has priced in a 16% stock price volatility post-earnings, and the pre-market stock price has surged over 10%. The release of the HBF standard, sold-out capacity, and pre-market rally — tonight will be a "big test" for the storage sector.
---
① Market Expectations: $8.39B Revenue, $33 EPS
Wall Street consensus expects SanDisk's Q4 revenue to be about $8.39 billion (up 41% quarter-over-quarter), with EPS around $33 (up 43% quarter-over-quarter). The company's own guidance is $7.75-8.25 billion. The market consensus is above the company's upper guidance, meaning expectations have been pushed quite high — merely "meeting expectations" might not be enough.
Goldman Sachs previously raised SanDisk's target price sharply from $1200 to $2200, citing the structural demand of the AI storage supercycle. The forward order book size has reached $41.6-42 billion, providing visibility for revenue in the coming quarters.
② HBF Standard Release: The "New Weapon" for AI Inference
On the day before the earnings report (August 4), SanDisk and SK Hynix jointly released the first standard specification for High Bandwidth Flash (HBF).
HBF is positioned between HBM and SSD, targeting AI inference scenarios, balancing high bandwidth and large capacity. It features 8-layer and 16-layer NAND stacking, up to 512GB capacity, with bandwidth covering 0.4-3.0TB/s. This standard is released through OCP (Open Compute Project), aiming to establish an open standard to solve memory bandwidth and capacity bottlenecks in AI inference.
③ Sold-Out Capacity: 2027 Inventory Already Sold Out
Industry chain sources indicate that Samsung, Micron, and SK Hynix's DRAM and HBM capacities for 2027 have all been fully allocated. SanDisk's NAND Flash full-year capacity has also been pre-sold. SK Group Chairman Choi Tae-won expects AI semiconductor demand in 2027 to increase by 60%-100% compared to 2026, with overall storage demand rising 50%-60%, facing the most severe supply-demand imbalance in history.
SanDisk and SK Hynix aim to launch HBF samples in the second half of 2026 and AI inference products in early 2027.
④ Earnings Highlights
Whether revenue can surpass $8.39 billion — if it only "meets expectations" or is "slightly above," the market may not reward much. The key is data center revenue; Zacks consensus expects data center revenue of $2.714 billion (up 85% quarter-over-quarter). NAND pricing trends and enterprise storage demand are core observation indicators. The 2027 outlook and HBF commercialization timeline will determine market confidence in the long-term narrative.
⑤ My Judgment
SanDisk's short-term trend depends on whether tonight's earnings are "above expectations" or "not sufficiently above." If revenue approaches $9 billion, data center revenue beats expectations, and the 2027 outlook is optimistic, the stock price could break through $1,500 or even higher. If it only "meets expectations," the pre-market 10% gain may already be fully priced in.
The fundamental logic of the storage sector remains unchanged — AI storage demand is real, and capacity has been sold out through 2027. But market concerns about "good news already priced in" are also real. Tonight's data will decide whether SanDisk returns to an upward trajectory or continues to digest its valuation. $SNDK Recently, as discussions on the U.S. digital asset regulatory bill continue to advance, market expectations for regulatory implementation have been steadily rising. However, until the final outcome is clear, listed companies related to stablecoins and crypto infrastructure still face a higher policy risk premium than Bitcoin. Recent market performance shows that BTC has generally remained relatively stable, while some crypto concept stocks, stablecoin industry chains, and DeFi sectors have seen significantly greater volatility, indicating that capital is reassessing the potential impact of regulatory changes on business models. Meanwhile, U.S. tech stocks remained strong, Asian semiconductor sectors continued to rebound, and overall risk asset sentiment improved. Spot Bitcoin ETFs have maintained net inflows for several consecutive days, providing some support for BTC. However, the market still finds it difficult to determine whether the price increase is due to improved macro liquidity or regulatory expectations. On the other hand, stablecoin business remains an important source of income for some trading platforms. If the U.S. introduces stricter regulations on stablecoin incentives, compliance requirements, and consumer protection in the future, related companies' profit models may face new adjustment pressures. Therefore, compared to BTC, the stablecoin ecosystem and DeFi sector remain more sensitive to policy changes. In the short term, the market will continue to focus on U.S. regulatory developments, Federal Reserve policy expectations, ETF capital flows, and earnings reports from large tech companies. These factors will continue to jointly determine the flow of funds and risk appetite in the next phase of the crypto market. $BTC $ETH $SOL $COIN $CRCL#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？
SpaceX's first earnings report numbers are indeed impressive, but the stock fell 8% after hours.
Revenue was ¥7.814 billion, up 92% year-over-year, exceeding the expected ¥6.9 billion. Net loss narrowed from ¥1 billion to ¥541 million. All three major business segments exceeded expectations, and the performance itself is flawless.
However, AI spending was ¥15.8 billion, total capital expenditure was ¥18.37 billion, up 550% year-over-year. The pressure on free cash flow immediately surfaced. During regular trading hours, the stock price first rose 9.4%, then within 15 minutes after hours, it dropped sharply from 131 to 114, following a very typical pattern of fully priced-in good news.
The bigger variable is tomorrow. On the second trading day after the earnings release, up to 911.5 million shares held by insiders will be unlocked. Currently, only 640 million shares are circulating, so the unlocked volume is 1.4 times the current float.
Short sellers have already gathered in advance. There are 219.3 million shares shorted, accounting for 34% of the float, even more than Tesla. Short sellers have unrealized gains of ¥7.3 billion and are still adding to their positions.
On the other side, 27 institutions have given "buy" ratings, with an average target price of $236-$239. Morgan Stanley's target is $300, with the highest target at $800. Both sides are betting real money.
My judgment: The earnings report itself is very good, but in the short term, pricing power is not in the performance but in the unlocking. The potential selling pressure of 900 million shares, combined with the already substantial unrealized gains of short sellers, means any rebound could be crushed. I will neither chase longs at this position nor short. Chasing longs is not worth the risk-reward ratio because the selling pressure from unlocking has not yet materialized; chasing shorts is risky because short positions are already extremely crowded. On Facebook's unlocking day in 2012, the stock price rose 12.6%, and a short squeeze would be disastrous.
Wait for the unlocking to settle and the chips to change hands before assessing the real supply and demand. A good company does not equal a good price, and a good price does not equal a good timing. #AMD财报超预期，增长已被透支？
AMD delivered an almost flawless report card—record revenue of $11.536 billion, data center revenue doubled, and earnings per share exceeded expectations. Yet, the stock price fell as much as 9% after hours.
It's not that the performance was bad; the market just felt it was "not good enough."
---
① Data: Flawless
AMD's Q2 revenue was $11.536 billion, up 50% year-over-year, a historic high; non-GAAP EPS was $1.66, beating the expected $1.62; gross margin was 56%, net profit $2.76 billion.
The standout was the data center business. Revenue was $6.7 billion, up 107% year-over-year, accounting for 58% of total revenue. Server CPUs set records for five consecutive quarters, with cloud and enterprise sales both growing over 70% year-over-year. PC client business revenue was $3.1 billion, up 23% year-over-year. The AI rack system Helios began shipping, securing major clients like OpenAI, Meta, Anthropic, and Microsoft. CEO Lisa Su expects the data center business to "more than double" by 2027.
The only weak spot was gaming, with revenue of just $779 million, down 31% year-over-year.
② Why the drop?
Q3 guidance didn’t deliver a "surprise." The company expects Q3 revenue of $13 billion (plus or minus $300 million), above analysts’ average estimate of $12.5 billion but below some aggressive buyers’ expectations of $13.5–14 billion. The stock has more than doubled this year, and the market’s tolerance for "exceeding expectations again" is very low; the midpoint of $13 billion only means "in line with expectations," not a "surprise."
Capital expenditures surged. Q2 capex was $808 million, compared to just $282 million the same period last year. Free cash flow dropped sequentially to $1.56 billion. Money has been spent, and the market is waiting for faster returns.
SpaceX dealt a blow. On the same day, SpaceX announced that future AI computing power will exclusively use NVIDIA GPUs, with Elon Musk calling NVIDIA’s Vera Rubin architecture "the best AI computer currently." This was a heavy psychological blow to AMD—the moat around NVIDIA’s ecosystem continues to widen.
③ Is it valuation overstretch or short-term sentiment?
AMD’s problem isn’t performance; it’s that expectations have run too far ahead. The market has already labeled it as the "NVIDIA challenger," and the stock price has doubled. When you’re expected to keep breaking records at the peak, any "not quite surprising" result can be magnified. Capital expenditures are soaring, competitive pressure is increasing, and the market no longer wants just "growth" but "growth that exceeds expectations."
Lisa Su said on the earnings call, "We are still in the early stages of a multi-year AI adoption cycle." The long-term narrative remains unchanged, but short-term valuation needs time to digest.
The stock fell despite beating expectations not because AMD performed poorly, but because the market wants more. When a company is already burdened with the high hopes of "challenging NVIDIA," every report card becomes harder to pass.
$AMD $NVDA Alt season, selective capital rotation phase The part already reflected in the price and the variables not yet reflected can be divided first, and the current market clearly shows that it is not a rise across the entire range but a selective rotation of capital. The key facts observed in the original text are as follows. The majority of altcoins are consolidating, and a pattern is confirmed where capital is concentrated only in certain assets such as $JTO $JELLY $BTC $OPG $BTCSLX $LAB $BSB $ALLO $CHIP. On the other hand, $BEAT $EDGE $COAI $TRUMP $RAVE $SPACE $SOPH $IP $AVNT $ZAMA $OFC $PIEVERSE $VIRTUAL $ACU $H $MEGA are classified as having already lost momentum. Additional watchlist assets mentioned include $MEME $EDEN $HUMA $ZKP $METIS. Reading this situation from the perspective of derivative risk, it shows not just a matter of asset selection but the market's leverage distribution and liquidation risk structure. Conditions for rise, invalidation conditions, tail risk, Italy's UniCredit Sanpaolo Bank reduced its IBIT common shares from 646,809 to 40,723 in Q2, a decrease of 606,086 shares, or 93.7%. The reported market value also dropped from $24.85 million to $1.356 million.
Looking at this data alone, it’s easy to conclude that "Italy's major bank is exiting BTC." However, when the entire 13F holdings are examined, the actual moves resemble product switching and risk restructuring.
The bank still held 3.4736 million shares of ARKB during the same period, only down 3.7% from 3.6076 million shares in Q1, with an end-of-period reported value of about $67.63 million. ARKB remains its largest disclosed crypto ETF holding by scale, approximately 50 times the value of the remaining IBIT common shares.
Therefore, reducing IBIT does not equate to a significant cut in total Bitcoin exposure. More accurately, it significantly lowered BlackRock's IBIT position but retained the ARK 21Shares Bitcoin ETF as the main BTC allocation.
On the ETH side, there was a clear increase.
BlackRock's staked Ethereum ETF ETHB holdings rose from 116,200 shares to 349,600 shares, an increase of 200.9%; the reported value increased from $3.148 million to $7.097 million. ETHB provides ETH price exposure while including some staking yield, adding an extra layer of income compared to simply holding a spot ETF.
However, this is not yet an "ETH replacing BTC" scenario. ETHB's reported value is only about 10.5% of ARKB's, and overall BTC-related holdings remain significantly larger.
Derivatives changes are even more noteworthy. IBIT call options dropped sharply from 2.4965 million shares to 18,000 shares, a 99.3% decrease; meanwhile, a new put option corresponding to 500,000 IBIT shares was added.
But the 13F does not disclose strike prices, expiration dates, premiums, or the full portfolio relationships of these options, so this 500,000-share put exposure could be protective hedging or part of a more complex strategy, and cannot be directly interpreted as the bank being fully bearish on BTC.
This filing does not reflect a simple "sell BTC, buy ETH" move, but rather four simultaneous actions: compressing IBIT common shares, nearly clearing IBIT call options, retaining the core ARKB position, and increasing ETH exposure with staking yield.
The next 13F will need to be watched for two ratios: whether ARKB also begins to decline significantly, and whether ETHB’s share of the crypto ETF portfolio continues to rise. Only if both occur simultaneously can it be considered a true shift from BTC to staked ETH.
#意大利大行减IBIT普通股94%，加仓质押ETH #闪迪财报前夕，HBF与存储紧缺引发热议
$SNDK This SanDisk grid trade has been running for most of the day, entered at 1391, now at 1465, with a floating profit of 17%. The grid has been triggered 509 times. I plan to stop it by 11 PM.
SanDisk's earnings report comes out early tomorrow morning; holding the position now is essentially betting on the direction. Grid trades are suitable for sideways markets, not for betting on a one-sided move. It's better to take profits before the earnings and decide again once the direction is clear.
Lately, the market is very familiar with the "beat expectations but price drops" scenario—$AMD and SpaceX are precedents. If SanDisk's earnings explode but the price drops after hours, the grid will be directly broken below the lower bound. If earnings miss expectations, a sharp drop is inevitable. In either case, holding a grid through earnings is not a good choice. BTC has returned to 64,000, but only 28 coins are seeing inflows. Is this a sign of recovery, or are big players quietly switching seats? The current market looks more like a quiet rotation game rather than a broad rally celebration. I've been watching the capital flow chart for a long time, and my biggest impression is: the money hasn't disappeared, it's just become very selective. BTC has a net inflow of $170 million; ETF money has indeed returned, but the volume didn't keep up when the price broke through, indicating that big money is buying, but not frantically hoarding—more like slowly accumulating at low levels while observing. ETH, on the other hand, has a net outflow of $11.9 million; institutions are quietly reducing risk by borrowing BTC's strength. This "big brother leads, little brother retreats" rhythm is often not a signal for a full-scale attack. What really draws attention are two very clear trends: AI computing power and U.S. Treasury RWA. ONDO saw an inflow of $11.5 million, PENDLE $12.6 million, and ENA $8.3 million. The funds are not scattered randomly but are precisely flowing into sectors that can tell a "yield" story. On the other side, TRX took in $16.7 million riding the RWA narrative, even veteran public chains are starting to leverage this tailwind. Meanwhile, MEME, old public chains, and GameFi have become hard-hit areas. DOGE had an outflow of $36.8 million, ADA $21.3 million, and ARB dropped by $12.4 million due to unlocking pressure. The withdrawal from these sectors is not accidental; the market is abandoning "pure sentiment" and " Italy's UniCredit Sanpaolo Bank reduced its IBIT common shares from 646,809 to 40,723 in Q2, a decrease of 606,086 shares, or 93.7%. The reported market value also dropped from $24.85 million to $1.356 million.
Looking at this data alone, it’s easy to conclude that "Italy's major bank is exiting BTC." However, when the entire 13F holdings are examined, the actual moves resemble product switching and risk restructuring.
The bank still held 3.4736 million shares of ARKB during the same period, only down 3.7% from 3.6076 million shares in Q1, with an end-of-period reported value of about $67.63 million. ARKB remains its largest disclosed crypto ETF holding by scale, approximately 50 times the value of the remaining IBIT common shares.
Therefore, reducing IBIT does not equate to a significant cut in total Bitcoin exposure. More accurately, it significantly lowered BlackRock's IBIT position but retained the ARK 21Shares Bitcoin ETF as the main BTC allocation.
On the ETH side, there was a clear increase.
BlackRock's staked Ethereum ETF ETHB holdings rose from 116,200 shares to 349,600 shares, an increase of 200.9%; the reported value increased from $3.148 million to $7.097 million. ETHB provides ETH price exposure while including some staking yield, adding an extra layer of income compared to simply holding a spot ETF.
However, this is not yet an "ETH replacing BTC" scenario. ETHB's reported value is only about 10.5% of ARKB's, and overall BTC-related holdings remain significantly larger.
Derivatives changes are even more noteworthy. IBIT call options dropped sharply from 2.4965 million shares to 18,000 shares, a 99.3% decrease; meanwhile, a new put option corresponding to 500,000 IBIT shares was added.
But the 13F does not disclose strike prices, expiration dates, premiums, or the full portfolio relationships of these options, so this 500,000-share put exposure could be protective hedging or part of a more complex strategy, and cannot be directly interpreted as the bank being fully bearish on BTC.
This filing does not reflect a simple "sell BTC, buy ETH" move, but rather four simultaneous actions: compressing IBIT common shares, nearly clearing IBIT call options, retaining the core ARKB position, and increasing ETH exposure with staking yield.
The next 13F will need to be watched for two ratios: whether ARKB also begins to decline significantly, and whether ETHB’s share of the crypto ETF portfolio continues to rise. Only if both occur simultaneously can it be considered a true shift from BTC to staked ETH.
#意大利大行减IBIT普通股94%，加仓质押ETH BICOUSDT (Strong rebound pullback for dip buying)
- Entry range: 0.0205~0.0210
- Stop loss: 0.0190
- Take profit level 1: 0.0248
- Take profit level 2: 0.0272
MVLLUSDT (Small to mid-cap rebound long)
- Entry range: 26.8~27.3
- Stop loss: 25.2
- Take profit level 1: 31.2
- Take profit level 2: 33.5
SNXXUSDT (Leveraged index trend long)
- Entry range: 12.6~12.8
- Stop loss: 11.9
- Take profit level 1: 14.5
- Take profit level 2: 15.6
KAITOUSDT (AI sector pullback dip buy)
- Entry range: 0.86~0.875
- Stop loss: 0.81
- Take profit level 1: 0.99
- Take profit level 2: 1.06
BTCUSDT (Large cap rebound dip buy)
- Entry range: 62000~62400
- Stop loss: 60800
- Take profit level 1: 65600
- Take profit level 2: 67200
ETHUSDT (Large cap rebound dip buy)
- Entry range: 1790~1810
- Stop loss: 1730
- Take profit level 1: 1920
- Take profit level 2: 1990
SOLUSDT (Leading public chain dip buy)
- Entry range: 70.5~71.5
- Stop loss: 67.8
- Take profit level 1: 78.2
- Take profit level 2: 82.0
ZECUSDT (Privacy coin trend dip buy)
- Entry range: 488~495
- Stop loss: 468
- Take profit level 1: 538
- Take profit level 2: 565Many people associate SanDisk with being an old manufacturer of USB flash drives and storage cards. Now, with its stock price soaring, I will explain the underlying logic thoroughly and also discuss the linked impact on the crypto market.
First, the spin-off listing completely sheds the burden.
Previously, it belonged to Western Digital, and the mechanical hard drive business always dragged down its valuation. After the independent spin-off in 2025, it purely focuses on NAND flash and enterprise-grade SSDs, with no outdated business holding it back. Wall Street immediately revalued it, opening up a new starting point.
Second, AI brings solid explosive demand, which is the core driver of the price surge.
Everyone focuses on GPUs when hyping AI but neglects storage. Large model inference and RAG retrieval cause AI servers to consume several times more NAND flash than traditional servers. Big companies like Microsoft and AWS are aggressively expanding data centers, and enterprise SSD orders are booming. As a leading supplier, SanDisk’s data center business has grown several times year-over-year, with gross margins reaching 78%. The performance explosion is not just a story; the financial reports truly deliver.
Third, NAND flash supply and demand are tight, and product prices continue to rise.
Major manufacturers prioritize capacity for high-margin HBM, and NAND expansion can’t keep up with the incremental demand from AI. Spot and long-term contract prices keep climbing. There’s no worry about selling inventory, and they hold the pricing power, directly amplifying profits.
Fourth, business model upgrade locks in long-term revenue.
They have signed long-term volume-lock supply agreements with leading overseas cloud providers, securing orders worth hundreds of billions in advance. Even if the storage cycle fluctuates, the company has guaranteed minimum revenue. The market no longer treats it as a purely cyclical stock with big swings and is willing to assign a higher valuation.
Fifth, institutional funds are banding together, and investment banks are collectively raising target prices.
Institutions like Goldman Sachs remain bullish, continuously raising target prices. Incremental funds keep flowing in, further boosting the stock price. But objectively, after such a huge rise, the valuation is already very high. If future AI capital expenditures fall short of expectations or storage supply releases, the correction could be very painful.
Regarding the connection to the crypto market, here’s the straightforward truth:
1. SanDisk represents the AI storage mainline, which continues to strengthen, indicating the market’s ongoing bet on AI capital expenditure. The overall risk appetite in the AI sector warming up will indirectly boost crypto market sentiment.
2. However, it belongs to the US semiconductor stock trend and will not directly drive a unilateral surge in Bitcoin. The US tech stock rally can provide emotional support for BTC but cannot determine its direction.
3. Conversely, if SanDisk’s future performance disappoints and valuation is cut sharply at high levels, it will drag down the entire tech growth sector, and Bitcoin will also come under pressure. This risk cannot be ignored.Is a $3 trillion market cap just the beginning? Amazon's $50 billion bet has already won the crucial first step.
Amazon officially reached a $3 trillion market cap, becoming the fifth global tech giant to hit this milestone. The recent surge in stock price is no longer driven by traditional e-commerce business, but by AWS cloud computing delivering a financial report far exceeding market expectations.
In Q2, AWS revenue surged 37% year-over-year, marking the highest growth rate in nearly 18 quarters, solidly confirming that enterprise AI computing demand is still expanding explosively. The biggest market concern before was whether Amazon's continuous heavy investment in expanding AI data centers and massive capital input would be difficult to yield returns. The latest financial report provides a clear answer.
The company directly raised its full-year 2026 capital expenditure to $220 billion, an increase of $20 billion from the original plan, with the vast majority of funds invested in AI computing infrastructure, server, and storage hardware procurement. Interestingly, despite the huge spending, the market chose to vote with a new stock price high. More and more institutional investors have reached a consensus: this massive investment should not be simply viewed as a cost but as an early lock-in of core infrastructure chips for the AI era in the coming years.
Management also sent a strong signal: even maintaining the current high-intensity construction pace, the tight computing power situation will continue, with some customer demand already booked through 2028. AWS has nearly $500 billion in long-term orders on hand, showing very strong demand certainty.
This upstream transmission along the industry chain directly benefits the storage chip sector. The large-scale expansion of AI data centers continues to drive memory and storage chip procurement, validating the positive outlook for SNDK and SKHYNIX.
Risk warning: high capital expenditure will continue to suppress free cash flow. Future performance depends on whether orders can continue to convert into actual profits. After computing power and storage supply ease, price competition pressure will also arise.
$AMZN $SNDK $SKHYNIX
#AmazonMarketCapBreaks3Trillion, $50BBetWinsFirstRound #AMD财报超预期，增长已被透支？
AMD's biggest problem is not that it isn't profitable, but that it has spoiled its investors.
Scoring 90 points the first time made people think it was really impressive;
later, everyone expected you to score above 90 every time.
So when you scored 90 this time, it actually felt a bit average.
$AMD This earnings report is just like that.
Revenue grew 50%, data center surged 107%, which would have been a bombshell in the past.
But now the market isn't looking at whether you are growing, but whether you can keep surpassing expectations and keep everyone amazed.
The AI market has entered the second half; investors are buying not today's profits, but the miracles of the coming years.
So AMD's stock price falling doesn't mean the market is pessimistic about it, but that the capital market is moving from believing the story to checking for fulfillment.
In the past, being excellent was enough to drive the stock up.
Now, excellence is just the entry ticket; you have to be outstanding, even astonishing.
To keep rising, you must prove time and again that you can still create miracles.
Otherwise, the stock price will definitely keep falling.
📌 Live trading statement
Current position: BTC spot long-term + index fund regular investment. No contracts, leverage, or futures.
This batch of news does not change the current position plan—DEX growth and RWA migration are long-term trends and will not affect BTC dollar-cost averaging in the short term.
---
1. Power Transition Between DEXs and CEXs: Trading Volume Shares Break 24% for the First Time
In July, the spot trading volume ratio between DEXs and CEXs exceeded 24% for the first time, compared to just 17% a year ago. The growth rate far exceeded most people's expectations.
A more critical signal is Uniswap's move: it is developing the token issuance platform "Pools" on Robinhood Chain. There are two models—Crowd Launch (4-hour auction, 50% auction + 50% v4 liquidity pool, migrating only after meeting the $50,000 FDV target) and Instant Launch (Bonding Curve model, 80% tokens traded on a curve). Hundreds of auction events have been spawned by related contracts, and infrastructure has been deployed.
What does this mean? Both transaction and token issuance gateways are migrating on-chain. DeFi is no longer just a stock TVL game. When DEXs start competing for CEX's "token issuance rights," the core moat of CEXs—the right to list tokens and liquidity distribution—is being dismantled from the root by on-chain protocols.
Uniswap uses a more institutional auction mechanism + Robinhood Chain's compliant entry point, setting it apart from pump.fun and other launch platformsPositive narratives cannot withstand real pressure; a comprehensive analysis of SPCX's financial report, cash-burning crisis, unlocking risks, and space computing power outlook. The first earnings report opened high but then declined; SPCX now faces the risk of falling below $100 (a double boom between bulls and bears). 1. First Listed Financial Report, Impressive Surface Performance Q2 SpaceX Revenue Reached $7.8 Billion, a 92% year-on-year increase, far exceeding market expectations; Net loss narrowed to $541 million, adjusted EBIT-DA rose 192% year-on-year, and the profit base continues to improve. The structure of three major business segments is very clear: 1. Starlink is the sole cash cow: revenue of $4.3 billion, operating profit of $1.66 billion; user base exceeds 12 million. To capture overseas markets, it launched low-priced packages, resulting in a decline in per-user revenue and explosive growth in user base. 2. AI business experienced rapid expansion: revenue tripled year-on-year, with a large number of new cloud service orders; Today, xAI, Grok, X social platforms, and ground computing clusters have formed a complete AI landscape. 3. Starship development is progressing steadily, with V3 version preparing for test flights, and a new generation of Starlink satellites to be launched soon. Musk updated his strategic plan externally, moving up the goal of achieving $1 trillion in annual revenue from 2031 to 2030; At the same time, it was officially announced that all AI computing power equipment will use NVIDIA GPUs, and plans to deploy orbital space data centers will utilize natural space cooling and solar energy resources, envisioning the long-term vision of a second-level Kardashev civilization. 2. The Core Pain Point Behind the After-Market Crash:#AMD财报超预期，增长已被透支？
Did AMD's earnings exceed expectations? The numbers did, but the market had already priced in those expectations.
Revenue of 11.5 billion, with data center revenue at 6.7 billion, growing 107%—sounds impressive, right? Yet the stock still fell 8% after hours.
The issue isn't the performance, but the expectations. Investors were looking for 13 billion, a record-breaking battle crushing Nvidia. Instead, AMD only hit 11.5 billion, and the Q3 guidance is just 13 billion—barely matching what others had forecasted, not a penny more. Gross margin is stuck at 56%, showing no sign of improvement. How can that justify a higher valuation?
But the drop isn't undeserved.
The Helios rack-level system will officially ship in Q3. This is AMD's first time going head-to-head with Nvidia with a complete solution. The gross margin hasn't moved because volume hasn't ramped up yet. Once Helios takes off, profit improvement is inevitable.
This drop is due to retail investors' expectations being too high and institutions taking profits. Helios hasn't powered up yet; once real data comes out, valuations will need to be recalculated.
The drop is handing chips to those who understand, making AMD more dazzling than ever for those waiting to buy the dip. Caught late by my boss this morning, I was instead awakened by a message on my phone: The S&P 500 surged past 7700 points, with a market cap breaking 70 trillion yuan!
This morning, as I rushed into the company elevator, I bumped right into my boss.
He glanced at his watch and was about to speak when my phone suddenly popped up a market alert:
The S&P 500 has surpassed 7700 points, hitting a new all-time high!
At that moment, I didn’t even have time to explain being late.
What does 7700 points mean?
Based on this level, the total market capitalization of the S&P 500 has reached an astonishing scale of about 70 trillion yuan. Some are excitedly proclaiming a "new round of wealth feast," while others are starting to warn:
The higher it rises, the harder it falls—could this be another bubble inflated by capital?
This is definitely not a broad bull market where everyone can easily make money.
1. The index is rising fiercely, but it’s not just driven by hype.
Many see 7700 points and immediately think: "Valuations have gone through the roof."
But to judge if the market is in a bubble, you can’t just look at the index level; you have to see how much profit companies are actually making.
The underlying driver of this rally isn’t just chasing momentum, but some core companies’:
- Continuous growth in earnings per share (EPS);
- Consistent realization of free cash flow;
- Genuine revenue growth;
- AI-related investments beginning to convert into commercial returns.
In the past, the market’s AI hype was more about imagination; now, some leading companies have started turning computing power, models, and data center investments into real revenue.
The story is shifting from "potential future profits" to "already starting to make money."
As long as profit growth can keep up with valuations, 7700 points can’t simply be equated with a nonsensical bubble.
Of course, current prices still include optimistic expectations for rate cuts, AI growth, and corporate earnings. If actual results fall short, market volatility will quickly amplify.
2. The real drivers pushing the index up are not all companies, but a few super giants.
The biggest feature of this boom is not "broad-based growth," but capital highly concentrated in top-tier assets.
The top ten weighted stocks account for nearly 40% of the market cap. On the surface, this is risk concentration; from another perspective, these companies are becoming "safe-haven hard currency" in the eyes of global capital.
The reasons are straightforward.
They often have:
- Strong balance sheets;
- Ample cash reserves;
- Global revenue sources;
- Extremely high industry barriers;
- Pricing power in an inflationary environment.
In uncertain economic times, capital doesn’t necessarily leave the market entirely but withdraws from weak assets and shifts to super companies with stable cash flow and strong profitability.
So, what we see is not all stocks partying together, but capital making a harsh choice:
Whoever can weather the cycle, money flows to them.
This also explains why the index keeps hitting records, but many ordinary investors’ accounts don’t necessarily reach new highs.
A rising index doesn’t mean every stock is rising; a booming market doesn’t mean everyone is making money.
3. The truly scary factor isn’t on-exchange funds, but the "ammunition depot" off-exchange.
Another often overlooked factor enabling the market to keep rising is off-exchange capital.
When interest rates were high, a large amount of capital stayed in money market funds and cash products because investors could earn relatively decent returns without doing anything.
But if the rate-cut cycle continues and cash asset yields keep falling, these waiting funds will start recalculating opportunity costs:
- Holding cash yields less and less;
- Entering the market risks buying at too high a level.
So, this money may not rush into stocks all at once but is likely to enter in batches during every noticeable pullback.
This means there could be a "liquidity cushion" formed by huge off-exchange capital beneath the market.
As long as profit expectations aren’t completely destroyed, every dip could attract bargain hunters.
This is also the typical operation mode of a structural long bull market:
Not a straight line up, but gradually raising the bottom through doubts, pullbacks, and oscillations.
But don’t get too excited: 7700 points doesn’t mean the risk has disappeared.
I am optimistic about this structural boom, but this optimism must be conditional: whether AI investments continue to convert into revenue. We shall wait and see. 🪝 Recently, a hot speculation has been circulating in the community: OKX's strategic focus is fully shifting toward X Layer and OKB, and in the future, will they gradually distance themselves or even abandon $CORE? Panic continues to spread, and many holders are becoming anxious. Setting aside sentiment rumors, we objectively analyze the cooperation history of both parties, their track strategies, and exchange operation logic. 1. First, review: The real foundation of cooperation between OKX and CORE Many people don't know that OKX was among the earliest leading exchanges to deeply support CORE. 1. Spot trading will be launched on the mainnet during its initial launch, continuously providing liquidity support; 2. The two parties have reached an official strategic cooperation, with OKX launching staking products on the CORE platform and opening up user participation in on-chain verification channels; 3. OKX Web3 wallet natively supports Core public chains, collaborating to launch NFT minting and community collaboration activities; 4. OKX has long served as a Core network validator node, directly participating in public chain security operation and maintenance. Most importantly: CORE focuses on BTC native non-custodial staking, and the track is the currently hot BTCFi. OKX has long been involved in the Bitcoin ecosystem, continuously funding underlying Bitcoin developers. From a track perspective, the two have a natural alignment and are not two unrelated tracks. 2. Why have rumors emerged in the market that "OKX is about to abandon CORE"? The rumors originated from OKX's own strategic adjustment: OKX concentrated resources on its own Layer 2 network, X Layer, empowering OKB with full platform resources,#EarningsObserver: AMD and SpaceX Report, Circle Closes the Show
The first three earnings reports of the week are done, and the results are quite divergent.
I'm Nanyan Gu
Palantir surged 12%, revenue up 93%, and raised its full-year guidance. The market is willing to pay a premium for a simple reason—it has proven that demand for AI application layers is real and accelerating.
AMD's data is solid, with revenue of 11.5 billion and data center revenue of 6.7 billion, up 107%. But it fell 9% after hours because its performance was "good," while the market demands "perfect." AMD's results represent the midstream of the chip sector, where upstream and downstream pressures are squeezing its profit margins.
SpaceX reported revenue of 7.8 billion, up 92%, with narrowed losses, but a 100 billion lock-up release looming the day after tomorrow weighed on it, causing a 2.7% drop after hours.
A pattern emerges from these three earnings reports—exceeding expectations has become the entry ticket; the next questions are how long growth can sustain, when profits will materialize, and whether valuations have already priced in these factors. Palantir wins on incremental expectations, while AMD and SpaceX lose due to higher thresholds.
Today before the market opens, Circle closes the show.
For the crypto market, this is what truly deserves attention.
Coinbase revenue fell 18.5%, Robinhood's crypto revenue dropped nearly 40%, and USDT growth has almost stalled. The crypto sector has already submitted three reports, all pointing in the same direction—the liquidity base is contracting.
If USDC circulation is still rising, it means money hasn't left but just shifted from USDT to USDC. If it is also falling, that is a real warning sign that the stablecoin market is shrinking overall.
USDC reserves had already shrunk to 72.06 billion USD by the end of July. Whether the volume gap can be filled by interest rates is the core suspense of Circle's earnings report. USDC reserves had already shrunk to 72.06 billion USD by the end of July; whether the volume gap can be filled by interest rates is the core suspense of this report.
My judgment is simple—Circle's answer will directly affect the market's assessment of liquidity across the entire crypto industry. What it validates is not just one company's performance but the direction in which the underlying liquidity structure of the whole market is moving.
Nanyan Gu has finished speaking, what do you think?
$BTC $SNDK $SOL#标普500首次站上7700点，创历史新高
If you want to make money in US stocks, you need to change your mindset. Even if you trade US stock contracts, you have to pick those with trending volatility.
Before, it was blindly buying AI.
Buy chips, buy storage, AI servers are hot, buy the entire industry chain.
It really made money; anyone related to AI started with a tenfold increase.
Now it’s not the case, because semiconductors are cooling down, but the funds haven’t left US stocks.
Where did the money go?
It went to cloud storage, software, consumer sectors, even traditional companies like Apple and Coca-Cola.
What does this indicate?
It’s not that the market doesn’t believe in AI anymore, but the market is starting to calm down.
GPU, chips, server valuations are getting higher and higher, soaring to the sky, so you start to think:
Who can truly use AI to improve efficiency?
Who can turn AI into profit?
Who has stable cash flow?
So you find: chips adjust, cloud computing takes over;
Tech stocks diverge, consumer stocks warm up.
This is actually a very normal step in a bull market:
From speculating on expectations to focusing on performance.
In the future, US stocks may no longer be a "buy the right sector and everyone goes up" market.
It will only be index fluctuations, sector rotations, and stock divergences.
If you want to make money, you have to learn; it’s not about always chasing the hottest concept, but understanding where the funds will go next when the market switches.
The big direction of AI may not be over, but the era of "blindly buying AI" is over.
What’s next is about understanding the industry, not chasing hot spots, so forget what you already know about US stocks.
Go learn about truly valuable undervalued companies. I suggest you look at $AMZN, $GOOGL, and the like; they are definitely more worth researching than storage.$SPCX
SpaceX's earnings far exceeded expectations, but the stock plunged 7% after hours: this script is quite different from what was imagined
Last night, SpaceX delivered its first earnings report since going public. Revenue was 7.8 billion, a year-over-year surge of 92%, far exceeding the expected 6.9 billion; adjusted EBITDA reached 3.5 billion, while the market expected only 2 billion. Looking at these numbers alone, it's a clear domination.
However, the stock price dropped nearly 7 points after hours, currently around $117.
The core reason can be summed up in three words: heavy spending.
Capital expenditures soared to 18.4 billion in Q2, significantly higher than the expected 13.2 billion, with 15.8 billion invested in AI business. Although Starlink contributed 4.3 billion in revenue and is the only cash cow, the AI side is still burning money, with the AI business alone losing 1.26 billion.
Musk also said that capital expenditures will continue at this scale in Q3 and Q4. The market fears: when will this money-burning stop?
Additionally, there is a major event tomorrow — lock-up expiration. Starting August 6, over 900 million insider shares can be sold, doubling the float. Shorts have already accumulated positions worth 24.6 billion dollars, with unrealized gains of 7.3 billion and are still increasing their bets. Both bulls and bears are waiting for this earnings report as an excuse.
How to operate?
For going long: don't rush at the 117 level. Wait for a pullback near 110-112 to try in batches, with a stop loss at 105. If the lock-up expiration tomorrow triggers a panic low, that could actually be an opportunity. The target is first 125-130.
For going short: don't chase the short either, as the stock has already dropped after hours. Wait for a rebound to 122-124 before acting, with a stop loss at 128 and a target of 110-112.
If the selling pressure after the lock-up expiration is not as big as expected, shorts might get squeezed.
$SPCX$SPCXB#SpaceXAI spending drags down first earnings report #SpaceX's first lock-up period expires August 6 #SpaceX to announce Q2 earnings #SpaceX stock rose 9.8% before earnings#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？
Q2 revenue reached $7.81 billion, a year-over-year surge of 92%, far exceeding Wall Street's general expectation range of $6.8-6.9 billion. Net loss narrowed sharply from $1 billion in the same period last year to $541 million, with a loss per share of only $0.09, better than the market's previous estimate of $0.24-0.26. Adjusted EBITDA hit $3.5 billion, up 191% year-over-year. All three major business segments exceeded expectations:
Connectivity (mainly Starlink) contributed $4.29 billion (+66%), with an operating profit of $1.66 billion; the AI segment generated $2.56 billion (+247%); the Space segment brought in $962 million.
The numbers look impressive. However, after-hours stock price once gave back nearly 8-9%, as the market's real focus did not stop at the word "outperformance." The business structure has completely flipped!
SpaceX is no longer just a rocket company.
Connectivity remains the cash cow. Starlink users doubled year-over-year to 12 million, with a net increase of 1.7 million in a single quarter, a record high. Enterprise and government business revenue more than doubled, with over $6 billion in new multi-year U.S. government Starshield contracts, plus new airline orders. Operating margin reached 38.6%, making it currently the only segment consistently contributing positive profits.
The AI segment is both the biggest surprise and the biggest concern in this earnings report. Revenue surged nearly 2.5 times year-over-year, mainly from newly signed cloud service agreements (total contract value of $14.1 billion). Computing capacity has reached 1.4GW. But capital expenditures soared to $18.4 billion in a single quarter, with about $15.8 billion related to AI, far exceeding market expectations of around $13 billion. The CFO clearly stated that capital expenditures will remain high for the next two to three quarters.
The Space segment continues to fund Starship. R&D investment remains high, operations are still at a loss, but two successful V3 test flights have brought the fully reusable goal one step closer.
From the business structure perspective, SpaceX has completed its transformation from a "launch service provider" to a dual-engine driven company focused on Starlink + AI infrastructure. After IPO fundraising and bond issuance, cash and marketable securities on hand total about $100 billion, with an order backlog of $47.5 billion, ammunition is ample.
The real variable starts with the unlocking beginning Thursday.
The earnings beat is just the appetizer!
According to the lock-up schedule, starting August 6 (Thursday), the first batch of about 911 million shares held by insiders and early investors will be allowed to circulate. At current prices, the potential sellable scale exceeds $100 billion, several times the IPO issuance volume. Larger unlocking windows will follow.
This is the first real stress test SpaceX faces after going public. Historically, the first large tech stock lock-up expirations often come with significant selling pressure, especially when valuations remain high and capital expenditure expectations are aggressive. The 9% pre-market rally driven by earnings expectations quickly gave back after hours, signaling this.
Market divergence is clear:
• Optimists value Starlink's ongoing cash generation ability, the speed of AI contract implementation, and the "vertical integration + scale effect" ultimately driving up gross margins.
• The cautious side watches whether the huge capital expenditures can convert into considerable profits within a reasonable timeframe, and whether the supply shock from unlocking will push the stock price back to a more reasonable range.
SpaceX's current P/E ratio remains one of the highest among Nasdaq 100 components. The high-growth narrative can support a high valuation, but the premise is that growth must continue to materialize, and the capital return cycle cannot be indefinitely extended.
First principles view of this game:
SpaceX's core competitiveness has never been a single technology but the cost and speed advantages brought by extreme vertical integration. Starlink has proven this path effective in consumer and enterprise markets; AI infrastructure is replicating the same capabilities in the computing power domain. Cursor's $60 billion acquisition intention, cooperation with Nvidia, and the release of Grok 4.5 all reinforce the narrative that "we are not just launching rockets, but building the next-generation computing and communication foundation."
But the capital market prices predictable free cash flow, not an indefinite investment story. Once the lock-up window opens, early investors' monetization needs will be real. If AI revenue and gross margin improvements in the next few quarters do not keep pace with capital expenditures, or if Starlink user growth slows, the selling pressure from unlocking will be amplified.
Conversely, if AI contracts continue to be implemented, computing power utilization rises rapidly, and Starlink maintains high double-digit growth, the supply from unlocking will be absorbed by demand, and the stock price has a chance to stabilize on a new platform.
Currently, the three most critical observation points are:
1. The actual selling pressure scale and absorption strength starting August 6;
2. Whether next quarter's capital expenditure guidance remains high;
3. The speed at which the AI segment's operating loss narrows.
The earnings report itself has proven the business flywheel is turning. But the real test after going public has never been a single quarter's impressive numbers, but whether long-term capital is willing to vote with their feet to support this rocket + Starlink + AI super complex amid the unlocking flood.
Unlocking is the main variable in the coming days to weeks. Good numbers do not mean the stock price path is already determined. AMD fell, but AI did not follow: What truly ended in July was not the market, but the panic
1. Core conclusion: Around July 29, the first phase trading bottom of the AI hardware sector was formed
I believe that around July 29, the AI hardware sector had already formed the first phase trading bottom. From last Thursday until now, the market is shifting from indiscriminate deleveraging to repricing based on fundamentals. However, this cannot yet be called a full confirmation of a new major uptrend; more accurately, the most dangerous main downtrend may have ended, and memory chips, semiconductors, and optical modules are entering a valuation recovery phase after bottoming.
What ended in July may not necessarily be a correction, but most likely the most irrational sell-off has ended.
2. What exactly fell in July: Not orders, but valuation, positions, and sentiment
Many think the tech stock crash in July was due to AI demand collapsing, compute investment failing, or memory cycles peaking, but the truth is quite the opposite.
AMD’s data center revenue is still growing rapidly, cloud providers’ capital expenditures have not suddenly disappeared, and industry demand for DRAM, NAND, and high-speed interconnects has not vanished overnight. What truly collapsed in July was not orders, but valuation, positions, and sentiment.
In previous months, the market had priced in growth for the next two to three years in advance, with hedge funds, quant funds, and trend funds all crowded into the same batch of AI assets. Once prices weakened, everyone rushed for the same exit; they had no time to judge whose orders were better and could only sell all AI stocks together.
What the market sold off in July was not artificial intelligence itself, but the illusion that "artificial intelligence will never pull back."
3. Why last Thursday was an important turning point: The bottom is not a single-day surge, but sustained non-decline
A true bottom is never a sudden surge on one day, but a surge that is not immediately reversed. Last Thursday, the Philadelphia Semiconductor Index rebounded sharply and did not experience a typical "one-day wonder" but rather a continuous oscillating recovery. Even with intraday pressure in recent days, memory chips and optical modules have not collectively plunged again like in July.
More importantly, the leading range began to expand: initially, core heavyweights like Nvidia and Microsoft stabilized first, then funds started flowing into Micron, SanDisk, Marvell, Broadcom, and optical module companies. This indicates the market is no longer content to just hold the safest leaders but is beginning to seek assets that were mistakenly sold off in July and have greater resilience.
A true bottom is not just one leader surviving, but more and more stocks refusing to hit new lows.
4. Is this rally just short covering or new money entering: Both, but the capital structure has improved
Is this rally merely short covering, or is new capital entering? The answer is both, but the capital structure has clearly improved.
Since the end of July, major semiconductor ETFs like SOXX and SMH have seen significant net inflows, totaling several billion dollars; simultaneously, subscriptions for high-elasticity semiconductor long products have surged, while inverse semiconductor ETFs like SOXS have experienced continuous net outflows. This means two things are happening simultaneously: some funds believe prices are cheap enough and are actively bottom-fishing; some shorts find prices won’t fall further and are forced to cover.
Pure short covering usually comes and goes quickly, but this rebound has lasted multiple trading days and has gradually spread from heavyweight stocks to memory, network chips, and optical communications. This shows the market is not just shorts fleeing but also funds rebuilding long positions. Short covering can only create a single big bullish candle; sustained diffusion can create a true bottom.
5. Why AMD’s post-earnings performance is the most critical signal of bottoming
AMD’s earnings were not bad; revenue, data center business, and guidance all grew, yet the stock still plunged after hours. The reason is simple: AMD had risen too much before, and the market demands not just excellence but perfection. AMD’s drop was not due to collapsing performance but because the performance wasn’t strong enough to meet the most aggressive expectations.
What’s truly worth noting is that after AMD’s drop, Micron did not plunge, SanDisk did not plunge, and Marvell and other AI hardware stocks were not dragged down. This means the market is finally starting to differentiate the industrial logic of different companies: AMD trades on GPU competition and gross margin; Micron trades on DRAM and HBM; SanDisk trades on NAND supply and demand; Marvell and Broadcom trade on custom chips and high-speed interconnects.
The market no longer drags the entire AI supply chain down just because one company misses expectations. AMD’s drop reflects AMD’s earnings gap, not a death sentence for the entire AI industry.
6. Has the technical side confirmed a reversal: The first leg of the bottom is clear but still needs confirmation
Has the technical side confirmed a reversal? Not fully yet, but the first leg of the bottom is becoming clearer. The Philadelphia Semiconductor Index still needs to break through and hold above the mid-term moving average; some oversold stocks have already rebounded significantly in the short term, so a pullback is very normal.
A truly healthy trend is not continuous daily surges but volume contraction during pullbacks without breaking last Thursday’s launch zone, followed by another volume surge to confirm the second leg. The key going forward is not whether the index can surge another day but who is still willing to sell on the next decline. As long as pullbacks don’t hit new lows and earnings disappointments remain isolated to individual stocks, the sector has a chance to upgrade from oversold rebounds to mid-term recovery.
A rebound shows how strong the rise is; a reversal is judged by whether the next drop can still scare people away.
7. Conclusion: The AI market in August will no longer be a broad rally but enter a phase of differentiation
Therefore, I believe the AI market in August will no longer see all stocks rising together as before but will enter a harsher phase of differentiation: core assets like Nvidia and Microsoft will stabilize the index; among memory, optical modules, network chips, and semiconductor equipment, companies with real orders, profits, and earnings delivery ability will receive valuation recovery; but those relying only on concepts, without cash flow, and purely driven by sentiment before July, even if they rebound, may only provide an exit opportunity for trapped investors.
The market will no longer trade on "whether AI is still viable" but on "who was truly oversold and who was just overhyped." Broad rallies reward courage; differentiated markets reward insight.
$AMD #AMD财报超预期，增长已被透支？ $BTC SPCX no longer allocates capital like a launch company with 86% of its $BTC 18B quarterly capex flowing into AI compute infrastructure that generated $BTC 2.6B in revenue.
Starlink remains the financial engine with $4.3B in revenue and $2.6B in EBITDA while the Space segment produced less than $1B and remained unprofitable.
SpaceX increasingly operates as a profitable communications network and rapidly scaling AI platform with rockets serving as the strategic moat connecting everything together.
#EarningsRealityCheck
#SpaceXBeatEstimates
#AMDBeatsButDrops 📉AMD earnings explode! All data beats expectations, but after-hours plunge 9%? What happened🤯$AMD
On the surface, AMD's earnings report looks all positive✅
But after-hours it plunged sharply, with a maximum drop close to 9%💸
A typical case of good news being cashed out crazily by funds. Let's dig into the truth behind this earnings report👇
📊Core earnings highlights✨
✅Total revenue: $11.536 billion｜beat market expectations
✅Non-GAAP EPS: $1.66｜slightly beat expectations
✅Gross margin: 56%｜slightly above expectations
⚠️Q3 revenue guidance $12.7–13.3 billion, midpoint $13 billion
👉Exceeds market average expectations but falls short of optimistic explosive numbers dreamed by some institutions!
Breakdown of four major business segments🔍
💻【Data Center｜AI core ace】$6.718 billion, up 107% YoY🔥
Accounts for 58% of total revenue! EPYC server CPUs + AI GPUs both booming $AMD
Already entered supply chains of Meta, OpenAI, AWS
Helios rack system ships end of Q3, official volume ramp in Q4📈
Management guidance: Data center revenue to double by 2027!
🖥️【PC Client】$3.1 billion, up 23% YoY👍
Ryzen chip demand rebounds, this segment firmly supports performance
🎮【Gaming business】$779 million, down 31% YoY😮💨
Weakness exposed! Console business declines, consumer gaming GPU underperforms
🔧【Embedded business】$977 million, up 19% YoY📌
Stable and steady performance, no errors
💣Hidden major negative: Capital expenditure explodes💥
Q2 capex $808 million! Market originally expected only $299 million, nearly 3 times higher!
Burning huge cash to expand AI chip capacity, free cash flow will be severely consumed😰
Also an $800 million one-time cost related to regulatory inventory
😱Why the 9% plunge after hours despite data beating expectations? 3 truths!
1️⃣Expectations were already sky-high📈
Stock rose 7% during regular trading, many profit-takers waiting to exit🚶
Guidance just meets average level, no explosive data bulls dreamed of, so facts were sold off!
2️⃣Capex surge hits institutional pain points⚠️
Now AI chips are judged not only by revenue but also by cash flow!
Crazy spending to expand production, market worries profits will be eaten by investments, long-term valuation pressured.
3️⃣External news adds insult to injury❄️
After-hours rumors say SpaceX prioritizes Nvidia GPUs, AMD's AI replacement narrative takes a hit
Plus gaming business continues to underperform, no positive offsets
Key point⚠️: Fundamentals are not broken! It's just that overly high prior expectations were corrected, AI business growth remains strong!
✨Reasons to still be bullish on AMD
▪️Data center business doubled YoY, server market share keeps rising
▪️Helios rack delivery in second half, competing with Nvidia full systems, strategic upgrade new story📝
▪️PC business rebounds, supporting the company's base
▪️Q3 still maintains double-digit sequential growth, business not weakening
⚠️Risks to watch
🔹If capex stays high, cash flow will remain pressured
🔹Helios rack shipment ramp slower than expected is biggest variable
🔹Gaming GPU remains weak, consumer side drags
🔹Strong Nvidia competition limits AI chip market ceiling
🔹Stock price has priced in a lot of AI growth already, future gains require continuous beats
👀Key things to watch next
1. Actual revenue contribution from Helios rack in Q4
2. Whether capex will continue at high levels
3. AI chip major customer order fulfillment
4. Whether gaming business can stop declining
#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？
#AMD财报超预期，增长已被透支？
#SpaceX首份财报超预期，解禁仍是关键变量 #AMD财报超预期，增长已被透支？
Earnings exceeded expectations, so why did the stock price still fall?
Last night, AMD and SpaceX both released their reports, with very similar results:
Earnings surpassed market expectations, yet the stock price was pressured in after-hours trading.
This perfectly validates a commonly overlooked rule during earnings season:
The market never trades on "good or bad," but on "whether it’s good enough to exceed expectations."
1️⃣ $AMD: Great results meet even higher expectations
AMD’s quarterly revenue was about $11.54 billion, up 50% year-over-year, with adjusted EPS of $1.66, both above market expectations.
Data center revenue more than doubled year-over-year, and next quarter’s revenue guidance also exceeds analysts’ general estimates.
Looking at the report alone, there are hardly any obvious issues.
But the stock price remains under pressure because the market has already priced in not just growth, but a sustained, significant acceleration in AI business.
When valuations have already priced in a "perfect score," ordinary beats aren’t enough.
2️⃣ SpaceX: Revenue surges, but capital faces selling pressure first
SpaceX’s revenue reached $7.8 billion, up over 90% year-over-year; quarterly loss was about $541 million, also clearly better than market expectations.
However, the company is investing heavily in AI, and some internal shareholders are about to enter lock-up expiration.
This means investors face not only company growth but a more realistic question:
Will there be more stock supply coming?
So SpaceX rose about 9% pre-market but gave back most of the gains after the earnings release.
3️⃣ Circle closes the night, also watch out for “good numbers but not surprising enough”
Circle is about to release its earnings, and the market will focus on:
• Whether USDC circulation continues to grow
• Whether reserve interest income can be maintained
• Whether interest rate cut expectations will depress future earnings
• Whether the payment network can bring new income beyond interest
If Circle only shows USDC scale growth without proving that its revenue structure is becoming more diversified, it may still face the same expectation gap.
My observation is:
The most dangerous phase of high expectations isn’t necessarily poor performance, but very good performance that doesn’t exceed the script that capital has already bet on.
Before earnings, look at the story; after earnings, you must look at three things:
Whether guidance is raised, whether valuation is overextended, and whether the share supply loosens.
Which trend do you think Circle will replicate tonight?
A: Earnings beat expectations, stock price continues to rise
B: Good numbers, but benefits are realized
C: Fall first then rise, waiting for the call to confirm
#AMD与SpaceX财报 #SPCX #AMD #Circle #美股 #AI
For market observation only, not investment advice.Negotiation news is still being denied by both sides, but oil prices have already fallen first as a sign of respect—the market prices before the truth.
CNBC's headline today puts the contradiction directly on the table—the news about the Iran nuclear deal is chaotic, but the market is rising. U.S. stock futures, Asian markets, and oil prices are all pricing in "good news," but no one can clearly say whether the deal exists or what the conditions are. This is not pricing facts; this is pricing sentiment.
Interestingly, this sentiment-driven rebound exactly follows the line I was watching yesterday: stablecoins continue to flow out, chip stocks rally collectively, and crypto mining companies have no buyers. Now suddenly a story of geopolitical easing emerges, and risk appetite in capital will only move further outward.
Crypto hasn't failed to rise. $BTC moved, $ETH moved too, but the strength is far behind the amplitude of U.S. stock futures. If this were truly a systemic geopolitical positive, crypto should lead the rally, not trail behind. This indicates that capital does not see this as a crypto narrative, at least not yet.
What’s more worth watching is the oil price. If the Iran deal really materializes, Brent has room to fall. When oil prices drop, inflation expectations decline accordingly, which is good for all risk assets—but precisely in this scenario, BTC's "digital gold" logic would be weakened. Demand for safe havens decreases, traditional assets regain appeal, and crypto might actually get drained.
At this point, don’t rush to judge the direction. What’s truly worth watching is the combination of oil prices and Nasdaq futures over the next 24 hours: if oil prices hold steady without falling and Nasdaq continues to surge, then today is just an emotional chase, not a structural change. If oil prices plunge and Nasdaq remains flat or shows upper shadows, it means the market is repricing inflation expectations, which is a bigger variable for crypto.
Mark one question: after the U.S. stock market opens tonight, can Coinbase (a U.S. crypto exchange) outperform the Nasdaq? If not, then the attitude of capital is very clear—crypto is not on the beneficiary list of this round of geopolitical easing.
#美伊谈判推进，油价跌破80美元 75%→23%, the script has been rewritten.
Brothers, don’t just focus on the K-line.
What’s deciding the direction of Bitcoin now isn’t the long-short ratio, nor ETF net inflows, but the session of those guys in Washington.
On Polymarket, the passage rate of the "Clarity Act" was 75% in May, now only 23% today.
The Senate won’t reconvene until September 11, leaving only a 3-day window.
Bitwise executives said: "Zombie zone" — neither dead nor able to rise.
But what I want to say is another layer:
Even if the SEC bypasses Congress to push rules, the next administration can overturn them in minutes.
Uncertain clarity equals no clarity.
Why aren’t the whales entering the market?
They’re not waiting for price, but for rules. Without clear rules, hundreds of billions in funds just sit on the sidelines watching.
What’s trading in the market now isn’t Bitcoin, but the calendar.
Betting on the comprehensive bill by year-end, betting on SEC easing, betting on political timing.
My judgment: short-term chaos, mid-term looking at the year-end window, long-term I remain optimistic.
But at this point, don’t mistake a rebound for a reversal, don’t mistake a patch for an engine.
Only those who survive the zombie zone are qualified to welcome the next spring.
#BTC #ClarityAct #RegulatoryGame #PatientHolding #OKXPlanet
$BTC $PUMP PUMP Surged 15% to $BTC 0.00245, Short Squeeze Drives the Rally
The meme coin PUMP surged 15% today, currently at $BTC 0.00245.
The main reason remains the short squeeze – after a large-scale unlock, only 4% of the tokens actually went to exchanges, the expected sell-off failed, forcing short sellers to close their positions, pushing the price up. The ongoing buying and burning of Pump.fun tokens provides deflationary support, but this rise is more sentiment-driven than based on fundamental changes. Watch short position changes closely. Be cautious chasing high prices.
#EarningsRealityCheck
#SpaceXBeatEstimates
#AMDBeatsButDrops Five years ago, a BAYC monkey was bought for 85 ETH and sold today for 9 ETH
Today on-chain, there was a transaction that looks quite painful. Brother Maji Huang Licheng bought that BAYC monkey, number 5670, five years ago for 85 ETH, and today sold it for 9 ETH. That's a loss of 76 ETH, which at current prices is about $142,000, a loss of 89.4%.
What's more worth mentioning is where the money went. He didn't keep the money from selling the NFT; he directly added it to his ETH long position. That position now holds 3,450 ETH, nominally about $6.44 million, with a liquidation price at $1,839.37.
Putting these numbers on the table is interesting. ETH is currently hovering around 1,870, less than 2% above 1,839.37. In other words, he cut an asset held for five years down to just one-tenth of its value, in exchange for a leveraged long position that could be liquidated at any moment.
Looking at it in USD terms is even more intuitive. When he bought, ETH was priced in the thousands of dollars, so 85 ETH was in the range of two to three hundred thousand dollars; today, 9 ETH is worth less than $17,000. The same asset, over five years, has dropped to a fraction of its original price. And this is BAYC, once called the blue chip of blue chips, not some small quick-flip NFT.
Meanwhile, on the same chain, another address 0xd9a has opened 20 ETH short positions since July 27, all with 10x leverage, taking $600,000 in 10 days, with half the positions held for only 1 hour. The latest short position is 14,519.62 ETH, nominally $27.21 million, opened at an average price of 1,874.2.
On one side is the veteran player cutting losses on the NFT to add to longs; on the other is a machine-like operation repeatedly harvesting ETH intraday. Both watching the same price, but with completely opposite approaches—this is the truest reflection of the current market.
A subjective note: The 1,839 line is not just where Maji’s position is stacked; long liquidation often clusters just below round number support levels. If the price really drops below, it can trigger accelerated liquidations. For swing traders, this level is a key observation point—watching for volume spikes or quick wicks and recoveries nearby, rather than guessing direction.
Looking higher, BTC isn’t doing much better, stuck grinding between 63,000 and 64,000, with the 200-week average cost line right beneath. Without strength from the main coins, money from NFT assets will only exit faster, which has been happening over the past few months.
In the short term, ETH is squeezed by these high-leverage positions, causing volatility greater than spot alone would suggest. Long term is another story: this round of NFT liquidity drying up is real. A top blue chip losing 90% over five years, money flowing back from NFTs to main coins, shows the market is contracting and concentrating chips into the strongest assets.
So here’s the question: do you think Maji bought at the cycle bottom, or did he bet his last bullet on a position just 2% away from liquidation? How far is your own position from its liquidation price? Do you know?From the numbers, this SpaceX earnings report is flawless.
But the stock price fell more than 6% after hours.
The problem lies in two areas. One is the massive cash burn, with quarterly capital expenditures reaching 18.37 billion, more than six times that of the same period last year, of which 15.8 billion was invested in AI infrastructure. The market now reacts nervously at the words "AI capital expenditure," especially with the shadow of semiconductor crashes in the first half of the year still lingering.
The other is that on August 6 tomorrow, 911.5 million shares will be unlocked. Currently, SpaceX has only about 640 million freely tradable shares, and the unlocking scale is nearly twice the circulating shares, corresponding to a market value of over 100 billion dollars. What does this mean? It means the amount of SpaceX stock available for trading on the market will suddenly nearly double. The short position has already piled up to 24.6 billion dollars, accounting for about 34% of the circulating shares, even larger than Tesla's short position.
So the current situation is—the earnings numbers are indeed impressive, but there could be a sell-off worth hundreds of billions tomorrow.
My own judgment is that the quality of the earnings report itself is solid. Starlink's profitability is being validated, the AI business is reducing losses faster than expected, and the revenue growth rate is top-tier in any industry. But the unlocking is a real supply shock, and selling pressure will definitely exist in the short term.
The key points to watch are: first, whether the trading volume within three days after unlocking can absorb the sell-off. If it can, it means institutions agree on this price, and this level might be the bottom area. Second, if the stock price dips due to unlocking, whether long-term funds are willing to come in and buy.
My current attitude is to wait and watch, not to act. Wait for the unlocking to happen and the direction to become clear. There's no rush for one or two days; SpaceX is not going bankrupt tomorrow.
$SPCX #SpaceX首份财报超预期，解禁仍是关键变量 $NVDA $BTC #SpaceX首份财报超预期，解禁仍是关键变量
SpaceX delivered an impressive report card, but the market is no longer concerned with "whether there is growth."
Instead, the question is: can such strong growth sustain the increased selling pressure ahead?
In its first earnings report since going public, SpaceX posted revenue of $7.8 billion, a 92% year-over-year increase, and losses significantly narrowed compared to the same period last year. Starlink and AI businesses have become key growth drivers. Looking purely at fundamentals, this report indeed exceeded many market expectations.
However, interestingly, the positive earnings did not completely eliminate market concerns.
Because investors are now facing a classic question:
The company is getting stronger, but has the stock already priced in the future?
The lock-up expiration on August 6 may be the real short-term stress test. According to public information, the first batch of unlocked shares could reach about 900 million shares, and whether this new circulating supply will be quickly absorbed by the market will directly affect the stock price trend.
My judgment is that in the short term, I will not base direction solely on the earnings report.
If I were a shareholder, I would focus more on three signals:
First, whether there is sustained selling pressure after the lock-up expiration.
If many early investors choose to cash out, it means the market needs time to digest the valuation.
Second, whether the AI business can become a second growth curve.
SpaceX’s current potential is not just rockets and satellites, but AI infrastructure. If future AI computing and satellite networks can contribute higher revenue, the valuation the market is willing to assign will also differ.
Third, whether institutional funds are stepping back in.
Many high-growth companies go through this phase: performance proves the company is solid, but the share structure determines short-term price.
This is actually similar to the logic behind many AI assets this year.
NVIDIA, AMD, data centers, AI infrastructure — the debate is no longer about "whether there is demand," but "whether future growth can justify current valuations."
So this time, SpaceX is more like a market test:
The earnings report tells you it’s a good company;
The lock-up expiration tells you whether it’s a good price.
If I had to choose, I wouldn’t chase the stock after the earnings release but would wait for the capital feedback after the lock-up expiration.
The truly worthwhile opportunities often don’t arise when the market is unanimously optimistic, but when good companies are temporarily punished due to short-term share pressure.
Going forward, the focus is not on whether SpaceX can continue to grow, but on how much the market is willing to pay for it. $SPCX $ETH 1867, the borrowers are back.
The twelve-hour loan ratio has increased by 145%. The previous wave multiplied thirty-six times, and now it's back. Those borrowing to go long have maxed out their leverage again. But the price is still at 1867, unchanged.
All twelve three-hour funding bars are red, looking decent. But the fifteen-minute large orders have already turned negative, with over seventy thousand exiting in five days. Every time, the long-term data looks good, but the short-term is running away; chasing in means taking the bag.
The fee rate is still positive; the bulls are paying protection fees to the bears. The number of whale accounts is increasing, but their positions are decreasing; the rhythm of pulling up while selling off hasn't changed.
If you don't understand, check the historical posts; the same plot is playing for the third time. Truly unbeatable.
#eth $ETH #标普500首次站上7700点，创历史新高
The S&P 500 surpassed 7700 points for the first time, setting a new historical record. This reflects the market's revaluation of the resilience of the U.S. economy, the profitability of tech companies, and future expectations of interest rate cuts. The core driver of this rally remains the AI industry chain, with large tech companies continuously investing in artificial intelligence, cloud computing, and chips, driving market funds to concentrate increasingly on high-quality growth stocks. Meanwhile, expectations for Federal Reserve rate cuts have intensified, improving dollar liquidity, reducing corporate financing pressure, and boosting market risk appetite. For investors, new highs in U.S. stocks do not mean there is no risk; current market valuations are already at high levels, and further gains require more corporate earnings to support them. For the crypto space, the S&P 500 breaking new highs sends a positive signal, indicating a global rebound in risk appetite. High-volatility assets like BTC and ETH are expected to benefit, especially sectors related to AI, RWA, and computing power, which may attract capital attention. However, caution is needed: if the U.S. stock market undergoes a high-level correction, it could trigger short-term volatility in risk assets. Going forward, the market will focus on Federal Reserve policy changes, tech giants' earnings reports, and capital flows. The bull market continues, but trading will demand more attention to timing and risk management. $AMD's earnings report is good, but market expectations were too high, currently Sell the News down 6%, waiting for the conference call
Q2 exceeded expectations:
1. Revenue $11.54 billion (expected $11.31 billion)
2. Adjusted EPS $1.66 (expected $1.62)
3. Data center revenue $6.72 billion, up 107% year-over-year
4. Adjusted operating margin 27%
5. Gross margin 56%, in line with expectations
Q3 guidance continues to exceed expectations:
1. Revenue about $13 billion (expected $12.5 billion)
2. Gross margin about 56%
Lisa Su stated that the second-generation Helios AI server has entered full production, equipped with MI455X accelerators and Venice EPYC processors, and will begin shipping in the coming months. Meanwhile, the company expects the data center business to further accelerate in the second half of 2026.
Fundamentals are solid: revenue, EPS, data center business, and guidance all exceeded expectations. But the stock price fell about 7% in after-hours trading. The reason is not a bad earnings report, but that market expectations were too high. Before the earnings, AMD had already priced in growth expectations for MI450, Helios, AI GPU, and server CPUs.#SpaceX首份财报超预期，解禁仍是关键变量
SpaceX's first financial report is out, and all the key numbers look good.
Revenue reached 7.814 billion, up 92% year-over-year. Operating loss narrowed from 970 million to 143 million, with the loss improvement exceeding market expectations. They also partnered with NVIDIA on the Starmind AI satellite payload. The direction is right, and the data is solid.
But the stock price still fell after hours. Why? Because the market has already priced in this report. The real unpriced event is tomorrow—the unlocking of 911.5 million restricted shares, with a potential supply scale exceeding the current public float. No matter how good the financial report is, it can't stop the fact that hundreds of billions worth of chips are about to hit the market.
This script is very common in the crypto world—good news comes out, but the price falls instead of rising because the market is front-running, moving ahead before the liquidity shock. SpaceX's drop today isn't due to poor performance but because the batch of shares unlocking tomorrow hasn't found buyers yet.
For the crypto world, there are two signals worth watching.
First, SpaceX's unlocking is an observation window. If it can hold under the pressure of hundreds of billions being sold, it means current market liquidity is better than expected, indirectly supporting confidence in the crypto market. If it can't hold, then projects in the crypto market with upcoming unlock periods need to reassess their risks.
Second, the unlocking itself is not bad news. After the chips change hands, if they can be held, it's a fresh start; if not, the grinding continues. The direction isn't in the financial report but in where that batch of shares goes tomorrow.
My view is straightforward—the financial report is already priced in; tomorrow's unlocking is the real test.
That's also why I've always said don't rush to act. Wait for the liquidity shock to pass and see clearly the direction of chip turnover before making a move. No rush these next couple of days.
$ETH $SNDK $SOL #EarningsObserver: Mixed Results, Unlocking Approaching! What's Next for SpaceX?
I'm Nanyan Gu. SpaceX's first earnings report is out, with core metrics beating expectations, AI business losses narrowing better than expected, but AI capital expenditures exceeding forecasts. The stock price turned down after hours.
The earnings themselves aren't bad; the problem is tomorrow's unlocking.
The performance itself is not poor, but the market chose to sell after the earnings release, indicating that funds are preemptively avoiding tomorrow's $100 billion unlocking. About 911.5 million shares will be unlocked on August 6, accounting for 20% of the total unlockable shares. At $110 per share, this exceeds $100 billion, surpassing the current public float. Good earnings but high capital expenditures, combined with the pressure of the $100 billion unlocking, lead funds to choose to exit first.
Impact on SanDisk
SpaceX's stock fell after hours, AMD dropped over 9%, while Palantir rose nearly 30%, an exception. Tech stocks are diverging internally: those with confirmed earnings rise, those with high capital expenditures fall. SanDisk's earnings report is imminent, and market expectations are already high. If capital expenditure guidance also exceeds expectations, it may replicate SpaceX's trend. The storage sector has risen significantly recently; it's advisable not to hold heavy positions before earnings.
What to expect next
At this level, SpaceX faces real short-term unlocking pressure. But unlocking doesn't mean all shares will be sold; the founder's shares are locked until 2027, and early investors and employees are the main sources of selling pressure. The stock price is already nearly 20% below the IPO price; how many are willing to sell at this level is the biggest variable tomorrow.
That's all from Nanyan Gu. Think it over. $BTC $ETH $SPCX Behind Nvidia's HBM Capacity Cuts: It's Not Demand Shrinking, But a Pricing Power Battle and Industry Logic Restructuring
1. Market Misinterpretation: HBM Capacity Cuts ≠ Demand Decline
Nvidia's Rubin Ultra GPU was originally planned with 12 layers of HBM4E, but the evaluation changed to 8 layers of HBM4E, or even reverted to 12 or 8 layers of HBM4. This news triggered market panic, thinking AI's memory demand has peaked and the HBM super cycle is ending. But the real logic is the opposite: it's not that no one wants HBM, but that HBM is too scarce, too expensive, and too difficult to produce, forcing Nvidia to reverse-modify GPU designs to adapt to supply chain constraints.
2. Technical Breakdown: The Asymmetric Game Between Capacity and Bandwidth
The number of HBM layers mainly determines capacity, while bandwidth is jointly determined by interface width, transfer rate, and stacking quantity. Reducing from 12 layers to 8 layers cuts capacity by about one-third, but if still using high-speed HBM4E, bandwidth may not decrease by one-third correspondingly. More importantly, lowering the number of layers significantly improves yield, heat dissipation, and packaging difficulty.
With the same number of DRAM dies, producing 8-layer HBM can theoretically stack 50% more than 12-layer HBM. This means Nvidia is really calculating not "how much memory a single GPU loses," but "whether the same batch of HBM can deliver 50% more GPUs"—rather than filling 100 graphics cards with full memory, it's better to equip 150 cards first and connect them via NVLink into a high-speed computing domain to start producing tokens.
3. Architectural Revolution: From Single-Card Dependence to System Collaboration
The deep logic of Rubin Ultra is that it no longer sells a single graphics card but an entire computer. In the NVL576 architecture, 576 GPUs are connected via NVLink into a high-speed computing domain, with adjacent GPUs and dedicated caches working collaboratively. A single GPU has dozens of GB less capacity, but the whole system does not reduce usable capacity proportionally.
Nvidia is using rack-level architecture to eliminate the absolute dependence of a single GPU on ultra-large capacity HBM. This means the value logic of HBM is shifting from "single-card capacity" to "system bandwidth" and "delivery scale supported by yield."
4. Pricing Power Battle: What's Cut Is Not Demand, But Premium Rights
The real bearish factor is not total HBM demand but the scarcity premium of 12-layer HBM4E and the pricing power Nvidia faces from Samsung, SK Hynix, and Micron (the "Three Seas and America").
- SK Hynix: Orders may not decrease, HBM4 and HBM4E technologies remain leading, but after Nvidia accepts 8-layer HBM4E or even regular HBM4, its monopoly story of "only I can stably supply top-tier 12-layer products" is weakened, possibly facing a scenario of "continued performance growth but valuation discounted first."
- Samsung: Relatively benefits, as it has prepared 8-layer, 12-layer, and 16-layer HBM4E, along with a complete system of DRAM base dies, packaging, and wafer foundry. The more diversified Nvidia's configurations, the easier it is for Samsung to bypass the most difficult 12-layer product and enter the supply chain first with mature specifications.
- Micron: In the middle position, lowering configurations weakens the upgrade premium of HBM4E but can extend the lifecycle of existing HBM4, reducing the pressure of new product validation and mass production in 2027.
5. Industry Restructuring: From "Stacking the Highest" to "Supporting the Most"
In the past, memory manufacturers told Nvidia: to get stronger GPUs, you must buy the most expensive, highest-layer HBM. Now Nvidia says: I can unify memory pools and offload caches through NVLink, completing the same tasks with less capacity.
It breaks down a "must-buy top-tier product" into four interchangeable configurations, forcing the three suppliers to re-bid. This is not Nvidia being bearish on HBM but a reverse domestication of HBM suppliers.
Future judgments of the HBM industry should not only look at how many GB each GPU carries but also consider:
1. How many qualified stacks can be produced per wafer;
2. How many GPUs each DRAM die can support;
3. Who can provide the most effective bandwidth at the lowest cost.
This is not the end of the HBM super cycle but the HBM super cycle officially entering a new phase from "who stacks the highest" to "who can support more GPUs with the same wafer." $NVDA #$BTC $ETH August 4｜In-depth Analysis of the Full-Day Crypto Market
1. Core Market Status (Price + Volume + Liquidations)
BTC dipped intraday then rebounded, bottoming at $62,643. Driven by easing Middle East geopolitical news, it bounced back to test the $64,000 level, currently consolidating narrowly between $63,500-$63,800. It rose slightly by about 0.3% in 24 hours, with volume slightly increased but not breaking recent average volume, showing a typical "news-driven recovery with capital cautious and locked positions" pattern.
ETH showed weak and divergent movement, hovering slightly around $1,855, facing clear short-term pressure; major altcoins showed mixed gains and losses, AVAX led modest gains, SOL and XRP consolidated at lows, lacking collective strength.
Total 24-hour contract liquidations across the network reached $256 million, with short positions as the main force of liquidation (short liquidations $143 million). Over 67,000 traders were liquidated, indicating many bottom-fishing shorts were flushed out during the morning sell-off. Short-term bearish momentum has been released, but buying support from bulls remains insufficient.
The Fear & Greed Index is at 26, staying in the fear zone. Market funds are holding chips tightly, waiting for Friday's nonfarm payroll data, unwilling to bet on a one-sided direction prematurely.
Key Support/Resistance Levels
BTC
Resistance: First barrier at 64,000, strong resistance at 64,700 (50-day moving average resistance)
Support: Short-term defense at 62,800, strong support at 62,200 (low point of this rally)
ETH
Resistance: 1,890, 1,940
Support: 1,820, 1,780
2. Macro Driving Logic (Root Causes of Market Moves)
1. Positive catalysts: Expectations of reopening the Strait of Hormuz in the Middle East warmed, oil prices plunged 7%, inflation expectations cooled, 10-year US Treasury yield fell to 4.68%, and Fed September rate hike expectations dropped from 67% to 58.9%, providing emotional support for crypto assets and serving as the core trigger for this rebound.
2. Market decoupling: The US Nasdaq surged over 2%, the broader market hit record highs, but BTC's gains lagged far behind. The core reason is deep market fear; even with positive external news, large funds prefer to hold coins and watch rather than enter early.
3. Core focus of the game: The ultimate turning point is Friday night’s nonfarm payroll data. Strong data → rate hike expectations rise, price pressure; weak data → rate cut expectations rise, opening rebound space. Before that, the market will only range trade narrowly.
3. Capital and Institutional Analysis
1. ETF funds: Ended 7 consecutive days of net outflows, slightly returning $170 million. IBIT is the main buyer but only short-term bottom-fishing capital; institutions have not started sustained accumulation, buying strength is weak.
2. Futures market: BTC spot-futures basis sharply narrowed to 3%, below US Treasury risk-free yield, arbitrage funds exited, derivatives trading activity declined, indirectly confirming low market bullish sentiment.
3. Whale movements: MicroStrategy paused large-scale accumulation, only small position turnover; cold wallet theft incidents continue to ferment, raising large holders’ risk aversion, creating a clear short-term gap in incremental funds on the market.
4. Sector Performance Breakdown
1. Major coins: BTC alone supports the market; ETH and secondary altcoins show weakened correlation, with token unlock selling pressure priced in early, locking rebound space.
2. Meme and small coins: Rotation is very fast, targets that surged one day quickly fall the next. Pools are thin, LPs not burned, small coins carry high rug pull risk. Entry requires checking pool liquidity first.
3. Narrative sectors: Asset tokenization concepts showed slight movement; BlackRock’s new tokenized money market fund triggered pulses in related coins, but this is a short-term one-day play without mid-term continuity.
5. Market Outlook and Practical Strategy
Market conclusion
Currently, this is only a technical rebound after a decline, not a reversal of the downtrend. Prices are pressured below medium- and long-term moving averages, with heavy overhead supply. Market direction is fully tied to nonfarm data, and the consolidation pattern will persist until Friday night.
Practical reference
1. Contract trading: Strictly control 10-20% low positions, avoid high leverage, do not bet on one-sided long or short; short-term only do range high sell and low buy, reduce positions near 64,000 resistance, lightly buy rebounds near 62,800 support, set strict stop losses.
2. Spot holdings: Mainly hold and observe, no additional accumulation; take profits in batches on short-term profitable positions, prioritizing capital preservation.
3. Meme coins: Control impulse, avoid chasing highs, remember "withdraw principal first, use remaining positions to play the market," avoid thinly pooled small coins.
4. Mid-term layout: Wait for nonfarm data release and volume stabilization before re-planning positions. #财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？ #SpaceX首份财报超预期，解禁仍是关键变量 #AMD财报超预期，增长已被透支？ Inside the Federal Reserve, discussions shifted from cutting interest rates to three members publicly calling for a 25 basis point hike, yet the crypto market has not seen a full retreat for now. More notably, spot funds, stablecoin supply, and contract leverage are sending three different signals.
In July, the FOMC maintained rates at 3.50%—3.75% with a 9-3 vote. All three dissenting votes advocated for a rate hike rather than a cut. The policy path shifted from a one-way easing expectation to a scenario where "hold, cut, and hike" coexist, making the opportunity cost of dollar funds harder to decline quickly.
On-chain dollars have already begun to contract.
Currently, the total stablecoin market cap is about $300.25 billion, down $2.219 billion in the past 7 days, a 0.73% decrease. Among them, USDT is about $183.16 billion, down 0.40% weekly; USDC is about $72.24 billion, down 1.97% weekly.
Together, they account for about 85.1% of the stablecoin market. Based on current scale and weekly decline, USDT and USDC combined decreased by about $2.187 billion in one week, explaining approximately 98.6% of the net decline in the entire stablecoin market.
In other words, while BTC holds near $64,000, the on-chain dollars available for trading, lending, and allocating risk assets have not increased correspondingly.
However, spot institutional funds are still absorbing. The US spot BTC ETFs saw a net inflow of $170.1 million in the most recent full trading day, with IBIT inflows of $111.4 million and FBTC inflows of $33.4 million. Funds have not fully exited the crypto market but have concentrated more into BTC spot instruments.
On the contract side, there is no obvious overheating. CoinGlass snapshots show the average funding rate across the BTC market is about 0.0007%. Binance, OKX, and Bybit BTC/USDT funding rates are approximately 0.0032%, 0.0051%, and 0.0044%, respectively, which annualize to about 3.52%, 5.60%, and 4.80%.
These levels are significantly lower than the high long premiums common in extreme market conditions, indicating that the $64,000 level is not solely supported by crowded leverage. But the contraction in stablecoin supply also means the market currently lacks the new on-chain funds needed for a broad expansion.
As of 12:42 Beijing time, BTC is around $64,134, with a daily range of $63,403—$64,435 and a volatility of about 1.6%. Price stability, ETF inflows, and moderate funding rates represent a bullish side; the $2.2 billion weekly stablecoin reduction limits the space for the rally to spread to ETH and altcoins.
Going forward, two combined signals can be directly monitored:
If stablecoin supply grows again while ETFs continue net inflows, it indicates new funds are starting to spread from BTC to the entire market; if stablecoins continue to decrease but funding rates rise rapidly, then the rally is more likely driven by existing funds leveraging up, with significantly weaker sustainability.
After the Fed's divergence widened, the crypto space did not simply enter a bearish mode. What truly happened is that funds began selecting more certain entry points rather than lifting all crypto assets simultaneously.
#从降息到加息，联储分歧全公开 #美伊谈判推进, oil prices fall below $80 $CL. Recently, global financial markets have shown a complex pattern intertwined with multiple main themes: signals of easing in the Middle East have driven international crude oil prices down, boosting major global stock indices and generally rising; Meanwhile, the yen has strengthened significantly, and the market is closely watching whether the Japanese government will take further intervention measures in the currency market. 1. Geopolitical Situation and Crude Oil: Safe-haven premiums are clearing out, inflation expectations are cooling Transmission mechanism: Expectations of easing geopolitical tensions in the Middle East (especially Iran) have directly reduced market concerns about disruptions in the global crude oil supply chain, prompting the "Geopolitical Risk Premium" previously included in oil prices to be quickly squeezed out. Macro effect: As the "mother of industry," crude oil prices can effectively reduce pressure on upstream production costs (PPI) and downstream consumer prices (CPI). This provides the Federal Reserve and other major central banks with greater room to cut rates and policy flexibility, reducing the risk of global economic stagflation. 2. Stock Market Rebound: Cost Reduction and Risk-On Recovery Earnings Recovery: The decline in energy costs has directly improved the profit margin expectations of the vast majority of non-energy companies. Decline in Discount Rate: The decline in inflation expectations has put downward pressure on long-term Treasury yields. In the Discounted Cash Flow Model (DCF), the reduction in cost of capital raises the theoretical valuation ceiling for risk assets (such as tech stocks and growth stocks), thereby driving a rebound in stock indices.CRCL ($CRCL) Q2 Earnings Analysis｜Earnings to be released pre-market on 8-5
US Treasury yields surge, revenue growth VS market share decline, what’s your take?
Core business: USDC stablecoin issuance, main revenue from reserve Treasury interest; followed by payments, API, Arc network business
Q1 historical situation: EPS beat expectations, but revenue missed, stock price surged then fell after hours
Market consensus for Q2
- Revenue consensus: $730M-$745M, QoQ +5-7%, YoY +12%
- EPS consensus: $0.18-$0.19 (vs -$0.43 same period last year), institutional expectations have been lowered in the past 30 days
- Key real-world background:
1) USDC circulation was 77B in Q1, fell to about 73B in Q2, shrinking supply; but short-term US Treasury yields remain high, interest income partially offsets scale decline losses
2) Obtained US OCC national trust bank license, a major regulatory positive, but won’t immediately translate to revenue in short term
3) Intensified competition: Open-USD new stablecoin (Visa, BlackRock, Stripe involved) expected to launch, market worries about share erosion
4) Distribution cost pressure: renewal negotiations with Hyperliquid, Coinbase, distribution share increase will squeeze retained yield RLDC, management guides 38-40% for full year, Q1 actual was 41.4%, likely to fall this quarter
Four key metrics to watch in earnings (more important than EPS, revenue)
1. Total USDC market cap
✅ Beat expectations: stopped falling and rebounded, >75B, indicating stablecoin share stabilization;
❌ Warning: continues dropping below 70B, meaning USDC is losing share to competitors, valuation logic impaired.
2. RLDC reserve retained yield (core profitability quality)
Represents the proportion of Treasury interest left to Circle after channel distribution shares, official guidance 38-40%
✅ Good: ≥39%, costs controllable;
❌ Bad: falls to 37% or below, meaning more share must be given to exchanges and on-chain channels, even if interest income is high, profits will be eaten up
3. Non-interest income (payments + API + Arc)
This is the second growth curve to break free from the "US Treasury yield cycle hostage"
✅ Beat expectations: non-interest income > $140M;
❌ Warning: almost no growth, all performance still relies on Treasury interest
If this doesn’t pick up, CRCL is essentially a US Treasury yield leveraged stock, and Fed rate cuts will directly pressure performance
4. Conference call guidance
1) H2 USDC circulation outlook;
2) Arc mainnet launch pace, institutional client adoption;
3) How trust license converts to business;
4) Management’s response to Open-USD competition
Four scenario simulations
Scenario A: All beats expectations
Revenue > $750M, EPS > $0.22; USDC inflow, RLDC ≥ 39%, non-interest income significantly rises
👉 Stock price relatively strong; but sector volatility is high, big gains often pull back
Scenario B: Revenue and EPS just meet expectations (most likely)
USDC slightly down, RLDC falls to 38-39%, non-interest income flat
👉 Data not bad, but lacks catalyst, prone to "sell the fact" and pullback. Market already worries about competition, hard to sustain rise without highlights
Scenario C: EPS meets target, but USDC keeps shrinking / RLDC drops sharply
Book profit looks good, but core business metrics worsen, a "fake beat", stock likely to plunge. Institutions most concerned about share loss risk.
Scenario D: Revenue misses expectations, EPS below 0.15
Double negative, combined with bearish stablecoin competition narrative, strong downside pressure.
Current market divergence
✅ Bull case: Obtained US trust license, compliant stablecoin leader, AI agents and on-chain payments bring USDC growth; Arc network opens new revenue streams
❌ Bear case:
1) USDC circulation stagnates, USDT squeeze + Open-USD new competitor entry, big share pressure
2) Profit highly tied to US Treasury yields, Fed rate cut cycle will directly hit reserve interest income
3) Most income from interest, real tech business monetization not yet proven
Institutional target prices vary widely: average $104-$120, pessimistic institutions near $50
#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？
#SpaceX首份财报超预期，解禁仍是关键变量
#AMD财报超预期，增长已被透支？ Five years ago, a BAYC monkey was bought for 85 ETH and sold today for 9 ETH
Today on-chain, there was a transaction that looks quite painful. Brother Maji Huang Licheng bought that BAYC monkey, number 5670, five years ago for 85 ETH, and today sold it for 9 ETH. That's a loss of 76 ETH, which at current prices is about $142,000, a loss of 89.4%.
What's more worth mentioning is where the money went. He didn't keep the money from selling the NFT; he directly added it to his ETH long position. That position now holds 3,450 ETH, nominally about $6.44 million, with a liquidation price at $1,839.37.
Putting these numbers on the table is interesting. ETH is currently hovering around 1,870, less than 2% above 1,839.37. In other words, he cut an asset held for five years down to just one-tenth of its value, in exchange for a leveraged long position that could be liquidated at any moment.
Looking at it in USD terms is even more intuitive. When he bought, ETH was priced in the thousands of dollars, so 85 ETH was in the range of two to three hundred thousand dollars; today, 9 ETH is worth less than $17,000. The same asset, over five years, has dropped to a fraction of its original price. And this is BAYC, once called the blue chip of blue chips, not some small quick-flip NFT.
Meanwhile, on the same chain, another address 0xd9a has opened 20 ETH short positions since July 27, all with 10x leverage, taking $600,000 in 10 days, with half the positions held for only 1 hour. The latest short position is 14,519.62 ETH, nominally $27.21 million, opened at an average price of 1,874.2.
On one side is the veteran player cutting losses on the NFT to add to longs; on the other is a machine-like operation repeatedly harvesting ETH intraday. Both watching the same price, but with completely opposite approaches—this is the truest reflection of the current market.
A subjective note: The 1,839 line is not just where Maji’s position is stacked; long liquidation often clusters just below round number support levels. If the price really drops below, it can trigger accelerated liquidations. For swing traders, this level is a key observation point—watching for volume spikes or quick wicks and recoveries nearby, rather than guessing direction.
Looking higher, BTC isn’t doing much better, stuck grinding between 63,000 and 64,000, with the 200-week average cost line right beneath. Without strength from the main coins, money from NFT assets will only exit faster, which has been happening over the past few months.
In the short term, ETH is squeezed by these high-leverage positions, causing volatility greater than spot alone would suggest. Long term is another story: this round of NFT liquidity drying up is real. A top blue chip losing 90% over five years, money flowing back from NFTs to main coins, shows the market is contracting and concentrating chips into the strongest assets.
So here’s the question: do you think Maji bought at the cycle bottom, or did he bet his last bullet on a position just 2% away from liquidation? How far is your own position from its liquidation price? Do you know?#财报观察员：业绩喜忧参半，解禁将至！SpaceX后续怎么看？
SpaceX (SPCX) just released its first quarterly report since going public (Q2 2026), which is indeed "mixed news," combined with this week's unlocking pressure, the stock price is under short-term pressure, but the mid-to-long-term story remains intact.
Key financial data (positive aspects):
Revenue: $7.81 billion, a year-over-year surge of 92%, significantly exceeding expectations (market forecast around $6.8-6.9 billion). Also a strong sequential increase.
Narrowed losses: Net loss of $541 million (about $1 billion in the same period last year), loss per share $0.09, better than market expectations of $0.23-0.26 loss.
Adjusted EBITDA: About $3.5 billion, a year-over-year jump of approximately 191%, with impressive profit margins.
Business highlights:
Connectivity (Starlink): Revenue $4.29 billion (+66%), operating profit $1.66 billion (+79%), still an absolute cash cow. Users doubled to 12 million, with a record 1.7 million added in a single quarter.
AI: Revenue $2.56 billion (about +247%), still operating at a loss of $1.26 billion, but significantly narrowed and better than expected; adjusted EBITDA turned positive at about $1.15 billion. New cloud service contracts total $14.1 billion, with rapid computing power expansion.
Space (Rockets): Revenue $962 million (+29%), still loss-making, but progress made on Starship V3 test flight.
The company holds about $93.5-100 billion in cash (post-IPO and bond issuance), with unfulfilled orders around $47.5 billion, providing ample financial buffer. Musk reiterated the 2030 revenue target of $1 trillion (possibly earlier).
Concerns (market worries):
Capital expenditure surged: Quarterly Capex reached $18.37 billion (+550% YoY), with about $15.8 billion in the AI segment. Expected to remain high in coming quarters, free cash flow still significantly negative. This is a typical "burn cash to grow" model.
Still unprofitable overall: Operating losses have narrowed significantly, but only Starlink among the three major segments is consistently profitable; space and AI are still in investment phases.
Valuation and stock price reaction: Stock price surged about 9.4% to around $125 before the report but plunged 7-9% in after-hours trading. The market's bar for "exceeding expectations" is very high, combined with high valuation (still far above most tech stocks), sensitive to Capex and cash burn.
Unlocking imminent: The biggest short-term pressure test is this Thursday (August 6), with the first major post-IPO unlocking: up to about 911.5 million shares (worth around $100 billion) of insider/early investor shares becoming tradable, far exceeding the current public float. This is the first step in a staggered unlocking, with multiple rounds from August to December. Musk himself is locked until mid-2027, but selling pressure from other shareholders may amplify volatility.
Historically, many high-valuation IPOs face pressure at the first unlocking, especially as the stock price has already fallen significantly from the post-IPO peak. What about SpaceX going forward? Short term (next few weeks to Q3): unlocking + high Capex concerns likely increase stock volatility, possibly continuing to test support. If Starlink users continue to exceed expectations and AI contracts are fulfilled smoothly, some selling pressure may be absorbed. Watch management's guidance on Capex return cycles (currently claiming AI computing investment payback under one year) and Starship progress. Mid-to-long term: business flywheel is forming—Starlink continues generating cash → supports Starship R&D + AI computing expansion → feeds back into larger-scale deployment. If Starship achieves high-frequency reuse and AI infrastructure truly becomes a "new growth pole," revenue scale could rapidly step up. Risks lie in execution (launch cadence, AI utilization, competition), macro interest rates/tech sentiment, and Musk's personal factors.
Overall, this is a "strong growth but aggressive investment" report, consistent with SpaceX's transformation from a "rocket company" to a "space + communications + AI infrastructure" platform. Unlocking is a short-term disturbance; the real determinant of stock price is whether execution can meet high expectations.
Investors should note the high volatility nature and manage positions carefully. The above discussion is based solely on public information and does not constitute any investment advice.BlackRock has launched two new tokenized funds on Ethereum, and it seems like no one cares at all
because the decline assumes Wall Street has lost interest in $ETH. But on the contrary, they are building and even accelerating.
BlackRock's new BSTBL and BRSRV funds went live on Ethereum on August 3, aiming to qualify as reserve assets for stablecoin issuers under the GENIUS Act.
This is no coincidence. This is their entire strategy.
BlackRock is building the actual infrastructure for how stablecoins get backed, and it is first landing on Ethereum
But this is just the first wave. BlackRock's flagship BUIDL fund is live on 8 different blockchains, while Ethereum still leads with $1.1 billion, accounting for BlackRock's total $2.93 billion in on-chain tokenized assets.
Among all chains, Ethereum wins
Zooming out, it gets even bigger.
The entire tokenized real-world asset market has tripled in the past year, reaching $32 billion (up from $11.8 billion), and only 10% of that is even connected to DeFi
Wall Street is just beginning to move its balance sheets on-chain
Billions will soon turn into trillions on Ethereum
I believe that a year from now, everyone will be chasing $ETH just like they are chasing AI now.