Orbit Post Sitemap
美股存储芯片板块隔夜集体回调，闪迪（SNDK）重挫6.58%，SK海力士跌5.80%，美光跌5.32%，西部数据跌4.18%，希捷跌2.47%。导火索是摩根士丹利7月21日发布的最新报告——由AI驱动的半导体存储狂欢正接近拐点，内存合同价格预计将于2026年第四季度见顶。 大摩亚洲及欧洲科技研究主管Shawn Kim给出两个早期预警信号：一是市场对存储制造商的净盈利上调率已从92%的峰值回落至77%，盈利升级周期正在失去动力；二是第二季度DRAM和NAND的库存水平出现回升，主要由内存模组厂商驱动。大摩的判断是：本轮存储周期将走向"拉长"而非直接崩溃，但周期变化率正在见顶。 值得玩味的是大摩内部的分裂立场。 覆盖美光的美国半导体分析师Joseph Moore仍维持看多，而Shawn Kim则被称为"韩国半导体死神"——他曾精准踩中2021年起的半导体下行周期，2024年又发布《Winter looms》HBM供过于求报告（事后承认预测偏差）。这一次看空NAND合约价，正值SK海力士ADR赴美IPO、大摩独家落选承销商之际，市场不乏"唱空动机论"的质疑声。 多空双方的核心分歧在于：AI存$CORE. Current progress of the Los Angeles itinerary 1. Connecting entities: North American traditional asset management, crypto funds, leading custodians The core goal of this Los Angeles project is to promote cooperation between Bit Grid + lstBTC dual staking institutions. Continuing the strategy of the Shanghai business meeting, it aims to supplement North American capital channels, explaining the underlying architecture of CORE BTCFi to overseas large BTC holders and asset managers, and exploring institutional large-scale access to staking channels. Currently, the company is in the in-depth negotiation phase, with the multi-party cooperation framework being refined, and no official signing announcement has been made yet. 2. Two major consensuses initially reached (1) Institutional recognition of the track logic: Bitcoin's native Layer+ liquid staking track has long-term value, and the narrative of BitGrid infrastructure is optimistic; (2) Future plans include small-scale joint testing to open up non-custodial BTC staking access channels for overseas whales. 3. Core Challenges to Address (Focus on Objective Viewing) North American compliance terms, token liquidity solutions, and institutional risk control standards are still in a tug-of-war. North American capital acts with extreme caution, business negotiation cycles are lengthy, and meetings ≠ to directly land large deals in the short term—there's no scenario of a direct surge immediately after landing. 2. The two most concerned online rumors are being identified ❌. Rumor 1: Los Angeles negotiations finalized a huge amount of funds entering the market. Conclusion: False expectations, no official signing documents yet, framework negotiations do not mean funds will enter immediately, so do not overdraw expectations prematurely. ❌ Rumor 2: Negotiations are stalled, overseas capital is not optimistic about CORE. Conclusion: One-sided pessimistic interpretation. The person in charge personally traveled to North America for face-to-face communication#美联储周四凌晨公布利率决议
芯片股一夜崩了
费城半导体跌2.23%，英伟达跌5%，ASML跌5.8%，闪迪跌11%，数字不算离谱，但逻辑变了
以前芯片股跌是因为业绩没到位，这次业绩还在超预期，跌的是估值逻辑
英伟达给OpenAI俄亥俄数据中心提供2500亿美元担保，潜在再加3500亿融资支持。这条消息没被市场解读成利好，而是被读成AI循环融资到顶的信号。芯片公司给数据中心做财务担保，再融资买自己的芯片，这是信贷扩张逻辑，不是产业逻辑
CDS数据最直接，英伟达5年期CDS盘中涨14个基点至82，是这批合约活跃以来最大单日波动。甲骨文、亚马逊、Meta、博通的CDS同步创历史高位。债券端在重定价，这不是股票市场的情绪波动
我觉得这次的本质是，市场开始质疑AI资本开支的可持续性。过去两年的叙事是科技巨头买算力、英伟达利润爆炸、循环自我强化。现在英伟达要给买家直接做融资，说明需求端在用杠杆支撑，不是自有资金
利率是另一个隐患，10年期实际收益率已到2023年以来最高，30年逼近3%，历史上这个位置只有金融危机期间短暂突破过。如果名义10年期国债冲到5%，美股压力会明显放大
长鑫上市加了存储板块的变量，ASML的传闻是另一根稻草，都不是主因，但市场脆弱的时候任何不确定性都会被放大
我的判断，这不是回调，是定价框架在切换，从业绩驱动切向信贷风险重定价。周三FOMC加财报是短期关键，如果鲍威尔不加息、微软Meta资本开支指引继续强，会有一波修复。但CDS已经动了，不会因为一个财报季就消失
这周等待，不追高。等财报确认需求端真实性，等方向清楚了再说。$FET Halved in one week, tell me this is a pullback? This is the collapse of faith! Loudly exposing SK Hynix's 13% plunge—don't think this has nothing to do with the crypto world. This marks a major short-term turning point in the entire AI narrative. Concerns about overcapacity in Korea are already written on the surface, and with China's CXMT storage chips about to launch, high-end storage has been turned from a "scarce" to a "mainstream commodity." How are they supposed to play like this? Even more shocking was the news coming from OpenAI. Sam Altman invested tens of billions of dollars in self-developed chips, and this money was originally meant to flow into Nvidia and SK Hynix's pockets. Now the boss cooks himself, and everyone in the supply chain is going hungry. Doubts about AI capital expenditures are growing louder. How much real income has been converted into all this money burned? No one could answer. The situation is even worse in the crypto world. $FET, $RNDR, $AGIX these AI little brothers have always been following the big US AI brothers; if the big brother sneezes, they're directly in the ICU. Last week, $FET was hovering around $2.8, but in the past two days it has dropped below 1.4—a pace even more thrilling than the LUNA crash. But I think there's a chance here. When panic reaches its peak, the chip exchange is often at its most intense. Look at on-chain data: big players placed massive buy orders at the 0.8 level, which is not something retail investors can publish. AI narratives won't die; they just need to get rich by replacing a new generation. Most of the funds drained from CXMT's IPO are traditional capital, while crypto AI projects follow a decentralized computing power path and hardware inventory cycles#韩股重挫8%，长鑫首日登顶A股 全球芯片股暴跌原因找到了#全球芯片股为啥突然崩了？一句话：市场被“三重恐惧”吓到了。
周一晚上，美股芯片板块集体跳水，英伟达跌5%，闪迪跌11%，连光刻机老大阿斯麦都扛不住。表面看是技术调整，背后其实是三把刀：
第一刀：美联储“加息悬疑”——周三开会，市场居然押注35%概率加息，这谁受得了？加上Meta、微软、苹果、亚马逊扎堆发财报，资金不敢动，先跑为敬。
第二刀：AI烧钱太猛，市场开始“恐高”——英伟达被传要帮OpenAI搞2500亿美元数据中心融资，市场非但不兴奋，反而觉得这是“AI融资狂热顶点”的信号。更吓人的是，英伟达的债务违约保险成本单日暴涨14个基点，创历史纪录，连甲骨文、谷歌的CDS都飙到新高——说明投资者真在担心科技巨头借钱搞AI会不会爆雷。
第三刀：中国存储芯片厂长鑫科技上市暴涨，加上阿斯麦被传闻带崩，让本就脆弱的板块雪上加霜。
高盛交易员直言：现在芯片股业绩再好也是“利好出货”，估值已跌到近10年平均线以下，但没人敢抄底。对冲基金忙着砍仓位，散户机构只看不动，再加上伊朗局势添乱——一句话，信心没了，钱就怂了。 这波调整，可能还没完。$SNDK #韩股重挫8%，长鑫首日登顶A股
你想想一个5000万人口的国家，怎么诞生出1亿个活跃股票账户的，而且每5个账户就有2个是借钱在炒股
这不是炒股，不是投资，是全民赌局
既然是赌那就是会输，现在韩国股市输了，多次下跌熔断，输在产业周期见顶，输在杠杆泡沫破灭，输在外资撤离，也输在韩指结构权重缺陷
高度依赖海力士和三星等半导体科技股，结果因为长鑫上市并且可以生产光刻机分走份额，美股下跌，韩股下跌
单一领域就能影响整个大盘系统性下跌，形成下跌-强平-在下跌是死亡螺旋
我不知道海力士还能不能反弹，但是我知道，只要反弹到10%，可以重点空他
他没有纳指基本盘强，比美光 闪迪杠杆资金重
1180-1230之间只要能再到，就是我做空的时机，不可能因为我做空海力士就能改变韩股下跌的命运的$SKHYNIX A comprehensive decline—what signal is this?
By morning, US stocks, crypto, gold, and crude oil were all falling, and the reverse linkage in oil prices had disappeared
The key point is that although oil prices have plummeted this time, there has been little news of friendly negotiations between the two sides.
Yesterday, Trump said negotiating with Iran would lead to good outcomes, but just around the corner, Iran slapped him in the face and denied the talks
This wave of decline was not only triggered by a tacit ceasefire between both sides but also triggered a stampede down.
Crude oil is such a macro risk asset that gets caught by news
As mentioned above, as long as Iran and Trump are both rekindled wars under pressure, full-scale war will not break out
So once there is news of a stop, oil prices will cool down quickly.
At present, this is indeed the case
But now, it has not yet entered a definite stage of negotiation and ceasefire
Instead, it was a spontaneous, tacit ceasefire between both sides, which could be a tactical ceasefire
Once the fire is restarted, oil prices will rebound once more
Technically, it is also close to the key support level of 79-77, so attention can be paid to the stop-decline here. If the Middle East makes a big splash, combined with Thursday's rate decision or Walsh taking a dovish move, a rebound will follow
#停火预期兑现, WTI crude oil futures fell 8.68% in a single day
$CL $STORJ Due to entity applications for Chapter 11, they face serious tail repayment risks. The core contradiction lies in the extremely low legal priority for token holders and the risk mismatch between short-term game funds entering the market.
After Storj Labs officially filed for Chapter 11 bankruptcy restructuring, the market experienced a rapid loss of liquidity and a price plunge. Although network nodes continue to operate, the debt crisis of the operating entities has directly broken through the original valuation framework of the decentralized storage sector.
The driving factors driving the current trend are ranked as follows: First, token holders are ranked lower than creditors in legal repayment procedures; Second, the market's reassessment of the operational compliance and profitability of DePIN sector entities; Third, the high uncertainty of the rumored token swap plan.
The trigger for the upward scenario is the presence of specific token rights protection clauses in the restructuring bill, or the presence of a highly qualified external restructuring party to cover the debt. The variable to watch is the details of the official debt restructuring agreement disclosed by the restructuring court. If there is a clear debt extension and guaranteed token performance, the price may undergo a phased squeeze recovery; If the restructuring party explicitly refuses to take over the token debt, the upward logic fails.
The downside scenario triggers the condition for the creditors' committee to lead the liquidation flow or judicial proceedings explicitly excluding token assets from the repayment sequence. The variable to watch is the court's priority ruling on repayment during the restructuring period. If creditors accelerate the freezing of operating entity funds, the market will enter a second stage of decline without support; If the operating entity receives independent funding, the downside scenario becomes invalid.
When Storj Labs completes asset isolation during debt restructuring and its commercial storage business can achieve independent profitability without entity subsidies, the negative pressure on token prices caused by entity bankruptcies will completely fail.
The debt breakdown of real economy entities is driving rapid capital withdrawals from decentralized storage sectors, while small and mid-cap peers are experiencing a contraction in risk appetite. This position exodus triggered by the real economy crisis is difficult to digest in the short term through a purely technical rebound.
The most important variable to watch in the next seven days is the official ruling at the first hearing of the Chapter 11 Restructuring Tribunal regarding creditor repayment priorities and the scope of funds frozen by the operating entity.
#交易之声: Your experience deserves to be heard #韩股重挫8%, Changxin topped the A-share #RWA永续月交易量4700亿美元 on its first dayAccording to real-time data from OKX, $XSKHY is currently quoted at $136.57, down 16.50% in 24 hours, with an intraday high of $164.63 and a low of $136.34, amplitude showing 0.0%, and turnover at 0.0B. The data itself reveals a cold signal: the coin has suffered nearly a 17% plunge with almost no turnover, liquidity is nearly frozen. From a professional trading perspective, this volume-price structure is more alarming than a simple sharp drop. From the K-line profile, the market shows a vertical one-sided slide downward, giving bulls almost no breathing room. The price was suppressed from $164.63 all the way down to $136.34, indicating that short-term selling pressure is not dense, but the buy orders are extremely thin, allowing bears to push the price into deep waters with minimal chips. If yesterday's high and low are viewed as a range, the current price is clinging to the intraday low edge, technically standing "on the edge of a cliff." The key support relies only on the $136.34 low; if a strong rebound cannot form here, the first downward target is likely to hit the $130 integer level, which is the center of a previously dense weekly-level chip zone. On the resistance side, $146 to $150 has become the recent suppression zone. $146 is the area near today's opening price where buy orders briefly gathered, and $150 is the psychological integer level coinciding with the hourly moving average death cross. In the short term, the 4-hour RSI has dropped to around 22, entering the extreme oversold zone, but the MACD's DIF and DEA lines are still diverging downward below the zero axis, and the green bearish energy bars show no obvious contraction, meaning the downward trend momentum continues, and any rebound may become a target for bears to increase their positions again. Unless the daily close stands back above $146, the technical outlook should not easily turn bullish. It is worth noting that $XSKHY's turnover shows 0.0B, combined with 0.0% amplitude; this data combination usually points to two possibilities: one is market makers retreating, widening bid-ask spreads, and real transactions nearly halted; the other is intentional market control, with order book depth showing only a bluff. Either way, retail participation risk rises exponentially. Also stuck in the liquidity drought zone are $SAFE, quoted at $0.0835, down 16.15%, with zero turnover; $XLITE at $687.31, down 12.65%; and $WCT at $0.0365, down 9.31%. The four are uniformly shrinking volume and falling, forming a collectively unsupported decline picture. The market language resembles a grand epic, with towering price peaks and valleys, behind which the narrative of the digital future is being re-priced by the liquidity retreat. In short-term strategy, if $XSKHY can stabilize with shrinking volume near $136.34 and show a 15-minute level bullish divergence, there may be a light position trial opportunity, targeting a rebound toward $146, but stop loss must be kept below $134 to avoid being dragged down by liquidity traps. If it breaks below $136.34 with volume, abandon any left-side bottom-fishing thoughts and wait for a lower structure the next day. The above is only a technical deduction based on market data and is not investment advice; entrants must assess the dual risks of extreme volatility and liquidity shortage themselves. Last night, the entire storage sector collapsed, with $FIL dropping over 18%, $AR falling more than 15%, and $STORJ also plunging 12%. The panic in the market was triggered by ChangXin Memory Technologies' explosive surge on its first day listing on the A-share market, raising hundreds of billions in capital and planning to fully expand DRAM and high-end storage production. Foreign institutions quickly reassessed valuations: the global storage market is expected to shift from a triopoly dominated by Samsung, SK Hynix, and Micron to a four-player competitive landscape. In recent years, the three major manufacturers have maintained high prices and ultra-high gross margins by actively controlling capacity. Once ChangXin's capacity ramps up, overseas manufacturers' pricing power will inevitably be weakened, putting the sustainability of future storage price increases to the test. Even more critical is the internal structure. The storage sector has seen astonishing gains since the beginning of the year, with $FIL's highest annual increase exceeding 250%, and $AR doubling, resulting in extremely crowded positions. High-level sectors inherently rely on new capital inflows to continue; once negative news emerges, concentrated profit-taking by floating positions can easily trigger a cascade of selling. Investment banks like Morgan Stanley have previously warned that the current DRAM/NAND spot price surge may slow and peak in the fourth quarter. End-user demand for PCs and smartphones remains weak, and relying solely on server demand is insufficient to sustain continuous price increases. The capital market fears a marginal decline in the industry’s prosperity, with capital outflows accelerating beyond expectations. Current sentiment has sharply shifted from optimism to pessimism. The turning point in the storage cycle is becoming increasingly apparent. Short-term rebounds require stronger catalysts, such as AI demand exceeding expectations or ChangXin's capacity ramp-up falling short of projections. However, without clear signals, chasing highs carries significant risk, and holders must accept a norm of high volatility. $FIL #Stor I’ve stopped trading every Washington headline like it changes the trend overnight. ⚠️🇺🇸
Senator Dave McCormick is pushing Senate leaders to bring the CLARITY Act to a floor vote and force every senator to take a public position.
That matters but a floor vote is not the same as final passage. The deeper catalyst is whether the bill can define SEC–CFTC oversight clearly enough for exchanges and institutions to deploy capital without guessing the rules.
Until that probability rises, expect headlines to move price briefly and sweep liquidity on both sides not automatically create a lasting trend. 👀
#CXMTDebutShockwave #FOMCRateWatch #AIEarningsWatch $BTC 7月26日晚上，一个朋友给我发来截图。
STORJ，0.06美元。
“我0.18买的。”
去年10月，Inveniam收购Storj Labs的时候，他兴奋地跟我说：“去中心化存储赛道要起飞了，AWS的对手来了。”
结果呢？
10个月，亏了66%。
7月26日，Storj Labs自愿向美国西弗吉尼亚州北区联邦破产法院申请Chapter 11破产重组。
消息一出，STORJ代币24小时内暴跌约20%，跌至0.06美元左右。
团队自己都承认：代币交易长期“低迷且量少”。
你以为这是个孤例？
太天真了。
Storj不是一个人在战斗。
7月，加密行业一周内倒了四家——
Movement Labs申请Chapter 11，账面资产只剩10万到50万美元，负债高达1000万。
BitMEX，那个曾经的衍生品交易所巨头，宣布9月永久关闭。
BitMart，清算交易业务。
一周四家。
这不是寒冬，这是物种大灭绝。
但Storj这个案子，跟其他人不一样。
最狠的地方在于——它撕开了“去中心化”最脆弱的那层窗户纸。
Storj Labs是去中心化存储协议Storj背后的公司主体。公司申请破产了，但官方说：网络继续运行，服务不受影响。
软件工程总监说：“公司基本面强劲，制约发展的是早期遗留负债。”
意思是——
公司欠债，但网络没事。代币还在，但公司要重组。
听懂了吗？
这里有一个核心问题，也是所有去中心化项目持有者必须面对的问题：
代币和法人实体，到底能不能真正脱钩？
Storj说能。网络是去中心化的，节点还在跑，存储服务还在提供。
但市场说不能。
消息一出，代币跌了20%。
因为投资者心里清楚：代币的价值，从来不只是“网络效用”。它还包含了团队的执行力、项目的品牌信誉、持续的开发投入。
这些东西，全绑在那个申请破产的法人实体上。
更有意思的是，Storj抛出了一个“大饼”——
公司表示，计划在重组过程中提出机制，让STORJ代币持有者参与重组后公司的股权。
翻译成人话：
你手上的代币，可能能换成新公司的股票。
但问题是——资格要求、结构、条款，统统没确定。
而且破产法规定，债权人在清偿顺序上享有优先权。
代币持有者？排在最后面。
官方自己都说了：“我们承诺给你一个席位，但不承诺任何结果。”
翻译：画个饼，能不能吃到，看命。
说句扎心的话——
去中心化存储，是Web3最性感的叙事之一。
“抗审查”、“永不宕机”、“数据主权”……每一个词都让人热血沸腾。
但Storj用一张Chapter 11的申请表告诉你：
性感归性感，公司该破产还是得破产。
2014年成立，2017年上线，熬过了两轮牛熊，被Inveniam收购，最后还是走进了破产法院。
去中心化的网络，救不了中心化的债务。
最后，说点实在的——
如果你持有STORJ，现在怎么办？
我不知道。但我知道几个事实：
第一，团队说重组期间不评论价格。这意味着短期内没有任何来自官方的“护盘”动力。
第二，一周倒了四家加密公司。这不是个别现象，是行业性出清。
第三，STORJ从历史最高3.82美元跌到0.06美元。跌了98%。
98%。
那些在0.18美元“抄底”的人，以为自己在捡便宜。
现在发现，底下面还有底。
Storj的案例，给所有“价值投资”Web3项目的人一个提醒——
你看好的是去中心化网络，但你持有的代币，绑定的是一家中心化公司的命运。
网络可以永远跑下去。
公司可能撑不过这个夏天。
“去中心化”四个字，从来不是免死金牌。
它是叙事。
而叙事，救不了资产负债表。
$STORJ $ETH $BTC
#Storj Labs申请Chapter 11破产重组，STORJ暴跌 AI storage fever does not mean that all "storage" assets will benefit.
Demand growth is real. IDC data shows that in the first quarter of 2026, global external OEM storage system spending will be about $9.9 billion, a year-on-year increase of 22.9%. AI training, inference, and unstructured data activation are turning storage from backend costs into part of computing infrastructure.
NVIDIA has repeatedly emphasized that when storage capacity can't keep up, expensive GPUs will sit idle waiting for data; During training, the model also requires the system to continuously retain and quickly restore the state.
But "AI storage" is not a market.
HBM and DRAM are responsible for high-speed memory close to computation; Enterprise-grade SSDs and high-performance file systems are responsible for feeding data to GPUs and storing checkpoints; Object storage, cold storage, and long-term archiving are the areas where decentralized storage has a better chance to enter.
Raising HBM prices isn't a benefit for projects like FIL and AR; there are several layers of business logic involved.
The value that decentralized storage can offer is well deserved. Publicly or authorizing training datasets can be used for content addressing and provenance verification; Model weights, version history, and inference logs can be archived long-term; Data used by multiple organizations does not have to be fully controlled by a single cloud provider; AI Agents' long-term memory, cold backup, and disaster recovery may also become new demands.
The problem is that delivery still has a long way to go.
For example, content proofs can answer "whether data has been saved or modified," but cannot achieve good low-latency retrieval, stable throughput, or enterprise-level SLAs. Training and inference require data to be close to computational resources, as well as handling encryption, permissions, keys, deletion requirements, data cross-borders, S3 compatibility, and bandwidth costs.
Permanent storage is suitable for model tracing and public archives, but may conflict with deletion rights or copyright disputes.
Filecoin's 2026 strategy has changed: the network will have exbibyte-level capacity, focusing on shifting from continued stacking supply to paid on-chain orders, network profitability, and flagship customer adoption.
In other words, hard drives have been around for a long time; what is lacking now is people who keep paying.
This is also the most common confusion when assessing decentralized storage opportunities. Capacity does not equal demand, the number of bytes deposited does not equal recurring income, tokens can pay fees or participate in staking, but that does not mean the value will necessarily return to the holder.
For the opportunities brought by AI to materialize, the market needs to look at paid storage orders, retrieval fees, active customers, renewal rates, protocol revenue, and how much of this income comes from real usage rather than token subsidies. Whether the protocol can integrate with existing cloud tools and AI workflows is more important than "how much total capacity there is."
AI will generate massive amounts of data, but for decentralized storage to catch this boom, it must evolve from a cheap hard drive market into a verifiable, callable, and continuously paid data service
$AR $FIL $MU EVERYTHING IS CRASHING...
US chip stocks crashed overnight.
$NVDA fell 4.4%.
$MU dropped nearly 5%.
$SNDK crashed more than 10%.
Now Asia is collapsing too.
Japan's Nikkei is down over 4%.
South Korea's KOSPI crashed 10%, triggering another circuit breaker.
Bitcoin also crashed below $63K
Here's why:
China has started producing its own advanced chipmaking machines, reducing its dependence on $ASML and threatening the global chip supply chain.
At the same time, $NVDA's $750 billion deal wave is raising fears that $AI companies are financing the same customers buying their chips.
And the biggest risk is still ahead.
The Fed meets tomorrow, with rate-hike odds surging from around 16% to nearly 38% in just one week. Bitcoin is already reacting.
$BTC crashed below $63,000 as traders priced in a much higher chance of another rate hike.
$MSFT , $META , $AAPL and $AMZN are also reporting earnings within the same 72 hours.
Citadel Securities is now going further, calling for a surprise Fed rate hike this week, arguing Chair Kevin Warsh will move to strengthen his inflation fighting credibility even as most economists still expect a hold.
China, the Fed and Big Tech are all hitting the $AI trade at once.
The next 72 hours could decide whether this is just a correction or the start of a much bigger crash.
Buckle up.#韩股重挫8%，长鑫首日登顶A股
Damn! As soon as Changxin went public, the Korean stock market was directly hit with a circuit breaker, Samsung and SK Hynix suffered double-digit crashes, US memory stocks all fell to their knees, a global memory earthquake!
That Chinese company called Changxin went crazy right after listing on the STAR Market, soaring over 400 at the open, its market cap immediately taking the top spot in A-shares, with trading volume hitting historic highs. Money poured in like a flood, as if the entire country's capital suddenly woke up collectively, determined to raise the banner of domestic DRAM to the sky.
Frankly, this is all nonsense. These people have been playing memory stocks way too hard over the past year. With AI booming, everyone treated that kind of high-end memory like a money-printing machine. Hynix made a fortune from it, and its stock price was inflated like a balloon.
Ordinary people borrowed money and leveraged desperately to buy, and leveraged funds were everywhere, leaving the Korean stock market almost supported only by Samsung and Hynix, with the rest just tagging along. Prices were pushed so high that at the slightest hint of trouble, the profit-takers all rushed out, causing a stampede and the index was smashed to pieces.
Changxin is still mainly mixing in general DRAM; it can’t reach HBM yet. Technical barriers, customer certifications, advanced processes—those hard skills can’t be caught up overnight just by throwing money at them.
SK Hynix’s moat in high-end products hasn’t collapsed in the short term. But the market insists on rewriting the story from "AI shortage, crazy price hikes" to "China-Korea showdown, market share battle," making it seem like Changxin could push Koreans off the table tomorrow.
Traders and analysts on X (formerly Twitter) have sharply pointed out: the Korean market concentration is abnormally high, with over half the market cap tied up in just two stocks. When the AI narrative shifts, the whole national stock market spasms, and retail investors’ pensions get slaughtered.
Some mock that export controls have forced China into a closed market, but Korean products are being repriced instead, and the US AI supply chain is shooting itself in the foot. Even harsher critics blame the past crazy rise, saying now is the time to let it out, and to stop using Changxin as a scapegoat.
In the short term, the memory sector will continue to fluctuate. Keep a close eye on DRAM spot prices—that’s the real indicator of the market’s health. This AI revolution won’t die, but the list of winners will definitely be reshuffled. Domestic substitution dividends are attracting capital on one side, while overseas high-level chips are fleeing on the other; differentiation is the norm.
Stop fantasizing about a broad rally. Opportunities come from declines, but don’t expect to get rich overnight. Prepare for five or ten years of endurance, or you’ll just be cannon fodder carrying others’ sedan chairs.
In the memory business, who laughs last depends not on who lists first, but on who can truly produce chips, sell them, and sell them at a good price. Everything else is just noise.Early this morning, $LA surged. At that time, I posted that I couldn't short for now. At that time, contract data indeed did not support short selling, but after I posted, it did rally again. In the new wave of gains, the bearish forces in the market have overtaken the bulls, and only then has $LA shown signs of decline. Based on the data from that time, my judgment wasn't too wrong; I just didn't expect the world to change so quickly. Life is unpredictable! Alright, back to the main topic: is $LA worth bottom-fishing now? To answer this question, we need to rely on some data. —————————————————— Let's take a look at its contract data. We need to pay close attention to two points in this data. The first point is that although its open interest is declining, it hasn't dropped back to its previous level. The second point is that its long-short ratio is rising, having already surpassed its previous level. Personally, I think this shows that many people were bottom-fishing during this round of crashes, which is why the drop in contract open interest was not as significant. Let's extend the timeline of the long-short ratio for a longer period. It can be seen that the absolute value of the long-short ratio in $LA is now almost identical to its value on July 23. What was the situation at that time? It can be seen that $LA also experienced a sharp drop in mid to late July, and July 23 marked the start of the rebound after this crash. So, I boldly speculate that now is the time for it to start rebounding. —————————7.28 Gold midday strategy and market analysis
The Federal Reserve's July policy meeting (July 28-29) has officially begun, with the market awaiting the rate decision and overall sentiment of waiting. Geopolitical uncertainties and central bank gold holdings increased by 14.93 tons, providing bottom support for gold prices; however, the market expected the Federal Reserve to maintain a hawkish stance, so the US dollar remained relatively strong, and buying at high levels was insufficient. In the early session, gold prices fell under pressure from their highs, hitting a low near 4042. Judging from the market rhythm, it is highly likely that the afternoon will show a pattern of rebound under pressure and high-level fluctuations and downward movement.
On the 4-hour period, a large bearish candlestick pulled back and broke through MA5 (4062), MA10 (4072), and the Bollinger middle band (4069). The short-term moving averages formed a death cross pattern, reappearing a downward channel.
On the daily chart, gold prices are trading below the MA5 (4070) and the Bollinger Middle Band (4073), highlighting weak market characteristics. The market may again test the 4000-4020 support range.
Operational advice
Kong: Rebound to the 4055-4080 range is under pressure, target 4030, after a breakout target 4000
(Personal advice is for reference only; remember to bring proper stop investments and maintain risk control.) #美联储周四凌晨公布利率决议 On Sunday night, you slept soundly.
My phone buzzed—Storj Labs filed for Chapter 11 bankruptcy reorganization.
Rubbing your eyes, you open the market app; STORJ has already dropped nearly 20%, at $0.06.
The group exploded.
"Damn, what about the storage nodes I staked?"
"Has the token gone to zero?"
"Wasn't it supposed to be decentralized? What does company bankruptcy have to do with networks? ”
You stare at the screen, with only one question in your mind:
A decentralized storage project has its parent company go bankrupt—Is my coin still here?
Let's first see what the official company has to say.
On July 26, Storj Labs voluntarily filed a Chapter 11 application with the U.S. Bankruptcy Court for the Northern District of West Virginia.
The company said, "The network continues to operate normally, and the service will not be interrupted." ”
Parent company Inveniam said it supports the restructuring. The engineering director said this was a "decisive and proactive step."
It even says—token holders may acquire equity in the restructured company.
Sounds pretty nice, right?
But the market only took 20 minutes to crash STORJ by 20%.
Why? Because everyone knows a harsh truth—
Chapter 11 is bankruptcy court, not a wishing pool.
Storj attributed the reason to "historical debt left over from early business operations."
The company said, "The core business is strong and appropriately scaled." But the problem is—
If the core business is truly that strong, why would bankruptcy be needed to "clean up" past bad debts?
What's even more heartbreaking is that STORJ has dropped from $0.1872 at the time of Inveniam's acquisition in October 2025 to just over $0.06 now.
In ten months, it has dropped 68%.
This is not "debt clearance"; it means the debt has finally been cleared onto token holders.
But what truly makes this case worth pondering is a more fundamental question:
If the parent company of a decentralized storage network files for bankruptcy, can the network still be "decentralized"?
Storj said the network continues to operate and token utility remains unchanged. But think carefully—
Who issues node rewards? Who supports the development team? Who maintains the S3-compatible gateway?
All this "infrastructure" is supported by Storj Labs.
The company went bankrupt, and you say the network runs independently?
It's like a landlord saying, "The house collapsed but the lease remains valid."
It sounds like comfort, but in reality, it's just an empty check.
Storj said it may allocate equity in the restructured company to token holders.
But the officials themselves have said—"This is not a promise, just an intention."
Translated into plain language: "We want to pay, but the court decides, and the creditor is ahead of you." ”
What is the order of repayment in Chapter 11?
Banks→ suppliers→ employees→ creditors→ shareholders→ token holders (if any).
You're at the very bottom.
Equity is just a dream; liquidation is the reality.
And Storj is not an isolated case.
In July 2026, the crypto industry is experiencing a wave of shutdowns:
BitMEX—the pioneer of perpetual contracts—announced its official closure on September 23
Movement Labs—Applied for Chapter 11 on July 15, with assets down to $100,000 to $500,000
BitMart — announced it would cease operations
RootData statistics show that by 2026, 99 crypto projects have entered a "dead" state.
Storj is just the latest domino.
To be honest—
Decentralized narratives cannot save a centralized balance sheet.
Storj Labs went bankrupt, and STORJ fell. It's that simple.
You can say, "The network is still running," "The nodes are still running," "The technology is still running"—
But the market only recognizes one thing: who's burning cash, who's making money, and who's struggling.
08.
Finally, here are three questions for you to consider yourself:
First, if Storj Labs is ultimately liquidated, how much can STORJ token holders get back?
The answer is most likely: close to zero.
Second, for a decentralized project whose parent company has gone bankrupt, would you still dare to store data there?
Do customers dare? Do companies dare? Will the next major client dare to sign a contract?
Third, do you believe the promise of "tokens for equity"?
In court, the creditor's lawyer is a hundred times harsher than you.
Storj's story tells us one thing:
Stop treating "decentralization" as a get-out-of-jail-free card.
The project team's balance sheet determines your fate more than the private keys you hold.
The internet can run, but people have to eat. If the company goes bankrupt, who will write code for you?
STORJ went from $0.187 to $0.06 in just ten months.
From $0.06 to $0, you may only need a court judgment.
$ETH $STORJ $FIL
#Storj Labs files for Chapter 11 bankruptcy restructuring, STORJ plunges Chip stocks crashed overnight
Philadelphia Semiconductor fell 2.23%, Nvidia dropped 5%, ASML fell 5.8%, SanDisk dropped 11%. The numbers aren't outrageous, but the logic has changed
Previously, chip stock declines were due to underwhelming earnings, but this time earnings still exceeded expectations; what's falling is the valuation logic
Nvidia provided a $250 billion guarantee for OpenAI's Ohio data center, with potential additional financing support of $350 billion. The market did not interpret this news as positive but rather as a signal that the AI financing cycle has peaked. Chip companies providing financial guarantees for data centers and refinancing to buy their own chips is credit expansion logic, not industrial logic
CDS data is the most direct indicator. Nvidia's 5-year CDS rose 14 basis points intraday to 82, the largest single-day move since these contracts became active. Oracle, Amazon, Meta, and Broadcom CDS also hit record highs simultaneously. The bond market is repricing; this is not just stock market sentiment volatility
I believe the essence this time is that the market is starting to question the sustainability of AI capital expenditures. The narrative over the past two years was tech giants buying computing power, Nvidia's profits exploding, and a self-reinforcing cycle. Now Nvidia has to directly finance buyers, indicating demand is leveraged rather than funded with own capital
Interest rates are another hidden risk. The 10-year real yield has reached its highest since 2023, and the 30-year yield is approaching 3%. Historically, this level was only briefly surpassed during financial crises. If the nominal 10-year Treasury hits 5%, pressure on US stocks will significantly increase
The listing of Changxin added variables to the storage sector, and rumors about ASML are another straw, neither being the main cause, but any uncertainty is amplified when the market is fragile
My judgment is that this is not a correction but a shift in the pricing framework, from earnings-driven to credit risk repricing. Wednesday's FOMC and earnings reports are short-term keys. If Powell doesn't raise rates and Microsoft and Meta continue strong capital expenditure guidance, there will be a recovery. But CDS has already moved and won't disappear just because of one earnings season
Waiting this week, no chasing highs. Wait for earnings to confirm the authenticity of demand, and wait until the direction is clear
DYOR Not investment advice
#财报观察员：OKX大师课今晚开播，带你看懂四大科技巨头财报 Market sentiment remains in fear at a score of 29, with BTC dominance at 56.38%. In this cautious environment, $SOON leads the pack among USDT coins with a solid score of 68, closely followed by $BULLA and $KAITO. The Privacy Blockchain sector is also gaining traction, averaging a score of 40. Keep an eye on the best Smart Setup, $PUMP, which boasts an 8.9/10 confidence level for potential opportunities. 📊🔍 #CXMTDebutShockwave #FOMCRateWatch #AIEarningsWatch $BTC 📉 昨晚美股科技板块整体下挫,半导体内部却走出独立行情,这种分化挺值得聊。
🔸 板块速览
隔夜走的有点撕裂。SPCXB -6.66%、TSLAB -5.52%、PLTRB -5.28% 排跌幅前三,跟宏观情绪走弱直接挂钩。另一边,SOXLB +4.75% 一枝独秀,AVGOB +3.28%、AMDB +3.18% 紧跟,NVDAB +2.51%,几家头部半导体硬是拉红。MEME 类的 MUB、SNDKB 反而温吞,几乎平盘。微软、英特尔、ARM 这类老牌科技,跌幅也控制在 2% 以内,没跟跌。
🔸 我的看法
分化背后逻辑其实清晰。AI 算力链条(AVGO / AMD / NVDA)凭业绩预期撑着,跌不动;SOXL 这种 3x 杠杆 ETF 把涨幅放大到近 5 个点,资金正在用杠杆博弈半导体反弹。另一边,TSLA 跟宏观周期贴得紧,首当其冲;SPCXB、PLTRB 高位回调也很正常。币圈这边得看 NVDA 脸色,NVDAB 守住 211 一线,山寨 AI 链才能缓口气。
我会等 SPCXB 回踩 110 附近再观察,SOXLB 这种加速段不追。$BTC 今日加密要闻
1）比特币一度跌破6.4万美元。伊朗局势升级推高油价和美债收益率，风险资产同步承压，宏观重新成为短线定价核心。
2）美国现货比特币ETF连续第7个交易日净流入，最新单日约6900万美元，阶段累计接近10亿美元；机构资金仍在承接，但力度尚不足以抵消宏观压力。
3）AFX与Verus运营的跨链桥数小时内接连发生安全事件，涉及约3160万美元资产；风险集中于桥的签名与密钥体系，Arbitrum原生桥未受影响。
结论：ETF资金提供边际支撑，但油价、收益率与安全风险仍占上风，短线宜控制杠杆，等待宏观波动收敛。$BEAT From last night to today's lowest point, it has dropped nearly 50%, which is quite an astonishing decline. So, now the question is, can this coin be bottom-fished? Personally, I think in the short term, the probability of this coin rebounding is very high. Why do I say this? Let's take a look at some data. —————————————————— First and foremost, naturally its near-term contract long-short ratio. We can see that as $BEAT plunged, its contract open interest dropped rapidly, even lower than before its surge. At the same time, his contract long-short ratio is also rising. What does this mean? This shows that last night's drop not only drove out short-term bears but also caused many previously trapped shorts to leave. Many short positions have exited, which means that the bulls have become stronger in relative terms. Moreover, judging from the later data, many bulls have indeed entered to bottom-fish. Let me further extend its recent contract long-short ratio. It can be seen that its contract long-short ratio has already reached the data from July 21. What was the situation on July 21 at that time? It is clear that July 21 marked the starting point of this major surge. —————————————————— After saying so much, does that mean $BEAT will definitely rebound now? I don't think so. The current price level should be the support level for this time, but it's also possible that it could fall to the previous support level#韩股重挫8%，长鑫首日登顶A股
Korean stocks plunged 8%, Changxin Technology topped the A-shares on its first day: Behind the semiconductor frenzy, is the market repricing?
Recently, an interesting contrast has appeared in the market:
On one side, the Korean stock market dropped 8% in a single day, with the semiconductor sector taking a heavy hit; on the other side, Changxin Technology surged on its IPO debut, becoming the focus of the A-share market.
On the surface, this shows a divergence in market sentiment.
But what it actually reflects is that the global semiconductor industry is entering a new phase of competition.
The core reason for the sharp drop in the Korean market is not a lack of demand for semiconductors, but that the market’s expectations for the AI storage cycle had been overly optimistic.
Over the past year, HBM, high-end storage, and AI servers have been the hottest sectors.
Capital continuously pushed up valuations of leaders like Samsung and SK Hynix, with the trading logic shifting from "earnings growth" to "unlimited AI growth."
However, the capital market has a rule:
When everyone believes a direction can’t be wrong, risks often start to accumulate.
Once the market realizes:
AI demand growth may be slower than expected, or storage price increases cannot be sustained, the first reaction of capital is to reduce positions.
So this decline is essentially not an industry collapse but a correction after overvaluation.
The enthusiasm for Changxin Technology’s IPO represents another trend:
The global storage industry is moving from competition dominated by a single giant to a multipolar competitive stage.
The DRAM market has long been dominated by Samsung, SK Hynix, and Micron.
Now, with Chinese companies accelerating their entry, not only is the supply chain landscape changing, but future storage price cycles will also be affected.
For the industry, this is a double-edged sword:
Increased competition benefits industry maturity;
But in the long term, it may also compress the high-profit cycles of the storage sector.
What’s the outlook for the semiconductor market?
My judgment:
There will still be intense volatility in the short term.
The AI trend is not over, but the market has shifted from "speculating on the future" to "watching for realization."
The real winners in future gains won’t be all companies with AI labels, but those:
That can convert AI capital investment into real profits.
If storage prices continue to rise and AI server demand keeps releasing, semiconductors still have a chance to rebound strongly.
But if there is:
Slowing AI capital expenditure + profit realization below expectations + overvaluation,
Then the market may face a deeper adjustment.
The biggest change in this cycle is:
Previously, the market asked:
"Who owns AI?"
Now the market is asking:
"Who can make money from AI?"
Stories can create bull markets.
But what ultimately determines stock price heights is always profit.
The semiconductor story is not over; it’s just moving from a phase of mass euphoria to one of truly selecting winners.#Korean stocks plunge 8%, Changxin tops A-shares on debut This may not be the end for Korean memory, but the beginning of a new era in global memory.
Changxin Technology surged 465% on its first day of listing, with its market capitalization briefly surpassing ¥3.28 trillion, becoming one of the highest-valued companies on the A-share market. Almost simultaneously, market sentiment quickly spread overseas: SanDisk fell more than 11% in a single day, followed by a sharp correction in the Korean stock market, with the KOSPI dropping about 8%, SK Hynix $SKHYNIX down about 11%, and Samsung Electronics falling over 9%.
In just 48 hours, the global capital markets completed a revaluation of the memory industry landscape.
Many believe this reflects market concerns over the weakening competitive advantage of Korea's two memory giants.
But I think it is still too early to say "the Korean giants are in danger."
What Changxin truly changes is not whether Samsung and SK Hynix can make money today, but that the capital markets are seriously starting to consider:
In the future, the global DRAM market may no longer be dominated by just three giants.
For the past two decades, the global DRAM market has been almost exclusively led by Samsung, SK Hynix, and Micron, with a relatively stable market structure that has enjoyed a high valuation premium.
Changxin’s listing means stronger support in capital, R&D, and production capacity.
Although Changxin still has a significant gap compared to the two Korean manufacturers in fields like HBM and high-end AI memory, for the capital markets, what truly impacts valuation is often not today but the competitive landscape over the next three to five years.
This is why the Korean stock market reacted so dramatically.
However, I believe this decline reflects more of a valuation reassessment rather than a deterioration in fundamentals.
AI server demand remains strong, HBM is still in short supply, and the main profit sources for Samsung and SK Hynix have not changed in the short term. If this week’s earnings reports continue to show strong HBM orders, ASP (average selling price), and capital expenditure guidance, market sentiment is likely to recover.
What is truly worth watching is not who fell the most.
But three signals over the next six months:
* Whether Changxin’s advanced process development progresses faster than expected;
* When HBM technology can form real competitiveness;
* Whether Samsung and SK Hynix adjust capital expenditure and product strategies due to competitive expectations.
I have always believed that the biggest opportunity in the AI era is not in the stock price of any single company, but in the entire infrastructure.
GPUs need computing power, computing power needs memory, and memory determines the efficiency of AI systems. Whether it’s robotics, autonomous driving, or Physical AI in the future, this industrial chain is indispensable.
So with Changxin’s listing, what I see is not the birth of a new company, but the global memory industry officially entering an era of "four-way competition."
As for whether the Korean giants will lose their advantage because of this, I think the answer depends on technology, products, and customers over the next few years, not on 48 hours of stock price movement.Bitcoin and Ethereum: The wealth effect in the crypto world has clearly diminished. The stories of ETFs, policies, institutional entrances, and the president's trade calls are all over; incremental funds have not arrived, and the market has entered a stock game of stock. Simply put, without new money entering the market, the foundation of a bull market is empty. Bitcoin: It is now the "digital gold" position for institutions, but the price has been pegged too high. Without sufficient buying support, it can only absorb the market through fluctuations. If the US stock market crashes, it will also fall. Don't expect to go all-in after drawing a big cycle bottom; the reality is: without incremental funds, it's just a large liquidity pool. Ethereum: Weaker fundamentals. There are few highlights in ecosystem innovation, severe L2 diversion, and staking yields are declining. The ETH/BTC exchange rate is still hovering at a low level. It requires breakthroughs in technology and applications, and currently, no new narrative is seen to attract large-scale funding. My judgment: both are tools for stock market competition. Don't expect trend-driven bulls in the short term; it's only suitable for swing trading or oversold investments. Control your position size; don't heavily bet on direction. Cash is king; wait until real panic strikes to grab a bargain.BITCOIN LEVERAGE IS BEING FLUSHED.
Bitcoin’s recent pullback looks different from previous sell-offs.
The reason?
Open interest is falling alongside price.
Total Bitcoin open interest now sits at:
749.32K BTC
$47.42B in open positions
Down 1.79% over the last 24 hours.
This tells us one important thing:
Traders are reducing risk.
We’re not seeing aggressive new shorts piling in.
We’re not seeing leverage building into weakness.
Instead, positions are being closed.
That’s a healthier type of correction.
Exchange data shows the same:
CME open interest: -3.30%
Binance: -0.80%
Bybit: -2.85%
Leverage is coming out of the market.
The only notable exception is OKX, where open interest increased 1.17%.
This lines up with what we saw in the liquidation heatmaps.
A lot of downside liquidity around $63,000 has already been cleared.
The market has already forced out a large amount of overleveraged positioning.
Now the question becomes:
Do buyers step in after the reset?
What I want to see next:
• Bitcoin holding above $63,000
• Open interest rebuilding alongside price
• Funding staying controlled
That would suggest real demand returning.
The risk scenario:
Bitcoin continues lower while open interest starts rising.
That would show traders are opening new positions into weakness, increasing the chance of more volatility.
For now, this looks less like a market breakdown and more like a leverage reset.
Bitcoin doesn’t need more leverage right now.
It needs buyers to step back in.
$BTC This week's Federal Reserve interest rate decision is coming: How will $BTC respond?
This week #FederalReserve will announce the interest rate decision.
Currently, the market expects about a 65% probability of maintaining the interest rate unchanged, but a September rate hike has been heavily priced in by the market (probability over 50%).
Impact on $BTC:
• Interest rates remain high → risk assets under pressure, the US dollar and US Treasury yields likely to strengthen
• If the statement is hawkish (implying a high probability of a September rate hike), a short-term pullback may be triggered
• If the tone is neutral or emphasizes "data dependency," risk sentiment may benefit amid falling oil prices
Key, key, key! My trading strategy:
1. Mainly light positions and wait-and-see, reduce leverage, avoid event-driven shocks
2. Focus on Powell/Walsh's wording at the post-decision press conference
3. Key support at $62,000-$63,000 area, consider positioning at $60,700 for medium to long-term spot; if it holds above $67,000, consider going long
4. If a September rate hike happens, it is expected to be a "buy the rumor, sell the fact" scenario, so pre-positioning for a pullback might be better
The high interest rate environment remains a medium-term pressure on BTC, but geopolitical easing and liquidity expectations are still present. Buying the dip in batches remains the main theme. AI stocks are under pressure again. Japan's Nikkei fell 4.1% after reports of a China chip breakthrough. AI-related stocks like SoftBank, Kioxia, and Tokyo Electron dropped over 6%. South Korea was hit even harder, with the Kospi down 7.9%. The market isn't rejecting AI. It's rejecting AI stock valuations.#CXMTDebutShockwave #FOMCRateWatch #AIEarningsWatch $ESP （Espresso）
The decline in ESP is a typical casualty of a sharp contraction in macro market risk appetite, representing a systemic sell-off of "no nest left intact."
The direct trigger was the Fed's July meeting minutes, which sent unexpectedly hawkish signals. The minutes show nonfarm data exceeding expectations and CPI showing stickiness, indicating that the Fed is far from ready to cut rates. This immediately triggered a broad plunge in risk assets: the S&P 500 fell 1.8% in a single day, marking its largest drop in nearly three months, commodities crashed simultaneously, and Bitcoin was swept through from its highs.
The high correlation between ESP and macro factors makes it impossible to remain unaffected. Data shows that over the past 15 days, ESP's rolling correlation coefficient with Nasdaq reached 0.73, and its correlation with the US Dollar Index was -0.65. This shows that ESP is essentially trading macro logic rather than its own fundamentals. On the quantitative side, key support levels have been broken, and Goldman Sachs models predict a wave of risk asset clearing. At the same time, the funding rates for cryptocurrency perpetual contracts across the market have turned negative, signaling active deleveraging — all assets are being sold off to reduce risk exposure.No major negative news, so why did SanDisk still drop 15% intraday?
$SNDK plunged from $1436.56 last night down to $1222.01, nearly hitting the intraday limit down, and finally closed at $1278.23, down 11.02% for the day.
The company had no major negative news or operational setbacks announced.
In the last quarter, SanDisk reported $5.95 billion in revenue, a gross margin of 78.4%, and data center revenue surged 233% quarter-over-quarter. The fundamentals look ridiculously strong.
But that’s exactly the problem.
The market is no longer satisfied with "great performance"; it’s starting to ask: how many more quarters can a nearly 80% gross margin really hold?
SanDisk’s stock has surged too much this year, with the price already factoring in NAND price hikes, AI storage demand, and high gross margin expectations. Once expectations are maxed out, it doesn’t take real bad news—just someone starting to worry about the storage cycle peaking, and profit-taking will collectively rush out.
The IPO of ChangXin is more like a fuse igniting emotions.
ChangXin focuses on DRAM, while SanDisk is in NAND; they’re not direct competitors. But China’s expansion of storage capacity will force investors to reassess the industry’s future supply, pricing power, and profit margins.
Short term, watch the $1220–$1250 range.
Holding here might just mean a cooldown of the high valuation; breaking below could mean the market is not just correcting, but that the storage boom is peaking.
The real answers will come in the August 5 earnings report:
📌 Can the gross margin stay high?
📌 Can data center revenue continue to grow rapidly?
📌 Will the order and price guidance for fiscal 2027 show any signs of weakening?
No matter how good the data looks, if the guidance isn’t explosive enough, people will still sell off.
After such a big run-up, the market is just like this.#财报观察员：OKX Masterclass premieres tonight, helping you understand the earnings reports of the four major tech giants
The big show is about to begin these days!
Apple, Microsoft, Meta, Amazon—four tech giants with a combined market value exceeding one trillion dollars—will take turns releasing their earnings reports over the next two days. The U.S. tech stock sector looks just like parents waiting outside the college entrance exam room: calm on the surface, but their knuckles gripping water bottles have turned white.
I suggest wearing a qipao to the event for a victorious start!
Is AI truly a world-changing revolution, or just the most expensive money-burning game in history? Tonight, the answers will start to come in.
Here’s my personal judgment: Apple leads off, likely to be lukewarm, AI progress still slow, but with the iPhone’s solid foundation and stable cash flow, no major issues are expected; it’s actually the safest. Microsoft and Meta are uncertain. If Azure’s growth falls short of expectations or capital expenditures show no signs of "contraction," a post-market drop is quite likely. Meta is even riskier—the biggest question is whether ad revenue can outpace capital expenditure growth. Amazon closes the show; if AWS growth can’t exceed 33%, the $200 billion infrastructure story loses credibility. The four giants face three fates: Apple steady, Microsoft and Meta uncertain, Amazon at fate’s mercy.
I consider the three most important indicators:
1. Capital expenditure guidance, the market’s critical point. Google just stepped on a landmine last week—its capital expenditure guidance was hundreds of billions higher than expected, causing its stock to plunge nearly 5% after hours and resulting in the first-ever quarterly negative cash flow since listing. The market is now highly sensitive to "money burning": spending is acceptable, but it must be "controlled." Meta raised its full-year capex to 145 billion, Microsoft expects $190 billion infrastructure investment in fiscal 2026, and Amazon announced $200 billion. If anyone dares to say "keep increasing with no limit" tonight, the stock price will likely turn hostile immediately.
2. Cloud business growth rate, direct evidence to test input-output ratio. Money is spent, servers are bought, AI runs, but someone has to pay the bill. Can Microsoft Azure withstand base pressure? Can AWS achieve growth above 33%? Stable cloud growth means the AI commercialization card can still be played; if growth drops, all the previous hype must be recalculated.
3. Cash flow, the ballast stone. Google has already taught the market a lesson—free cash flow dropped from 25.7 billion to negative, causing an immediate market backlash. Microsoft slid from 25.7 billion to 15.8 billion, Amazon from 25.9 billion to just 1.2 billion, a decline so fast it’s chilling. Meta relies on ad revenue but burns over 30 billion each quarter; its cash reserves will run dry sooner or later. If this area falters again, the market won’t just sell off; it will directly question whether the entire AI narrative still holds.
By the way, about storage: SK Hynix $SKHY, SanDisk $SNDK, Micron $MU, and Samsung have all been battered these past two days. On the surface, it’s a supply panic triggered by ChangXin Technology’s IPO, but fundamentally it’s the "collateral damage" of the AI chain—the market worries that if cloud giants’ capital expenditures peak, how can the storage supercycle continue? Tonight’s three reports are the storage sector’s "lifeline." If capital expenditure exceeds expectations, HBM and DDR5 prices can still rise; but if any one says "pull back," the storage big three could still be halved.
After the U.S. market closes tonight, once the data is out, OKX will immediately enable trading of tokenized U.S. stocks like XMSFT, XMETA, XAMZN—convenient, fast, worry-free, so I’m increasing my position in $OKB!
If these three numbers are right, AI can still hype for another quarter, possibly driving BTC and ETH to rise in tandem; but if any one of the three falters, the entire tech sector will tremble.
Data doesn’t lie. When the tide recedes, you’ll know who’s been swimming naked. 睡醒天塌了😭
$ETH 浮亏2600多U
老贝子又把3310枚BTC和28370枚ETH转进Coinbase Prime
接近2.72亿美元的筹码进场
市场第一反应肯定是砸盘
但严谨一点说
转入Coinbase Prime不等于已经卖出
也可能是ETF申赎结算和资金调度
狗庄想靠一条链上消息骗走我的筹码
没那么容易
$BEAT
24小时下跌18.17%
成交额约6698万美元
日内波动区间2.48—4.63
放量下跌说明多空正在激烈换手
更麻烦的是8月1日预计解锁2125万枚BEAT
约占流通量6.87%
项目一周销毁约79.7万枚
解锁量远高于销毁量
短线2.48附近是防守位
重新站回3.3—3.5才算止跌
否则反弹更像解锁前的诱多
#LAB
LAB目前还在0.14—0.15附近挣扎
七天跌幅约13%
三十天跌幅接近99%
这已经不是普通回调
而是流动性和筹码结构重新定价
项目官方回购页面显示
累计回购3284万枚LAB
花费约1713万美元
但最近几次日常回购只有几千美元级别
想靠这点买盘直接扭转趋势很难
0.14是第一防线
下面看0.125附近
反弹压力集中在0.16—0.17
LAB可以突然拉针
但没放量站稳之前
只能先按超跌反抽看
$SNDK
闪迪昨晚又杀疯了
现货价格约1278美元
单日下跌约11%
盘中最低打到1222附近
成交量超过2100万股
这次不是闪迪单独暴雷
长鑫存储上市首日暴涨466%
市场开始担心中国存储厂商带来的竞争压力
美光和SK海力士也同步下跌
本质还是高位科技股集体去风险
1220—1200是短线防线
反弹先看1400—1450
基本面逻辑还在
但前面涨幅太夸张
现在杀的是估值和获利盘
别把SNDK当普通山寨币硬扛
它疯起来比币圈还狠
嘴上可以继续硬刚狗庄
仓位可扛不住一根插针
#韩股重挫8%，长鑫首日登顶A股
#美联储周四凌晨公布利率决议 How many hours has F&G been suspended on the 29th today? I didn't count. Anyway, he hasn't moved at all since morning.
Here's the question: can a "fear" index that remains unchanged for several hours still be called fear?
True fear is active—panic selling, a series of stop-loss explosions, exchange crashes. The fear of 29 is an emotion of "Oh, it dropped, so let's just wait and see." It doesn't secrete adrenaline; what it secretes is indifferent.
Let's look at today's data:
- The price-to-decline ratio is 1:13, with 13 out of 14 coins declining
- Trading volume -97.5%, almost no one is trading
- Funding rate -0.0016%, neutral enough to be ignored
- Holding 106,200 BTC, unchanged
- No news, no events, nothing worth FOMO or FUD
This isn't called a bear market; it's called the market taking sick leave.
And to be honest, the 1:13 ratio of daily gains to losses is quite exaggerated—out of 14 stocks, only 1 is red, and the rest are all green (oh, according to crypto terms, rising is red, falling is green, don't get confused). At times like this, no one posts saying "Ox returns, quick return," which shows how cold the market is.
Back to my own two holdings:
The PUMP stayed untouched for 19 hours, and the AEON stayed untouched for 5 hours. In a market with -97.5% trading volume, you don't need to keep an eye on the market—even if you do, you won't get any real rewards. What you need is patience and a chair comfortable enough.
This is very likely not the bottom. The real bottom is often accompanied by a panic-driven volume drop, knocking out the last holders. Today's combination of reduced volume down 97.5% plus unchanged prices feels more like a middle ground where "nobody wants to play anymore, but no one wants to leave."
The event that made everyone "want to leave" hadn't arrived yet. Before that, the fear index of 29 was just a facade—the market wasn't afraid at all, it was just too lazy to move.【比特币现货ETF单日小幅净流出，短线偏谨慎但信号强度有限】
比特币短线偏谨慎，但单日ETF资金流的信号强度仍然有限。美国现货比特币ETF出现净流出，说明边际配置需求暂未形成一致扩张；不过流出规模不大，且资金集中于少数产品，尚不足以单独定义更长周期的资金趋势。
Farside Investors数据显示，昨日美国现货比特币ETF总净流出1160万美元，其中BlackRock的IBIT净流出880万美元，Fidelity的FBTC净流出280万美元，其余产品当日资金流均为零。同期以太坊现货ETF净流入1170万美元，但两类产品的资金变化应分别理解，不能简单互相替代。
值得关注的是，比特币ETF当日并非出现多产品同步大额赎回，而是有限的负流量集中在两只头部产品。这意味着市场更像处于增量买盘不足的状态，而不是已经出现明确的系统性撤离。对于短线定价，持续性通常比单日数字更关键，连续流出才会更直接地影响市场对机构需求的预期。
后续应观察头部ETF是否继续净流出、其他产品能否出现承接，以及净流量变化是否与现货市场的成交和波动相互验证。若后续迅速回流，本次数据更像短期扰动；若负流量扩大并扩散至更多产品，谨慎情绪才可能进一步强化。
以上仅为个人观点分享，不构成任何投资建议。市场瞬息万变，交易盈亏自负。#停火预期兑现，WTI原油期货单日跌8.68%
I think the market might be celebrating too early
WTI crude oil plunged 8.68% in a single day, closing near $82.62/barrel, down about 12% from the previous peak of $93.83/barrel. Meanwhile, the forecast market's probability of a US-Iran ceasefire before the end of August has risen to about 75%, with gold, US stock futures, and crypto market risk appetite all warming up simultaneously. $BTC
Many interpret this oil price drop as "war risk removed," but I believe the market is trading more on expectations than outcomes.
In the past two weeks, the core reason for the oil price rise was not a sudden surge in global demand but the geopolitical risk premium caused by the escalation in the Middle East. Now that the US military has paused airstrikes on Iran, the market begins to believe the situation may cool down, so this risk premium is quickly squeezed out, naturally causing a sharp correction in oil prices.
But there is a detail worth noting.
Pausing airstrikes does not mean a ceasefire agreement has been reached.
Both sides still retain the possibility of resuming military actions, and risks to shipping in the Strait of Hormuz and the Red Sea have not been fully eliminated. If subsequent negotiations falter, oil prices could quickly rebound. In other words, the market is currently trading "the best-case scenario" in advance, not that the risk has completely disappeared.
What I am truly focused on is the FOMC.
The oil price decline means energy price pressure on inflation has eased, which indeed gives the Federal Reserve more room to maintain dovish rhetoric. For this week's rate decision, the market is not only watching whether rates will be held but more importantly whether Powell (or the Fed statement) will release more information about the future path of rate cuts and inflation assessments.
Therefore, I will not chase risk assets just because of a one-day plunge.
I prefer to wait for two signals to appear simultaneously: first, the ceasefire expectation continues to be realized with no new geopolitical escalations; second, the FOMC releases a signal more dovish than the market expects.
If both conditions are met, I believe AI, US tech stocks, and the crypto market could see a window of risk appetite recovery.
What will truly determine the market's next phase direction may no longer be oil prices but whether the Federal Reserve takes over the baton after the oil price decline.[DeFi risk appetite temporarily treated with caution, macro and AI variables still awaiting validation]
The direction of DeFi-related risk appetite is currently unclear and is being treated with caution. As the Federal Reserve policy, AI capital expenditure, and geopolitical situation all enter a validation phase, the core market issue is not to find a single catalyst but to judge whether the high volatility environment will continue to compress the valuation tolerance for high-risk narratives.
The material mentions that the market generally expects the Federal Reserve to hold steady, but oil price volatility, unresolved Middle East tensions, and weakened forward guidance all increase the importance of policy communication. Meanwhile, the capital expenditure and expansion pace of the AI industry also face scrutiny, meaning risk assets are influenced not only by interest rate expectations but also by the repricing of growth narratives.
For DeFi, liquidity conditions and risk appetite are often more important than any single positive factor. If the market begins to accept a pricing model where "policy no longer provides clear answers, data determines direction," capital is more likely to prioritize allocations with good liquidity and lower volatility, while narrative-driven and high-beta sectors will face more selective capital inflows. Conversely, only if uncertainty decreases can the repair of risk appetite be discussed.
Next, attention should be paid to the actual market reaction after the policy meeting, whether oil prices and geopolitical risks ease, and whether AI capital expenditure expectations can receive new confirmation. Expectations alone do not equal trends; if macro volatility continues to amplify, the DeFi sector still needs to be wary of the magnifying effects caused by liquidity contraction.
The above is only a personal opinion sharing and does not constitute any investment advice. The market changes rapidly, and trading profits and losses are at your own risk. [Semiconductor sector is cautious in the short term; after concentrated sell-offs, first check if risk pricing is complete]
The global semiconductor sector is cautious in the short term, and it is currently better to observe whether risk pricing is sufficient, rather than rushing to treat sharp drops as low-level opportunities. Nvidia and Asian AI hardware assets are under simultaneous pressure, indicating that transactions are shifting from a single company logic to a revaluation involving multiple variables such as interest rates, capital expenditure, geopolitical factors, and industry competition.
Materials show that Nvidia closed down about 5%, the Philadelphia Semiconductor Index dropped 2.23%, followed by continued declines in AI hardware assets related to South Korea, Japan, and Hong Kong stocks; South Korea's KOSPI fell more than 10% intraday and triggered circuit breakers twice, while SK Hynix and Samsung Electronics plunged sharply. The market attributed the sell-off to multiple factors, including the upcoming central bank interest rate decision, rising Nvidia CDS, and progress in China's semiconductor sector.
The risk of such concentrated adjustments is that growth expectations that originally supported high valuations will be tested by both discount rates and competitive landscapes. CDS changes may not necessarily mean worsening fundamentals, but they increase investors' sensitivity to risk premiums; When cross-market leveraged products decline further amplify, it is also likely to reverse pressure on spot risk appetite.
Going forward, attention should be paid to the performance of risk assets following the rate decision, whether semiconductor leaders can halt their relative weakness, and whether liquidity in Asian markets has returned to normal. If the sector still fails to form stable support after digesting the negative news, it suggests the adjustment may not be over yet; If volatility converges and differentiation occurs, it will be closer to fundamentals regaining dominance in pricing.
The above is just a personal opinion sharing and does not constitute any investment advice. The market changes rapidly, and trading profits and losses are borne by the buyer.[Discussion on Compensation for Single-Stock Leveraged ETFs in South Korea is Cautious, Product Regulatory Risks May Be Repriced]
Narratives around South Korea's single-stock leveraged ETFs are cautious, and policy discussions are unlikely to provide direct short-term benefits. If investor losses begin to be included in discussions of state compensation and regulatory responsibility, the market will first need to reassess not the underlying companies themselves, but the rule stability, risk disclosure, and future expansion potential of these high-leverage products.
Kim Eun-hye, a member of South Korea's People Power Party, is collecting investor opinions, investigating the scale of losses, and studying whether to file a state compensation lawsuit against the government for losses related to Samsung Electronics and SK Hynix single-stock leveraged ETFs. The core controversy lies in whether the Financial Services Commission conducted sufficient risk assessments and investor protection arrangements when it previously amended regulations to allow these products to be launched.
The risks of leveraged ETFs come not only from directional judgment errors but also from daily rebalancing, volatility decay, and liquidity changes. In a highly volatile environment, even if investors are correct about the medium- to long-term trend of individual stocks, the product's net asset value may significantly deviate from intuition due to path dependency. Therefore, if compensation discussions continue to escalate, issuers, market-making arrangements, and regulatory approval standards may all face stricter scrutiny.
The key going forward is whether regulators respond to issues of product design and risk warnings, and whether discussions will lead to specific institutional adjustments. If it remains at the level of individual case relief, the impact may be limited; if it touches on leverage multiples, entry thresholds, or approval mechanisms, the liquidity and valuation discounts of related products will need to be reexamined.
The above is only a personal opinion sharing and does not constitute any investment advice. The market changes rapidly, and trading profits and losses are at your own risk. [Changxin Technology's Sentiment Overheated After First-Day Surge, Short-Term Trading Not Advisable Based Solely on the "Lei Jun Concept"]
Trading sentiment around Changxin Technology is overheated in the short term, but high volatility risk is rising simultaneously. The significant surge on the first day of listing quickly amplifies the resonance between scarce shares and market narratives. However, the "related party's allocated unrealized gains" is more likely to become an emotional label and may not directly translate into a sustained revaluation of the company's fundamentals.
Disclosed information shows that Xiaomi Technology's wholly-owned subsidiary Wuhan 1810 was allocated 18.2448 million shares of Changxin Technology, corresponding to about ¥158 million at the issue price; calculated at the first-day closing price, the unrealized gain is about ¥736 million. The market thus associates this with Lei Jun, while Xiaomi emphasizes that company investments and personal wealth should not be conflated. The key to this response is to clarify the boundaries of the entities involved, preventing the market from simply projecting company-level asset allocation as personal investment decisions.
For trading, the first-day increase of 465.82% already implies that the market has given a very high immediate valuation to subsequent growth expectations. Those benefiting from the narrative spread are early allocators and low-cost holders; subsequent participants face higher valuation tolerance requirements and uncertainties brought by rapid changes in share supply and turnover structure in the early listing phase.
Going forward, it should be observed whether the enthusiasm can settle into sustained pricing based on business, production capacity, and profit expectations, rather than remaining focused on shareholder relationships. If trading volume continues to expand but price volatility intensifies, it indicates that short-term speculative weight remains high, and the risk of chasing a single label will significantly increase.
The above is only a personal opinion sharing and does not constitute any investment advice. The market changes rapidly, and trading profits and losses are at your own risk. $SNDK 当前呈现出明显的单边下跌趋势。价格在触及 1,518.74 的高点后一路下行，最低插针至 1,205.00。目前价格维持在 1,209.80 附近。
趋势判断：整体处于下降通道中，短期均线（如 VWAP 1,257.54）在价格上方形成压制。
短线试多。计划在前低支撑位附近博弈超跌反弹。
入场价：1212～1225
止损价：1185
目标价1：1245
目标价2：1270
目标价3：1300
耐心等待价格在 1,205 上方出现明显的止跌信号（如收出实体较小的阳线或十字星）后再考虑进场。Has China really made lithography machines? Why did US chip stocks plunge collectively?
Following news of mass production of Chinese DUV lithography machines, US chip equipment stocks plunged across the board, with ASML at one point falling more than 7%. In fact, this is just immersion DUV, not the most critical EUV; chips below 7 nanometers remain a domestic weak point. Storage chips like Micron are more indirectly affected; fundamentals remain unchanged, short-term sentiment fluctuates, long-term pressure is limited. The market has somewhat overreacted. I decided to buy the bottom $MU right now Damn, I lost 1060U! The landlord said if I delay any longer, I'll move out. I just finished a meeting last night, glanced at my phone, and $BTC crashed another 3%. I stared at the short positions in my account, with an unrealized loss of 1059.67 USD, and my margin down to just over 6000. The landlord's WeChat message urging rent was still flashing. I said, "Pay next week," but he replied, "If you delay any longer, I'll move out." Damn, this market is really going against my wallet. 📉 Short positions stuck, suffering a 17-point loss. I hold $BTC short positions with 15x leverage, average opening price 62,715, current price 63,451. Held 1.438 contracts, margin 6014U, unrealized loss 1059U, a loss of 17.62%. A double kill for long and short positions? No, it's a one-sided kill on me. $BTC 24H, the lowest hit 63,055, the highest was 65,713, with trading volume soaring to over $400 million—double the previous 24 hours. Dog Farm is shaking out the market, with volume up 104%, which means some are bottom-fishing and others are dumping. My short position cost is low, and now the price has rebounded to 63,451, just over 700 U short of my opening price. The good news is that MA7 is still at 63,356, barely holding above the price; The bad news is that the MA30 is still at 64,607, with prices lying below and the bearish trend unchanged. 📊 All technical indicators collapsed, with only the RSI shouting for help. The moving averages clearly form a classic death cross pattern—MA7 at 63,356, MA30 at 64,607, a gap of 1,251 points. The price is in the middle, stuck between upper and lower positions. The Bollinger Bands are even worse, with the upper band at 65,792, the lower band at 62,741, and the middle band$BTC 韩股重挫 8%，长鑫首日登顶 A 股
韩股 KOSPI 单日暴跌超 8% 触发程序化交易限制，三星、SK 海力士领衔存储板块深度杀跌；另一边长鑫科技登陆科创板，首日大涨登顶 A 股市值，一冷一热的极端行情，映射全球存储格局重大拐点。
盘面分化并非单纯情绪巧合。韩国股市高度依赖存储双雄，资金定价底层逻辑出现扭转：长鑫完成巨额募资，产能扩张通道彻底打通，全球 DRAM 正式从三家寡头垄断迈向四方竞争，市场开始折现韩厂长期定价权削弱的预期。叠加前期韩存储板块累积巨大涨幅、本土高杠杆 ETF 被动平仓，悲观预期集中释放，催生大幅回调。
但必须理清赛道结构性差异：长鑫主攻通用 DRAM，短期很难冲击韩厂把持的 HBM 高端赛道，二者属于错位竞争。本轮韩股下跌，交易的是通用存储远期供给过剩预期，而非 AI 高端存储景气终结。
资金视角出现明显赛道迁移：外资开始重新评估国产存储产业链价值，国内半导体设备、材料迎来长期增量空间。短期全球存储板块震荡会持续放大，后市重点跟踪长鑫产能释放节奏、HBM 订单落地以及现货价格变化。
（仅市场观点分享，不构成投资建议）#韩股重挫8%，长鑫首日登顶A股 $RE
1. 美债收益率上行 → 无风险收益提高，资金撤出高风险投机资产，BTC、ETH承压，山寨跌幅更大。
2. 美元走强，以美元计价加密资产承压；美股科技股（MU、SNDK）与加密高度联动，股市杀跌会传染加密情绪。
3. 杠杆负反馈：预期偏鹰，合约多单清算放大涨跌，山寨因为流动性差波动会成倍放大。
4. 结构规律：利空环境资金回流BTC避险，ETH/BTC汇率下跌，山寨集体跑输主流；只有宏观转暖，山寨才会迎来修复行情。#Storj Labs files for Chapter 11 bankruptcy reorganization, STORJ plummets
STORJ dropped 17% overnight! Storj Labs filed for Chapter 11; even the veteran decentralized storage project can't hold on?
On July 26, Storj Labs voluntarily filed for Chapter 11 bankruptcy reorganization in the Northern District Court of West Virginia.
Note: This is a reorganization, not liquidation. The official statement says the storage network, customer business, and node services are running normally. The parent company Inveniam also supports this, aiming to clear old debts from early acquisitions and return to pure storage business.
But the market is not buying it—before the news, STORJ was around $0.074, then within 24 hours it dropped to about $0.060, a decline of approximately 17%–20%. Daily trading volume surged more than 8 times, indicating panic-driven turnover.
The key point to watch: the company said it is considering allowing STORJ token holders to exchange tokens for equity after reorganization, but there is no snapshot, no ratio, no lock-up details. Under bankruptcy law, creditors have priority over token holders, so how much they can actually get is unknown;
Node operators' unpaid rent before bankruptcy has entered the creditor queue, and subsequent payments require court approval, increasing the risk of node loss;
Filecoin / Arweave are unaffected, but the sector sentiment has been dragged down, so don't blindly catch falling knives in alt storage tokens in the short term.
Currently, STORJ's technical support is at the previous low of $0.058–0.060; if it breaks below, it will likely drop to the $0.05 round number; the rebound target is $0.068 as a recovery level, failure to hold there indicates a weak rebound.
Do you hold STORJ? Are you planning to wait for the token-to-equity swap, or cut losses and switch to mainstream tokens? Let's discuss in the comments.Before the Fed's rate decision in the early morning, the market priced in hawkish expectations in advance, $BTC directly broke through the 65,000 mark, tested a low of 63,011, and is currently fluctuating weakly around 63,461. The amount of liquidation in 24 hours exceeded $610 million, with over 160,000 positions wiped out. My short position entered at price 65014, with floating profits exceeding 2100 points. This was not due to luck but a grasp of the three-layer resonance logic. First, the macro aspect tightened. The probability of a 25 basis point rate hike before the FOMC meeting soared from 13% to 36.3%, with Castle Securities even predicting an unexpected hike. Meanwhile, the wave of AI hardware sell-offs spread to the crypto market, with SanDisk falling over 11%, SK Hynix falling below its IPO price, and Nvidia down nearly 5%. Funds shifted from AI infrastructure to software platforms, causing Bitcoin, as a high-beta risk asset, to crash. Second, geopolitical risk premiums rapidly faded. Trump paused his strikes on Iran and signaled negotiations, causing oil prices to plunge 8.68% in a single day, putting pressure on safe-haven assets simultaneously. The previously accumulated premiums were quickly digested, $BTC losing support. Third, the bulls stamped through chain liquidation. ETFs saw outflows of over $465 million on July 23 and 24, ending a seven-day streak of net inflows. The 65,000 level has shifted from support to pressure; even 64,000 cannot hold steadily, and bears cannot even organize defense. My short stop loss was set above 66,500, and the take-profit was executed in batches. #停火预期兑现, WTI crude oil futures fell 8.68% or #韩股重挫8% in a single day, while Changxin topped the A-share market for the first time ETH was the right direction, but the profits were almost eaten up by his own greed.
I opened long near $1850, and when it rose above $1900, I didn't reduce my position. I always felt that continuous ETF inflows and breaking through $2000 were only a matter of time. But after a single pullback, the unrealized profit shrank by more than half, and in the end, only small profits were taken.
ETH's recent advantage is the rebound of institutional funds, with L2s like Robinhood Chain continuously contributing trading volume; The question is whether ecosystem prosperity can translate into ETH fees, burning, and holding requirements, but the market remains skeptical.
Next, I will focus on the 1840 to 1800 USD range. If it holds, I can try low leverage and go long. If it falls below 1800, I should withdraw first; If the volume above $1920 to $2000 doesn't increase, I won't celebrate too early.
The most common mistake when playing ETH contracts is using long-term faith to hold short-term positions. Have you ever turned a profitable long ETH position into a stop-loss position?
#ETH #Ethereum #合约交易
This does not constitute investment advice.When crypto moves this uniformly, it is not a market story, it is a macro one. BTC, ETH, and SOL all down in the same 3-4% band without any chain-specific catalyst tells you correlation has tightened. The same institutional flows that built the TradFi-into-crypto narrative are now applying TradFi's defensive logic with equal efficiency. FOMC uncertainty and AI earnings anxiety are doing the work here. The market is trimming high-beta exposure broadly, and crypto qualifies. That is not necessarilCurrently, the AI hardware sector is simultaneously playing out five chilling “industry ghost stories” that burst the bubble:
1. Hidden worries about capital expenditure from giants have already been planted
At present, the capital expenditure data of major US internet companies remains stable, but their cash flow statements are continuously deteriorating. The ability to generate cash on the books is shrinking, and the market has begun to price in long-term risks: whether these giants can continue to pour massive funds into computing power infrastructure is now a big question mark, and the fundamental demand base for AI hardware is starting to loosen.
2. The logic of domestic computing power alternatives has officially landed
Liang Wenfeng, founder of DeepSeek, clearly stated at an internal investor meeting that relying on Huawei Ascend 950 super-node cluster architecture, the current stage in actual business scenarios can fully replace Nvidia’s GB200 and GB300 series high-end computing chips. The ecological barrier built by Nvidia CUDA is rapidly dissolving, and now the only bottleneck limiting the volume of domestic computing power is production capacity supply.
3. Kimi K3 shatters the “computing power stacking omnipotence theory”
Upon its release, Kimi K3 shocked the global industry. It relies on MoE sparse architecture and efficient computational scheduling optimization to achieve first-tier large model capabilities at a computing cost far lower than overseas giants. The market finally realizes that top-tier large model achievements do not necessarily require infinitely stacking computing hardware, and the rigid narrative of computing power as an absolute necessity is directly weakened.
4. ChangXin enters the capital market, reshaping the global storage landscape
ChangXin Technology, the world’s fourth-largest storage manufacturer, officially went public, delivering a strong fundamental shock to Korean giants Samsung and SK Hynix, who have long monopolized the DRAM market. The Korean storage sector weakened accordingly, and the pricing power of overseas storage oligopolies is now facing real challenges.
5. Breakthrough news from lithography machines, US hard tech stocks plunged last night
The market reported that domestic immersion DUV lithography machines have entered small-batch trial production, causing ASML, Micron, Lam Research, and other US semiconductor stocks to plunge overnight. Once mature process lithography equipment achieves independent breakthroughs, the manufacturing costs of computing chips and storage chips will be greatly reduced, and the premium bubble of high-priced hardware will ultimately be completely burst.
The entire AI hardware rally narrative is being dismantled layer by layer; the once unbreakable price increase logic is now full of reverse bearish signals. The safest time is when bad news no longer causes a drop. Seize this big plunge, and the second half of the year will be better than a prosperous year!!!