Orbit Post Sitemap
Core Judgment: U.S. stocks have not entered a full-scale risk-averse phase but continue internal repricing: easing Middle East tensions have caused crude oil risk premiums to shrink rapidly, with the energy sector leading the decline; Nvidia could potentially bear massive AI project financing risks, triggering semiconductor valuation adjustments; The S&P 500 was basically flat, but the Nasdaq weakened for the fourth consecutive day; The SPCX hit a new low again, indicating the market is still trading early on initial earnings reports, capital expenditures, and unlocked supply. The current main theme is not an immediate economic recession, but investors beginning to distinguish: who can support growth with operating cash flow and who needs to rely on financing to keep the narrative going. Market Overview: US Stock Market | Index Stable, Internal Divergence Continues. SPY is at $739.09, nearly flat. The S&P 500 rose 0.02%, the Dow Jones rose 0.51%, the Nasdaq fell 0.18%, and the Russell 2000 gained about 0.6%. There was no systemic sell-off at the index level, but funds clearly rotated from high-valuation tech stocks to small-cap stocks, defensive consumer stocks, and some traditional industries. A stable market does not mean the valuation pressure on growth sectors has ended. SPCX | Continued to underperform the broader market. SPCX is quoted at $113.50, down 1.41%. The intraday low reached $108.68, setting a new low since its listing. Against the backdrop of SPY remaining basically flat, the continued decline in SPCX indicates that its price pressure mainly comes from the company itself: initial earnings reports, capital expenditures, and potential solutions半导体板块回调背后，信用市场已经亮起红灯
这一轮半导体集体回落，不只是股票市场单方面的情绪宣泄，债务衍生品市场已经提前释放风险信号。甲骨文、SpaceX、Alphabet、亚马逊、Meta、博通的CDS（信用违约互换）近期报价全部刷新历史新高。
CDS可以理解为企业债务的风险保险，价格持续走高，代表债券交易市场正在不断抬升对这批科技巨头的风险定价，机构愿意花更高成本去对冲潜在违约风险。
市场真正焦虑的点，已经不再局限于芯片出货量、服务器订单这些表层数据。越来越多投资者注意到一条值得警惕的业务模式：英伟达的角色正在发生变化，它已经不只是单纯出货芯片赚取硬件销售收入。市场传闻，企业还会通过放贷、信用担保等各类金融手段，帮助合作客户筹措资金，用来采购自家芯片。
这套模式逻辑十分直白：借助头部企业的高信用，帮客户拿到大额融资，客户拿到资金之后反向采购芯片，直接推高账面营收。但隐患同样突出，一切运转完全建立在AI业务持续盈利的前提之上。
一旦下游AI项目收益不及预期，客户无力偿还债务，原本表外的担保敞口就会反噬上游芯片厂商，整条AI产业链的信用链条会面临断裂风险。股票市场看到的是业绩增长，而债务市场已经开始给这种循环融资模式定价风险。
免责声明：本文仅为市场现象解读，信息来自公开市场传闻，不构成任何投资建议。Hynix's move was ruthless: $SKHX on Hyperliquid jumped from $1,065 to $1,120 in one second, instantly rebounding 5.2%. A company with a scale close to a trillion dollars has managed to break away from the Meme coin flavor.
The first reaction was indeed like a "spike in the pun for a huge overload," but candlestick charts alone cannot conclude that someone is manipulating the liquidation. What is traded here is not native shares from the Korean exchange, but perpetual stock contracts deployed by TradeXYZ on Hyperliquid, tracked by oracles and anchored by order books, funding rates, and arbitrage funds.
The problem lies in this structure: when the Korean stock market is closed, native spot cannot provide timely price discovery, while on-chain contracts continue to be traded 24 hours a day. Once the long crowd is crowded and leverage too high, selling a single break through a weak level may trigger a deep needle pattern of "contracts falling first, oracles correcting later."
Currently, SKHX's open interest is about $384 million, with a 24-hour turnover close to $915 million, and leverage up to 10x. If 1065 fell to 1120 and then pulled back, spot shareholders may not feel anything, but high-leverage longs on the chain may have already been forced out.
This needle seems more like the result of liquidity and liquidation mechanisms working together, and is not enough to prove the platform intentionally overcharged. But it reminds everyone involved in stock perpetual trading: just because the underlying stock is a large-cap stock doesn't mean the contract is also a large-cap stock.
So when making money, you must be careful with market trends. Do you short SK Hynix stock? #海力士 #新手必看: Everything you need is here Only losing allows people to think calmly
Winning only makes people arrogant and gives up thinking
At 2 a.m. on the 30th, Bitcoin$BTC and Ethereum $ETH
Sudden surges and crashes have reappeared
Interest rate decision
Looking at the forecast market, the probability of a rate hike is very high
The long-term bearish outlook remains unchanged
In the past couple of days, Ethereum has surged 100 points and plunged 100 points, which are just minor skirmishes
One day is east of the river, the other is west of the river
Anything obtained by luck
They would always return the same way due to insufficient strength美光的财报为什么总让人同时看到春天和冬天？
存储芯片很奇怪。需求旺时，市场会相信供不应求还能持续很久；价格下跌时，又像全世界再也不需要更多存储。几轮周期看下来，我最大的感受不是行业难预测，而是人们总愿意把眼前的价格，误认为永久的趋势。
美光做的产品并不神秘：数据要被暂时处理，也要被长期保存，背后离不开内存和闪存。手机、电脑、汽车、服务器都需要它们。可需求广泛并没有自动带来稳定利润，因为存储产品标准化程度高，供给稍微超过需求，价格就可能迅速松动。
这个行业真正残酷的地方，是扩产需要时间。景气好时，厂商看到高利润，开始增加资本开支；新产能真正落地时，市场环境可能已经变了。等大家同时收缩投资，库存逐渐消化，下一轮短缺又在不远处酝酿。每个人单独看都理性，合在一起却形成周期。
有人说：“治愈高价格的最好办法，就是高价格本身。”因为高价格会刺激供给，也会压制一部分需求。反过来，低价格迫使厂商减产，又让产品进入更多应用。用这句话看存储行业，比用一条笔直增长线更接近现实。
所以看美光财报，我先看平均售价和出货量的组合。收入增长如果主要来自价格修复，利润弹性会很大，但也要问修复能走多久；如果出货量增长来自真实终端需求，质量通常更扎实。两者同时改善当然最好，也最容易让市场过度兴奋。
库存是第二个重点。美光自己的库存下降，不代表全行业库存已经健康。还要看客户手里有多少芯片、渠道是否继续去库存，以及客户是因为真实订单补货，还是担心涨价提前采购。补库存能推高一两个季度，却不能代替最终消费者。
AI服务器带来了新的想象空间，尤其是高带宽内存。它对性能、封装和良率要求更高，单位价值也更可观。问题是，热门产品并不等于轻松利润。先进产能要投入大量资本，客户认证周期长，竞争对手也不会站着不动。订单很重要，产能爬坡和良率同样重要。
我会特别留意一个问题：AI相关的强需求，能否抵消手机、个人电脑等传统市场的波动？如果高端产品很强，普通存储仍供过于求，公司整体利润可能没有叙事听起来那么顺滑。投资者喜欢一个统一故事，工厂面对的却是多个产品、多个节点和不同客户节奏。
资本开支更像行业的体温计。单家公司削减投资，有助于控制未来供给；但所有厂商都害怕错过技术升级，又不能完全停下。少投会损害竞争力，多投可能加剧过剩。管理层真正的考验，不是能否喊出需求前景，而是在最乐观的时候仍能克制投资冲动。
地缘与供应链风险也绕不开。半导体设备、材料、生产基地和终端客户分布在多个地区，政策变化可能影响销售，也可能增加建厂成本。补贴能降低部分投入，却不会消除运营复杂度。新工厂从破土到稳定量产，中间隔着人才、良率和供应商协同。
再说毛利率。存储价格一旦上涨，新增收入很容易流向利润端，于是毛利率改善速度惊人；下行时同样如此。看到利润快速反弹，我不会立刻把峰值当常态，而会估算一个更保守的周期中枢。周期股最危险的估值，往往建立在“这一次不同”上。
当然，这一次也可能真的有不同之处。行业参与者减少、资本纪律改善，高端存储的技术壁垒上升，这些都可能让未来周期比过去温和。但“可能更温和”和“周期消失”是两回事。只要供给决策分散、需求又会波动，价格就不会变成直线。
我想看的，不只是管理层给出的下一季指引。我更关心客户预付款有没有变化，高带宽内存的良率是否提升，传统产品库存是否回到健康水平，以及资本开支增长有没有跑在真实需求前面。这些细节比一句“AI需求强劲”更能说明利润质量。
对普通投资者来说，美光最难的地方不是看懂芯片参数，而是管理自己的情绪。行业低谷时，坏消息到处都是，估值看起来也未必便宜；行业高点时，利润暴增，市盈率反而诱人。你是在买未来，还是在给刚刚发生的繁荣付钱？
我不会否认AI给存储带来的长期增量，也不会因为周期二字就拒绝研究。恰恰相反，周期让研究更有意义。春天会让人相信花永远开着，冬天又让人忘记种子还在土里。美光的真实答卷，往往写在两个季节之间。
本文仅供信息与教育用途，不构成任何投资建议。数字资产价格波动较大，请独立判断并注意风险。#$BTC Why does Mastercard's business look like a toll station, yet can't be valued solely by toll stations?
In that very second of swiping the card, we could barely sense what was happening behind the scenes. The cashier rang, and a notification popped up on his phone—the transaction was complete. But in just a few seconds, issuing banks, acquiring institutions, merchants, card organizations, and risk control systems have already exchanged a wealth of information. The most fascinating thing about Mastercard isn't how many plastic cards it has, but that it stands right in the middle of a global payment network.
Many people researching this company for the first time say it "hardly assumes credit risk." This statement is generally correct, but it can easily make people let their guard down. Mastercard is usually not a bank that lends money to consumers; it mainly handles connections, authorization, and clearing, so it does not swallow large-scale bad debts like credit card issuers do. But not taking on the same risk does not mean there is no risk.
I prefer to think of it as a set of trust infrastructure. Consumers trust that payments will succeed, merchants trust that money will be received, and banks trust transaction information to be verifiable. The wider the network and the more participants, the higher the value of new user access. This is the classic network effect, but network effects are not talismans; they must prove themselves every day through stability, security, and acceptance.
"The best business is one where customers don't have to reconsider every day whether to use it." This saying isn't a cure-all, but it fits well in the context of payment networks. People don't study the underlying clearing route every time they buy coffee, and merchants are reluctant to frequently switch to mature systems. Habits and compatibility together form stickiness, which ultimately manifests in transaction volume and service revenue.
So, what should you look for first in an earnings report? I first look at the payment amount and cross-border transactions, not just how many cards have been issued. The number of cards may increase, but activity may not keep pace; Transaction amounts can more directly tell us whether the network is actually being used. Especially for cross-border consumption, the fee structure is usually better, but it is also more affected by tourism, exchange rates, and economic cycles.
Why is cross-border business important? When someone sends a card in their home country, they are simply moving funds within a familiar financial system; When spending abroad, currency conversion, fraud detection, and coordination between different institutions are more complex, and the value provided by the network is more apparent. But this part of the income looks great when it's good, but it quickly stalls when travel cools down. Directly extrapolating cyclical highs is often the most hidden pitfall in valuation.
I also look at value-added services. Identity verification, data analysis, anti-fraud, and cybersecurity—these may not be as intuitive as card services, but they may determine the quality of the next phase of growth. Payment rates are subject to regulatory and customer bargaining pressure, while security and data services address constantly evolving new challenges. As long as fraud escalates, customers have a reason to keep investing.
Don't forget to supervise. The payment network exists between consumers, merchants, and banks, with each party seeking lower costs. Debates over exchange fees, routing, and market competition will not disappear. The higher Mastercard's profit margin, the more regulators ask: Is this a reward for high efficiency, or the result of too strong market forces? This issue cannot be brushed off with a single phrase like "the moat is deep."
New payment methods are also a practical challenge. Account-to-account transfers, instant payments, digital wallets—could they bypass card networks? I don't think the answer is simply "yes" or "no." The wallet has superficially changed its entry point, and the underlying source of funds may still be a card; Instant payment is cheaper in some scenarios but requires handling refunds, dispute resolution, and fraud prevention yourself. Technology substitution usually doesn't happen overnight, but rather gradually cuts away from the weakest profit link.
There's another detail that's easy to overlook: Mastercard's clients are also negotiators. Large banks, fintech platforms, and giant merchants all have their own scale and won't unconditionally accept higher fees. Network effects give Mastercard bargaining power, but customer concentration limits that power. Truly healthy growth should come from expanding transaction and service value, not simply tightening rates.
When it comes to valuation, I fear one thing the most: "This is a good company, so I can buy at any price." "There is a buying price between a good company and a good investment. Payment networks are asset-light and have strong cash flow, so the market is naturally willing to offer a premium; But when valuations have been tailwinding for years, even if the performance only shifts from excellent to normal, the stock price may still be disappointed.
I'll break down the question simply: Is the increase in transaction volume coming from real consumption, rather than one-off inflation? Is cross-border business at an abnormally high level? Can value-added services continue to grow without relying on M&A packaging? Has regulatory cost started to change the business model? Is the buyback a reasonable price to reduce share capital, or is it a high price to maintain per-share figures?
Mastercard's greatest strength lies in its presence in everyday life. The more insensitive the infrastructure, the easier it is to be used long-term; The higher the profitability of the infrastructure, the more likely it is for competitors and regulators to target it. Only when both statements are valid together is the complete answer.
Investment isn't just about labeling a company as "great." What's truly interesting is the continued question: How much value has this network created for all participants, and how much value has it taken away? As long as the former is longer than the latter, it has room to continue expanding; Once the order is reversed, no matter how deep the moat, someone will start looking for a bridge.
This article is for informational and educational purposes only and does not constitute any investment advice. Digital asset prices fluctuate significantly; please make independent judgments and be aware of risks. #$BTC Last night and this morning, global capital markets underwent a rare and troubling "Great Cleanup." US stocks, crypto, gold, crude oil—asset classes that should have diverged and hedged against each other—unexpectedly experienced uneven collective declines within the same time window. Even stranger, the once-tried-effective "inverse oil price linkage" logic completely failed: crude oil prices plummeted, but the stock market did not rise due to expectations of cooling inflation; instead, it followed the decline. This is by no means an ordinary pullback, but a piercing alarm. The market is voting with real money, telling everyone a harsh truth: it no longer cares about the melodramatic geopolitical dramas; it fears only one thing—a hard landing for the global economy. Trump's "one-man show" and Iran's cold attitude As the most sensitive indicator of macro risks, the direct trigger for this round of crude oil plunge ostensibly stems from "expectations of a ceasefire." Yesterday, Trump loudly declared that negotiations with Iran would bring good news, and the market briefly priced in peace. However, before he finished speaking, Iran flatly denied the existence of negotiations, coldly exposing this "political smoke screen." This tactic of "leaking first, creating momentum, suppressing oil prices, and seizing the initiative in negotiations" is Trump's usual extreme pressure tactic. But this time, the opponents did not cooperate. As a result, the market fell into an unprecedented awkward situation: both sides were in a state of "spontaneous tacit ceasefire," with neither agreement nor guarantees. This tactical pause could be halted at any moment, and once the fire is restarted, the geopolitical risk premium should immediately return. However, the reality is—the market has not responded to this. Oil prices伯克希尔最值得看的，真的是账上的现金吗？
每次伯克希尔披露财报，市场都会盯着那个越来越醒目的现金数字。有人把它理解成巴菲特看空，有人把它当成下一次“大象级收购”的弹药。可我越来越觉得，只盯现金，很容易把这家公司看窄了。现金是结果，不是答案。真正的问题应该是：一家规模已经如此庞大的公司，为什么仍愿意为“不犯错”支付这么高的机会成本？
这听起来不够性感。牛市里，持有现金甚至显得有些笨。别人讨论的是哪只股票又大涨，伯克希尔却像一个提前到车站的人，坐在长椅上慢慢等。可投资最难的地方，不就是在热闹中承认“我现在没有特别好的主意”吗？
芒格说过一句很朴素的话：“知道自己能力圈的边界，比能力圈有多大更重要。”放到伯克希尔身上，巨额现金更像边界感的价格。它不代表悲观，也不自动代表聪明；它只说明管理层不愿为了看起来积极，就把股东的钱塞进回报率不够的项目。
当然，现金也不是免费的午餐。假如市场持续上涨，而伯克希尔长期找不到足够大的投资标的，这部分资金就会拖累整体回报。公司越大，问题越现实：一个十亿美元的好机会，对普通基金很重要，对伯克希尔可能只是水面上的一圈涟漪。规模带来安全感，也会吞掉灵活性。
所以我看伯克希尔，第一眼看现金，第二眼一定看保险浮存金。保险业务像一台常被低估的发动机：保费先收进来，赔付以后发生，中间这段时间形成可投资资金。只要承保纪律没有松，浮存金的成本足够低，它就不仅是负债，更是一种长期资本来源。
问题也恰恰在“纪律”两个字。保险公司最怕什么？不是某一年遇到大灾害，而是在竞争激烈时为了抢规模，给风险报出过低的价格。短期保费增长很好看，几年后账单才寄到。真正该盯的，往往不是新闻稿里的增长速度，而是综合成本率、准备金变化，以及管理层有没有开始用含糊的语言解释承保结果。
再看铁路和能源，它们不太容易制造社交媒体上的兴奋，却构成了伯克希尔的地基。铁路要持续维护线路、机车和设备，能源要不断投入电网与基础设施。这些生意资本开支重，回报不会突然冲上天，但只要监管框架稳定、需求长期存在，就能把大量资本以相对可预测的方式重新投入。
这也解释了伯克希尔的矛盾感：外表像一只股票组合，骨子里却更接近一个资本配置系统。保险产生资金，成熟企业贡献现金流，铁路与能源吸收长期资本，剩余的钱再用于买股票、回购或等待收购。单独看每块都不神秘，难的是几十年不把齿轮装反。
那回购该怎么看？我不喜欢把“公司回购”天然等同于利好。只有当回购价格低于管理层对内在价值的保守估计，而且不妨碍公司的安全垫时，它才真正增厚每股价值。高价回购只是把现金换成掌声，低价回购才是在替长期股东买便宜货。
还有接班问题。巴菲特的个人判断当然无法复制，但伯克希尔真正需要传承的，也许并不是某一套选股口诀，而是三件事：不追逐短期排名，不用高杠杆逼迫自己做决定，以及愿意在没有好机会时保持沉默。制度能不能守住这种克制，比猜下一位投资经理会买什么更重要。
我对伯克希尔最朴素的观察是，它的优势从来不是“每次都买在最低点”。它也会错过，会买贵，会看错行业。真正稀缺的是，犯错之后公司仍有足够的现金流、信誉和时间继续留在牌桌上。复利并不要求每一年都精彩，它更怕一次无法恢复的重伤。
所以，下一次再看到那个巨额现金数字，我不会急着把它翻译成多空信号。我更想问：保险定价有没有变松？非保险业务的资本回报是否稳定？回购有没有坚持价格纪律？管理层面对没有答案的问题时，是否还愿意说“我们不知道”？
投资里最贵的往往不是错过，而是为了害怕错过，硬把等待变成行动。伯克希尔这份答卷未必让每个人满意，但它提醒我一件很有人情味的事：承认暂时没有好机会，也是一种能力。市场每天都在催人表态，可真正的长期主义，有时只是允许自己不急。
本文仅供信息与教育用途，不构成任何投资建议。数字资产价格波动较大，请独立判断并注意风险。#$BTC TL; DR · Nvidia's 5-year CDS surged intraday on July 27, prompting credit investors to reassess potential AI infrastructure obligations. Oracle's 5-year CDS was around 1.25% over the same period, higher than Nvidia, indicating that credit markets have priced in the expansion of AI cloud infrastructure. Guarantees and partnership frameworks can lock in forward demand and may also pass on customer financing risks back to Nvidia. Related stocks: Nvidia (NVDA), Oracle (ORCL), SK Hynix, Broadcom (AVGO), TSMC (TSM), Microsoft (MSFT), Amazon (AMZN). According to Bloomberg citing ICE Data Services, Nvidia's 5-year default protection cost peaked at about 0.82% intraday on July 27, rising about 14 basis points in a single day, marking the largest single-day increase for the contract since active trading began in November 2025. CDS can be understood as 'default insurance' for company debts. A price increase does not mean the market believes Nvidia will face trouble soon, but rather that credit investors are demanding higher risk compensation. For a company that was just upgraded to AA by S&P Global in June and still has strong cash flow, the absolute level is not high, but the signal is worth watching. On the same line, Oracle was earlier used by credit markets to observe AI infrastructure financingMicrosoft's $40 Billion Quarterly Capital Expenditure: Distinguishing GPU, Facilities, and Finance Leasing
Before the release of the final quarter of this fiscal year, the market most frequently cited management's previous quarter statement that "Q4 capital expenditure will exceed $40 billion." This is a large figure, but without breaking down asset lifespan, payment timing, and finance leasing, it's easy to confuse capital expenditure with expenses for the quarter. The Q4 results will be officially released after market close on July 29; for now, we are only establishing a repeatable table reading sequence.
FY2026 Q3 capital expenditure was $31.9 billion, with about two-thirds spent on shorter-lived assets like GPUs and CPUs, and the remainder on long-lived assets that management says can support monetization for over fifteen years. Finance leasing for the quarter was $4.7 billion, mainly for large data center sites; cash payments for property, plant, and equipment totaled $30.9 billion. These three figures answer different questions: capital expenditure reflects asset acquisition, finance leasing reflects commitments without immediate full cash payment, and cash purchases directly enter the cash flow statement.
Management's Q4 outlook also stated that of the over $40 billion capital expenditure, about $5 billion comes from rising component prices, with the short-lived asset mix expected to be similar to Q3. These are still forward-looking statements, not actual results. After the official financial report, actual capital expenditure, finance leasing, and cash purchases should be verified first, then depreciation, cloud gross margin, and operating cash flow examined. Using a single total to infer "demand explosion" or "return deterioration" is premature for both conclusions.
On the demand side, verification should cross-check Azure and other cloud service revenues, Microsoft Cloud revenue, and remaining performance obligations. Q3 Azure constant currency growth was 38%, Microsoft Cloud revenue grew 25%, including OpenAI's commercial RPO reaching $627 billion; however, the average RPO term is about two and a half years, with only about one-quarter expected to be recognized in the next twelve months. Long-term contracts cannot be directly offset against this quarter's capital expenditure, as their timelines differ.
I divide the results into three stages: "capacity formation, revenue conversion, and cash recovery." GPU and CPU deployment belong to capacity formation, Azure usage and revenue to conversion, and operating cash flow minus capital expenditure approximates recovery. All three stages must improve simultaneously to support investment efficiency; if capacity arrives first and revenue is recognized later, multiple consecutive quarters need observation, and a single quarter's free cash flow cannot be used for a final judgment. The FY2027 capital expenditure or demand descriptions added in the earnings call will be clearly marked as management forward-looking statements and will not be mixed with actual Q4 values.
Another checkpoint is depreciation lifespan and capacity utilization. Short-lived chips depreciate faster, while long-lived facilities spread costs over a longer period; both affect gross margin and cash flow at different rates. If Microsoft only provides directional guidance without precise breakdowns in the call, the article will maintain these limitations and not assume GPU quantities or unit costs. Management's description of supply constraints will also serve only as one piece of demand evidence, which still needs verification through revenue and billing data.🌍 Why did $BTC suddenly come under pressure and fall back? It lost the 64,000 level in early trading!
This time BTC dropped from the high of 65,750 to around 63,055 (as of early trading, ETH also fell nearly 3%, Nasdaq futures followed down), which is the result of the combined effect of macro sentiment, industry dynamics, and technical factors:
1. Macro sentiment and rising risk aversion (core external factors)
Recently, global macro uncertainty has increased, and market risk aversion has clearly intensified. The repeated geopolitical tensions between the US and Iran and the upcoming Federal Reserve meeting on 7/28-29 have made funds more cautious about risk assets; South Korea's KOSPI fell 7% intraday triggering a circuit breaker, US AI/semiconductor sectors led the decline (Nvidia -5%), and global risk assets retreated in tandem. Notably, Citibank raised its short-term gold target price from $4,000 to $4,500 (currently about 4,045), which, although still below the January historical high of 5,600, is a clear signal of rebound compared to the current price below 4,000 in June—traditional safe-haven assets are endorsed by institutions, reinforcing expectations that some funds will shift from high-risk assets like BTC to gold.
2. Short-term market play triggered by industry dynamics
The US Senate shelved the CLARITY Act this week, with the earliest vote expected before the August recess, making the prospect of passing it this year unclear; spot BTC ETFs saw a net outflow of about $465 million on 7/23-24, ending seven consecutive days of inflows. The lack of policy catalysts plus weakening ETF funds limited on- and off-exchange buying enthusiasm, and large holders are more likely to sell off rather than catch falling knives amid a bearish macro environment.
3. Technical profit-taking and leverage liquidation
BTC previously rebounded to 65,750 but failed to break through further, with consecutive bearish candles on the 4-hour chart breaking below MA5 (64,000)/MA10 (64,574)/MA20 (64,452), and SuperTrend turning bearish; daily candles also broke below MA5 (64,193) and MA10 (64,882). Short-term profit-taking accumulated at the rebound highs plus high-leverage long positions were liquidated en masse after moving averages were breached, triggering a cascade of liquidations that accelerated the drop to 63,055, approaching the psychological 63,000 level.
Key support: 63,000-63,300 (4H support 63,318 + intraday low 63,055), stabilization here suggests consolidation; a volume-driven break below 63,000 targets 62,000-61,800.
Key resistance: 64,500-65,000 (near daily MA20 64,458) is strong short-term resistance; a volume-backed recovery above this is needed to re-enter a bullish trend.
Friendly reminder: Volatility will increase before the Fed decision; strictly control position sizes and avoid high leverage to prevent stop-loss spikes.
⚠️ Disclaimer: The above market analysis and interpretation are for reference only and do not constitute any investment or trading advice. Cryptocurrency markets are highly volatile; investing carries risks, and decisions should be made cautiously. #韩股重挫8%，长鑫首日登顶A股
On its first day of listing, Changxin surged 465%, marking not only a highlight moment for the A-share market but also a shift in the global memory chip landscape.
My judgment: The pricing power of Chinese memory assets is returning, and the valuation premiums of the two Korean giants will be forcibly compressed.
The reason is simple: the market is no longer willing to pay solely for overseas monopolies; domestic substitution has shifted from a "backup" to a "main force." Capital is voting with real money, confirming Changxin's position as a new benchmark.
The data is most intuitive: Changxin's first-day turnover exceeded 140 billion, with a market cap soaring to 3.28 trillion; meanwhile, South Korea's KOSPI index plunged 8%, with SK HYNIX and SAMSUNG both dropping over 9%.
This rare "one rises as the other falls" linkage effect indicates that global capital is recalculating the value of Chinese memory manufacturers, completely breaking the previous high-valuation logic given to Korean companies.
Next steps: Do not chase highs in the short term; wait for Changxin to stabilize after a pullback. Focus on Samsung and Hynix's earnings reports this week, which will be the definitive proof of whether they are truly panicking.韩国综合股价指数KOSPI已经正式击穿6500这一机构重点关注的关键技术支撑位。
此前这一位置被不少机构视作重要底部，包括高盛在内多家券商都把6500标记为强支撑，指数两次回踩该点位，都迎来抄底资金进场，走出明显反弹，也让不少市场参与者把这里当成安全的防守区间。
但如今支撑宣告失效。韩国市场充斥着大量散户杠杆仓位，指数有效跌破关键技术关口之后，杠杆风控机制会被动启动，接下来很可能迎来一轮规模不小的强制平仓踩踏。一旦平仓盘集中涌出，又会进一步向下拖拽指数，形成下跌和爆仓互相强化的负向循环。
现在盘面已经不再只是简单的技术破位，高杠杆带来的连锁风险开始浮出水面。
免责声明：仅为盘面现象客观解读，不构成投资建议。Yesterday, Panda Bro (@0xCryptoChan) selflessly shared the BTC LTH Market Cap 365D-MA Recovery Structure. I call it the BTC bear market exit confirmation model. This model does not predict a bear bottom, but rather confirms: after the bear market bottom is formed, whether the market has entered a recovery phase.
I spent an entire night doing historical backtesting, and the results are as follows:
🔹 2015 cycle
After the bottom, a recovery breakout appeared, and after confirmation on the 30D, a long-term recovery began.
🔹 2018 cycle
After one failed test, it stood back up, and after 30D confirmation, entered the recovery phase. (Later hit again by the COVID black swan shock)
🔹 2022 cycle
After a successful breakout and confirmation at 30D, a new cycle begins.
The first three cycles show:
LTH365D-MA Recovery Breakout has good confirmatory significance for a bear market exit.
However, this round has seen a situation that has never happened before:
May 12, 2026:
BTC has completed confirmation of the LTH365D-MA 30D recovery.
According to historical samples:
This usually means the cycle bottom has most likely formed, and the market has entered the bear exit phase.
However:
Just 20 days later,
June 1, 2026:
BTC has once again fallen below the LTH365D-MA.
This is a structure that did not appear in the previous three cycles.
Current Status:
BTC Market Cap
≈ LTH365D-MA -2.82%
Currently:
It has not yet regained its position on LTH365D-MA.
My understanding:
This model remains a very valuable framework for confirming long-term bear market exits.
But 2026 is becoming an important stress test for it.
If BTC rebounds back to the LTH365D-MA and completes another 30-day confirmation:
So this round might just be an abnormal pullback.
If it rises again in the future but still repeatedly breaks below the previous level:
So the explanation is:
LTH365D-MA Recovery Structure needs to be adjusted for the new cycle.Last night, I was just 20% short of being liquidated
That injection from the dog farm almost sent me out on the spot
Now, $ETH has fallen back down to around 1870
Short positions finally pulled some distance from the edge of the cliff
But I remain bearish
The Federal Reserve's interest rate decision was scheduled for early Thursday morning
Currently, the mainstream expectation is still to keep interest rates unchanged
$SNDK The current interest rate range is 3.50%—3.75%
The market estimates the probability of an unexpected rate hike around 30% to 40%.
So rate hikes are not the main plot
But it's definitely not zero
$BEAT What matters most is not raising interest rates but sending hawkish signals
Just keep emphasizing inflation and energy prices
Risk assets may still be the first to crash
The resolution will be announced at 2 a.m. Beijing time and Taiwan time on Thursday
A press conference was held at 2:30 a.m
Korean stocks plunged more than 10% intraday today
The decline then narrowed to around 7%.
Samsung Electronics and SK Hynix both suffered heavy losses
KOSPI and KOSDAQ even triggered the Sidecar mechanism to restrict programmatic selling
This time, it's not just South Korea causing problems
The core is still the collective cooling of global semiconductor and AI high-valuation sectors
Funds began actively reducing risk exposure
This is bearish for ETH in the short term
Because the sharp drop in Korean stocks will continue to suppress risk sentiment in Asia
If US chip stocks and the Nasdaq continue to catch up,
It's hard for the crypto world to remain completely unaffected
But this is more like a concentrated crushing on the tech sector
For now, it cannot be directly defined as a systemic financial crisis
Technically, ETH has already fallen below several moving averages within one hour
From 1895 to 1905, it became a pressure zone again
As long as you can't recover 1900
The bearish structure was not truly destroyed
Let's first look at 1860
If it falls below it, look for around 1850
But a 100x position doesn't offer real peace of mind
A single message needle can still take profits and positions with you
#韩股重挫8%, Changxin topped the A-share market on its first day
#美联储周四凌晨公布利率决议 On July 28, the Asia-Pacific market experienced an extreme sell-off, with South Korea's KOSPI plunging and triggering its 8th circuit breaker of the year. Storage giants plummeted, and overnight U.S. semiconductor stocks sharply declined across the board. Coupled with the dual pressures of the Federal Reserve's policy meeting and tech earnings reports, BTC and AI altcoins simultaneously came under pressure. The complete market situation and trading logic are summarized as follows: 1. Asia-Pacific markets crashed across the board, South Korean stock market triggered a second circuit breaker 1. South Korean market: The KOSPI index fell more than 8% intraday, triggering a circuit breaker that paused trading for 20 minutes. After resuming, the decline widened to 10%; major storage leaders plunged: SK Hynix dropped 13%, Samsung Electronics fell over 12%, completely abandoning the previous trillion-level AI supply optimism. 2. A-share market: The three major indexes all opened lower and declined unilaterally. At midday, the Shanghai Composite Index was down 0.98%, the Shenzhen Component Index down 3.42%, and the ChiNext Index plunged 5.37%; computing power, semiconductors, and precious metals all declined across the board. 3. Overnight U.S. stocks diverged, AI hardware suffered a collective bloodbath: Apple and Google hit new highs against the trend; Nvidia fell nearly 5%, the Philadelphia Semiconductor Index dropped over 2%; SanDisk fell 11%, SK Hynix ADR dropped 7% and fell below its IPO price, Western Digital, Micron, and optical communications all weakened. 2. Two core underlying logics behind this round of sharp decline 1. AI capital expenditure cash flow panic is fermenting Nvidia, Microsoft, and SK Hynix reached a trillion-level computing power supply agreement, but the market worries that huge investments will overdraw corporate cash flow, significantly increasing supplier financing risks; funds are frantically fleeing from overvalued storage and computing power stocks, causing a crowded trade stampede. According to internal sources, the Federal Reserve's July meeting has decided to raise interest rates by 100 basis points. #美联储周四凌晨公布利率决议 #$BTC $ETH The latest news is that WTI crude oil is rapidly giving back premiums caused by geopolitical conflicts. The WTI mapped contract on Hyperliquid (xyz:CL) is currently quoted at $80.91, down 5.2% in 24 hours. From the stage high of $93.44 on July 24, it has fallen 13.4% cumulatively, putting the $80 mark on the brink of collapse again.
During this decline, one whale made quite comfortable profits. Address 0x60a8 shorted 171,900 CL at 2x isolated margin positions, with a position value of about $13.91 million, an average opening price of $91.57, and a liquidation price far above $133.53. Currently, the floating profit is about $1.833 million, with a return rate of 23.3%. For now, there are no orders to increase or decrease positions, indicating they plan to keep holding on.
The core driver behind this drop in oil prices is the sudden cooling of the US-Iran situation. The U.S. has paused its military strikes against Iran, and Iran has simultaneously halted its retaliatory actions. Diplomatic signs of easing, combined with expectations of a gradual recovery of Middle Eastern energy transport, are rapidly squeezing out the previous wave of panic premiums.
Currently, CL contracts have a 24-hour turnover of about $320 million, with open interest nominal value of about $161 million. $CL #停火预期兑现, WTI crude oil futures fell 8.68% in a single day The Korean stock storage sector has weakened again, and the underlying cyclical logic is not complicated.
The current round of price correction in the storage sector is not because the industry is about to fall into losses immediately, but because the market is pricing in changes in the long-term supply pattern in advance.
As early as the end of 2022, the storage industry had already fallen into a cyclical trough. The three leading manufacturers—Samsung, SK Hynix, and Micron—chose to proactively control production by reducing capacity to withstand downward price pressure. From 2024 to the first half of 2025, the industry's overall capital expenditure sharply contracted, and the companies coordinated to control output, directly causing a persistent tight supply of general-purpose memory like DDR4.
After 2025, the demand for high-margin HBM exploded, and the three major manufacturers further adjusted their production line allocations, shifting a large amount of capacity toward AI high-bandwidth memory. The capacity for traditional DRAM and NAND flash was passively squeezed, resulting in fewer ordinary storage chips available on the market, which directly pushed up the industry's overall gross margin, making corporate financial performance look quite good.
While the oligopolies enjoy the cyclical dividends, the industry landscape is facing new variables. ChangXin Memory has grown into the world's fourth-largest DRAM manufacturer, holding abundant capital reserves and continuously advancing its expansion pace. For a market originally dominated by the three overseas giants, this means a considerable amount of new supply will enter, directly impacting the existing supply-demand balance.
The capital market is not looking at current financial reports but at the future. Even though storage product prices remain high now, the market has begun to worry that once domestic capacity is gradually released, the high-price dividends of traditional DRAM will likely be quickly diluted. This is the core reason for the recent sell-off in the Korean stock storage sector.
Disclaimer: The above is only an analysis of industry phenomena and does not constitute investment advice. Before Changxin's listing, the price was already "spoiled" on-chain: the pre-market pricing on crypto was more accurate than brokerage research reports
On July 27, Changxin Technology debuted on the STAR Market.
It opened at ¥49.5, soaring 471.59% from the issue price of ¥8.66. It closed at ¥49, with a daily turnover of ¥141.2 billion — the first A-share stock in history to exceed ¥100 billion in single-day trading volume. The total market value reached ¥3.28 trillion, surpassing Industrial and Commercial Bank of China to become the "big brother" of A-shares.
But what really sent chills down my spine wasn’t these numbers.
It was another number: 5.4 times.
Two weeks before the listing, on July 14, the crypto platform Trade.xyz launched Changxin Technology’s pre-market perpetual contract on Hyperliquid, ticker xyz:CXMT.
The initial reference price was $5. Within hours of going live, millions of dollars in buy orders appeared on the order book, pushing the price up to $8.64. Converted at the exchange rate, the implied on-chain stock price was about ¥58.5. The price then retreated and stabilized between $6.1 and $6.4 before the listing, equivalent to about ¥41-43.
The on-chain pre-market contract priced about 5.4 times the initial reference.
The A-share first-day close was 5.66 times.
Almost identical.
This is no coincidence.
Most domestic brokerages previously expected Changxin’s first-day valuation to be between ¥2 trillion and ¥3 trillion. Huaxi Securities’ research report gave a neutral estimate of ¥2 trillion to ¥3 trillion, with an optimistic scenario reaching ¥4 trillion.
The result? The on-chain pre-market pricing of ¥3.5 trillion was more accurate than the vast majority of brokerages.
At the moment of opening, the A-share price wasn’t "discovering" a new price — it was "converging" to the anchor already set on-chain.
What was the error margin? The A-share opening price was ¥49.5, about $7.31 at the exchange rate; at the same time, the Trade.XYZ contract quoted $7.12 — an error of less than $0.2.
Think about how absurd this is.
Retail investors in A-shares cannot participate in pre-market trading. Institutions face compliance restrictions. Want to go long? T+1 rule means you can’t sell on the same day. Want to short? STAR Market stocks cannot be shorted via margin.
But on-chain contracts have none of these restrictions.
24/7 trading, long and short positions, anywhere in the world, participation with just stablecoins. A pricing venue free from time zone and access restrictions bypassed all institutional barriers.
This was the first time an on-chain pre-market contract targeted a STAR Market IPO.
The story gets even more intense.
On Changxin’s listing day, the global memory chip market was bloodied.
US stocks: SanDisk plunged 11.6%, Micron dropped over 4%, market cap fell below $1 trillion. The Philadelphia Semiconductor Index closed at its lowest since May 19.
The next day, South Korea’s KOSPI index dropped 8%, triggering a circuit breaker. SK Hynix fell over 10%, Samsung Electronics dropped more than 8%. Nikkei 225 fell 4%, Kioxia once dropped 18%. Taiwan stocks Nanya Tech, Winbond, Phison, and Macronix all hit limit down.
One A-share IPO tore through the global memory supply chain.
The reason is simple: Changxin Technology’s global DRAM revenue share jumped from about 3% in Q1 2025 to 8% in Q1 2026, ranking fourth worldwide. Based on Q4 2025 sales, the share reached 7.67%. Multiple institutions predict monthly production capacity will reach about 350,000 wafers by the end of 2026, just 25,000 wafers less than Micron’s 375,000.
The valuation premium narrative of the "Korean giants" now has a clear challenger.
To be blunt:
People used to think on-chain contracts were just speculative tools. Now it turns out they might be the world’s most accurate IPO pricing machines.
Changxin is not an isolated case. In May this year, before AI chip company Cerebras’s listing, Hyperliquid’s pre-market contract differed from Nasdaq’s opening price by only 1.3%; on SpaceX’s IPO day in June, on-chain contracts traded $1.38 billion in a single day.
Traditional investment banks’ pricing models are being crushed by on-chain order books.
A few final words —
Changxin Technology’s freely tradable shares on day one accounted for only 6.73% of total shares. A tiny float, no price limits for the first five days, plus the price anchor already set on-chain — this formula made a surge on day one inevitable.
But the bigger question is: will more large A-share IPOs be "pre-priced" on-chain in the future?
How will regulators respond to this cross-border, cross-market price transmission?
One thing is certain —
The on-chain market is no longer just crypto "self-entertainment." It is becoming the pricing vanguard for global assets.
The A-share opening price was merely "converging" to the answer written on-chain two weeks earlier.
$SKHY $MU $SAMSUNG
#韩股重挫8%，长鑫首日登顶A股 A strange atmosphere is now spreading in the market—the money hasn't slipped away, but everyone has chickened out. Bitcoin flopped to $63,300, and Ethereum and Solana also lay flat along with it.
The most ironic part is the flow of funds. From July 20 to 22, the US Bitcoin ETF was still booming, drawing in nearly 500 million yuan, but in the following three days, it immediately turned hostile, giving up 477 million yuan—almost every bit of money was dumped out. The stablecoin pool remains completely unchanged, holding tightly around $308 billion with no signs of growth.
What does this indicate? The money didn't leave at all; it just shrank by the shore and watched. The culprit is most likely the Federal Reserve—the possibility of a rate hike still hangs overhead, and the dollar has climbed to a nearly one-month high. Who would dare to rush in at this critical moment?
The next window of observation is clear: after the Fed meeting, keep a close eye on ETF capital flows. If money flows back but Bitcoin still looks like a weakling, that's the real warning sign. What they fear isn't a drop, but that even when someone buys, they still can't get it up. $BTC $ETH $SOL #交易之声: Your experience deserves to be heard $JMKE Three whales rushed to grab $1,498, only 58% of the shipments — an ultra-early sniper window, but the dev is a seasoned veteran
🕵️ $JMKE Three Smart Wallet wallets collectively bought $1,498, MCap only $16K, current selling pressure ratio 58%, not yet cleared. Bundler reset to zero, addresses clean. But one dev has issued 1,363 tokens, 11 migrations, and 1 rug—not a beginner, just an assembly line operator.
💰 Chart Crushing Highlights: $EPIK continued to dominate today, +895%, MCap broke through $10.76 million. Three chart-crushing wallets made $52K, $21K, and $32K respectively from EPIK, with one address achieving a 100% win rate. Smart money acted very coordinated on this stock.
In the new session, $POCK (Pock.chat) has official websites and X accounts, MCap $48K with only 3 positions, 0 bundlers, 0 snipers, which are rare serious project fundamentals. $JACARE Although three whales accumulated $1,029, 51 bundler traces and bundler ATH 48%—clear signs are organized, so buying should be cautious.
$JMKE This position is advantageous because it was early and clean, but dev's record of 1,363 token issuances is an unavoidable credit deficit. If you're a PvP-type sniper, you can try bottoming out 0.5-1 SOL to see if dev is pulling the market; otherwise, looking at ultra-early $POCK with fundamental support is safer.
#暗影萨满#OKX生态#聪明钱#Solana#韩股重挫8%，长鑫首日登顶A股
Korean stocks plunge 8%, Changxin tops A-shares on debut
On July 28, the Asia-Pacific capital markets witnessed the most ironic scene: South Korea's KOSPI index plummeted over 8% intraday, triggering the eighth circuit breaker this year. Samsung Electronics and SK Hynix, the two major memory leaders, each dropped more than 9% in a single day, dragging down the entire market; meanwhile, just across the sea, the A-share market saw Changxin Technology's market value stabilize at 3.28 trillion yuan, immediately dethroning Industrial and Commercial Bank of China to claim the top spot in A-share market capitalization.
Both belong to the memory chip sector, yet one side is stampeding to exit while the other is celebrating wildly. These two extreme market conditions collided on the same day, revealing the most authentic global capital pricing logic for semiconductor assets.
1. What crashed in Korean stocks was not performance, but shattered sky-high expectations
Many think the Korean stock plunge was due to deteriorating fundamentals; on the contrary, SK Hynix's second-quarter HBM shipments continued to rise, and performance did not collapse. What truly broke was the previously hyped "AI memory perpetual motion machine" expectations.
Over the past six months, the market treated HBM as a cyclical-free growth sector, driving SK Hynix to its highest-ever valuation. Retail investors leveraged up, foreign capital clustered, pushing these two memory heavyweights to dominate half of the market. But when Google's earnings came out showing cloud providers slowing capital expenditure growth and AI computing power procurement cooling off, the market suddenly realized: no matter how high-end HBM is, it is still fundamentally a memory chip, and the cycle is unavoidable.
Coupled with the Korean stock market's fragile structure of high leverage and large foreign ownership, any slight disturbance triggered a cascade of forced selling. The so-called golden sector, once expectations are maxed out, falls harder than any other.
2. Changxin's rise is not about the present, but the newly opened imagination space
On the other hand, Changxin surged 465% on its first day. Looking only at the current 7.67% market share and peak-cycle profits, a 3 trillion yuan market cap is obviously not cheap. But the market's frantic buying logic is simple: this is the first truly globally competitive DRAM IDM leader in A-shares, and the domestic substitution story is just beginning.
Previously, speculation on domestic memory was about "whether it can be made"; now Changxin has delivered a full series of mass production, full customer coverage, and a quarterly profit of 24.7 billion yuan. The story has shifted to "how much market share can be captured." The mid-end market voluntarily ceded by Korean manufacturers, the rigid demand of the domestic supply chain, and expansion expectations after fundraising all provide long-term growth logic to investors.
Simply put, A-shares have lacked hardcore tech leaders for too long. Finally, a rare target with technology, performance, and a promising sector emerges, naturally attracting concentrated capital to max out expectations at once.
The essence is all cyclical, just at different stages
Though one surges and the other plunges, the underlying logic is highly consistent:
- Korean memory giants are at a cycle peak and expectation peak stage, with capital cashing out and exiting;
- Changxin Technology is at a cycle upswing and early growth stage, with capital entering and pricing in.
There is no forever rising sector, nor forever undervalued asset. Today's Korean stampede may be a rehearsal for Changxin's future cycle downturn; today's Changxin celebration has been experienced by Korean manufacturers over the past two years. The iron law of the memory industry for thirty years has never changed: supply and demand determine the cycle, the cycle determines valuation, and all emotional premiums will eventually be erased by time.
A reality check
A single-day plunge in Korean stocks does not mean Korean manufacturers are declining; Samsung and SK Hynix still hold significant advantages in high-end HBM and advanced processes, difficult to shake in the short term. Changxin's market cap topping does not mean it has surpassed others yet; process generation gaps and high-end product shortages remain obvious, and the breakthrough path is still long.
But what is certain is that the era of three-way division and ironclad dominance in the global memory industry is over. From the day Changxin officially entered the capital market, the industry gained an unignorable player, and competition in price, technology routes, and market share will comprehensively upgrade.
For investors, don't be dazzled by single-day rises or falls: memory is always a strong cyclical industry, soaring in upcycles and revealing true strength in downcycles. Whether overseas leaders or domestic manufacturers, how far they can go ultimately depends on technology, cost, and the hard strength of customers. AI这波半导体回调的背后，不只是二级市场的洗盘，而是债务市场的风暴预警——甲骨文、SpaceX、Alphabet、亚马逊、Meta、博通的 CDS（信用违约互换，即违约保险）价格，近日均冲上了历史新高！
做交易和研究的朋友注意了：股市看的是 AI 订单增速，但债市已经在拷问“到底是谁在为这波增长买单”。
市场上一次看到类似的资金链运作，还是 2000 年千禧年电信泡沫时期 Lucent（朗讯）采用的“卖方融资（Vendor Financing）”模式。债市大佬们集体下场避险，核心担心这 3 点：
1️⃣ 芯片巨头演变成“兜底银行”：英伟达不再只是卖芯片收现款，而是开始为 OpenAI、SK 集团等合作方提供数百亿美元级别的融资担保与合作额度。本质上变成了“我替你做担保/借钱，你拿钱买我的 GPU”。
2️⃣ 大厂自由现金流开始失血：AI 基础设施建设（CapEx）耗资极其惊人。Alphabet 甚至出现了上市 20 多年来首次单季度自由现金流转负；甲骨文因数百亿级别的数据中心砸钱扩张，标普评级直接被下调至 BBB-（只比垃圾债高一级）。
3️⃣ “循环融资”放大系统风险：如果 AI 上层应用的变现造血能力，追不上底层硬件的高额折旧与债务到期速度，这套自我循环的金融杠杆一旦松动，风险会沿着链条迅速传导。
判断与提醒：
在 AI 故事最红火的时候，往往是固定收益与信用债券市场最先发现水下的暗礁。无论是炒美股、做 Web3 链上宏观对冲，还是关注 AI 产业链的开发者，别只看 EPS 利润表，接下来一定要盯紧大厂的自由现金流（FCF）与 CDS 违约风险溢价。
链条绷太紧时，别急着盲目梭哈，学会看债务指标能帮你躲掉不少大坑。I've been watching $RE for a few days and feel it hasn't bottomed out yet
I've been monitoring RE's market these past two days; it has dropped steadily from around 0.68 at its peak to about 0.44 today. Many people in the group are shouting "bottom fishing, bottom fishing," but honestly, looking at this trend, I'm a bit hesitant to make a move.
Looking at the data, on July 23rd, an analyst posted a report on Gate Plaza saying the coin was around 0.628 at that time. Although it had risen nearly 30% before, indicators showed it was heavily overbought, with the KDJ J value shooting above 100. In such an extreme overbought state, a correction is almost inevitable.
And indeed, the correction came, and so far, there’s no sign of stabilization.
The coin’s fundamentals are actually decent. Re Protocol does on-chain reinsurance, using stablecoins to underwrite real-world insurance business and earn premium income. This sector is quite interesting, with low correlation to the crypto market, and its revenue comes from the real reinsurance market. Coinbase Ventures also invested, indicating that big institutions are optimistic.
But a coin is a coin, and a project is a project. This project had its TGE on June 18th, with a total supply of 1 billion tokens, about 160 million circulated at TGE. The key point is that investors and the team have a 12-month lock-up, with ongoing unlocking pressure ahead.
The current trend gives me the feeling: it hasn’t dropped enough yet.
There might be a short-term rebound since it has fallen so much and shorts need to cover. But if I had to pick a direction, I’d probably wait for it to stabilize before making a move. Bottom fishing like this is risky; entering on the left side can easily get you trapped.
#波动雷达：币种异动观察 ——$RE Market Midday Review | Changxin is not "directly crashing the market," but rather causing the market to start repricing memory chips
Today, the semiconductor sector weakened, with storage chain companies like SK Hynix and Micron under pressure. Many people's first reaction was: Did Changxin's listing scare global memory stocks?
My judgment is: it is related, but not a simple cause-and-effect like "once Changxin listed, Hynix was immediately crushed." More precisely, Changxin's IPO and its subsequent strong performance, combined with market concerns about the improvement of China's memory competitiveness, jointly triggered a global capital repricing of the memory industry. Reuters reported that the current downturn in Asian chip stocks includes the listing of Chinese memory manufacturer CXMT (Changxin), market worries about intensified competition from China, and a cooling of AI trading itself.
First, let's look at Changxin itself. CXMT is already the world's fourth-largest DRAM manufacturer. This time, it raised 5.792 billion RMB in Shanghai, with its stock price soaring over 400% on the first day of listing, quickly becoming a market focus. Reuters also mentioned that its high market value and strong start reinforced investors' expectations that "China's memory power is on the rise."
Why does this affect companies like Hynix and Micron? Because memory chips are not about "who tells the best story to rise," but are a typical cyclical industry. The market's biggest fear is never a single company listing, but that future supply continues to increase while demand growth slows. Reuters noted that analysts focus not on how much CXMT is earning now, but whether it can rapidly expand production in the future to further squeeze global DRAM prices and market share.
This is also the core logic behind today's pressure on Hynix and Micron: capital is starting to trade ahead on "intensified competition" and "valuation re-rating." Especially for products like HBM and DRAM, which are highly related to AI, the market assigned a high growth premium over the past year. Once new competitive variables appear, capital will withdraw first rather than wait for earnings to be disproved. SK Hynix itself previously emphasized that AI-driven memory shortages may last until 2030, but market trading is often faster and more emotional than company guidance.
Therefore, I prefer to define today's decline as a rehearsal for the future competitive landscape of the memory industry, rather than a single-point shock caused by Changxin's listing itself. Changxin did not suddenly change the industry but made the market more clearly see one thing: global DRAM competition may be shifting from "dominated by three" to "a more intense four-player battle." In this case, the valuations of Hynix, Micron, and Samsung may all be re-examined.
But this does not mean there is no opportunity in the memory chain. Reuters also pointed out that AI demand still drives the long-term prosperity of HBM, DRAM, and NAND. The core contradiction in the memory industry remains "strong demand, slow supply, and high technical barriers." In other words, short-term stock prices may fluctuate due to competition expectations, but the long-term logic may not be destroyed.
My conclusion
Changxin's listing is not the sole reason for today's decline, but it is indeed a very important catalyst.
It has made the market seriously consider three questions:
First, will the expansion speed of domestic DRAM be faster?
Second, will global memory prices be pressured as a result?
Third, can the high valuations of leaders like Hynix and Micron continue to be maintained?
So, the insight I gain from this is not "memory chips are no longer buyable," but rather: in the future, when looking at memory chips, we must not only look at AI demand but also at new supply and competitive landscape. Whoever can continuously lead in technology, yield, customers, and capacity is more qualified to survive the cycle.
💬 Do you think this is a short-term emotional valuation kill, or is the memory industry really entering a more intense competition phase? Bitcoin is $63,320, down about 49% from $126,300 in October 2025. On July 23 and 24, spot ETFs saw a combined outflow of over $465 million, ending seven trading days of net inflows.
What makes this round special is that there is no single black swan. Selling pressure comes from ETF redemptions, real interest rates, tech asset deleveraging, regulatory delays, and Strategy's cash flow constraints—all five factors overlapping over the long term.
Without a single piece of bad news, there will never be a day when all bad news is out. Those waiting for a V-shaped reversal will have to wait a long time. This is exactly what the FalconX trader meant.
The next visible point is whether the ETF has had consecutive net outflows for more than five trading days. $ETH $BTC 英伟达未来可能出现的债务违约风险？信用违约掉期（CDS）创纪录飙升!
英伟达（NVDA）今天的大跌主要源于市场对其潜在巨额财务承诺和债务风险的担忧升温。
根据最新的市场消息，引发此次抛售和恐慌的主要原因包括：
巨额AI基础设施投资计划：市场传出英伟达正在洽谈总规模超过7500亿美元的人工智能基础设施合作项目。
为OpenAI提供天价担保：据报道，英伟达正与OpenAI洽谈一项融资安排，计划为金额最高达2500亿美元的美国数据中心项目提供担保，以帮助OpenAI租赁算力资源。这可能成为英伟达历史上规模最大的客户融资合作之一。
与SK海力士的深度绑定：英伟达近期还宣布，与韩国芯片制造商SK海力士母公司合作推进一项总规模超过5000亿美元的AI基础设施计划。
信用违约掉期（CDS）创纪录飙升：随着AI基础设施投资规模的极速扩大，分析人士和投资者开始担忧英伟达未来可能承担的庞大融资和担保义务。这直接导致英伟达的五年期信用违约掉期（CDS）价格在周一创下有记录以来的最大单日涨幅。
美股投资网分析：CDS的走高意味着投资者为了对冲英伟达未来可能出现的债务违约风险，需要支付明显更高的成本。虽然这并不代表英伟达马上会违约，但反映出在AI投资狂潮下，资本市场对其过度扩张和财务杠杆产生了强烈的警惕，从而要求更高的风险补偿，进而引发了股价的剧烈震荡。
$nvda $amd $MU
#美股ETF funding divides, crypto awaiting change
Recently, the crypto market has entered a typical tug-of-war of volatility, with bullish and bearish tug-of-war continuing to intensify. The previously recovering Bitcoin spot ETF saw divergent capital, with several consecutive days of net inflows halting and a single day seeing large net outflows. The split data clearly shows that capital flows are not one-sided; leading products still have funds entering the market, redemption pressure is concentrated on established trust products, and internal market divisions have significantly amplified.
Macro expectations continue to weigh on the market. The market has once again fueled expectations for Federal Reserve interest rates, and the high real interest rate environment suppresses risk asset valuations, putting Bitcoin's "digital gold" narrative to the test repeatedly. BTC repeatedly tested support within key ranges, while mainstream coins like Ethereum and Solana followed the broader market movement. The profit-making effect of altcoins continued to weaken, and the concentration of funds in leading assets became increasingly apparent, with Bitcoin's market cap share steadily rising.
Regulatory tracking remains the long-term core theme. EU MiCA regulation fully implemented, with many non-compliant platforms gradually exiting the market; Hong Kong's stablecoin ecosystem continues to advance, with the development of Hong Kong dollar stablecoins and the RWA tokenization asset sector attracting ongoing institutional attention. The global crypto industry has officially entered a compliance reshuffling cycle, with speculative projects without compliance support experiencing continued liquidity.
Another noteworthy signal on the market: signs of sector rotation are beginning to appear. Some funds have diverted from Bitcoin to invest in Ethereum and mainstream public chain ecosystem ETFs, but a broad rally has yet to form. On-chain data shows that medium- and long-term holdings have not loosened significantly, and short-term fluctuations mostly come from the back-and-forth of derivatives leveraged funds. Frequent liquidations across the internet have further amplified losses caused by chasing gains and selling lows.
The biggest contradiction in the current market is that long-term institutional allocation logic still exists, but there is a lack of strong short-term catalyst and positive effects. Until a new narrative ignites the market, the range-bound consolidation pattern is likely to continue. Traders need to be wary of sudden spikes caused by macro news and approach high-leverage positions with caution.
There are two main points to watch going forward: first, the linkage between US and Treasury yields; Second, whether Bitcoin ETF funds can return to a sustained inflow channel. Only if one of these shows a clear turn is there hope to break the current dull consolidation.
In terms of sector opportunities, the medium- to long-term focus remains on three main areas: compliant stablecoins, tokenized real-world asset RWA, and on-chain institutional services; In the short term, before the trend becomes clear, blindly bet on one-sided movesIn recent weeks, the market has been trading one thing—the escalation of the Middle East conflict. Oil prices have been rising, gold continues to strengthen, and risk aversion is rapidly intensifying, suppressing global risk assets. But just today, as expectations of a ceasefire heated, WTI crude oil plunged 8.68% in a single day, marking one of the rare single-day declines in recent years. Many people think this is a matter for the crude oil market, but I believe what truly deserves attention behind this is the repricing of global capital risk appetite. Why are oil prices falling so quickly? Because the market has never traded about the war itself, but whether the war continues to expand. Previous oil price increases were mostly due to risk premiums such as "possible impact on the Strait of Hormuz," "supply chain disruptions," and "increased transportation costs." Now that a ceasefire is expected, these risks are being quickly eliminated by the market. As risk premiums fall, oil prices naturally fall rapidly. Meanwhile, US Treasury yields have also started to stabilize, demand for US dollar safe-haven assets has cooled, and sentiment in US stock futures has clearly improved. These signals together actually indicate one thing: funds are flowing back into risk assets. For digital assets, this is often more important than the price of oil itself. In recent days, Bitcoin has maintained high-level fluctuations without any obvious panic selling. This shows that real big money hasn't left the market, but is waiting for a new catalyst. Now, as geopolitical risks ease, market attention has shifted back to several core events: first, this week's Federal Reserve interest rate decision. Second, the latest financial reports from major tech companies. Third, U.S. figures$BTC $XRP ..... Đúng như dự đoán trước đó, thị trường đã phản ứng với "tin đồn" FED "có thể" tăng lãi suất vào thứ 5 tới đây.
Nhưng ......
Hãy nhìn vào thực tế, xung đột ở Trung Đông đang hạ nhiệt và "rất có thể" sẽ đạt được 1 thỏa thuận "tốt nhất" dành cho các bên. Giá dầu vẫn tiếp tục giảm, nó phản ánh "đúng" những gì "sắp diễn ra".
FED "sẽ tăng lãi" ??? KHÔNG.
Dưới thời tân chủ tịch, FED dường như không còn đi theo xu hướng cũ. Mặc dù các chỉ số kinh tế vẫn ảnh hưởng đến các quyết định của FED, nhưng sẽ không còn là những yếu tố đặc trưng. FED sẽ "đi theo hướng tương lai" nhiều hơn là "thực tế". Vậy nên, 95% FED vẫn "giữ lãi suất" để hướng đến kết quả tốt hơn thay vì làm rối loạn thị trường.
Clarity Act: Vẫn còn là câu hỏi lớn. KHI NÀO và BAO GIỜ ?
Triển vọng một cuộc bỏ phiếu tại thượng viện vào đầu tháng 8 là rất "khả quan". Sự "nhượng bộ" trong các điều khoản giữa 2 đảng đang đạt được các bước tiến rõ ràng, hướng đến 1 lợi ích chung cho toàn nước Mỹ.
Những "khởi sắc tuyệt vời nhất" sẽ bắt đầu ... $AAVE
Price action is trading around 98.15, holding strong above dynamic MA5 (96.51), MA10 (94.97), and MA20 (94.58).
EP
96.00 - 98.00
TP
102.48
105.00
110.00
SL
93.50
Following a solid recovery off the 82.73 low, price broke higher to print a top at 102.48. Holding above the dynamic MA5 baseline keeps the market positioned for another attempt at the highs.
Let's go $AAVE
#NvidiaBacksOpenAI #NvidiaBacksOpenAI #StorjChapter11 A Tale of Two Extremes! On one side, a new stock frenzy; on the other, a major index plunge—marking a watershed moment for the storage sector
On the same day in the capital markets, two extreme trends collided, leaving observers deeply moved.
China's domestic storage leader Changxin debuted on the A-share market, igniting market sentiment on its first day of listing. The stock surged from the opening bell and closed with an astonishing gain, with its total market value topping the A-share market and daily trading volume setting a new record for new stocks. Countless funds poured into the domestic semiconductor main theme, all betting on the long-term story of domestic substitution in storage chips.
Everyone understands that Changxin's rise is not just a company going public; it signifies a genuine local challenger entering the DRAM market, which has been monopolized by overseas giants for many years.
But turning to the South Korean market, the picture is completely opposite. The KOSPI index plunged over 8% intraday, triggering a circuit breaker at one point.
South Korea's stock market heavily relies on semiconductors, with Samsung and SK Hynix shares under simultaneous pressure. The concern among investors is straightforward: as domestic production capacity continues to ramp up, competition in the global storage market will intensify, and Korean manufacturers, who have held their market share for years, will face strong challengers.
The market dynamics go beyond mere industry competition. Many South Korean retail investors habitually use leverage to speculate in the chip sector, so when the market turns, forced liquidations amplify panic. Coupled with the market reassessing the supply-demand outlook of the storage cycle, many funds choose to exit early to avoid risk.
A very realistic phenomenon has also emerged: even within the storage industry chain, there is a clear divergence in capital allocation. A-share funds are frantically embracing the leading new stock, while many smaller follow-up stocks face profit-taking; overseas funds, however, worry about intensifying competition and continue to sell off established semiconductor companies.
One IPO event has stirred the valuation logic of the global storage sector. The old pattern, where a few companies controlled market supply, is slowly changing. #韩股重挫8%，长鑫首日登顶A股
What do you think? Can this trend of strong and weak divergence in domestic and overseas storage assets continue?#英伟达拟为OpenAI提供2500亿美元担保 If this news finally materializes, I believe the market will see far more than just a round of financing.
It means the AI industry has begun to shift from "technological competition" to "capital competition."
In recent years, everyone has been competing on models, parameters, and computing power.
But now it's becoming increasingly clear that the real competitive threshold has become:
Whoever can continuously obtain funds can keep purchasing computing power; Whoever can continuously purchase computing power has a better chance of training next-generation models.
From this perspective, Nvidia's willingness to provide massive guarantees to OpenAI is essentially not a simple financial partnership, but a deeper integration of the entire AI ecosystem.
NVIDIA needs OpenAI to continuously expand its computing power demand, and OpenAI needs NVIDIA to keep supplying the most advanced GPUs.
The two form a mutually reinforcing flywheel:
More funding → more GPUs → stronger models → more commercialization → more cash flow → more computing power.
Many people believe this is a major positive for Nvidia.
But as a trader, I won't jump to conclusions so quickly.
Because what the market trades is never about the event itself, but about the gap between the event and expectations.
If investors had long expected the two sides to continue deep cooperation, then even after the news was released, the stock price might not continue to rise.
Conversely, if the market starts to worry about overvaluations, rapid capital expenditure, or even a longer AI investment return cycle, then no matter how strong the positive development, it could see a "positive news realization" trend.
Therefore, when analyzing this type of news, I pay more attention to three questions:
* Did the market anticipate this in advance beforehand?
* After the news is released, will funds continue to flow in or be cashed out?
* Does the candlestick confirm this logic?
Over the years, trading has made me increasingly believe in one saying:
What really drives prices is not the news, but how the market interprets the news.
News provides stories, while funding determines price.
And price is the only language the market never lies.
So whenever I face any major news, I always observe how the funds vote before deciding whether to follow suit.
Respect price, not blind information; Respect the market, not stick to your own views.
This is also, in my opinion, the most important underlying logic for long-term trading.#停火预期兑现, WTI crude oil futures fell 8.68% in a single day. The market gave all traders another lesson today:
What truly determines the price is not the news itself, but whether expectations have been fulfilled.
WTI crude oil plunged 8.68% in a single day, and many people's first reaction was: "A ceasefire is bearish for crude oil." ”
But I think that's just the surface.
What truly drove today's sharp drop was not the ceasefire itself, but the market's start to reprice future supply and demand.
During the war, the largest premium on crude oil comes from geopolitical risks.
Once ceasefire expectations strengthen and concerns about supply disruptions ease, risk premiums will naturally be quickly squeezed out.
So the price drop isn't just oil, but risk expectations.
This also reminds me of a frequently overlooked issue in trading:
Is the news more important, or is the expectation more important?
My answer has always been the latter.
The market won't wait for news to materialize before acting; funds always position when expectations are formed and exit when expectations are fulfilled.
So many people notice a phenomenon:
When good news is announced, prices actually fall;
Negative news materialized, but the market started to rise.
It's not that the market is wrong, but that the funds have already completed the trade ahead of time.
That's why I spend less and less time predicting news and more observing prices.
If everyone knows something, it's very likely already reflected in the price.
What truly deserves attention are the expectations in the market that have yet to reach consensus.
Therefore, when I trade now, I prefer to ask myself three questions:
* Has this matter already been traded in advance by the market?
* Does the current price reflect the facts, or are they expected to look ahead?
* If everyone stands on the same side, who will continue to push prices?
By the end of the trade, I increasingly believed one saying:
Prices are not the result of news, but of expected changes.
Macroeconomics, wars, and policies all affect the market, but what truly determines profit or loss is often not what you know, but whether you understand which news the market has already priced in in price.
Respect expectations, respect capital, and finally respect candlestick charts.
This is, in my opinion, the most important trading principle for navigating any news market.#美国暂停预测市场州级禁令
This came quite suddenly.
Minnesota's previous law that classified operating prediction markets as a felony has been directly halted by a federal court. Judge Katherine Menendez issued a preliminary injunction, reasoning that the state law likely violates the federal Commodity Exchange Act (CEA).
Simply put—federal law takes precedence over state law, the CFTC has exclusive regulatory authority over prediction markets, and states cannot create their own separate systems.
This case is actually more than just a win or loss for one state. The CFTC has previously sued multiple states including Arizona, Connecticut, Illinois, New York, and Massachusetts. Michigan even tried to directly demand Kalshi to cancel completed trades, which CFTC Chair Selig characterized as an "unprecedented move" that "could shatter public confidence."
The judge's ruling is quite critical—she stated that prediction market contracts structurally fall under the CFTC-regulated "swap" products category, and the plaintiff "is likely to succeed at trial," and that not issuing the injunction would cause "irreparable harm" to Kalshi and Polymarket.
Minnesota is a major agricultural state where farmers could hedge risks through weather and crop contracts, but the state law directly criminalized such transactions. This created a direct conflict between the CFTC's nationwide regulatory framework and the state-level ban.
The biggest suspense now is whether Kalshi and Polymarket can leverage this precedent to open up the entire U.S. The federal court has already confirmed CFTC's priority jurisdiction in one state, and theoretically, bans in other states can be challenged on the same grounds. The joint opinion letter from the CFTC, HPC, and Multicoin essentially pits federal regulation against state bans, with HPC explicitly advocating "rules over uncertainty," and that prediction markets should be uniformly regulated federally rather than legislated separately by each state.
The federal court's preliminary injunction has opened a new direction in this tug-of-war. As long as the precedent holds, the blockade by states will have to loosen. #韩股重挫8%，长鑫首日登顶A股 The most noteworthy thing today is not the plunge in the South Korean stock market, nor Changxin Technology topping the A-share market, but the global storage industry's power dynamics being redistributed.
The South Korean KOSPI sharply dropped intraday, with heavyweight stocks like SK Hynix and Samsung Electronics falling significantly. On one hand, this was influenced by the global semiconductor sector adjustment; on the other hand, the market began reassessing the impact of China's rising storage industry on the long-term competitiveness of South Korean leaders.
Meanwhile, Changxin Technology's stock price surged over 460% on its first day of listing, with a total market value surpassing ¥3 trillion and trading volume setting a new A-share record, instantly becoming the A-share market's largest company by market cap.
Many might interpret these two events as:
South Korea falls, China wins.
But I believe the market is truly trading not today's profits, but the industrial landscape of the coming years.
The scarcest resource in the AI era is not just GPUs, but also storage.
With model parameters growing larger and inference demands increasing, whether HBM or DRAM, they have become part of AI infrastructure. Capital's willingness to assign Changxin a higher valuation essentially bets on the future growth potential of China's storage industry, while the adjustment in the South Korean market reflects investors' growing concerns about whether traditional advantages will continue to shrink.
However, as a trader, I would not conclude the entire industry is entering a bull market just because one company surged after listing.
The biggest trap in the market is confusing long-term logic with short-term price movements.
Even the best companies can pull back due to overvaluation; even the biggest negative news might have already been priced in.
So my trading logic remains unchanged:
Look at the price first, then find the reason; trust the candlestick, not the story.
Macro trends, industry shifts, policy changes determine the long-term direction; what truly decides whether I trade is whether the capital has already voted with price.
In trading, I increasingly believe this:
The market does not reward those who know the news first, but rewards those who can quickly adjust their understanding when the world changes. Solana never closes. Its liquidity does.
We measured 90 days - 2.77B trades, $857B:
- Peak: Tue–Wed 13:00–17:00 UTC (6–10am PT), up to 132% of avg
- Thinnest: Saturdays and the 23:00 UTC hour (−13%)
- The pattern held through a 60% volume drawdown
Time your entries. Don't trade on vibes.In the same storage sector, completely opposite market trends have emerged on both sides
On July 28, South Korea's KOSPI index plunged 8% triggering a circuit breaker, with SK Hynix falling about 11% and Samsung Electronics dropping over 9%
Meanwhile, on the other side, Changxin Technology surged 471.59% on its first day of trading, with its market value briefly exceeding ¥3.3 trillion, topping the A-share market
This contrast does not mean the global storage landscape was rewritten in a single day; it more likely reflects capital recalculating competition and valuations
The crypto market is also cooling down simultaneously, with BTC dropping to about $63262 and ETH falling to $1874
If Asian tech stocks continue to face pressure, short-term risk sentiment may still impact the crypto market $ETH $BTC
#韩股重挫8%，长鑫首日登顶A股 🚨 Major signal: Ripple CEO Brad Garlinghouse bluntly stated that the CLARITY Act is expected to completely remove the biggest policy barriers to institutional-level adoption of XRP. Once approved, a clearer regulatory framework will significantly boost institutional confidence and accelerate deeper market participation.
But clear regulation has always been a double-edged sword. On the positive side, it opens the door to compliance and attracts massive capital; On the flip side, the industry will face stricter oversight, higher compliance costs, and more intensive audit inspections. This is an unavoidable cost.
📊 The core question is no longer "whether institutions will enter," but rather: after institutions enter, how will XRP's liquidity, price stability, and long-term volatility model be reshaped? This is the underlying variable that will determine XRP's future trajectory.
👀 If the CLARITY Act proceeds smoothly, it is highly likely to become one of the most critical catalysts for XRP in the coming months. Market sentiment has begun to heat up, and fundamental narratives are being reshaped.
NFA，DYOR。 Stay hungry and keep a close eye on progress.
#FOMCRateWatch #CXMTMemoryIPO #AIEarningsWatch$BTC
Price action is trading around 63,324.5, sitting right below dynamic MA5 (64,179.0), MA10 (64,848.6), and MA20 (64,395.3).
EP
62,500.0 - 63,500.0
TP
64,395.3
64,848.6
66,928.0
SL
61,500.0
Price encountered resistance around 66,928.0 and pulled back slightly below the short-term moving average cluster. Holding above local support near 62,000 keeps a relief attempt back above 64,800 intact.
Let's go $BTC
#CeasefireHitsCrude #AIEarningsWatch #CXMTDebutShockwave Macroeconomic Liquidity + Jensen Huang's Entry into AI: What's the Logic Behind It?
1. The Real Impact of the Macroeconomic Environment on ETH
The biggest recent market variable is the sudden intensification of the Federal Reserve's rate hike expectations. In just two weeks, the probability of a July rate hike surged from 13% to 36.3%, meaning the market suddenly started pricing in the risk of "liquidity tightening."
ETH is a typical high-volatility risk asset, very sensitive to US dollar interest rates and US Treasury yields. Rising rate expectations mean higher risk-free returns in the market, so funds are unwilling to stay in high-risk markets like crypto and stocks, causing overall incremental capital to shrink.
Although ETH's fundamentals are not bad, with nearly 28% of the total network staked and locked, and exchange reserves continuously decreasing, making tokens increasingly scarce, macro sentiment outweighs short-term fundamentals. Before the Fed's decision lands, funds dare not enter aggressively, so ETH can only passively follow the broader market's oscillation and pressure.
2. The Real Underlying Logic Behind Jensen Huang Joining Twitter
Many think Jensen Huang joined Twitter to ride the hype, but it's actually the opposite—this is a very precise business strategy.
NVIDIA is the absolute leader in global AI computing power. What it most wants to see is not just one AI company making money, but the entire AI industry flourishing.
Recently, he publicly supported open-source AI, with the core purpose of lowering the barriers to AI entrepreneurship. After open-source models become widespread, countless small and medium enterprises and developers will deploy locally and fine-tune models, all of which ultimately require GPU computing power support.
Simply put: the more popular open-source becomes, the more graphics cards NVIDIA sells.
His joining Twitter is to control the narrative, guide industry trends, influence regulatory attitudes, and secure the long-term basic demand for computing power.
3. Impact on the AI Sector and Crypto AI Track
This wave of positive news has indeed revitalized the entire AI sector, especially targets related to computing power, storage, and AI infrastructure, which have collectively rebounded.
The logic is straightforward: open-source AI explosion → surge in computing power demand → persistent tightness in HBM and storage demand.
Transmitted to the crypto market, narratives like AI Agents, automated settlements, and on-chain AI interactions are being picked up again by capital, which is a long-term positive for the ETH ecosystem and AI concept tokens.
4. Key Point: Why This AI Wave Is Only an Expectation-Driven Market, Not a Trend Market?
Many wonder why, despite such strong logic, the price doesn't rise strongly or sustain.
First, the macro liquidity environment does not support a major bull market.
The Fed's rate hike expectations are still uncertain, and the market is overall in a low-volume wait-and-see mode, with no continuous incremental funds. Without a steady inflow of money, even the best logic can only drive a short-term rebound, not a sustained main rise.
Second, it is all expectation speculation with no actual performance realization yet.
Jensen Huang's open-source AI logic is an industry trend for the next six months to a year, but currently, there is no actual revenue or performance fulfillment. Markets driven by imagination and expectations are purely expectation-driven. Once sentiment cools and news fades, funds immediately exit.
Third, the sector is extremely fragmented, not a broad resonance.
Targets with real computing power and storage businesses are resilient, while most AI coins riding the concept are purely one-day wonders.
Only localized clusters exist without an overall broad rise, which is a typical feature of expectation rotation.
In summary: The long-term AI logic is sound, but short-term it is suppressed by macro factors and can only experience expectation-driven rebounds. After the Fed's decision on Thursday, the market will truly choose the next trend direction. #美联储周四凌晨公布利率决议 #美国禁止开源AI的预期大幅回落 #Korean stocks plunge 8%, Changxin tops A-shares on debut
A new king rises in China, the old king collapses in South Korea. In 48 hours, the global pricing system for memory chips has been completely rewritten.
🇨🇳 On the China side: Changxin Technology, the new king of A-shares
On July 27, domestic DRAM leader Changxin Technology debuted on the STAR Market. Issue price was ¥8.66 per share, opening at ¥49.5, soaring 471.59%. The intraday high reached ¥55.03, closing at ¥49. Market cap reached ¥3.28 trillion, surpassing Industrial and Commercial Bank of China to become the top A-share.
The total turnover for the day was ¥141.187 billion, the first A-share stock in history to exceed ¥100 billion in single-day trading volume. Turnover rate exceeded 66%.
Another intriguing detail: before listing, the on-chain pre-market contract pricing was about 5.4 times the issue price. The closing price on the first day was 5.66 times the issue price. The on-chain pricing and reality almost perfectly matched, a precision worth noting.
🇰🇷 On the South Korea side: circuit breakers, price drops, double whammy
The day after Changxin’s listing, July 28, the Korean KOSPI index plunged over 8%, triggering a circuit breaker. This was the eighth full-market circuit breaker this year.
SK Hynix fell over 11%, Samsung Electronics dropped over 9%. The Nikkei 225 also fell over 4%.
Even worse, SK Hynix’s US ADR fell as much as 10% to $139.01 on July 27, closing at $143.02, below its IPO price of $149 on July 9. It broke below IPO price in less than three weeks.
One weekend, two markets, two different outcomes.
🤔 Why now?
First, Changxin is no longer a "follower" but a "price setter." With 8% global DRAM share, it has completed in 10 years what Micron took 30 years to do. The ¥3.28 trillion market cap implies expectations that it is moving from global fourth place into the top tier.
Second, the valuation premium of the two Korean giants now has a clear challenger. Samsung 36%, SK Hynix 29%, Micron 24%, Changxin 8%. Previously three players split the market; now four are competing. Every 1% share Changxin gains is taken from others.
Third, the AI narrative itself is loosening. Google and Tesla earnings have proven the market is starting to question "when will the money burn turn into profit?" Memory chips are the most sensitive link in the AI hardware chain. When the belief that "AI is never enough" begins to waver, the first to be hit are the stocks that rose the most early on.
💎 What does this mean for the crypto market?
Memory chips and the crypto market share the same macro backdrop—AI narrative, liquidity expectations, risk appetite. Changxin’s listing itself is positive, but the chain reaction it triggered is that global capital is reassessing the valuation of the entire AI hardware chain. KOSPI circuit breaker, SK Hynix breaking IPO price, semiconductor index down 2.2%—these signals combined indicate declining risk appetite. As institutions withdraw from overvalued semiconductors, crypto assets will find it hard to remain unaffected.
The memory chip playing field has been reshuffled. It used to be three players missing one; now four players are at the table. Whoever falters first will be the first out.
$EWY $SKHYNIX $SAMSUNG $ACH What is the next step for the dog farm?
Short-term (pre-FOMC): Prices are likely to fluctuate within the 0.0040-0.0050 range. The July 29 FOMC meeting is the biggest variable—if it leans hawkish, small-cap counterfeit ACH will fall harder than anyone else.
The last two FOMC scenarios:
· Scenario 1 (dovish / rate maintained): ACH may break through 0.0050, targeting 0.0055-0.0060.
· Scenario 2 (Hawkish / Rising Rate Hike Expectations): ACH is highly likely to fall below 0.0040, or even 0.0035-0.0037.
Mid-term: The biggest variable is whether compliance expansion can convert into real users and revenue. Alchemy Pay's fundamentals are indeed improving—Mastercard endorsement, 18 state licenses, 173 countries coverage—but ACH has dropped 97% from its peak. Whether this is a rebound or a reversal depends on whether the payment network can truly generate economies of scale. Some analysts have pointed out: "What will truly drive large-scale adoption in the crypto industry in the future is not just faster blockchains, but more convenient fiat currency deposits and withdrawals and improved global compliance capabilities." ”
The final heartfelt words:
ACH today is $0.0043, Mastercard, 18 state licenses, 173 countries—the fundamentals are indeed holding back a big move. But the price difference across different exchanges is several times higher, the price has dropped 97% from its peak, and the FOMC meeting is imminent—all three major mines are right there. For those chasing the highs now, think about whether you can withstand the sudden 20% drop from the dog farm. Stop the action, wait until the FOMC boots fall on July 29, and wait until the direction is clearer before taking action. Remember, staying long in crypto is ten thousand times more important than making a lot of money! Meeting adjourned!The Korean stock market hit another circuit breaker today! It's only been seven months since this year, and this is already the eighth circuit breaker...... Korean retail investors really have big hearts.
Many people think it's because the Korean economy is failing, but that's not the case. The essence of this crash is: global AI bubble sentiment + foreign ATMs + collective crushing by Korean retail investors' high leverage.
💡 Let's talk about a few plain truths:
1️⃣ The index is held hostage by the "Twin Stars": Chip giants Samsung and SK Hynix account for more than half of the market. Overnight, when US tech stocks fell, foreign investors immediately started dumping on these two stocks, dragging the market indices down in an instant.
2️⃣ Korean retail investors are bold in gambling: Young Koreans, in order to change social classes, are extremely fond of speculating on 2x/3x leveraged ETFs and chasing hot topics (AI, batteries). When prices rise, it's a perpetual motion machine; once a pullback triggers strong liquidation, it turns into a vicious cycle of "automatic sell-off."
3️⃣ Foreign ATM: Korean stocks have excellent liquidity; whenever there is a global disturbance (such as Fed rate cut expectations or geopolitical risks), foreign institutions are the first to sell Korean stocks to cash out and hedge risks.
📌 Summary and insights:
This year, the Korean stock market is far from a normal bull-bear market, but rather a "high-leverage + high-concentration" sentiment amplifier. When prices rise, it's exhilarating; when prices fall, liquidity is instantly shattered.
Beware of those inflated themes that rely on retail investor sentiment and leverage. When the tide recedes, only the core leaders with HBM chip orders can survive. Small concept stocks entering are buying stocks.Today's crash is equivalent to every South Korean losing 13%, not including leverage.
The South Korean KOSPI index fell 8% today and triggered a circuit breaker, officially breaking below the key support line of 6500 previously given by Goldman Sachs. SK Hynix dropped more than 12%. According to data from South Korean brokerage Naver, the current average holding price for retail investors in SK Hynix is 1,815,000 KRW, meaning every South Korean has lost 13% on Hynix.
On the surface, this crash is a chain reaction of capital stampede, but the core reason is that the global semiconductor pricing power is being reshaped.
First, domestic chips are accelerating their momentum. ChangXin Technology successfully went public and secured substantial financing, coupled with breakthroughs in domestic lithography machines, China will inevitably expand production and improve self-sufficiency. The memory duopoly can no longer monopolize the market, so expected profits are naturally discounted.
Second, the market is beginning to worry that big companies won’t recoup their investments. Previously, giants like Nvidia poured money into building data centers, which boosted the entire semiconductor sector. Now investors are concerned about how long it will take to recover these huge investments, so they choose to take profits at high levels.
In short, the big companies that used to make money through monopoly are now facing competitors and have spent too much on equipment. Going forward, it depends on the real performance of each major company, relying on their own capabilities.
#韩股重挫8%，长鑫首日登顶A股
@OKX中文 @OKX星球 1) Today's Market Overview Today's market did not fully strengthen but rather resembles a repricing after the weekend's risk appetite was released. BTC has now returned to around $63,300, down about 3% from the previous trading day; ETH is around $1625, and DOGE has also pulled back about 3.7%. With the FOMC meeting approaching, short-term funds have become noticeably more cautious, and in recent days, sentiment chasing highly elastic assets has begun to cool. However, the market has not fully shifted to safe-haven mode. In the previous U.S. stock trading day, Strategy rose about 7.6%, Coinbase gained about 5.9%, and Robinhood closed slightly higher. Coin prices pulled back, but crypto concept stocks remained relatively strong, indicating that the stock market is still trading ahead of expectations of industry revenue recovery, institutional capital returning, and regulatory environment improvement. 2) Popular token movements: The most intense community discussion these past two days has still been on the MEME sector, but today it has shifted from collective chasing to a clear divergence. SHIB had previously surged continuously, with trading volume once expanding from about $50 million to over $500 million; Currently, the price has pulled down about 13% in a single day, but still rose about 9% over seven days, and the 24-hour trading volume has also decreased by about 57% compared to the previous day. This indicates that the hype hasn't completely disappeared; it's just that the most intense chasing phase has passed, and funds are starting to cash out. MemeCore also failed to maintain its previous strength, falling about 3% in the past 24 hours, with a turnover of approximately $10.7 million, down about 11% from the previous day. It still has a high market cap and community recognition, but today's data seems more like a cooling of high hype,$ALLO 狗庄下一步怎么割？
短期：价格大概率在0.35-0.42区间震荡。有分析直言：“这波拉升找不到明确催化，更像是低流通盘被资金推着走，缺乏基本面支撑，情绪一冷就容易反向”。
中期：最大变量是代币解锁和DeAI赛道热度。Allora基本面确实硬——Polychain等机构3500万美元投资、沃达丰合作、Forge平台——但流通盘小、解锁压力大是硬伤。ALLO历史高点约0.92美元——从高点跌了60%以上。这波是反弹还是反转，得看解锁压力消化和DeAI赛道热度能不能持续。
最后一句掏心窝的话：
ALLO今天0.35美元，正好卡在多头最后防线。Forge平台、沃达丰合作、DeAI叙事——基本面确实硬。但7月11日刚解锁1725万枚、流通盘小深度差、FOMC在即——三颗雷全摆在那。0.35这位置，多头怕跌破继续寻底，空头怕狗庄借利好拉盘。有分析说得透彻：“别把小市值币的波动当趋势来做”。现在抄底的老铁，想想自己能不能扛住狗庄突然砸到0.32。管住手，等0.35确认支撑、等解锁压力消化、等方向明朗再动手。记住，在币圈活得久，比赚得多重要一万倍！散会！Last night, US tech stocks continued to drop valuations, looking quite alarming. Many opened high at three or four points, plunging deep into the waters, and then dropped more than ten points.
This is the case with US stocks. When they fall, they don't hold back, but there is always a limit. A single wave of 20 or 30 points can basically hold their ground. A few that have dropped sharply, like SanDisk, clearly saw major capital stop losses today, with two stop-loss orders close to 500 million at the bottom. However, many tech stocks have stopped hitting new lows, especially the Big Seven companies, which have mostly rebounded from their lows, with some even hitting new highs.
Don't guess the bottom—just walk out of it. At least the Dow will still be bright red
$SNDK SK Hynix experienced a flash crash of over 10% today, with the semiconductor sector continuing to face adjustment pressure. $SKHYNIX
On July 28, 2026, the semiconductor sector in the South Korean stock market suffered another heavy blow. SK Hynix's stock price plunged more than 10% intraday on the Seoul market, while Samsung Electronics also fell over 8%, dragging down the overall tech stocks. This volatility continues the intense fluctuations seen in the Korean stock market and the global memory/semiconductor sector since July.
Market performance overview today
SK Hynix**: Intraday plunge of over 10%, continuing the recent correction trend.
Samsung Electronics**: Simultaneously dropped more than 8%.
Other AI-related stocks such as Samsung SDI and LG Innotek also saw significant declines.
This trend is directly related to the weakness in the US semiconductor sector in the previous trading day, showing a high correlation between Asian tech stocks and US AI trade.
As a leading global supplier of high bandwidth memory (HBM), SK Hynix's stock price is highly sensitive to AI server demand expectations. This decline reflects market concerns about a slowdown in hyperscaler (large-scale cloud providers) spending or valuation adjustments.
Background since July: "Deleveraging" and valuation reappraisal driven by multiple factors
SK Hynix is not an isolated case but part of the ongoing adjustment in the Korean stock market and global memory sector since July:
Leveraged ETFs amplifying volatility**: Korean regulators previously approved several double-leveraged single-stock ETFs targeting Samsung and SK Hynix, attracting massive retail capital inflows. The leverage rebalancing mechanism caused severe price swings, triggering multiple KOSPI circuit breakers, with daily declines often reaching 8-15%.
Valuation correction and arbitrage pressure**: SK Hynix briefly surpassed Samsung as South Korea's largest market cap in June, but its stock price sharply retreated from the peak (significant cumulative monthly decline). After Nasdaq ADR listing, a premium existed between Seoul shares and US depositary receipts, increasing arbitrage selling pressure.
Mixed fundamental signals**: On one hand, AI memory demand is promising long-term (SK Hynix CEO warned of supply tightness in 2027); on the other hand, short-term profit forecast downgrades, capacity adjustment rumors, and global discussions about AI infrastructure oversupply (e.g., Meta-related news) triggered profit-taking.
Macro and geopolitical factors**: The "triple threat" of high oil prices, high interest rates, and a strong dollar, combined with Middle East tensions, also affected risk sentiment.
The KOSPI index in South Korea has fallen sharply cumulatively since July, triggering multiple circuit breakers, with SK Hynix and Samsung as heavyweight stocks being the main drags.
Market interpretation and outlook
This "flash crash" is more a result of leverage reduction, profit-taking, and sentiment transmission rather than a comprehensive deterioration of fundamentals. The storage industry remains in a high-cycle phase, and the tight HBM supply-demand pattern is unlikely to change in the mid-to-long term, but high valuations and leverage amplification cause severe short-term volatility.
Investor focus shifts to:
Upcoming SK Hynix earnings report (with prior expectation adjustments).
Global AI capital expenditure trends.
Further regulatory tightening on leveraged products (South Korea has raised margin requirements).
Summary: Today's SK Hynix flash crash is the latest chapter in the semiconductor sector adjustment in July, reflecting the AI theme's transition from extreme overcrowding to risk release. Short-term volatility may continue, but the long-term logic still depends on actual AI implementation and supply-demand dynamics. Market participants should be cautious of leverage risks and monitor fundamental signal validation.
(Data based on public market reports; investing involves risks; this article is for reference only.)