Orbit Post Sitemap
The S&P 500 is getting remarkably close to 8,000
The S&P 500 closed at a fresh record on August 13, 2026, reaching 7,798.99, up 0.65% on the day.
That leaves the index only about 201 points, or 2.6%, below 8,000.
The bigger story isn't simply the round-number target.
The index has already gained 13.9% year-to-date, according to the latest market data, while the Nasdaq is up 15.3% and the Russell 2000 is up roughly 23%.
The latest push higher also came alongside softer inflation data.
July PPI, released on August 13, was flat month-over-month and increased 4.7% YoY, below the previous month's 5.5% reading. That helped ease concerns about another near-term rate increase and supported equities.
But 8,000 is still a psychological level, not a fundamental valuation target.
The important question is whether earnings and economic growth can continue catching up with the market's expectations.
A market trading at record highs can keep going higher, but the margin for disappointment becomes smaller.
For now, the setup is clear:
7,800 → record territory.
8,000 → psychological milestone.
Above 8,000 → markets will need fresh earnings and macro support to justify the next leg.
I'm watching breadth, earnings revisions, Treasury yields and upcoming inflation data more closely than the number 8,000 itself.
$SOL $BTC $ETH
#DailyOrbit SanDisk just gave investors a much bigger long-term AI-storage story
SanDisk ($SNDK) made a major announcement at its Investor Day on August 13, 2026, and the numbers immediately caught the market's attention.
The company expects mid-to-high-teens annual revenue growth from fiscal 2028 through fiscal 2030.
But revenue growth wasn't the only headline.
SanDisk is targeting approximately 80% adjusted gross margins and around 75% adjusted operating margins over that period.
Those are extremely ambitious numbers for a memory business traditionally known for strong cyclicality.
The thesis is increasingly tied to AI infrastructure.
AI data centers require enormous amounts of storage, and demand for NAND memory is expanding as AI workloads become more storage-intensive.
SanDisk is also trying to reduce the traditional boom-and-bust nature of memory through longer-term customer agreements. The company already has eight customers under its new business model agreements, with those contracts expected to cover roughly two-thirds of its memory capacity by FY2028.
The market reacted immediately.
On August 13, SNDK shares jumped more than 15% during trading as investors digested the long-term framework.
But this is where I think investors need to separate AI demand from stock expectations.
The long-term opportunity looks strong, but an 80% gross-margin target and mid-to-high-teens revenue growth are forecasts, not guarantees.
Memory remains cyclical, supply can change quickly, and a lot of optimism is already reflected in the stock.
So the real question for $SNDK isn't whether AI needs more storage.
It clearly does.
The question is whether SanDisk can turn that demand into sustainable margins, predictable contracts and strong free cash flow through 2030.
$SNDK $BTC $FIL
#DailyOrbit 🔥$SNDK is crazy again: after a two-day surge, I'm actually not chasing it!! Stay rational! Don't forget the painful July!!!!
SNDK has now surged to around $1617, with an intraday high of $1666.
Investor Day delivered exactly what the market wanted to hear:
80% gross margin
75% operating profit margin
50% free cash flow margin
More importantly, SNDK has signed long-term NBM agreements with 8 customers, expected to cover about 2/3 of shipments by FY2028.
So now the hype is not just about NAND price increases, but:
The market is starting to believe SNDK's high profits may last longer.
💰 My buying points
1570–1590: 30%
1520–1550: 40%
1470–1500: 30%
I won’t chase above 1610 now.
🔥 How to view the above?
1660–1700: core resistance zone
If it breaks and holds above 1700 with volume, I will continue to watch:
1750 → 1800 dollars
But if it fails to break 1700 and falls back to 1550, I’m more willing to buy in.
My judgment:
The logic for SNDK is stronger than two days ago, but the stock price has already run ahead.
The most comfortable strategy now is not to chase the rise, but:
Follow the breakout at 1700, buy on the pullback near 1550.
I’ll wait in the middle range.
#SNDK #AIStorage #NAND昨晚真的是被狠狠打脸了，我还在茶馆里跟伙计们吹牛说SanDisk这波高位要回调，转头就挂了个空单。结果呢？美盘直接一根大阳线冲天，最高干到+17%，收盘还稳稳站在13.67%的涨幅上。我现在的心情，就像抄底抄在半山腰的兄弟，看着账户发呆，嘴里只剩一句：这熟悉的味道，是韭菜的芳香。 说实话，存储板块这波集体暴动，不是没由头的。SanDisk在投资者日上掏出了一沓厚厚的长期指引，直接把市场情绪给点燃了。人家说了啥？三年后的营收要维持两位数高增长，长期毛利率目标干到80%，自由现金流利润率奔着50%去。最狠的是第三条，承诺产能投资搞完后，剩下的现金流全部分给股东，回购加分红，一样不少。这一下就把市场最担心的“赚钱之后乱扩产”的恐慌给打掉了。还有那个AI的故事，人家说接下来AI的重心要从训练切到推理，闪存的需求要大爆发，连HBF高带宽闪存的技术路线都给你画好了。这一套组合拳下来，做空的兄弟们都傻眼了。 再说说市场这个大赌场的底层逻辑。存储股前面跌得太惨了，财报出来后一堆人喊着周期见顶，空头攒了一大堆，跌过头了。这时候好消息一来，空头们就像挤公交一样扎堆往回买，踩踏式回补，直接把涨幅给推高了。再SanDisk just gave investors a much bigger long-term AI-storage story
SanDisk ($SNDK) made a major announcement at its Investor Day on August 13, 2026, and the numbers immediately caught the market's attention.
The company expects mid-to-high-teens annual revenue growth from fiscal 2028 through fiscal 2030.
But revenue growth wasn't the only headline.
SanDisk is targeting approximately 80% adjusted gross margins and around 75% adjusted operating margins over that period.
Those are extremely ambitious numbers for a memory business traditionally known for strong cyclicality.
The thesis is increasingly tied to AI infrastructure.
AI data centers require enormous amounts of storage, and demand for NAND memory is expanding as AI workloads become more storage-intensive.
SanDisk is also trying to reduce the traditional boom-and-bust nature of memory through longer-term customer agreements. The company already has eight customers under its new business model agreements, with those contracts expected to cover roughly two-thirds of its memory capacity by FY2028.
The market reacted immediately.
On August 13, SNDK shares jumped more than 15% during trading as investors digested the long-term framework.
But this is where I think investors need to separate AI demand from stock expectations.
The long-term opportunity looks strong, but an 80% gross-margin target and mid-to-high-teens revenue growth are forecasts, not guarantees.
Memory remains cyclical, supply can change quickly, and a lot of optimism is already reflected in the stock.
So the real question for $SNDK isn't whether AI needs more storage.
It clearly does.
The question is whether SanDisk can turn that demand into sustainable margins, predictable contracts and strong free cash flow through 2030.
That's the part worth watching.
#SandiskInvestorDayRally $SNDK $BTC $FIL #OKXTraderVoices #OKXOrbitTopics $SNDK has been skyrocketing these past two days, driving people crazy, damn it,
right after the investor day ended, it surged, a two-day rally. The company told the story so fully, the hype was just perfect.
From 2028 to 2030, revenue is expected to grow in the high double digits, gross margin is locked tightly at 80%, operating margin at 75%, and free cash flow margin directly hitting 50%. All extra money will be returned to shareholders. Sounds like bragging, but the market buys into it.
What really crushed the shorts were those long-term agreements, already signed multi-year contracts with 8 clients, covering half of shipments in 2027 and nearly two-thirds in 2028. There are floor prices and guarantees. It’s like locking in part of the revenue in advance. Suddenly the market feels: this thing really doesn’t want to be a cyclical stock anymore.
Shorts got collectively blown out. There were liquidations of two to three million USD worth of short positions within half an hour, the largest single one on Binance blew up over two million. Those who previously shouted “too high, short it” have basically gone silent these days.
Many got trapped too. From over 2,000 in July down to just above 1,000, many couldn’t hold and cut losses. Now it’s pulled back near 1,500, some are cursing themselves in chat groups, some are secretly adding positions, and some are still asking “can I get in now” even after such a rise.
There’s a saying that’s pretty accurate: those who believed bought at 1000-1100, those asking at 1500 probably didn’t believe before. Now those chasing are basically betting the sentiment can push one more wave.
This rally is about expectations, not actual performance. Whether the long-term contracts can really suppress the cycle, whether the 80% gross margin can hold for years, we’ll see in the next few quarters. As long as sentiment is there, volatility will be high, and those chasing highs better think carefully about how much drawdown they can endure. The S&P 500 is getting remarkably close to 8,000
The S&P 500 closed at a fresh record on August 13, 2026, reaching 7,798.99, up 0.65% on the day.
That leaves the index only about 201 points, or 2.6%, below 8,000.
The bigger story isn't simply the round-number target.
The index has already gained 13.9% year-to-date, according to the latest market data, while the Nasdaq is up 15.3% and the Russell 2000 is up roughly 23%.
The latest push higher also came alongside softer inflation data.
July PPI, released on August 13, was flat month-over-month and increased 4.7% YoY, below the previous month's 5.5% reading. That helped ease concerns about another near-term rate increase and supported equities.
But 8,000 is still a psychological level, not a fundamental valuation target.
The important question is whether earnings and economic growth can continue catching up with the market's expectations.
A market trading at record highs can keep going higher, but the margin for disappointment becomes smaller.
For now, the setup is clear:
7,800 → record territory.
8,000 → psychological milestone.
Above 8,000 → markets will need fresh earnings and macro support to justify the next leg.
I'm watching breadth, earnings revisions, Treasury yields and upcoming inflation data more closely than the number 8,000 itself.
#SP500Nears8000 $SOL $BTC $ETH
#OKXOrbitTopics #OKXTraderVoices $SNDK SanDisk｜Surge Analysis: Is the rally driven by speculation on expectations or by early discounting of future valuations?
On August 13 Investor Day, SanDisk unveiled its FY2028-2030 long-term targets: mid-to-high double-digit revenue growth, 80% gross margin, 75% operating margin, and 100% excess cash returned to shareholders after business investments.
Market reaction: The stock surged 13.7% that day, then stabilized above $1600, initiating a market value re-rating.
Core market divergence
Is this rally driven by trading on AI storage plus high cash return expectations, or has the valuation already priced in the fulfillment of long-term targets?
✅ Partially priced-in logic
1. AI inference drives incremental NAND storage growth; AI storage growth logic is recognized by capital
2. 100% excess cash flow return attracts institutional funds, enhancing holding safety margin
3. NBM long-term contracts weaken the cycle, gaining some premium, but actual effect has not fully maximized valuation
⚠️ Not fully priced-in parts
The 80% high gross margin and others are long-term targets, not current performance. The market is willing to believe the story but does not assign full valuation based on perfect outcomes. Future fulfillment will continue to lift valuation; underperformance will cause premium retraction.
Summary
This rally mainly speculates on AI storage and shareholder return expectations; the high profit margin for 2028-2030 has not been fully discounted.
Risk warning: Long-term targets carry uncertainty, and peer capacity expansion may bring potential pressure. (For review and reflection only, not investment advice) #闪迪投资者日后股价大涨，长期目标待验证 Heima (HEI) +15.18% to $0.141 in 24h, clearly outperforming the flat-to-down market.Main driver:
Social media hype from influencers. FussLeah claimed HEI rose over 30% after their post (promoting a paid group), while others like Marry__255 called for a breakout. This sparked retail inflows and volume exploded +356% to $58.6M.No fundamental catalyst:
No news, partnerships, or project updates. Moved independently of Bitcoin. Extremely high turnover ratio (5.09) confirms thin, retail-driven trading.Short-term outlook:
Hold above $0.12 support → possible challenge of $0.15–0.16 resistance.
Break below $0.12 → quick reversal risk toward $0.10. Sustainability depends entirely on continued social momentum.Classic influencer-driven retail pump. Positive but highly fragile.Not financial advice. High volatility and reversal risk. Only risk what you can afford to lose. DYOR.
#HEI $HEIGold at $4380 $XAU, are you chasing it?
Global central banks net purchased 289 tons of gold in Q2, setting a historical record for the same period. China bought another 20 tons in July, with reserves exceeding 2340 tons. Poland, Turkey, India—all are buying.
First glance: a rise followed by a pullback, bulls and bears battling.
Today, gold rebounded strongly from around 4310 to 4380, with a slight increase over 24 hours, about 3% up in the last 5 days, and over 8% gain on the monthly chart. After crashing from the year’s ATH of 5500-5600, it has rebounded from 4000 to 4450 in July-August, currently undergoing profit-taking correction. A shooting star formed near 4450, volume shrinks on the pullback, indicating a need for consolidation.
First thing: If you think gold is rising because of "chaotic times," you’re wrong.
Global central banks net purchased 289 tons of gold in Q2, setting a historical record for the same period. China bought another 20 tons in July, with reserves exceeding 2340 tons. Poland, Turkey, India—all are buying.
After the new Fed chair took office with a hawkish stance, once CPI and PPI data came out, the market’s expectation for a September rate hike dropped directly to 35%. The US dollar index retreated from highs, and the 10-year US Treasury yield fell accordingly. Gold immediately rebounded.
Second thing: If you think gold is a "safe haven," you’re wrong; it’s the "US dollar trust crisis index."
Gold generates no interest, and holding it incurs opportunity cost. Why do central banks buy it even at the cost of losing interest?
Because the US dollar settlement system is being challenged by the de-dollarization wave. China, Russia, BRICS countries—all are seeking alternatives. Gold is the only reserve asset without "counterparty risk."
Third thing: A technical signal has appeared that must be taken seriously.
The upward channel from the July low remains intact, volume shrinks on the pullback, RSI has moved from overbought back to neutral-bullish, MACD momentum slows but remains in positive territory. This is a typical "healthy correction after overbought."
Key levels
Resistance above: 4430-4450 → 4500 → historical highs
Support below: 4310-4330 → 4287 (channel lower boundary) → 4155-4200
Trading strategy
Short-term traders:
Buy on a pullback to 4310-4330 with stop loss at 4250, target 4430-4450. Partial profit-taking near the high at 4450.
Swing traders:
Break above 4450 with volume to chase longs, target 4500-4600, stop loss below 4400.
Shorts/Hedgers:
Light short positions only if price clearly breaks below 4287 with volume, target 4155, stop loss above 4330.
Long-term believers:
Dollar-cost average in the 4000-4300 range. Central banks are buying, will you follow?Reasons to be bullish on OKB (upward drivers)
1. Major changes in the tokenomics model
One-time burn, total supply permanently fixed at 21 million, minting rights canceled, extremely scarce supply side, aligned with Bitcoin's total supply narrative. This is the core logic behind the current hype.
Value is shifting from exchange fee rights to being the sole Gas token of the X-Layer public chain, binding OKB to the Layer2 public chain ecosystem, attempting to avoid the US Howey Test securities classification risk.
2. Strong fundamentals of OKX
Global trading volume remains among the top, Web3 wallet is a significant differentiator; ICE institutional investment, US business is progressing, market speculation on the "going to the US" narrative, with OKB becoming the only hype target in this story.
3. Narrative expectations
If the X-Layer ecosystem takes off and RWA (real-world assets) are implemented, OKB will have real on-chain consumption; if OKX achieves breakthroughs in the US business, it will bring emotional speculative space. $OKB $BTC $ETH #加密估值转向收入，BTC如何定价？ Many people only focus on the price of BTC, but recently I've been more interested in watching one thing: stablecoins.
The reason is simple. Where $BTC rises indicates what has happened in the market, but where USDT and USDC flow often indicates what the market is preparing to do next. Especially when the market hasn't shown a clear direction, stablecoins make it easier to see whether funds really want to enter or are just watching from the sidelines compared to candlestick charts.
For example, in a market cycle, if BTC rises but the purchasing power of stablecoins on exchanges doesn't significantly increase, and the overall demand for USDT and USDC doesn't expand accordingly, I tend to be a bit cautious about such a rise. Because the price might just be driven by ETF buying, short covering, or existing funds pushing it up; there isn't a large influx of new ammunition within Crypto. In this kind of market, BTC itself can be strong, but the widespread rally you expect in $SOL, DOGE, and a bunch of altcoins may not happen.
Conversely, if BTC hasn't moved much but stablecoins start showing obvious changes, I become more interested.
Because stablecoins in Crypto are somewhat like chips already exchanged at the casino entrance. When money moves from bank accounts into USDT or USDC, it doesn't necessarily mean it will immediately buy BTC, but at least it shows that this portion of funds has completed the preparation to enter the market. It can go to BTC, to SOL, or suddenly rush into some Meme. Once real risk appetite arises, the movement speed of these funds is very fast.
This is also why I think just watching inflows into $BTC and $ETH is no longer enough.
ETF money and native Crypto funds are fundamentally different. One hundred million dollars buying BTC through an ETF might just quietly sit there after purchase; one hundred million USDT entering trading platforms might first buy BTC, then switch to SOL after profits, then from SOL to DOGE, and finally rush into some new hotspot. The same one hundred million dollars creates completely different wealth effects on the entire market.
So if $BTC suddenly breaks out in the future, my first reaction won't be to chase immediately.
I want to confirm what kind of rise it really is: whether institutions are slowly allocating BTC as digital gold, or whether stablecoin funds, spot trading volume, and high Beta assets are all becoming active simultaneously.
The former market might belong to BTC alone.
The latter market is more like the entire Crypto market's liquidity really starting to rise.
Many people guess daily where BTC's next candlestick will go, but sometimes the money has already told you the answer in advance.
Price is the result; liquidity is the fuel.
What’s truly exciting is never just that $BTC has risen, but that when BTC rises, more and more money outside the market is converting into USDT and USDC, lining up to come in.
#BTC #Bitcoin #USDT #USDC #SOL #DOGE #Stablecoins #Crypto #Bitcoin #OKXPlanetETH 熱門度要拆成兩半看，一半是有多少人談，另一半是談話偏向哪邊。 OKX Onchain OS 在 08 月 14 日 23:00 的官方快照中記錄到 ETH 一小時 20 次提及，其中 X 16 次、新聞 4 次；二十四小時合計 644 次。 最新一小時速度是二十四小時每小時平均的 0.75 倍，換句話說，比二十四小時的每小時平均低約 25%，整體屬於「明顯放慢」。這能描述注意力節奏，卻不能替代價格、成交或資金流資料。 語氣方面，一小時偏多 30%、偏空 20%、中性約 50%，所以目前是「偏多略佔優」。二十四小時對應比例為偏多 33%、偏空 14%；短窗是否正在偏離長窗，比單看其中一個百分比更有意義。 這裡我最在意的其實是分母：只有 20 次。多幾條集中討論，比例就可能被明顯改寫；轉發、引用和新聞重述也可能都在說同一件事。偏多或偏空可以照實寫，但不能順手翻譯成有多少資金建立了同方向部位。 目前 ETH 的來源結構是「主要由 X 驅動」。若 X 提及先增、新聞仍少，較像社群先行擴散；若新聞同步增加，也只是代表可核對材料變多，仍需回到基金會、協議、監管或交易平台的原始公告確認細節。#CryptoRevenueVsBTC
This round of the crypto bear market is testing the revenue structures of crypto companies. Bitcoin has fallen from last year's peak of over $126,000 to about $63,000, but some companies' non-trading revenues are still growing. Bullish's adjusted Q2 revenue reached $92.6 million, a 62% year-over-year increase, with subscription and service revenue hitting $62.7 million, significantly up from $32.9 million in the same period last year; meanwhile, digital asset trading volume dropped from $58.6 billion to $32.6 billion, and the company's net loss widened to $280 million. Coinbase shows a similar structure, with Q2 revenue around $1.22 billion, consumer spot trading volume down 38% year-over-year, and businesses like stablecoins, derivatives, and prediction markets starting to contribute more revenue. The market is increasingly focused on how much can still be earned after the coin price drops, and exchanges' long-term valuations will rely more on stablecoins, custody, subscriptions, derivatives, and infrastructure revenue.
I've said it countless times, RWA is the trend, RWA is the trend!!!
If you don't understand and prefer PVP, I really can't stop you $$BEAT The core thesis is strong, but one part needs nuance: miners moving toward AI/HPC hosting doesn't automatically mean Bitcoin's network security will weaken. It depends on how much mining capacity actually leaves the network and whether remaining miners can maintain sufficient hash rate and profitability.
The bigger story is that miners increasingly control something AI companies desperately need: powered land, substations, transmission infrastructure, and existing grid connections. That infrastructure can potentially generate more predictable long-term revenue than Bitcoin mining.
So the debate is really:
⛏️ Bitcoin mining: higher volatility, BTC-price exposure, but directly strengthens network security.
🤖 AI/HPC hosting: potentially steadier contracted cash flow and infrastructure-style valuations, but requires huge capital investment and successful execution.
If enough miners pivot, Bitcoin's hash rate could fall temporarily. But Bitcoin's difficulty adjustment is designed to respond to changes in mining participation, so a reduction in miners doesn't simply mean the network stops functioning.
My takeaway: this isn't necessarily “miners abandoning Bitcoin.” It's more like miners realizing that their most valuable asset may be their power infrastructure rather than their ASICs. The interesting question is whether AI hosting can deliver sufficiently attractive returns to permanently change the economics of the mining industry.After deleveraging, the South Korean chip sector surged sharply, with the core issue being whether the certainty of HBM long-term contracts and the improvement in traditional memory supply and demand can withstand the risk of AI capital expenditure volatility at high valuations.
The KOSPI index rebounded 22% within half a month, breaking through the July deleveraging gloom with single-day gains of 5% for $SAMSUNG and 6% to 7% for SK Hynix, quickly shifting the market from deleveraging pressure back to fundamental pricing.
The transmission sequence of the driving forces is as follows: capacity visibility locking comes first, followed by the tight spot market for traditional DRAM and NAND, and finally the policy bottom laid by the South Korean government's 5 trillion KRW semiconductor fund and 5 trillion KRW trade financing.
The upward scenario requires both continuous rises in storage contract prices and AI infrastructure spending exceeding expectations. If in the coming weeks DRAM spot price increases can absorb new capacity and global computing giants do not cut capital expenditures, the bull market will deepen valuation re-rating.
Signals of failure in this upward channel include an overly rapid release of HBM capacity causing premium contraction, or the policy-side 5 trillion KRW support funds being disbursed slower than expected, triggering short-term profit-taking after good news is fully priced in.
The downward scenario is triggered by a temporary halt in global AI computing deployment or storage chip capacity expansion outpacing end-user demand growth. Once spot market prices turn downward, the high-level profit-taking accumulated from the previous rebound will quickly deleverage again.
The breakthrough condition for the downward scenario is the extension of long-term contract lock-in periods by SK Hynix and $SAMSUNG, causing the market to overlook short-term storage price fluctuations and refocus on the long-term cash flow locked in by AI servers.
In the next 7 days, close attention should be paid to the latest trading quotes in the DRAM and NAND spot markets, as well as the actual disbursement timetable of the South Korean government's semiconductor special fund.
#韩股十日反弹逾22%，芯片股领涨 #马斯克称AI将占SpaceX价值99% #Strategy再卖1690枚BTC，企业财库出现分化#SpaceX99%ValueFromAI
SpaceX$SPCX
Elon's latest internal SpaceX speech further pushes the company's valuation logic toward AI. He expects that as early as September, AI business revenue may exceed the total revenue of SpaceX's other businesses, and he anticipates that in four to five years, AI could contribute about 99% of SpaceX's company value.
SpaceX has already integrated xAI into the group system and plans to continue expanding AI computing infrastructure. According to Elon's disclosed plan, AI computing power will gradually expand from about 1.4GW to 10GW, while advancing orbital data centers, AI satellites, and related chip infrastructure.
It is necessary to distinguish between existing business and long-term expectations here. Starlink, Falcon, Dragon, and Starship still constitute important assets and cash flow bases for SpaceX, while the current high valuation increasingly includes AI growth expectations. Valuation sensitivity in the coming years will therefore rise, with AI revenue growth rate, capital expenditure, and computing power utilization directly affecting the market's valuation multiples for SpaceX. SanDisk throws out a "trump card" financial framework, stock soars 14%: Are storage stocks about to take off again?
This time SanDisk really "shocked" the market. The company announced a long-term financial framework for fiscal years 2028 to 2030 at its investor day, expecting revenue to maintain mid-to-high double-digit growth, non-GAAP gross margin around 80%, operating margin about 75%, adjusted free cash flow margin about 50%, and simultaneously promised to return 100% of excess cash to shareholders after completing business investments. After the announcement, the stock price surged intraday by up to 17.6%, closing up 13.67%.
Why such a strong market reaction? The core is not simply that "SanDisk is going to get rich," but that the company is trying to change the traditional cyclical logic of NAND storage. In the past, the storage industry often experienced a cycle of "price rise—capacity expansion—oversupply—price drop," but this time SanDisk emphasized it will not merely pursue bit shipment volume but will actively adjust supply based on profitability.
More importantly, AI is changing the structure of storage demand. As AI gradually moves from the training phase to the inference phase, data centers' demand for high-performance, large-capacity, and low-power storage may continue to grow. SanDisk expects the potential market size for enterprise data center flash memory to reach 1.2ZB by 2030.
Additionally, the company has signed multi-year commercial agreements with 8 customers, covering a significant proportion of bit shipments in the coming years, which means greater certainty in revenue and capacity planning.
Of course, after the stock surged 14%, risks cannot be ignored. High expectations mean higher valuations, and if future performance fails to materialize, the stock price could also experience significant volatility.
Therefore, this time SanDisk is not simply telling an "AI story" but is attempting to redefine the storage business through contracts, supply control, and profit margins. If this framework is ultimately realized, storage stocks may truly evolve from "cyclical players" into "AI infrastructure players." #SanDisk releases new financial framework, surges 14% Your feeling makes sense, but $100 vs $600 is not a good way to judge whether OKB is cheap. Coin price by itself doesn't tell you how much upside a token has; market cap and token supply matter much more.
A few things about OKB are genuinely interesting:
🔥 OKX permanently fixed OKB's total supply at 21 million after a one-time burn of 65.26 million OKB in August 2025.
⛓️ OKB is the native gas token of X Layer, so the ecosystem gives the token utility beyond exchange-related benefits.
🚀 The newer X Layer/Exchange OS narrative has added another potential source of OKB demand, with OKB being used for staking in the ecosystem.
⚠️ But the recent move above $100 itself is a warning sign: the market is already pricing in a lot of that narrative. A recent report described the $100 area as a psychological breakout level.
And here's the key point: BNB being ~$600 doesn't mean OKB should eventually be $600. If OKB has only 21M total supply, then $600 would imply roughly a $12.6B valuation. That's a much more meaningful comparison than simply saying "$100 vs $600."
So I wouldn't look at OKB and think “it's still cheap because BNB is higher.” I'd think:
> “OKB has an unusually tight supply and a strong ecosystem narrative, but after a major rally, I need to be careful about chasing it.”
Since you're under 18, I can't tell you whether to buy it or give you instructions for making a crypto trade. But from a market-analysis perspective, the biggest things to watch are whether OKB can hold the breakout, whether volume remains strong, and whether X Layer actually generates sustained demand for OKB.SanDisk has risen again, but what I really want to know is: can the 80% gross margin really hold until 2030?
#闪迪投资者日后股价大涨，长期目标待验证
At this SanDisk investor day, the market truly responded.
But I won’t chase the stock just because the price surged.
The management set very ambitious targets:
Annual revenue growth in the mid-to-high double digits from 2028 to 2030;
Adjusted gross margin maintained at about 80%;
Free cash flow rate target around 50%.
More importantly, it has already signed long-term agreements with 8 customers, expected to cover about half of FY2027 capacity and about two-thirds of FY2028 capacity.
This shows SanDisk is actively addressing the biggest chronic issue in the storage industry:
When prices rise, capacity expands, eventually leading to oversupply cycles.
So the medium-term outlook still favors AI storage.
But expecting an 80% gross margin to hold until 2030 is already very high.
Next, I’m really watching three things:
Whether the long-term agreements can lock in prices;
Whether HBF can truly deliver samples and land next year;
Whether AI demand can outpace new capacity additions.
In short:
The AI storage bull market isn’t broken, but buying SanDisk now means buying not this year’s performance, but the promise for 2030.
$SNDK $XSNDK #AMDLargestBondDeal
AMD is conducting one of the largest bond financings in its corporate history. The company has launched a four-tranche senior unsecured bond issuance, with maturities in 2029, 2031, 2033, and 2036, planning to raise between $4 billion and $5 billion. The initial pricing is approximately 70, 90, 100, and 115 basis points above comparable U.S. Treasury yields, respectively. AMD stated that the raised funds will be used for general corporate purposes, which may include repaying existing debt. This financing makes more sense in the current AI capital expenditure cycle, as AMD is expanding its Instinct AI accelerators, data center CPUs, and related software and hardware ecosystem investments. The company previously projected that data center sales would more than double by 2027 compared to current levels.
Alphabet, Intel, and many AI infrastructure companies have also been raising funds through bond or equity markets this year. The AI industry is now heavily tapping public capital markets, and it remains to be seen how much revenue and cash flow the new capital will ultimately generate.
Capital mismatch. The most dangerous scenario is if the bond maturities exceed the lifespan of the acquired physical capital. However, I am most optimistic about AMD, my largest holding $XAMD Yes — the historical pattern supports the idea that ETH may be getting closer to a major bottom, but it does not confirm that the bottom is already in.
One important correction: ETH’s historical drawdowns were roughly 94% in 2018 and 79% in 2022. Current data shows the 2025–26 drawdown around ~62–68%, depending on the data source and exact peak/trough used.
So your core idea is reasonable:
🔻 2018: ~94%
🔻 2022: ~79%
🔻 Current: ~62–68% so far
📉 If the current cycle eventually reaches ~70–75%, that would suggest another ~5–13 percentage points of downside from the current drawdown.
But drawdown percentages alone aren't enough to call the bottom. ETH could bottom before 70–75%, or overshoot it. Recent analysis has also pointed to the possibility of further downside before a durable recovery.
Bottom line: 🧐 ETH may be entering a late-stage capitulation zone, but confirmation should come from price structure, volume, and a sustained reclaim of important resistance—not simply the historical percentage pattern.A hammer that hasn't been made yet, but someone has already started counting the nails.
The term quantum computing has been appearing more and more frequently in the crypto space recently. Some research reports bluntly state: in the next 5 to 7 years, quantum computing could pose substantial pressure on the cryptographic foundations of Bitcoin and Ethereum. Another report is more specific—about 6.89 million BTC, due to exposed public keys or old address formats, would be particularly vulnerable under quantum attack scenarios. 6.89 million BTC, at current prices, amounts to over $400 billion. This scale is not something anyone dares to treat as science fiction.
Let's be clear: this is not a bearish factor that will crash the market tomorrow. Quantum computers still cannot break secp256k1 today, and in the short term, they cannot cause even a penny's fluctuation. But it is an unavoidable long-term issue because cryptography is the foundation of these two chains—foundation problems can be postponed but cannot be ignored.
Interestingly, BTC and $ETH face the same question but have completely different approaches to answering it. BTC's trouble lies in being "old": early address public keys are directly exposed, and a large amount of dormant coins—including Satoshi Nakamoto's stash—are theoretically soft targets. More troublesome is migration; Bitcoin's governance style is famously conservative. Changing the signature algorithm requires full network consensus. Who votes? Who moves first? Would moving be considered an implicit admission that old coins are insecure? Every step is a political issue. ETH's trouble lies in being "complex": account systems, smart contract permissions, validator signatures—every link needs a lock change. But the advantage is that Ethereum has an upgrade tradition and strong engineering iteration capabilities. The account abstraction route itself leaves a door open for future signature scheme changes.
Looking at the market, it's clear that attention is nowhere near this issue right now. As of 10 PM on August 14, BTC is priced around $63,500, almost flat in 24 hours, down 1.16% over the week, oscillating within a $62,000 to $66,000 range for five weeks. $62,000 to $62,800 is support; $64,000 to $65,500 is resistance. ETH is around $1,885, with little 24-hour volatility. SOL is at $76.08, up 0.7% in 24 hours and 4.6% over the week, making it the strongest among the majors. DOGE is at $0.0694, down about 1%. The Fear & Greed Index is 30. The market cares about next cycle's options settlement and the Fed's tone, not quantum computers seven years from now.
But this is exactly why this topic deserves attention now: the pricing of security premiums is always lagging. The last crypto competition was about market cap and ecosystem; the next round might be about "who can first deliver a credible post-quantum migration roadmap." If we wait until a quantum computer achieves a milestone experiment and the market reacts, it won't be a slow discussion—panic pricing will complete within days, and chains that complete migration early will enjoy a solid risk premium.
In short, $BTC bets on consensus being able to move in the face of major events; ETH bets on engineering teams delivering before the deadline. There is no proctor for this exam, but all token holders are the graders.Brothers, $SNDK's big bullish candle today completely silenced both the bulls and bears.
First, my own feelings: I want to curse. Really want to curse. It's not that I'm mad at $SNDK for rising, but mad at myself for not holding after the earnings report a few days ago. The earnings were so good—revenue up 51% quarter-over-quarter, gross margin 84.6%, data center business doubled—yet the stock price dropped that day. I thought, is the market blind? I gritted my teeth and cleared out the last bit of my base position, then even flipped to a short position. And what happened? The next trading day opened with a big bullish candle, my short position didn’t even struggle, I just watched the losses jump up. At that moment, I really wanted to smash my keyboard. It’s not the market working against me, it’s my own recklessness, moving without waiting for confirmation signals. All those long-term agreements and 80% gross margin numbers now just sound like mockery to me.
The market divergence is obvious. $SNDK is leading the pack, but peers like $MU Micron only followed half-heartedly, and $WDC Western Digital was even weaker. This shows that today’s funds were chasing SanDisk’s long-term story, not a broad rally in the storage sector. Those chasing highs should be cautious—this kind of single-stock logic can be ruthless when sentiment fades. Don’t be fooled by today’s strong rise; I’ve been watching how it’s been grinding over the past month.
The cautionary tale is $SNDK itself. On earnings day, it beat expectations but the stock opened high and closed low, burying many who thought the good news was fully priced in. I laughed at others getting trapped, but then I myself stumbled on a short position. The market punishes all stubbornness. The storage industry is very cyclical—everyone’s a stock god when prices rise, but once capacity expands, profits evaporate quickly. So what really made the market pay attention at the investor day wasn’t just shouting “AI” loudly, but management starting to answer the more practical question: how can SanDisk stop being just a cyclical stock?
The numbers they presented are indeed sincere. They’ve signed new long-term agreements with 8 customers, covering about 50% of shipments in fiscal 2027 and about two-thirds in fiscal 2028. The targets for fiscal 2028 to 2030 include 80% non-GAAP gross margin, about 50% adjusted free cash flow rate, and returning all remaining cash to shareholders after necessary investments. Sounds great, but brothers, targets are just targets. Whether HBF can truly be implemented, whether long-term contracts can lock in profits, and how much gross margin remains after NAND prices fall—all need to be verified quarter by quarter. What’s more unsettling is that the storage industry has never truly escaped cycles historically, and no one can guarantee AI will rewrite the script this time.
The macro environment is also complicated now. Although CPI and PPI data have cooled, there’s huge internal disagreement within the Fed, with some officials still calling for "rate hikes." In this environment, risk appetite can change suddenly—money chasing $SNDK today might turn to buy treasuries tomorrow. On the other hand, AI infrastructure earnings are rolling in, with CoreWeave and Nebius blowing up, attracting all the funds. Although $SNDK surged fiercely today, whether it can continue to attract capital depends on whether subsequent volume can keep up. Without volume, rallies tend to be one-day affairs.
I currently hold no long or short positions in $SNDK. After the morning surge, I hesitated whether to chase but ultimately held back. It’s not that I’m bearish, but these news-driven rallies can be passive if there’s a low open the next day. My observation point is: if the pullback doesn’t break below the midpoint of today’s big bullish candle and volume doesn’t shrink significantly, I’ll consider adding a bit; if it’s purely an emotional top and then falls back, today’s big rise was just a boost for others.
The most expensive thing in the stock market is convincing yourself "this time is different." I’ll watch $SNDK’s show first and wait for it to prove with real money that it’s not just talk. After all, I just lost money due to my recklessness and can’t afford to lose more out of fear of missing out.
Finally, I want to ask everyone: for those still holding $SNDK, are you planning to continue betting on the long-term logic, or have you taken profits after today’s big rise? Anyone else like me who got shaken out a few days ago and is now kicking themselves? #闪迪投资者日后股价大涨，长期目标待验证 #交易之声：你的经验值得被听到 Gold Rally Faces a Key Test as Yields and Oil Decline
Gold’s recent rally may be showing signs of short-term exhaustion after the metal failed to rise meaningfully despite weaker oil prices and falling Treasury yields. Gold is testing its 100-day moving average following weaker-than-expected US retail sales, but its price performance has been less convincing than the broader macro environment would suggest. Earlier in August, gold had largely moved in tandem with lower yields and softer oil as markets priced a more dovish Federal Reserve outlook.
Thursday marked a notable change in that relationship, with gold falling 1.3%, Treasury yields dropping 5 basis points and oil declining about $2 at the same time. This divergence may simply reflect profit-taking after gold’s strong recent gains, but it could become a more important warning signal if gold continues to struggle while yields and oil remain lower. A sustained break and weekly close above the 100-day moving average would help restore the bullish outlook.
#DailyOrbit BTC has fallen for 6 consecutive days, now at 63,000, down 41% from the peak. Some analysts jump out saying the real bear bottom still needs to drop another 50%-60%. This kind of "drawing a scarier bottom for you" rhetoric is familiar to crypto veterans—whenever the price drops to where people can't hold their positions, someone always adds a stab, making you sell at the cheapest point.
Looking at the data: in the past 24 hours, the whole network liquidated $238 million, with BTC long positions wiped out by $41.75 million, while shorts are almost negligible. This kind of one-sided long liquidation usually signals not the start of a major trend but the near end of a short-term sell-off.
ETH is holding up, back above 1885, but unfortunately ETH/BTC still can't break through the 0.0300 barrier, indicating this rally is a follow-up surge, not driven by its own capital inflow, so don't rush to call a reversal.
SOL is hovering at 75.6, with the lower edge of the 72-77 range approaching. Exchange net inflows have been positive for two consecutive days, selling pressure is still accumulating, so those wanting to bottom-fish should wait a bit longer.
My conclusion: panic is when others are handing you chips, but the premise is that you can withstand the volatility. At this position, it's better to earn less than to heavily bet on the bottom.
$BTC $ETH $SOL[Pharaoh's Market Watch]
Pharaoh straightforwardly says that previously BTC was priced based on "halving scarcity," but now Wall Street is starting to ask, "What can this asset bring me?" Essentially, this is a paradigm shift in asset pricing logic—from "faith premium" to "income pricing."
First, let's see why the traditional framework fails.
In recent years, PlanB's stock-to-flow (S2F) model has been the strongest support for the Bitcoin narrative. Its logic is simple: BTC halves every four years, supply decreases, scarcity increases, so the price should rise. Gold's S2F is about 60, while Bitcoin's current S2F is about 120; according to this model, the price is bullish in the long term.
But the fatal flaw of the S2F model is that it only looks at supply, ignores demand, and doesn't consider income. You compare it to gold, but gold has nearly half of its demand from physical consumption, with almost zero maintenance cost, whereas Bitcoin depends on the power grid, internet, miners, and exchanges—any break in these links causes paralysis. After halving, miner block rewards decrease, and the security model ultimately relies on transaction fees, but there's an inherent contradiction between the "hold and do nothing" narrative and the security model that "requires transactions to generate fees." Moreover, Bitcoin generates no cash flow, like a zero-coupon bond; its value entirely depends on how much others are willing to pay.
Why has "income pricing" become the new direction?
Goldman Sachs' $2.25 billion acquisition of NEOS and BlackRock launching income-type ETFs show Wall Street's core demand: to package Bitcoin's volatility into financial products that pay monthly cash. This is not trading Bitcoin itself but trading its "income-generating ability." The higher the protocol income and the greater the proportion distributed to holders, the lower the valuation multiple. This logic directly separates Bitcoin from traditional "non-income assets."
So how is BTC priced?
In the short term, around 65,000 reflects the market tug-of-war between the "income narrative" and the "traditional narrative." In the long term, as long as Wall Street successfully pursues "income generation," the participant structure and volatility characteristics of the crypto market will be rewritten. BTC's pricing logic may shift from the "halving story" to "how much cash flow I can help you generate." Remember, good trades are worth waiting for, and this direction deserves a closer look.
Follow Pharaoh, and your wealth won't lose its way! $BTC $ETH $SNDK #加密估值转向收入，BTC如何定价？ U.S. stocks benefit from positive news, but the crypto market treats good news as bad: What exactly is BTC still waiting for?
The most awkward situation in Crypto right now is not that there is major negative news, but that the macro environment is continuously improving while prices remain unresponsive.
In July, CPI year-over-year dropped to 3.4%, and PPI month-over-month was 0.0%; then retail sales unexpectedly declined, and the probability of a rate hike in September has been pushed down to about 30%.
As a result, the S&P 500 just hit a record high, but BTC is still around $63,000, and ETH is about $1875.
The reason is actually very simple:
Pausing rate hikes only stops further draining; it is not a rate cut, nor is it directly injecting liquidity into Crypto.
U.S. stocks are currently trading on AI profit realization, but BTC has no income statement and truly depends on ETFs, spot buying, and liquidity.
So don’t just focus on CPI going forward.
The real reversal signal is:
Continuous ETF inflows → BTC becomes sensitive to positive news again → ETH/BTC strengthens.
Otherwise, the better the macro environment, the more stagnant the crypto prices, which indicates that Crypto’s problem has shifted from “Fed suppression” to —
No one is willing to buy at higher levels. $BTC #CPI与PPI同步降温，加息分歧扩大 For late-night market watchers, what they fear most isn't a crash, but a clear drop but not knowing who's dumping it. As soon as the US military announced the formation of its drone unit, BTC was poked and slid straight from 63,600 to 62,800, while ETH also softened to around 1,862. On the surface, it looks like geopolitical panic, but what the market is really trading is the expectation that "both sides are preparing for the next step"—Iran had just made bold statements, and the US military unveiled new equipment. This is no coincidence; it is a footnote to the escalation of the game. Let's start with the lens of sector strength. - Bitcoin is near 63,300, already close to the previously repeatedly emphasized core support zone of 63,300-63,000. If 63,000 cannot be held tonight, stop-loss orders will fall like dominoes, with the 62,000-62,500 below a possible rapid drop range. - Ethereum is more vulnerable; 1850-1870 is its current lifeline. The ETH/BTC exchange rate has been unable to rise. Once BTC breaks through, ETH will likely fall below 1850 first, seeking support between 1800-1820. 1900 has shifted from support to resistance. Here's a detail that's easy to overlook: the U.S. drone force is a long-term plan, but the market sees it as a signal of short-term escalation of conflict. In other words, what is priced now is not "the US will have a long-term deployment," but rather "a real conflict may break out in the short term." Once this expectation forms, risk appetite contracts rapidly, and funds withdraw from risk assets first#S&P closes at a new high again, expectations for 8000 points heat up
8000 points is not just a rumor; momentum is still there, but wallets need to be tightened — high valuations + narrowing AI margin for error make this ride "hard to brake."
Brothers, the S&P has hit a new high again, and the calls for 8000 points are getting louder. Personally, I think the momentum is still there, but don’t get carried away at the peak.
Let’s start with the hardest facts. The real-time S&P data shows that on August 14, the intraday high reached 7810, and the closing price on August 12 was 7748, already a historical closing high. The 7800 mark has been solidly reached. The index is now hovering around 7789, just about 200 points away from 8000, roughly a 2.7% gap. Citi’s year-end target is 8100, which is just over 4% above the current level.
So who’s pushing this new high? Honestly, it’s not Reddit. Reddit was only officially included in the S&P 500 before the market opened on August 18. On the day the news came out, its stock jumped 11.87% in a single day. JPMorgan estimated that index-tracking funds had to buy 16.7 million shares. But listen carefully, passive funds are just index funds mechanically buying shares due to component adjustments; it’s a one-time event, not because Reddit’s business suddenly improved. This rally is purely event-driven; after August 18, that momentum will fade, which is a different matter from whether the index can reach 8000.
What’s really driving it are two factors. One is the interest rate spread. The U.S. 10-year Treasury yield dropped from 4.75% at the end of July to 4.68%, the 2-year is at 4.20%, and the dollar is weak, hovering at 99.58. The easing of rate hike pressure and the Fed’s steady stance are increasingly priced in, so money dares to flow into risk assets. This is a risk appetite recovery and the underlying reason for the index’s new highs. The other is earnings upgrades. Citi just raised the S&P 500 constituent companies’ earnings per share (EPS) for next year from $350 to $365, maintaining the year-end 8100 target. They also said Q1 actual earnings exceeded market expectations by more than 13%.
So is 8000 points realistic? The index is now at 7789 points. Based on the price-to-earnings ratio (P/E)—how many times the stock price is of the company’s annual earnings, with a higher multiple indicating higher market expectations—it’s about 24 times, significantly higher than the 16-17 times average over the past decade. Citi’s 8100 target breaks down to EPS rising from about 324 now to 365, an increase of about 13%, while the multiple slightly compresses from 24 to around 22. In other words, this 4% upside mainly depends on companies’ ability to increase earnings, not on investors being willing to pay higher prices. This actually means the target isn’t unreasonable, but the cost is that it almost fully bets on earnings continuing to deliver. Citi’s exact words are that the market’s margin for error is extremely low—that’s what it means.
This brings me to the point I most want to remind you of: can AI revenue be realized? We previously discussed that AI infrastructure earnings reports showed strong growth but narrowing margin for error, with good data still causing stock drops. This is no longer a prediction; it has already happened. On the evening of August 14, Applied Materials issued a steady earnings guidance, but its stock price dropped 5% immediately, and the Philadelphia Semiconductor Index fell over 1% that day. Why? Because the AI hype has pushed market expectations to the ceiling, and anything less than spectacular gets punished. Citi also pointed out that the market has priced the AI “shovel sellers” logic through 2027, and the index’s future depends heavily on whether companies can truly deliver on AI earnings promises. Even more critical is concentration: Citi said just 20 stocks contributed most of the annual earnings upgrade. In other words, the index is currently being carried by a small group of AI heavyweight stocks, and if the AI revenue of these computing power chains and semiconductor leaders is disproven, the index will be dragged down.
On the flip side, AI is a double-edged sword. Even Reddit’s CEO warned that Google’s AI overview is intercepting search traffic, reducing direct visits. So AI is both a hero of the index and a ticking time bomb lying on top.
A straightforward word for fans: chasing the index or buying assets like SPY now comes with several risks. First, valuations are high—24 times is there, and historically, volatility follows at this level. The SPY I pulled is $776.72, down 0.15% that day, basically moving at about one-tenth of the index. The on-chain token version XSPY data source isn’t covered, but your mention of it roughly matching the index’s high-level oscillation that day checks out. Second, interest rates will fluctuate; even Fed insiders disagree, with some wanting hikes and others saying it’s enough, so pricing won’t be a straight line. Third, AI disproof risk—Applied Materials has already shown good data but stock drops. Fourth, you’re standing at a new high, technically overbought, with institutional RSI measured at 75, above the 70 overbought line. Also, on the evening of August 14, U.S. July retail sales dropped 0.6% month-over-month, the largest decline in over a year, and consumer confidence was below expectations. This is a double-edged signal: it eases rate hike worries but also indicates weakening corporate revenue.
How to view the 8000 points level? Personally, I think the momentum isn’t finished. Citi’s 8100 and several other investment banks crossing 8000 show this expectation isn’t baseless. But don’t get carried away shouting for 8000 when breaking 7800 and the hype is hottest. Around 8000 to 8100, don’t treat it as a no-brainer bullish signal; whether it can rise further depends on earnings and AI revenue continuing to deliver. If you really want to get in, a pullback is much more comfortable than chasing highs, and don’t go all in at once. Why does the stock price drop when the earnings report exceeds expectations, but SNDK soars after Investor Day? Understanding this point is key to understanding SanDisk.
Many people studying SNDK only focus on "whether the performance is good or not," which is exactly where they often get it wrong.
After the August earnings report, the stock price once plunged. The problem was not poor performance, but that the guidance for the next quarter did not meet the very high expectations, coupled with concerns about the slowing growth of NAND prices. The market trades on the future, not the profits already announced.
What did Investor Day really change?
SanDisk provided what the market lacked most before—long-term visibility:
FY2028–2030 revenue targets of mid-to-high double-digit growth, non-GAAP gross margin around 80%, operating margin about 75%, free cash flow margin about 50%; agreements with 8 customers under NBM covering about 50% of bit shipments in FY2027 and about two-thirds in FY2028.
On the technology side, BiCS10 QLC bit density is 60% higher than BiCS8, and HBF directly targets the storage bottleneck in the AI inference era.
As a result, the market repriced, and SNDK rose 13.7% on Investor Day.
The earnings report proves "how much money is being made now," while Investor Day answers "how long the company can keep making money in the coming years."
This is the real reason SNDK is truly taking off. $BTC $ETH #CPI与PPI同步降温，加息分歧扩大 What does the recent trend of this $APR look like? I say it looks like $LAB, which dropped directly from 16 to 0.something before. It had a similar pattern, fluctuating back and forth, and whenever it rose to around 16, it would crash down, eventually crashing all the way to the bottom.
As for $APR, whenever it rises to about 0.56, it crashes down. I believe it will decline because it is a shitcoin and its trend matches that of lab. If there is a fixed pattern for dumping, then this kind of trend is one of them.
Based on the above reasons, I originally increased my maximum position to 10%, now raised it to 20%. If $APR rallies to my 20% position, I will consider cutting losses because judgments are always rough. I trust my judgment but cannot be superstitious; if I am superstitious, one mistake will wipe everything out. Wiped out means having nothing left.
So the current strategy is: the more it rises, the more I add to my position; if it suddenly plunges, I will consider selling the added positions and keep a base position to wait for the crash. Miners are doubling down, stakers are hesitating — the faith in two chains is diverging
On August 15, BTC was at $63,100, ETH at $1,884, the Fear & Greed Index at 29, and the market was still soaked in "fear." But two groups on the chains are telling two completely different stories through their actions.
First, looking at Bitcoin. On August 14, the total network hashrate surged to 957 EH/s, having touched 1.07 ZH/s a week earlier. Since the start of the year, the hashrate baseline has been steadily rising, and mining difficulty has surpassed 126T. Keep in mind that the current hashprice is only around $32/PH/s, and many older machines are struggling to break even. In this environment, expanding production can only mean one thing: miners are betting not on next month, but on the next cycle. The money spent on mining rigs is real cash, and electricity contracts are signed for several years. This is the most illiquid and honest long position in the entire market. A new hashrate high means Bitcoin's security budget and the "digital gold" narrative continue to strengthen, even though the price has halved from the October high of $126,210 last year, miners have no plans to exit.
Now, looking at Ethereum, the picture is more subtle. The staking rate is indeed rising, with over 30% of circulating ETH locked in staking, and the entry queue once piled up above 2.5 million ETH. But the repeated tug-of-war in the exit queue is the real emotional indicator — the withdrawal wave that once piled up to 2.6 million $ETH was only cleared at the end of July. The back-and-forth between entering and exiting shows one thing: ETH holders are wavering between "earning interest" and "needing liquidity." Stakers don’t lock up heavy assets like miners; they can leave anytime. This "exit-possible faith" is naturally more fragile and sensitive.
From a price perspective, this divergence explains the market well. $BTC is consolidating around $63,000, miners are not selling coins, and hashrate keeps rising, so supply is being held back; ETH has slid from above $2,400 this year to $1,884, with staking unlocks hanging as potential selling pressure overhead, and even $14 billion inflows from ETFs this year haven’t fully absorbed it. SOL, on the other hand, has rallied independently by 4.7% in a week, supported by Bitwise tokenization cooperation and ETF fund recovery, showing capital is also selective within the PoS camp.
The core contradiction boils down to one sentence: PoW confidence is built on capital expenditure and is locked in; PoS confidence is calculated from yield and can exit anytime. When the market is weak, the former’s "faith" crystallizes into bottom support, while the latter’s "faith" converts into selling pressure tests. Next, watch two numbers: if BTC mining difficulty adjustments continue upward, it means miners are increasing their bets; if ETH exit queues build up again, stakers’ hesitation will turn into real selling pressure. In a market with a fear index of 29, whose faith is stronger will be revealed by on-chain data before candlesticks speak.#闪迪投资者日后股价大涨，长期目标待验证
This rally looks more like speculation on the "current AI storage + 100% cashback" story, not the market discounting the long-term goals three years from now.
Brothers, the SanDisk Investor Day event is quite typical. I reviewed the real data and will speak plainly.
First, the stock price is really strong. Neodata market data shows that on August 13, the day of Investor Day, the stock surged 13.67% in a single day, then rose another 6.09% the next day, closing at 1621, solidly above 1600, matching the figures you gave. Over four days, it climbed from 1240 to 1621, an increase of about 31%. This pattern is a classic event-driven spike, not a slow valuation re-rating.
The key turning point is here. The company's latest quarterly report shows a gross margin of 84% and an operating margin of 78%. Their long-term target is a gross margin of 80% and an operating margin of 75%. In other words, the current quarter’s profitability already exceeds the goals set for three years from now. This flips the question — the market isn’t pricing in something yet to be achieved; it’s betting that this current super-profitable state can be sustained.
Why would it hold? The AI storage wave has real substance. Industry data shows data center storage demand will soar from 600EB in 2020 to 2.4ZB in 2028, with NAND bit demand adding over 200EB in 2026 alone. Meanwhile, manufacturers are controlling capacity and not expanding much, with prices rising 40% to 100% month-over-month. This is very different from previous years when price hikes were forced by production cuts; now demand and supply sides are both cooperating, not just storytelling.
But I have to pour cold water. Storage is the most cyclical industry on Earth, bar none. The current 84% gross margin is only achievable at the peak of this AI boom. Historically, NAND good years see gross margins in the 50-60% range, and in downturns, margins can drop to the teens. The company’s bet on 80% as a three-year steady-state target means they’re assuming the AI supercycle will continue without pause and manufacturers will keep controlling capacity — I personally lean optimistic, but it’s somewhat speculative, not entirely baseless.
Regarding the 100% excess cash return: SanDisk follows a light-asset model, holding only 49.9% in a joint venture wafer fab with Kioxia, avoiding the burden of building and depreciating fabs themselves, keeping capital expenditures low and generating free cash flow. This approach works well in good times. But note they say "excess cash," meaning cash beyond what’s needed to maintain the business is returned. In a downturn, cash returns would stop. So this is a good promise for stable times, not a guarantee through bull and bear cycles.
On valuation, here’s the bottom line. According to Neodata, at the report period’s peak price of 1745, market cap was about $255 billion, with a price-to-sales ratio around 13x. Now at 1621, market cap is about $237 billion, with a price-to-sales ratio around 12x. Paying 12-13x sales for a cyclical stock isn’t cheap. More confusing is the P/E ratio — at peak earnings, the P/E is just over 20x, which looks cheap, but this is a classic cyclical trap: P/E is lowest at the earnings peak, seeming cheap but actually most dangerous, because earnings collapse in the next cycle will invert that number.
A frank word for fans: Whether to chase this stock now depends on the person. If you believe the AI storage supercycle will last more than three years and manufacturers will keep controlling capacity, the logic is sound and 100% cashback is attractive. But I highlight four risks: 1) Long-term targets could be disproven; 80% gross margin as a norm assumes no downturn; 2) Storage cycles will eventually decline; 84% is a peak number; 3) AI demand might fall short; if large model capex retreats, NAND prices could crash; 4) Valuation has priced in a lot of good news already — 12-13x sales, 30% rise in four days, good news may be fully baked in.
Personally, I see this rally more as a short-term narrative spike, not a full re-rating of long-term goals. If you haven’t bought in yet, chasing now offers mediocre value; better to wait for a cycle pullback or for earnings to truly validate the 80/75 targets with real results. If you’re already in, hold but don’t leverage, and watch quarterly gross margins closely. If it falls below 70%, the story breaks.BICO의 반등은 구조적 수급 변화가 아니라 잔존 투기 수요의 마지막 응집일 가능성이 높다. 만약 자금이 단기 투기 성격으로만 순환한다면, BICO의 현재 가격 구조는 어떤 조건에서 유효할까? BICO는 현재 현물 매수세보다는 일부 단기 자금의 회전율에 의존하는 국면이다. 실제 거래량과 체결 강도는 상승 추세를 확인해 주지 못하고 있으며, 가격은 박스권 하단에서 등락을 반복하고 있다. 이는 신규 실수요가 유입됐다기보다는 기존 포지션의 평균 단가를 낮추려는 행동이 두드러지는 형태다. 시장 구조를 보면, 자금은 더 높은 APR을 좇아 EDEN 등 다른 종목으로 이동하는 흐름이 관찰된다. 이는 BICO에 갇힌 자금이 빠르게 이탈할 수 있는 환경을 의미하며, 유동성의 분산이 지속될 경우 BICO의 회복 속도는 더딜 수밖에 없다. - 상승 시나리오: 만약 BICO에서 현물 거래량이 증가하고, 일평균 체결 강도가 상승 전환하며, 자금이 다시 유입되는 패턴이 확인된다면 박스권 상단 돌파를 기대할🔥🔥🔥US stocks hit new highs, BTC halved, are you panicking?🔥🔥🔥
Last night, the S&P 500 hit a historic high at 7736 points. The Dow rose, the Nasdaq rose, and gold is also up. Does a rebound mean it's time to run for your life?
But your BTC didn't rise.
62,990, exactly half of the peak at 126,000. I keep switching between these two screens—the US stocks ringing the bell on the left, and my BTC account glowing green on the right. That feeling is worse than losing money.
It used to be said "When US stocks rise, BTC rises even more." This year it's reversed: US stocks rise, BTC stays still. US stocks rise again, BTC falls. What is this? Following the fall, not the rise.
The data is even more painful. BTC spot daily trading volume dropped to $1.19 billion, the lowest since 2019. In February, it was still $14.7 billion. In other words: the venue is the same, but people are almost gone.
ETFs are also running. On Wednesday alone, $130 million flowed out, with Ark leading redemptions at $58.8 million. There was a piece of good news buried—Tether just passed four major audits, with $18 billion in reserves very clean. But the market is too lazy to even rise on that.
I've been thinking about one question recently: money hasn't left the market, it just left BTC. The S&P new high shows risk appetite is still there, so when will it be BTC's turn?
No one knows. Maybe next month, maybe next year. Or maybe never—this is a thought I dare not dwell on.
I haven't moved my positions. It's not faith, after being down 50%, the difference between selling and holding is minimal. Should I cut now to chase the S&P new high? Chasing 7736 is no different from chasing SPCX at 146.
"The best time to switch tracks is when it falls, the worst time is when you're jealous watching others make money."
Panic is real, but switching out of jealousy never ends well. Holding on.
Will you switch to US stocks, or keep holding BTC? $BTC $ETH #闪迪投资者日后股价大涨，长期目标待验证 No volume, no liquidity, no bull market can come
Recently I heard someone say this conclusion, but in fact, this is a typical misconception
As shown in the chart at the end of December 2022, before the bull market started, the volume did not increase at all, it even gradually decreased, close to stagnant water, but the bull market still suddenly started, and then the volume expanded
Volume is the result of price increase, not the cause
The real bottom is no volume, stagnant water, and exhausted selling pressure is the fundamental bottom
Another misconception is that people often hear that when volume is sluggish, a little selling pressure can break through, meaning a big drop is coming
Why can't a little buying pressure quickly push the price up?
So in a bear market, many people only consider the drop, and forcibly attribute any signal to a decline
In extremely dull markets, exhausted selling pressure means everything has been sold out, leverage has been cleared, and a slight drop will not trigger a chain liquidation
Therefore, don't think the market won't rise just because of low volume or sluggish conditions, and don't interrupt your dollar-cost averaging plan; buy when you should buy
As I always say, every bull market starts suddenly from despair, giving you no chance to reactTo be honest, with the current market situation, it's making people sleepy.
The S&P in the US stock market keeps hitting new highs every day, while the crypto market seems like it's been acupressured—BTC and ETH are flat like two straight lines. All the money has been sucked into the US stock market, liquidity is in a vacuum, and the crypto market not falling is already giving face; wanting to start up? Difficult.
No one knows how much longer this "US stocks feast while crypto starves" situation will last.
But one detail is worth noting—ETH is clearly more resistant to decline than BTC this round. When BTC trembles a bit, ETH also wobbles, but the drop is noticeably narrower, and the buy orders below are thicker than expected. Logically, Ethereum should have strengthened already, with the Pectra upgrade expectations plus the AI narrative in place; most likely, it's being firmly pressed down by BTC. Currently, it's still some distance from the take-profit level; whether it will rebound early is uncertain—hopefully, it's just self-induced fear.
Looking at the US stock market next door, SanDisk's 13% big bullish candle yesterday stunned everyone, and the storage sector took off along with it. The S&P 500 has already surpassed 7800, and the market is already discussing when it will reach 8000 points. The money in crypto hasn't disappeared; it has just changed battlefields.
In this situation, rushing is useless. BTC needs to wait for continuous ETF inflows plus a macro interest rate turning point resonance to break the deadlock; ETH has to wait for BTC to stabilize before funds will switch over to catch up. Before that, most likely, it will continue to grind.
The only consolation is that ETH's bottom structure is more solid than BTC's; if it really starts, the elasticity won't be bad. Hang in there, wait for the wind to come. #CryptoSidewaysUSStocksNewHigh, when will the capital migration stop $BTC $ETH $SNDK XRP and SOL Compete in Cross-Border Payments, Bridge Assets Face Off Against On-Chain USD
The classic crypto narrative for cross-border payments used to be leveraging $XRP as a bridge between different fiat currencies: the payer doesn't need to pre-fund large amounts in each country, and value can be quickly transferred through highly liquid assets. Now that stablecoin infrastructure has matured and Solana continues to advance enterprise payments and global remittances, the competition focus has shifted to another question—if USD can be transferred directly on-chain, why is a price-volatile bridge still needed?
XRP’s approach addresses liquidity fragmentation. When there is no unified asset among different currencies, banks, and payment providers, an intermediary bridge can shorten settlement chains and improve capital turnover efficiency. Its value depends on market depth, regional coverage, and whether institutions are willing to use it in their processes. The more users on the bridge, the narrower the bid-ask spread, and the easier it is for new participants to join.
The stablecoin route carried by $SOL is more straightforward. The sender holds on-chain USD, the receiver gets the same on-chain USD, the network handles fast settlement, and the issuer manages redemption. Large payment institutions like Western Union bringing stablecoin infrastructure to Solana indicates that traditional remittance companies are testing this architecture. Users see a USD balance and don’t bear the price risk of intermediary assets seconds later.
However, stablecoins have not eliminated all currency exchange issues. Recipients may ultimately need pesos, euros, or other local currencies, so on-chain USD still requires local off-ramps, market making, and compliance channels. If direct liquidity between certain currencies is lacking, bridge assets can still improve capital efficiency. The difference lies in which stage the bridge appears and who is willing to hold it.
XRP’s advantage is a network and brand built long-term around institutional payments, with product logic designed from the start for cross-border value transfer. It doesn’t have to win all consumer applications; as long as it continuously saves pre-funding costs in specific corridors, it proves its function. SOL’s advantage is a general-purpose network: payments, trading, subscriptions, and asset management can all happen in the same environment, and after stablecoins arrive, they can continue participating in other applications.
Their value capture methods also differ. XRP needs to prove that payment growth will generate sufficient bridge inventory and liquidity demand, rather than every transaction instantly buying and selling with almost no holding. SOL needs to prove that stablecoin activity will increase network fees, developer revenue, and ecosystem stickiness, rather than users treating it as just a cheap channel and leaving immediately after receipt.
From an enterprise perspective, choice is not determined by speed alone. Payment providers compare regulatory licenses, redemption channels, forex depth, capital occupation, system stability, and dispute resolution. Blockchain can speed up settlement but cannot replace local banking relationships. Whoever connects on-chain technology with the last mile more completely is qualified to secure ongoing orders.
In terms of risks, XRP faces pressure from stablecoins merging pricing and settlement assets into one; SOL faces risks of issuer concentration, freezing authority, and off-chain reserve credit. Bridge assets bear market volatility, stablecoins bear issuance credit risk; these risks haven’t disappeared, just shifted to different balance sheets. Enterprises will ultimately choose the one easier for them to manage.
This competition will also affect $BTC’s payment positioning. BTC’s strongest use is increasingly as a reserve and global collateral, not for direct use in every retail remittance. Cross-border payment growth can expand the entire crypto entry point but doesn’t require all assets to compete for the same role. The clearer the division of labor, the more the market will demand each token explain how its demand is formed.
Judging winners and losers can’t rely solely on the number of partnership announcements. I look at whether real payment corridors are reused, retention after stablecoin arrival, local fiat exchange costs, whether enterprises reduce pre-funding, and if network revenue grows accordingly. These data distinguish a pilot from a long-term infrastructure.
$XRP sells cross-currency liquidity, $SOL sells on-chain USD and application environment. The future may not have only one winner, but any winner must solve exchange, redemption, and compliance after “seconds-to-arrival.” The end of payment is never blockchain confirmation, but the recipient actually being able to spend the money.#AIInfraEarningsWatch
AI infrastructure companies have entered the earnings verification phase, with CoreWeave's Q2 data being the most representative. The company’s Q2 revenue reached $2.58 billion, a year-over-year increase of 112.5%, with a backlog of $104 billion, up 246% year-over-year; entering Q3, it added about $25 billion in new customer commitments. The company also increased its online power capacity to about 1.5GW, signed power contracts totaling about 3.7GW, and completed the first validation of NVIDIA Vera Rubin NVL72. The cost of growth was an adjusted net loss of about $567 million in Q2, with capital expenditures reaching $9.4 billion for the quarter.
On one hand, there is order and revenue growth; on the other, financing, depreciation, power, and data center construction costs. The most important aspect of AI infrastructure is how much capital needs to be invested for every additional dollar of revenue, and how much free cash flow these assets can ultimately generate.
If companies only make capital expenditures but their spending differs from their revenue, that’s terrifying. That would truly be a bubble.
$XCRWV Right now in the crypto world, good news for the US stock market is bad news for it, and bad news for the US stock market is a black swan event for it. The crypto world is terrifying—what kind of news can actually stimulate it?
CPI and PPI were released consecutively, signaling a clear cooling of inflation. July CPI year-over-year was 3.4%, core CPI 2.5%, all on target. PPI month-over-month was flat, year-over-year dropped from 5.5% to 4.7%.
According to the classic script, the probability of a rate hike should decrease, and risk assets should rise. CME data confirms this—the probability of a rate hike in September dropped from 40% to 32%.
But the market split.
In the crypto world, Bitcoin $BTC is still hovering around 64,000, now almost like a stablecoin. It surged a bit before the data release, but immediately fell after the news.
What about Ethereum? $ETH has been fluctuating between 1,870-1,890. It spiked briefly after the data but then faded. Over 60,000 people were liquidated in the past 24 hours, ETF funds did not flow back, and 1,900 has become a short-term ceiling for ETH.
On the US stock side, it’s a completely different story. SanDisk $SNDK jumped 10% to 1,550, SK Hynix rose over 7%.
Same data, two worlds. This can’t be explained by economics alone.
Inside the Federal Reserve, there’s a big argument—Harker calls for a rate hike, Kaplan says wait and see. On the surface, it’s a data disagreement, but behind it are two political forces arm-wrestling. Whether to hike in September depends only half on economic data.
Crypto is stuck in an awkward position. Inflation is down, rate hike probability is down, so it should rise, but funds are stagnant. Because the market wants "rate cuts," not just "no rate hikes." No rate hikes just stop the bleeding; rate cuts are the transfusion. ETH has been stuck around 1,900 for almost two weeks; it gets sold off when it goes up—a typical wait for a catalyst. Once the rate cut expectation shifts from "whether to hike" to "when to cut," ETH’s elasticity will be much stronger than BTC’s, and a drop in staking yields will directly boost the ETH/BTC exchange rate.
SanDisk $SNDK’s 10% rise is superficially about AI, but behind it is the chip bill’s production shift expectations. The crypto world is still stuck in liquidity narratives, while the US stock market is already trading politics. The US stock market this week has indeed been quite aggressive.
On August 4th, the S&P 500 just crossed 7700 points; by August 13th, it had already surpassed 7800.
In just 7 trading days, it pushed up by 100 points like that.
The reason behind this is actually not hard to understand.
The latest PPI was lower than market expectations, further strengthening signals of cooling inflation, and the market's expectation for a policy shift in September has clearly heated up. Meanwhile, Citi raised its earnings forecast for the S&P 500 from $350 per share to $365 and set a target of 8100 points.
In other words, the market has simultaneously received three somewhat positive signals:
Inflationary pressure is easing, rate cut expectations are rising, and corporate earnings forecasts are still being revised upward.
When these three factors come together, capital naturally finds it easier to enter early rather than wait on the sidelines.
What’s even more interesting is the technology and memory sectors.
SanDisk $XSNDK continues to be strong, rising about 2.7% pre-market, reaching around $161. Investors’ expectations of high gross margins and continued shareholder returns are still being repriced by the market.
Memory-related companies like SK Hynix and Micron also followed suit and strengthened.
This indicates that capital is no longer just speculating on the performance of a single company, but on the improving outlook of the entire memory industry chain.
On the other hand, gold’s performance is somewhat intriguing.
Gold prices have fallen back from highs to around 4370.
By traditional logic, rising rate cut expectations should support gold, but this time gold did not continue upward and instead pulled back.
This may reveal a more important signal:
Risk appetite among investors is returning.
If the market believes inflation is cooling and the economy is not sharply slowing, then capital doesn’t need to fully retreat into safe-haven assets like gold.
So we see an interesting combination:
US stocks continuously hitting new highs, the memory sector strengthening, and gold retreating from its peak.
These three market moves actually point to the same macro logic — inflationary pressure is easing, but the economy is not yet bad enough to require full-scale risk aversion.
Now the S&P 500 has firmly stood near 7800, less than 3% away from 8000 points.
Adding the upward revisions to earnings forecasts and the potential passive capital demand from Reddit’s upcoming inclusion in the S&P 500, the market indeed still has momentum to push higher in the short term.
So I am not pessimistic about 8000 points.
But one thing I won’t change:
Being bullish does not mean chasing the top.
The current macro direction is indeed bullish, and risk appetite is improving, but prices have already moved a long way.
Especially at times like this, you must not forget that pullbacks are also part of an uptrend just because the market is strong.
My thinking is simple:
You can be bullish on direction, but there’s no need to rush on position.
If a normal pullback occurs later and key support is confirmed, the risk-reward ratio at that point might actually be more comfortable than chasing in now.
The market won’t go straight up just because everyone is bullish.
The truly comfortable opportunities often come not at the most crowded times, but after a pullback when the trend is reconfirmed.
$BTC $DOS
#标普收盘再创新高，8000点预期升温 #CPIPPIEaseFedSplit
US inflation data for July continues to cool down. PPI remained flat month-over-month, after a 0.1% decline in June; the year-over-year growth rate dropped from 5.5% to 4.7%. Commodity prices fell 0.7% month-over-month, largely due to declines in energy and food prices, while service prices rose 0.2%. Previously released July CPI showed a 3.4% year-over-year increase, with core CPI falling to 2.5%. Two consecutive softer inflation reports have clearly reduced market pricing for a September rate hike. Currently, the market estimates about a 68% probability that the Fed will maintain the 3.5%–3.75% interest rate range in September. The divergence remains focused on core PCE, with the market expecting core PCE to still be significantly above the long-term 2% target.
The July FOMC meeting itself saw a 9-to-3 vote, with three members supporting a rate hike. The core internal conflict within the Fed now is that inflation is declining, yet the current inflation level remains relatively high. August inflation and employment data will continue to influence the September decision.
Perhaps a rate cut is just around the corner. Haha, it's nice to fantasize.
But most likely, rates will be maintained The most dangerous moment on the chessboard is not when the king's wing is under a storm of attacks, but when you have carefully calculated a sacrifice twenty moves ahead, and your opponent resigns on the seventh move — this means all your deductions have turned into cold sweat in the palm of your hand in the face of reality.
Today's game opens with the opening moves posted on the financial report: Lumentum's revenue doubled by 9%, Coherent rose by 34%, Cisco increased by 18%, and Applied Materials also climbed by 25%. Each achievement is like a beautiful opening move in the Spanish Opening, with pawns directly targeting the center. However, the market's reaction is like an experienced opponent crossing out "beautiful" on the score sheet and writing "doubt" instead.
Chess players all know that the beauty of the opening is always just a facade. The real battle is whether your capital expenditures can be redeemed in the middle game without tears. When Cisco's full-year orders reach 9.3 billion, that's your promise to bring the bishop into Hegre's territory; but what the market sees is not the sharpness of the attack, but the empty pawn line on your kingside defense. The brighter the numbers, the thicker the clouds of doubt — just like when you push two pawns into the opponent's half to make way for your rook, but the castle behind you is left undefended.
The middle game has arrived. The stock prices of Coherent, Cisco, and Applied Materials collectively retreat; this is not a rout, this is redemption. The opponent voluntarily exchanges queens and drags the game into a prolonged battle. They are asking a brutally harsh question: how many pawns can your "growth" pieces capture and how many lines can they defend in the endgame? The answer is uncertain. But chess players never hand over their king to an unknown.
The guide is just the preset next move on the chess clock. Lumentum says next quarter will be between 1.23 billion and 1.27 billion; that is not a horn of confidence, that is your pawn in front of the king you have given up. Unfortunately, when you keep sacrificing pawns to maintain the position, a smart defender will not trade attacks with you; he will quietly count your remaining pawns, then retreat to the baseline, waiting for your offensive to exhaust itself before striking back.
Orders for AI infrastructure are a long asymmetric war. You will win many local battles, but the ultimate victory depends on whether you can move your forces to the opponent's secondary baseline without losing blood. What the market is doing now is dismantling every exaggerated long castling move: has your king really castled to a safe place? Are your rooks truly connected? Is your pawn formation still strong enough to support the endgame?
True grandmasters never care about how flashy your move is at any step. They watch like hawks the cost behind every move you make. They wait for only one moment: when you have to push the third pawn into the enemy camp to maintain the offensive, and your baseline reveals the only flaw — that is the moment of check.
On this chessboard, no one has truly reached that step yet. But the air is already filled with the sound of horses neighing. #AIInfraEarningsWatch #SandiskInvestorDayRally
In the short term, the core driver of SanDisk's recent rally is the Investor Day on August 13.
The company's FY2028–FY2030 long-term model is quite aggressive: revenue is expected to maintain mid-to-high double-digit growth, adjusted gross margin around 80%, adjusted operating margin about 75%, and free cash flow margin about 50%. More importantly, SanDisk aims to reduce the severe price cycles historically seen in the NAND industry through multi-year customer agreements. It has already signed such agreements with 8 customers, and by FY2028, these contracts are expected to cover about two-thirds of bit capacity. As a result, the market has begun to reassess the stability of SanDisk's future profitability. The stock price rose 13.7% in a single day after Investor Day, and increased about 25.8% over four trading days. Micron, Western Digital, and SK Hynix were also driven up in tandem.
The most important question is how long the approximately 80% gross margin can be maintained. Won't upstream costs rise? Is downstream bargaining power really that weak? Can profits be sustained?
$SNDK $BTC The biggest risk is not a drop, but "no rise despite good news": Tonight, $64,500 is the watershed Today BTC is trading around $63,000, with the Fear and Greed Index at only 37, still in the fear zone; more importantly, spot daily volume once dropped to about $1.19 billion, a low since 2019. The anomaly is: US inflation data is mild, US Treasury yields have fallen, and US stocks continue to strengthen—S&P 500 even hit a record high yesterday, but BTC did not follow. This indicates that theKorean stocks have rebounded over 22% in ten days, but I don't consider this a "technical bull market" confirmation of fundamentals.
Korean chip stocks still have mid-term potential, but this sharp rise is primarily a position recovery after deleveraging, and only secondly the realization of AI memory logic.
As of the close on August 14, the KOSPI stood at 6977.94 points, a cumulative rebound of about 24.8% from the low close on July 30; on that day Samsung Electronics rose 2.4%, SK Hynix rose 3.26%. The 20% threshold only proves that prices have returned, not that profits have caught up.
In the next phase, I’m only watching two things: whether HBM orders and DRAM and NAND contract prices can continue to push up profit expectations, and whether inventory will rebound after capacity expansion. SK Hynix has approved about 19.1 trillion KRW to build the M17 NAND factory, planning to start construction in February 2027 and activate the first clean room in December 2028. Supply will come sooner or later; the key is whether demand can keep up.
I don’t chase the 11th day; AI hardware total exposure does not exceed 5%, and Korean memory is initially set at 2%. Only if contract prices strengthen for two consecutive months and next quarter’s consensus profit expectations are raised simultaneously will I add 1% twice; if inventory rises and gross margin falls for two consecutive quarters, I will halve the position.
This is the same as in crypto: a 20% rise means funds have returned, but sustained gains depend on profits staying.
#韩股十日反弹逾22%，芯片股领涨
$SAMSUNG $SKHYNIX #CPI and PPI Cool Down Simultaneously, Interest Rate Hike Divergence Widens
July inflation data in the country both weakened: CPI year-on-year fell to 3.4%, core CPI dropped to 2.5%; PPI month-on-month remained flat, year-on-year slowed to 4.7%, all below expectations. After the data release, traders quickly lowered their bets on rate hikes this year—probability of no change in September rose to about 67%, with year-end tightening expectations only about 23 basis points, less than a full rate hike. The focus of monetary policy is shifting from "whether to raise rates" to "how long to maintain them."
However, the divergence has not narrowed but rather become more apparent. Cleveland Fed President Mester turned hawkish against the trend, stating inflation remains above 3% and widespread, calling for "rate hikes now." Last month, she and two other officials cast dissenting votes, the most since 2016. New Chair Waller's Jackson Hole speech, along with August PCE and employment data, will determine the tug-of-war between "dovish data" and "hawkish voices." Middle East oil prices and AI capital expenditure inflation remain variables hanging over the cooling narrative. $BTC $ETH $SNDK $SNDK surges, is it a bubble game or a starting point for valuation reappraisal?
$SNDK surges again. Many in the market are still judging this round of the market using the traditional storage cycle framework, but my core judgment remains unchanged: it cannot be simply defined as a cyclical stock; AI storage is reshaping the valuation system.
The underlying support for this round of rise comes from the continuous expansion of AI inference scenarios driving incremental enterprise-level NAND, the landing of long-term locked orders, and product structure optimization pushing the gross margin midpoint upward. The continuous strengthening of the stock price essentially reflects capital gradually pricing in this brand-new growth logic.
At present, I maintain a moderately bullish view, not simply gambling on sentiment. The short-term stock price rise is indeed quite fast, but I will not choose to exit solely based on the superficial reason of "excessive gains." The best strategy at this stage is to wait for earnings data to verify the industry logic. As long as AI storage demand continues to exceed expectations and long-term NAND orders steadily land, $SNDK still has room for further valuation reappraisal.
In my view, the biggest risk in the current position is not a short-term stock price pullback, but the inability of fundamental realization to match the market's continuously rising expectations. As long as the core mid-to-long-term logic is not broken, I maintain the original bullish judgment unchanged. The construction elevator at the White House East Wing is offering a direct route to the top-level design room to select traders for $100,000 a month. Ordinary investors stand outside the construction site's iron fence, waiting for the blueprints to be released next month—while those inside the elevator are already debating whether to raise the foundation half a meter higher.
I've been in construction for twenty-five years and have seen many bold blueprints. What truly keeps a building standing through an 8-magnitude earthquake are the dozens of invisible friction piles underground, the hidden steel plates inside the shear walls, and the 28 days of patience during concrete curing. Every tariff, war, and monetary policy statement from Trump is essentially a "design change notice" for the entire financial market. A single change order can force a complete recalculation of the building's load system and shift the original load-bearing wall by fifteen centimeters—while ordinary investors are still waiting for the construction team to hold a meeting, and haven't even entered the conference room.
What Truth API does is precisely turn this change notice into a real-time paid streaming service. Trading institutions pay $100,000 a month for millisecond-level access. What does this mean? The site is still piling foundations, but they already know which room will be converted into a vault and which wall will have a blast door installed. Why? Because their accounts have the highest permissions in the design institute's computer. Ordinary investors are still reading blueprints in the dust, while they are already standing on the tower crane directing the hoisting—even the concrete grade has been pre-approved by them.
As an architect, I deeply understand the fatal damage information asymmetry causes to structures. Once design drawings leak early to material suppliers, the entire project's bidding system becomes distorted. The market is the same. Presidential statements were once the drumbeats in a public space, but now they are muffled by low-latency APIs into secret whispers for a select few. Ordinary investors stare at candlestick charts doing technical analysis, but that's just construction debris spat out after being crushed by others—you think you're reading blueprints, but you're actually reading their dried wall plaster.
The XCH asset is very much like the inclinometer on a construction site. On the surface, it follows the steady subsidence of the US stock market, but the internal stress has long exceeded design limits. You think you're observing market linkage, but you're only observing the transmission path of information hierarchy within the structure. When paid APIs become part of the market's underlying protocol, price curves are no longer naturally grown bamboo but bonsai trees deliberately watered—no matter how beautiful the shape, the roots never reach the real soil.
Building codes emphasize that all load-bearing components must undergo equal strength testing. The existence of low-latency APIs allows a few participants to obtain un-cured test block strength reports. This is not data service; it's selling the mix ratio to the contractor before the load-bearing wall is poured. When policy statements are openly priced, the market's "public space" becomes a facade—the so-called transparent design review meetings are just model sand tables for outsiders.
If news about load-bearing walls can be priced by the millisecond, then I suggest you look up—the foundation of this skyscraper is actually hanging in the clouds. #trumptruthapilawsuit