
唐华斑竹
唐华斑竹:著名KOL,区块链研究者,财经学者。 2013年开始加密货币交易,拥有丰富的交易经验,左侧交易者,风格偏稳健,注重强安全边际,坚守既定策略和纪律,追求长期稳定收益和高胜率。 欢迎志同道合的朋友加入一起交流,一起成长!
1.1KFollowing
825followers
Feed
Feed
Still want to learn from Rich Dad? $1.2 billion in debt
On September 2, citing the New York Post, Robert Kiyosaki, author of "Rich Dad Poor Dad," recently revealed in an interview that he is carrying $1.2 billion in debt, quickly drawing market attention.
However, this approximately $1.2 billion debt actually comes from mortgage loans on apartment buildings, not from his long-promoted investments in Bitcoin and gold.
Kiyosaki's ex-wife and business partner further clarified that this debt is a loan shared with partners, involving about 1,500 apartment units, and Kiyosaki's personal financial exposure is far less than this apparent figure.
It is worth noting that although Kiyosaki has publicly advocated holding gold and Bitcoin as tools to hedge against inflation multiple times, these digital asset holdings have not been used as collateral for any loans.
In fact, while Kiyosaki heavily promotes Bitcoin, the real collateral is physical real estate and rental income, which also reflects that Bitcoin is not yet widely accepted as collateral in the real financial system like real estate.
However, critics point out that this high-leverage operation model can amplify gains during market upswings but will generate huge losses once the housing price rally ends. This risk is particularly worth attention in the current economic environment.
Interestingly, Kiyosaki has publicly warned that cheap credit will destroy the financial system, yet his own investment portfolio heavily relies on credit to sustain itself. This discrepancy between words and practice has also caused market observers to question his investment advice.
In fact, Kiyosaki has a prior bankruptcy record; his company filed for bankruptcy protection in 2012 due to a lawsuit loss. This experience shows that even authorities in financial education do not always make flawless personal investment decisions. bitcoin:native

Chengdu really has a lot of food, with spicy rabbit heads and fatty intestine noodles all over the streets. It made me want to eat so badly, but unfortunately my companion doesn't eat spicy food, so we had no choice but to settle for five bland dishes, costing less than 18U.
There are small taverns and bars everywhere, with live singers, folk music, and street performances, all quite distinctive. At night, we ate spicy hotpot skewers but didn't drink; we'll make up for it tomorrow.




Zhang Lei's "LONGi Moment": Billions in Losses, When the "Friend of Time" Is Devoured by Time
On the last day of August 2026, LONGi Green Energy's stock closed at ¥12.07, with a market value under 100 billion. Yet five years ago that winter, this company was the brightest photovoltaic star in the A-share market, with a market cap surpassing 540 billion. When Hillhouse Capital acquired 6% of shares via a ¥15.8 billion agreement, the market hailed it as "the best example of value investing in China."
Today, Hillhouse has disappeared from LONGi's top ten shareholders list. Market estimates put the actual net loss on this investment between ¥7 billion and ¥10 billion. Over six years, from a "textbook investment" to a "textbook exit," what exactly happened?
The Cruelty of Timing: Not "Overpaying," but "Buying at the Cycle's Peak"
One easily overlooked fact in reviewing this deal: Hillhouse's entry price of ¥70 per share came after LONGi's stock had already nearly tripled that year. Even accounting for two subsequent 10-for-4 stock splits, the adjusted cost basis was about ¥35.71 per share, and at the November 2021 peak, unrealized gains exceeded ¥10 billion.
But the problem was that the photovoltaic industry's profit peak occurred in 2022—net profit attributable to parent company was ¥14.8 billion, with a gross margin around 20%—and declined steadily thereafter. Hillhouse bought near the cycle top valuation, expecting LONGi to grow through cycles like a consumer sector leader. This is like buying down jacket futures in midsummer, betting on a colder, longer winter, but failing to anticipate the "warm winter" caused by capacity surges.
Why Was LONGi's "Moat" Vulnerable to the Cycle?
Zhang Lei emphasized structural competitive advantages in "Value." LONGi indeed has the world's lowest non-silicon cost for wafers, leading monocrystalline technology, and a solid financial strategy. But the nature of photovoltaic manufacturing means these advantages lack "pricing power" amid overcapacity.
From 2020 to 2023, the industry invested over ¥3 trillion in capacity expansion, causing silicon material, wafers, cells, and modules capacity to far exceed global demand. When module prices fall below cost, even the lowest-cost companies only "lose the least," not "make profits." LONGi's module and cell gross margins approached zero in the first half of 2026, and wafer business even showed negative gross margin—losing money on every wafer sold.
More critically, the wavering technology path hurt. The industry's shift from PERC to TOPCon exceeded expectations, while LONGi bet on the BC (back contact) route. In 2024, HPBC 1.0 underperformed in performance and cost, causing large inventory write-downs and production halts, directly dragging down results. Founder Li Zhenguo resigned as general manager in 2025, retreating to R&D—a "founder stepping back" signal more symbolic than any financial report figure.
Controversy over Securities Lending: From "Beacon" to "Under Investigation" and Reputation Collapse
In 2023, Hillhouse's HHLR lent 64.48 million shares through securities lending, reducing its stake from 5.85% to 5.00%, then further to 4.98% without announcement. This "precise calculation" triggered regulatory investigation, forcing Hillhouse to proactively repurchase shares above 5% in 2024 to comply.
But the reputational damage was irreversible. An institution once known for "long-termism" and regarded by countless retail investors as a value investing benchmark was suspected of exploiting regulatory loopholes to "secretly transfer shares." This not only cast doubt on Hillhouse's integrity but exposed private equity's fundamental dilemma: when LP redemption pressure, performance evaluation, and fund duration overlap, "long-term" often must yield to "liquidity."
Who Is Responsible for the Billions in Losses? A More Structural Perspective
Attributing it simply to "Zhang Lei's misjudgment" is too crude. In fact, Hillhouse's failure in LONGi is an intersection of three mismatches:
Mismatch of valuation and cycle: entering at the industry's peak but pricing with a "ten-year perspective."
Mismatch of asset nature and investment framework: valuing a manufacturing cyclical stock as a consumer or tech growth stock, overestimating the depth of the cost advantage "moat."
Mismatch of capital nature and holding period: private equity fund duration and exit pressure prevent truly waiting for the next cycle recovery.
A useful comparison is Hillhouse's investment in Gree—¥41.6 billion entry, similarly over ¥10 billion unrealized loss, but Gree still benefits from brand premium, dividend cash flow, and stable air conditioner replacement demand as buffers. Photovoltaic modules lack brand stickiness; downstream customers only recognize price, so when overcapacity hits, no leader can single-handedly resist industry-wide losses.
LONGi's "Post-Hillhouse Era": Structural Dawn and a Long Tunnel
After Hillhouse's exit, LONGi's 2026 interim report remains bleak: revenue ¥27 billion, down 17.6% year-on-year; net loss ¥3.68 billion, loss widened 43%; operating cash flow negative ¥5.8 billion. But behind these figures, two structural changes deserve attention:
BC module sales reached 19.55 GW, up 125% year-on-year, shipment share jumped above 65%—Li Zhenguo's six-year technology bet finally scaling;
Second quarter loss narrowed quarter-on-quarter, single-quarter gross margin turned positive at 3.13%, overseas module sales grew 26%, overseas revenue share exceeded 65%.
With over ¥50 billion in cash on hand, LONGi has a "thicker coat" than most peers in this industry winter. If BC technology continues to reduce costs and improve efficiency over the next two to three years, reaching or surpassing TOPCon competition, then the capital that exited today may indeed have sold at "the darkest moment before dawn."
The Friend of Time Ultimately Lost to Time
Hillhouse's LONGi case is essentially a stress test of "long-termism" on Chinese manufacturing cyclical stocks. The result: when industry cycle length exceeds fund duration, when technology uncertainty overturns cost advantages, when regulatory and public opinion compress operational space, even the smartest capital can only accept losses and exit.
But this does not mean "long-termism" itself is invalid. It reminds investors that the premise of long-termism is truly being able to wait for that "long term." For institutional capital with exit deadlines and performance pressure, discussing a ten-year vision at the cycle peak is an overly expensive luxury.
LONGi remains, Li Zhenguo remains in the lab, BC is scaling. Hillhouse has left, losses realized. The final answer on the "right or wrong" of this investment may only be seen when the next photovoltaic cycle arrives and we look back at today's ¥12.07 stock price—to know who truly was the "friend of time."

The biggest worry in the Chinese film industry has arrived. "Niu Lai" not only hasn't been taken down, but its run has been extended, and now the box office is close to 60 million.
What's even funnier is that more and more "Niu Lai" films are flooding into cinemas. For example, the third picture, "Unblinking," was entirely handled by one person, who maximized solo operation, even surpassing the original creators of "Niu Lai," who were two people after all. #NiuLai

Is JT's troll army making a move? Everywhere there are people bashing Justin Sun, saying he was put on some kind of fox-hunting list. I think that's impossible; he's in Hong Kong (Sun even posted photos holding the TRON blockchain hash with Victoria Harbour). If they wanted to arrest him, would they need a list? They would just intercept him directly. Hong Kong has already returned to China.
Think about it a bit: this time Hong Kong is hosting the Bitcoin Asia conference in a high-profile way. Which of the crypto big shots coming here doesn't have border control risks? They all dare to come to Hong Kong openly to participate in the crypto conference, openly appearing on camera. This already says a lot. These people are not stupid, right? The stupid ones are the rumor spreaders.
Yesterday I had dinner with a relative and we talked about the egg incident. She doesn't understand what cryptocurrency is but still cursed Sun's virtual currency as a scam. I told her to look at the data first: TRON, in its 9 years of existence, has increased 226 times; in the last 5 years, it increased 4.7 times; in the last 3 years, 4.5 times; and in the last year, 1.2 times.
Of course, I don't encourage everyone to play with virtual coins because the volatility is too high and most people can't bear it. But if you say Sun cheated you out of money, that's a bit far-fetched. You can't just talk about getting hit without mentioning the gains.
@justinsuntron @HTX_Molly #TRONEcoStar


Does everyone still remember when Lao Tang called to short Yushu Technology on its IPO day? Check the post below from that time. Unfortunately, Lao Tang's prediction came true: Yushu Technology's stock price halved in 9 days. Who pocketed 100 billion?
On August 31, A-share "the first humanoid robot stock" Yushu Technology plunged again by over 4%, once falling below ¥560.
From the peak of ¥1100 on the first trading day, August 19, in just 9 trading days, the stock price nearly halved, wiping out over 200 billion in market value.
On the day it opened at ¥1100, the highest profit for one lot was ¥470,000, causing a frenzy online. Looking back now, that ¥1100 opening price was the peak right from the start.
Who made the money?
200 billion in market value evaporated, but money doesn't just disappear; it shifted from one group of people’s pockets to another’s.
The biggest winner is, of course, founder Wang Xingxing. Even with the stock price falling to current levels, his personal holdings are still worth over 50 billion. On the IPO day, when market value surged to 444.9 billion, he briefly became the richest person born in the 90s, with holdings worth over 100 billion.
Early investors profited even more. Meituan’s holdings are worth about 16.4 billion, Sequoia China 13.4 billion, and Matrix Partners 9.2 billion. These institutions’ initial investment costs were far below the current stock price.
Underwriter CITIC Securities also made a fortune. Sponsorship and underwriting fees exceeded 140 million, its subsidiary Jinshi Investment’s holdings gained over 8.9 billion in unrealized profits, and combined with co-investment returns, total unrealized gains exceeded 9.3 billion.
Retail investors who won the online IPO lottery also got a piece of the pie; the winning rate was only 0.018%, setting a record low on the STAR Market.
So who lost?
Retail investors chasing the high at the open. Some bought heavily at ¥1100 and lost nearly 47% in 8 days. Some bought at ¥897 and lost ¥610,000 in 8 days. Others traded in and out three times within 8 days, losing a total of ¥760,000.
97% of holders are at a loss.
Who priced Yushu?
Yushu Technology’s issue price was ¥150.80, corresponding to a price-to-earnings ratio (P/E) of 219.23 times. The general equipment manufacturing industry it belongs to has an average P/E of only 38.56 times.
The issue price was 5.7 times the industry average.
This price was not set by Yushu itself but formed through institutional inquiry and bidding. 285 institutions participated in pricing, all raising bids to grab shares, pushing the issue P/E to 219 times, far exceeding the market’s previous estimate of about ¥104 and a 40 billion market cap range.
The IPO day was even more absurd, opening directly at ¥1100, with a dynamic P/E exceeding 800 times at one point.
Nomura Securities gave a "buy" rating with a target price of only ¥370. CCB International’s target price was ¥269. The lead underwriter CITIC Securities gave a reasonable market value range of 50.6 billion to 55.9 billion.
Yet Yushu Technology’s peak market value on IPO day was 444.9 billion.
Do institutions really think "buy" in their hearts?
Who is overfishing?
Back to a basic question: Is a company with 2025 revenue of 1.699 billion and net profit of 278 million worth 440 billion?
Yushu Technology’s issue P/E is 219 times; Changxin Technology’s issue P/E is 309 times. Numerically, Changxin is higher.
But these two numbers mean completely different things.
Changxin Technology’s 2025 revenue is 61.8 billion, net profit 7.144 billion. In the first half of 2026, revenue was 150.3 billion, and net profit excluding non-recurring items was 78.793 billion. It had losses for years before, just turning profitable in 2025, so the issue P/E looks high, but based on annualized 2026 H1 profits, the actual P/E has dropped to double digits.
Yushu Technology is the opposite. In 2025, revenue grew 332%, net profit excluding non-recurring items grew 753%, but from a very low base. By the first half of 2026, revenue growth slowed sharply to 48.5%, net profit excluding non-recurring items dropped 19.34% year-on-year, showing signs of revenue growth without profit growth.
More importantly, customer structure: In the first three quarters of 2025, 73.6% of Yushu’s humanoid robot revenue came from research and education scenarios, with industrial or commercial scenarios only about 5%.
Research purchases R&D platforms, not production tools, so they won’t have continuous large-scale repurchases like industrial equipment.
Based on 2025 net profit of 278 million, even if the stock price falls to ¥560, the P/E is still over 877 times. Even if profits double every year, it would take nine years to justify the current valuation.
Changxin Technology’s stock price has been relatively stable after listing, peaking at ¥61.8, closing at ¥58.6 on August 28, with no wild fluctuations. The reason is simple: the issue price of ¥8.66 corresponds to a 579 billion market cap, which is reasonable for a company earning 78.7 billion in half a year.
Yushu Technology earned 244 million in half a year, with an issue price corresponding to a 61 billion market cap, but was hyped to 444.9 billion on the first day.
This is not pricing; it’s pricing plus another round of hype.
Who bears the cost?
The company raised 6 billion in its IPO but carries a 440 billion market cap bubble.
The math is extremely unfavorable.
A small amount of financing money exchanged for a valuation over 100 times P/E, endless stock price volatility risk, and huge losses for retail investors in the secondary market. The company got a little money, early capital locked in huge profits, and retail investors took on all the bubble.
More realistic pressure lies ahead. On August 19, 2027, pre-IPO restricted shares held by original shareholders of Yushu Technology will be massively unlocked, expected to release 228.7 million shares, accounting for 56.56% of total shares.
Those early shareholders who bought at just a few or tens of yuan, what do you think they will do on the unlock day?
Is Yushu Technology a good company? Technically, yes. It is the world’s number one humanoid robot shipper, with core components self-developed and produced, a 60% gross margin, and one of the few domestic humanoid robot companies achieving scaled profitability.
But a good company does not equal a good stock, much less a good price.
The new stock pricing "overfished the pond," squeezing the growth space of the next ten years all at once, likely leaving the secondary market with a long road of value reversion.


If not shorted now, then when? The bubble stocks in the big A market, a remote-controlled toy company, it's impossible for it not to fall. Those who won the new stock subscription made over 400,000 yuan per share; are you going to cash out and keep it for the New Year? Where can you short it? Of course, on Gate! Look here, I've already shorted it, everyone do as you please:
Yushu surged 629% on its first day of listing: opening price 1100 yuan, market value approaching 450 billion yuan
Yushu Technology officially landed on the STAR Market, with an opening price of 1100 yuan on the first day of listing, a surge of 629.44% compared to the issue price of 150.80 yuan. Based on the total shares after issuance, the company's market value has surged to about 444.9 billion yuan.
Yushu's IPO raised about 6.1 billion yuan, with retail subscriptions exceeding 8000 times before issuance, setting a record on the Shanghai Stock Exchange STAR Market. The market had already bet on a big rise before listing, but the first-day performance still significantly exceeded off-market expectations.
Sometimes you just have to marvel at how bizarrely diverse people are, even though they are man-made. The differences between people are really huge, and even siblings can have completely different personalities.
For example, some people can't stay indoors for even a minute; they always want to go out hiking, crossing rivers, or exercising nonstop. Others can't understand this at all and prefer to stay at home, eating well, sleeping well, and playing online games. Actually, everyone just does their own thing, and there's no right or wrong.
The real problem is when two very different people insist on living together. That becomes exhausting, and usually one person ends up giving up their hobbies to accommodate the other, possibly for a lifetime. When young, you tolerate it because the other looks beautiful, but over time, as you grow old and aesthetic fatigue sets in, you no longer want to accommodate, and conflicts arise.
Therefore, I strongly oppose the idea of finding a partner who complements you. In reality, there is no such thing as complementarity; it's just one side accommodating the other. It's much simpler, more comfortable, and more harmonious for two similar people to be together. Neither has to make an effort to accommodate or please the other. You love diving and hiking, and I do too; you love staying home playing games, and I do too. That is true harmony and bliss.
So remember, when looking for a partner, be sure to find someone like yourself. Never look for a complementary person, absolutely not

Yesterday, I had the opportunity to watch this year's college student drama showcase live at the Beijing People's Art Theatre Small Stage. It was really great and exciting, with a full house and a very lively atmosphere. Notice the dining photo below; I think I saw a familiar face from the crypto circle. Can anyone recognize them?
The schools performing that day included Capital Normal University, Renmin University, Tianjin Normal University, Beijing Normal University, and the Central Academy of Drama. The young students' performances were passionate and full of infectious energy.
The Capital Normal University Drama Troupe's "Snow White Pig Head" left a deep impression on me. It is adapted from a short story by Su Tong, with a plot full of twists and turns, sincere and natural acting, and very professional directing, lighting, and scene transitions, truly capturing the essence of the original work.
I have to say, the performance effect at the Beijing People's Art Theatre Small Stage is amazing. The lighting makes it feel like watching a movie, with clear and bright facial colors, and the sound effects are even better. The actors perform without microphones, and every line is heard clearly throughout the venue. It was such an enjoyable experience. No wonder it is a top national stage located in the heart of Beijing. I recommend everyone to experience it if they have the chance



Grok Bot lowered to $20 Cursor Pro: Weekly quota reset for all users
Grok Bot has comprehensively lowered the usage threshold. The official announcement states that all SuperGrok and Cursor Pro subscription users can now use Grok Bot, and the weekly quota for existing users' Grok Bot has been reset. Cursor Pro currently costs only $20 per month.
Grok Bot is a set of AI Agents that can run long-term. Each Bot has its own cloud computer, capable of logging into websites, operating applications, handling files, and terminals. Even if the user turns off their computer, it can continue working and allows multiple Bots to work simultaneously in a division of labor.
On August 21, the official expanded Grok Bot to SuperGrok Plus, Cursor Pro+, and Cursor Teams; in less than a week, the minimum threshold was lowered again to Cursor Pro. Grok Bot itself has an independent weekly usage quota and does not directly consume the regular Cursor model quota. However, if it further calls the Cursor Cloud Agent, the latter will still count towards the normal Cursor quota.
Many Cursor Pro users believe this is equivalent to getting an additional long-term Agent set for free on top of the original programming package. However, Grok Bot is still in Beta, with some users reporting instability in web operations and login persistence, and some Pro users have not yet received the new permissions.

