I've always had a conflicted feeling about the CRV coin. Saying that no one uses Curve anymore isn't quite right. It still manages over a billion dollars in funds and generated about $2.06 million in fees in the last 30 days. Among a bunch of old coins left with only communities and roadmaps, this counts as a solid business. But to say CRV is severely undervalued, I don't dare to conclude that quickly. CRV is currently around $0.26, up about 27% in a week. Many people see that it used to be over ten dollars and think it's ridiculously cheap now. But a low coin price doesn't equal a low valuation. CRV currently has a circulating supply of about 1.547 billion, a total supply of about 2.409 billion, and a theoretical max supply of 3.03 billion. Comparing today's CRV price directly with historical prices under early supply conditions isn't very meaningful. I think the real problem with CRV has never been whether Curve has business. It's whether the money Curve earns can sustainably translate into CRV's value. Simply holding CRV doesn't automatically grant protocol revenue. You need to lock CRV into veCRV to gain governance rights and corresponding fees; the maximum lock period can be up to four years. The advantage of this design is that those willing to lock long-term are more tightly aligned with the protocol's interests. The downside is obvious: the mechanism is too complex, making it hard for ordinary people to have a "buy and hold" experience. Plus, CRV emissions have always existed; while the protocol generates income, it also needs to incentivize liquidity with tokens, which easily leads to a situation where the business is good but the coin price disappoints. So now when I look at CRV, I'm not too concerned about when it will return to $1. I'm more focused on three things: Whether Curve's real income can continue to grow; Whether crvUSD can become a stable source of business; Whether the growth rate of protocol income can gradually surpass token incentives and supply pressure. Also, security risks can't be ignored. In March 2026, the sDOLA/crvUSD LlamaLend market still experienced an oracle manipulation event, causing about 822,500 crvUSD borrower equity losses. My view is simple: Curve is not a dead protocol, and CRV is not a bargain you can understand at a glance. It has products, users, and income. But it also has emissions, lock-up thresholds, complex governance, and security risks. If in the future Curve can rely less on "issuing tokens for liquidity" and more on trading fees, lending, and crvUSD to sustain itself, CRV might truly see a value reappraisal. Until then, I'm willing to pay attention. But I won't automatically think that the current 26 cents is a golden opportunity just because it was over ten dollars before. The biggest fear for old projects isn't that no one remembers them. It's that everyone remembers how glorious they once were but no one seriously calculates how much they actually earn today. $CRV #Curve #DeFi #Ethereum #CryptoMarket

Brothers Recently, discussions about CORE's price have increased again. Some calculate market cap based on total supply, some benchmark against other public chains, and some have already planned the gains for the next few bull markets. These calculations look very professional, but they all share one common problem: They first decide on the desired price, then look for reasons to support that price. What I care about now are a few other things. First, how much sustained capital can BTC staking actually bring to CORE, not just how much is locked during the activity period. Second, whether the on-chain applications have real users. Not how many wallet addresses have increased, but whether a person will be willing to come back next month after using it once. Third, whether the fees and income generated by the ecosystem can ultimately be converted into actual demand for CORE. If these three questions gradually get answers, the price won't need anyone to predict it; the market will naturally reprice. If these data don't come out for a long time, then shouting $1, $5, or $10 is just giving emotions a number. I'm not pessimistic about CORE. Precisely because I am still paying attention, I am even less willing to numb myself with an exaggerated target price. First, see if the product can retain users. Then see if the capital can form a cycle. Only then is it time to discuss the price. Rising needs time to verify, but falling never needs advance notice. Count a little less on how much you can earn in the future. Think a little more about whether you can bear it if you are wrong. $CORE #CoreDAO #BitcoinEcosystem #CryptoMarket

I've been watching SPCX these past few days. Honestly, what surprised me the most wasn't that it went up, but that it didn't get crushed after the lock-up period ended. The market was waiting for this August lock-up release. The logic is simple: It was so hyped at IPO, it surged so much before, and now a large batch of shares can be sold—employees and early investors would want to cash out, right? But the market was expecting fireworks, and it turned out the shorts got blown up instead. SPCX jumped nearly 16% on Friday. I went back and re-read the financials and found this stock quite interesting now. Q2 revenue is about $7.8 billion, with adjusted EBITDA of $3.5 billion. But on the other hand? Net profit is still negative, losing about $540 million. So people buying SPCX now aren't really buying how much it earns today. They're buying the future. Starlink, rockets, satellite communications, plus AI—the market is already pricing it as a super infrastructure company. And that's exactly where the problem lies. I really like SpaceX as a company, but "liking the company" and "liking the price" are two completely different things. At $225, the market basically told all the stories it could. After it dropped to just over $100, people started seriously crunching the numbers. Now that it's pulled back to around $130, I think the truly interesting phase for SPCX is just beginning. Because the most panic-inducing lock-up expectations have passed for now, but valuation issues remain. Going forward, I won't care much if it goes up 10% or down 10% in a day. What I want to watch are three things: Whether Starlink can continue its rapid growth; Whether the AI business can truly contribute profits; And whether such massive capital expenditures can eventually turn into cash flow. If two of these three come through, SPCX might look expensive today but won't be in a few years. But if AI ends up just being a valuation story and Starlink's growth slows, then this price isn't cheap at all. So my current stance on SPCX is simple: I am very optimistic about the company, I'm starting to be interested in the stock, but I won't chase it just because it spikes one day. The biggest mistake SpaceX investors make is forgetting that stocks still need to be accounted for, because Musk, rockets, and Starlink are so sexy. A good company doesn't necessarily mean a good price. But when a good company drops to a good price, I will definitely take a serious look. $SPCX

AI demand is exploding? Nvidia's revenue grows 85%, Microsoft's AI business grows 123% The latest earnings report shows that AI demand has not cooled down significantly. Nvidia's Q1 revenue reached $81.6 billion, up 85% year-over-year; data center revenue was $75.2 billion, up 92% year-over-year, and the company expects next quarter revenue to be about $91 billion. What truly makes the market reassess the AI trend is not just chip sales. Microsoft's AI business annualized revenue has exceeded $37 billion, up 123% year-over-year; Azure demand still exceeds existing computing capacity. Microsoft expects capital expenditures of about $190 billion in 2026, while Amazon plans to invest about $200 billion to continue expanding data centers, developing chips, and AI infrastructure. In the past, many people simply understood the AI trend as "buy Nvidia," but now capital is reevaluating the entire industry chain: chips, optical modules, storage, power, cloud computing, and AI applications that can ultimately charge enterprises. However, the larger the capital expenditure, the greater the future depreciation pressure. Microsoft's capital expenditure for the quarter reached $31.9 billion, and the company's gross margin declined year-over-year, partly due to investments in AI infrastructure and rising costs of AI product usage. This means the next phase of AI stocks in the US market will no longer see all companies rising together. What the market really wants to see is who can turn expensive GPUs and data centers into sustained revenue, cash flow, and higher customer stickiness. Computing power proves AI demand exists; profits prove this round of investment is worthwhile.

Is OKB demand starting to materialize? X Layer's weekly fees increased by 66%, but chain-level Gas fees are still under a thousand dollars X Layer's recent on-chain data shows a clear rebound. According to DefiLlama data, the total fees in the X Layer ecosystem over 24 hours are about $4,474, with a 30-day cumulative total of approximately $106,400, representing a 65.8% week-over-week increase. Among these, Gas fees generated by X Layer itself are about $883 per day, higher than before but still relatively small in absolute terms. What truly deserves attention is not just the fee growth, but that OKB is shifting from being an "exchange platform token" to a system resource for X Layer. OKX previously fixed the total supply of OKB at 21 million tokens and clearly stated that OKB is the only native Gas token of X Layer. The latest Exchange OS plan also requires market deployers to stake OKB in order to create trading venues or list trading pairs. In the past, the market mainly valued OKB based on OKX exchange's brand and scale, but now OKX is trying to add a new value path: developers need OKB to deploy markets, users consume OKB by using X Layer, and trading, payments, and RWA applications collectively form on-chain demand. However, Exchange OS is still in the early stages of implementation, and a week's fee growth alone cannot prove the ecosystem has exploded. What truly determines whether OKB can be revalued are the number of third-party developers, the actual amount of OKB staked, on-chain fees, and whether users continue to use it without subsidies. Locked supply is just design; sustained usage is the product. If these data continue to grow for several months, only then can OKB truly upgrade from an "exchange token" to a "resource token of an on-chain financial operating system." $OKB #xlayer #ExchangeOS #OKB total supply fixed at 21 million