avatar
Fiona’s daily news
Follow
Posts
Signal Clone Analysis
avatar
Fiona’s daily news
04-01
The market always swings between the best and worst. Recent stock market performance and attitudes towards memory stocks seem to prove this. It reminds me of a chart I saw a long time ago: the fundamentals of a stock remain unchanged, but the price fluctuates wildly with changing market sentiment. Recently, memory stocks have plummeted, and many posts are calling memory a scam, like many other Ponzi schemes/memes. I find this ridiculous. Take Micron as an example; its latest quarterly revenue was $23.8 billion. SK and Samsung's quarterly revenue also reached $20-30 billion. This is real money; very few companies in the world can earn that much in a single quarter. Is this also a scam? Is this also a meme? The drop might be the original sin. The demand for memory is real, and cycles exist, but this round of cycle drivers has shifted more from consumer electronics to enterprise data center AI demand. Of course, this demand also has cycles. However, we can see from the Capex of various companies that the cycle is still in a period of frenzied investment and growth. More concerns lie with AI, the tightening of cash flow for various companies, the market disruption caused by war, the huge fluctuations in oil prices, and the withdrawal of funds from the Middle East. However, I am still betting on the future. If AI is real and represents a booming future, then I still believe in memory stocks this year. This is a personal decision, not investment advice.
SK
0%
avatar
Fiona’s daily news
03-25
A few words about Google's TurboQuant and Micron/memory. Following the release of TurboQuant today, all three stocks ($MU $WDC $SNDK) fell. The market consensus is that this new technology is impacting memory chip stocks. What is TurboQuant? It is an algorithm that compresses AI's "memory usage" by 6 times, increases inference speed by up to 8 times, and almost does not lose accuracy. Therefore, it can drastically change the cost structure of AI, enabling a significant breakthrough in long-context processing capabilities. This naturally means that what we previously perceived as a memory gap may be greatly reduced. Inevitably, this will have a huge short-term impact on memory stocks. However, in the long run, lower costs make a surge in applications possible. The decrease in unit memory demand coupled with an increase in total demand essentially solves the efficiency problem. $MU All items were sold out this year, and despite the overall growth in AI demand, memory demand continues to grow, at most with a slower growth curve. High profit margins, high growth, and a good price—I still bought them. I buy the dips some at the bottom to see how it goes 👀 The risk of this is still quite high, this is not investment advice.
TA
0.77%
avatar
Fiona’s daily news
03-18
Analyzing the Earnings Report with Key Questions | _2024111120230_ | What are the key takeaways from this earnings report? 1️⃣ Will management further quantify HBM order visibility? For example, the target range for 2026 shipment volume and revenue share? Will analysts press them on the specific progress of the NVDA platform during Q&A, thus signaling a "let orders speak for themselves" approach? 2️⃣ Has there been a significant change in HBM yield? When will HBM4 enter mass production? 3️⃣ Next quarter's guidance is more important than this quarter's results! What will management say about next quarter's revenue/GM/supply tightness? 4️⃣ Where did capital expenditure/CapEx go? Can it be converted into revenue? How long will this high-intensity investment continue? 5️⃣ Will HBM continue to sell-out? If Micron confirms that 2027 HBM orders are starting to lock in, the market will repric the memory sector. 6️⃣ What is the fundamental situation of DRAM/NAND? 7️⃣ Does HBM play the role of a "protagonist" or a "supporter" in the financial statements? What is HBM's contribution to the financial statements?
loading indicator
Loading..