$Applied Materials.US, I have such a love-hate relationship with this stock. What I love is that it's the most down-to-earth among this wave of "shovel" stocks in AI infrastructure. It's not as flashy as NVIDIA, doesn't spin fancy stories about quantum computing or robots, just quietly sells wafer fab equipment. A year's work resulted in a +159% return for shareholders. Every time I see that number, I want to slap myself for not holding on back then. What I hate is that every time I want to buy it, I think it's too expensive. I looked at it in March, thought I'd wait for a pullback, and then it went up another ~30% in 90 days. Chasing a bus you can never catch—that's exactly how it feels. The quarterly report just released a couple of days ago beat expectations again. Management locked the growth guidance for the semiconductor equipment business in 2026 at +20%. More importantly, that old thorn from the U.S. Department of Commerce is finally out—the regulatory investigation is closed. This has been weighing on AMAT's valuation for over a year. Now it's cleared, giving the shorts one less reason. But I still haven't moved. The reason is simple: For a stock with a P/E of over 40x, if there's even a tiny flaw in the Q2 earnings report, the market will instantly recalculate "excellent + expensive" into "okay + should drop." Chasing at this level, nine times out of ten, you'll get trapped before the earnings, then watch it drop -8% after-hours. May 14th, Q2 FY2026 numbers are out—that's the watershed moment. I'm just lying in wait, I'll decide after seeing the after-hours results. If the data is good, I'll chase and maybe make 5% less. If the data is bad, I'll pick it up cheap. Either way, I'm not betting on earnings, not betting on earnings, not betting on earnings. As for the $0.53 dividend on May 21st? For my small position, it's just a psychological comfort, not a reason to buy. In short, the company is fine, but I don't dare to jump in at this level. Waiting three more weeks.