Post

@amitisinvesting Me.

@amitisinvesting Yes

@amitisinvesting The money is going to flow into other sectors now.. leaving memory. $DRAM $SNDK

@amitisinvesting I am buying the dip y’all are just making stuff up just because they all dipped

@amitisinvesting that's a big red flag 🥴 what's the story behind it?

@amitisinvesting To be the best you can be in this game you have to be your own BULL and make your own decisions

@amitisinvesting that guy is a full on retard does ai and robots need memory less or more in the future?? more how long will more capacity take to get online? years demand exceeds capacity by 2-3x a year for many years micron is the NVIDIA of memory get used to it

@amitisinvesting My brain memory is keeping me reminded not to be in memory.

@amitisinvesting 99% of this is short term minded.

@amitisinvesting @grok who sold all of their memory stocks that Amit is referring to ?

@amitisinvesting The market loves to rotate from “memory is the bottleneck” to “optics is the new bottleneck” the second a credible narrative appears. Both can be true at different time horizons. Positioning for the rotation is different from abandoning the structural case.

@amitisinvesting $MU is still sold out through 2027 but price action clearly front-ran that story already.

@amitisinvesting Very interesting and smart I would assume!

@amitisinvesting I was very surprised by this post. Im still up by approx 400% on MU. Thank you Jukan for posting about memory last year. I don’t want to sell and pay capital gains. But betting against momentum is a recipe for disaster.

@amitisinvesting Bro, I’d say the 5yr hold for memory was 4 years ago when we first heard about chatGPT/whispers of AI, I think we’re close to the end of the memory cycle… neoclouds, well, that’s different. A newer concept built out of capacity constraints, I think neoclouds are the new 5yr

@amitisinvesting He’s a little late to that take

@amitisinvesting Are we sure he didn't panic sell when leopold aschenbrenner was liquidated?

@amitisinvesting Hyperscalers all day long. All the returns eventually come back to the Hyperscalers.

@amitisinvesting The 99% treating it as a trade is the real signal. When nearly the entire holder base has a short time horizon, any narrative shift can trigger forced selling. That explains the current move far better than a broken long term thesis.

@amitisinvesting I think it’s fair to say people are going to be more interested in the software play for a more exciting experience, but I think the solid play will always be the hardware at this stage of the AI rollout. As more physical implements get to market, there will be a growing demand.

@amitisinvesting Maybe he used gullible people as exit liquidity. Who knows 🤷

@amitisinvesting When everybody capitulate is always a good entre point

@amitisinvesting I just don't know.. I get it's a cyclical business but at the same time those businesses have never seen growth like this.. Surely they are fundermentally stronger companies now and deserve re-rating. And then perhaps it is different this time .. The world has changed.

@amitisinvesting Im not buying here too risky

@amitisinvesting Capitulation

@amitisinvesting Hard to argue with someone who’s been shouting from the rooftops and then quietly exits.

@amitisinvesting He does this after the 40% drop. Thats usually the best time to sell.

@amitisinvesting Probably his leveraged positions were liquidated

@amitisinvesting I think many software stocks will likely experience another slump in Micron's September earnings report. The market will probably be ashamed of itself, wondering what it's doing when it sees this profit continuing. $MU $IGV

@amitisinvesting Hell no I‘m not buying any of the memory stock. All-in for software, value stocks and strong long term fundamentals in beaten-down stocks

@amitisinvesting It’s because we all know that 85% margins can’t last forever hence the trade it story. I am long on the storage drive makers like WDC and STX rather than the DRAM makers MU, SKHynix and Samsung. DRAM needs may go up or down but storage there is no way around those.

@amitisinvesting I've called for this for a good part of the last 2-3 months. It's time to move from neocloud to memory to POWER stocks. No one can do anything with chips with POWER.

Would add more to my current $MU position at $650 ish level. Correct, w/o LTAs and share buybacks (which are both ongoing/coming) $MU is $300-400. With them and assuming strong AI demand until 2030 - $1400 (my PT). But yes, time will tell 🤗 $MU is ~2% of my total net worth (incl passive and active holdings) - so not stressed as 2x leveraged and on margin folks 🤣

So, I expect to see falling memory consumption per task -- because that is what evolution of technology needs. Waiting on memory, slow memory access is a real barrier to efficiency. So, this problem must be solved one way or another. But at the same time, as gains in efficiency rise, overall volume of consumption could also rise. The fundamental thesis here is that memory technology, though concentrated with a few is not as much a moat compared to CPUs/GPUs and other computing technologies. Its a combination of barrier to entry AND point in time demand. Too high a per unit cost of memory could hurt the industry even if it's a once in a lifetime opportunity to make some money.

@amitisinvesting bullshit...he is just simping his subs for the optical trade

@amitisinvesting Wow, what a gutsy call! At market tops too! Woohoo. 🥳

@amitisinvesting I believe it's FUD coming in. Even the most knowledgeable people in semis have some fear. Long term I believe more memory, CPUs and GPUs are needed for a wider purpose. Ex: autonomous vehicles, robotics, etc.

@amitisinvesting This will mark the bottom for memory

@amitisinvesting Memory cartel still has a lot of power pair that with CXMT not bending for $AAPL I think the memory trade still has legs in the short term

@amitisinvesting I’m like you, sitting on the sidelines and not chasing. But I am mega bullish everything in the AI supply chain. Memory, semis and everything else related isn’t going anywhere, this is just a fact! Computers and computing power will keep going up in demand.

I disagree as AI evolves the need for memory will only increase. And a few years from now even if their profits are halved they will still be the richest companies in the AI economy. This would be like getting out of oil shortly after Henry Ford massed produced cars thinking, there is no way this will continue….

@amitisinvesting I've bought some, and definitely see it as a long term investment. Supply is nowhere near meeting demand and I doubt it will for the next decade. LTAs and price plateau seem good for sustainability.

@amitisinvesting If you buy now, the FCF will cover the market cap within 5 years.

@amitisinvesting Nice write up , thanks

@amitisinvesting Buying every dip below 900, every day. Jukan snapped, I guess. Part of the process, weak hands got to go. Yes I will hold for 5 years.

@amitisinvesting I reduced my memory holdings from 70% to 30% selling with a profit (large drop in profit), I believe in their future for the next 2 years and I will wait and see.

@amitisinvesting @CaktusJxck Irrelevant to me

@amitisinvesting I selectively and conservatively sell cash-secured puts on SKHY and DRAM, and put credit spreads on MU. So far, so good!

@amitisinvesting I have $soxx and $smh and $qtum :)

@amitisinvesting Yes, as always it could go up or could go down. I think people are panicking way too much over a drawdown after an insane run. Memory demand remains sky high, and I trust MU to make great decisions with the extra cash they are getting from these margins. No selling for me 📈