I don't understand how Marvel has a solution for Data Centers and they are so far behind.
I think you mean you don't understand how MRVL has a solution for datacenters and SIMO is so far behind?
3 items to consider.
First, MRVL got to where they are buy acquiring lots of small semi companies that made "data center" semiconductors. So MRVL acquired lots of their data center technology, and just kept it rolling along to be where it is now.
Second, MRVL's data center semconductors target a variety of segments. I don't even know what most of them are. So it's not a good idea to compare MRVL's data center semiconductors to SIMO's Mon Titan (which targets only one segment). I don't have an in depth understanding of MRVL's data center semis, but I can say this. They almost NEVER discuss SSD NAND flash controller chips in their conference calls. For this reason, I don't think MRVL's main area of focus or MRVL's strength is in the area Mon Titan is pursuing. MRVL may have a great enterprise SSD controller but I've never heard them even mention it over the past two years on conference calls.
Third, SIMO has itself internally developed (starting from zero) a chip (Mon Titan) which is supposed to be a super fast, power efficient (low electricity useage) enterprise SSD controller. That's it, that's the only space in the data center that SIMO is pursuing. The chip that manages the NAND flash, nothing else. NAND flash is now and going forward going to be deployed in more and more areas in a data center, and SIMO wants Mon Titan to be the chip that manages that NAND flash chip. So SIMO's only competition with MRVL is trying to win design slots where the customer will choose SIMO, MRVL or (more likely) a controller developed by a NAND maker. How big is MRVL in data center flash controllers? I got no idea, but they don't pump it like it's a big deal for them.
I don't think SIMO is behind at all. We're just impatient. They said 18 months ago they have two customers that will have initial production now (that is happening) with larger sales in 2026. All they've said is the larger sales seem more like a late 2026 event rather than a Q1 2026 event. Ok. This segment is going to be SO BIG that if SIMO can get just 5% share, or just get a foothold, that it will grow for a decade, it will be a 60%+ gross margin chip, and a superior chip (fast, with lowest power consumption) will win MASSIVE sales volumes. The main thing is low power consumption, as once built a main expense of the data center is electricity. If SIMO's chip is 3% lower in power consumption than Samsung's equivalent, SIMO will win a ton of business.
Here's the thing. The two new PC chips (8 channel and 4 channel) are going to drive sales growth in 2026 AND 2027 by themselves. SIMO's market share in PC controllers now is 30%, those two - as they become mainstream - have 50% market share. So PC unit sales can be flat, and SIMO's sales in it's largest product area are going to grow. Then, hopefully, by 2027 or 2028, Mon Titan takes off in a serious way while the PC market share gain starts to flatten out, and Mon Titan drives further sales growth.
Is SIMO still a buy? Well, I don't think we're going to get the "straight up" trend which we had since the tariffs, but I think SIMO's sales are going to show annual growth each year through 2027 at least, and that's probably going to push the share price higher. It may have up and down quarters, because smart investors are used to selling SIMO at peaks, and they're now probably looking for a peak, but if you hold through end of 2027 I think you're going to be happy. What happens between now and then requires someone smarter than me to call the peaks and valleys, but the sales growth story is really good (with the offset that the valuation is now on the high end).
So maybe you can sell Monday at $99 and get a chance to buy it back at $78 in H1 2026. Maybe, I don't kniow. It also may be that SIMO starts moving up again right away as the strength of the two PC chips is recognized by the market. Hard to say. But over the mid to long term, yes, I think it's a great buy.
Their decision to build so much inventory makes me think they're going to have big sales growth, soon. They increased sales sequentially by $40m in Q3, and also increased inventory sequentially by $130m! That's amazing. They have an order book for H1 2026, it's gotta be great for them to build so much inventory now.
Also, as autos and eventually Mon Titan become meaningful, SIMO as an acquisition target just gets more and more attractive.
For valuation, lets says in 2027/28 SIMO does $360m per Q, $1.44 billion, with a 25% operating margin. That's $360m operating profit, taxed at 18% it leaves $295m net income, $8.70 EPS. With sales growth 2-3 years in a row and expansion of autos and Mon Titan, that will get at least a 20x PE, , so $170 share price by ... 2027? That's less than two years away. The PE could be higher. If gross margins tick up due to Mon Titan the operating margin could be higher.
So, I do think it's still a good buy here, but not a value stock. It's a growth stock, and it may require the growth to actually appear before the market accepts that. I'm holding, at least until the MXL settlement appears which perhaps is Q2 2026. I like the idea of learning why they ordered so much inventory. It's not because things are going to slowdown.
Yes, the Mon Titan story is pushed back another 6-12 months from the best we could expect, but luckily it's not necessary for growth because the new gen5 PC SSDs will carry the sales growth story for the next 12-24 months, enough time for Mon Titan to kick in (if it's going to). |
