I think others have answered the question.
No one has answered the question. "The memory market is growing faster than usual" doesn't answer the question at all. As long as the memory makers are making lots of money, they will over invest in order to gain share. That's how the memory market has always worked. If the memory guys expect the market to grow 30% per year for the next 5 years, they invest to grow 35% per year. If they memory market is expected to grow 5% per year for the next 5 years, they invest to grow 10%. This is basic business strategy in memory.
If memory makes good money, invest to gain share.
"The market is growing really really fast" doesn't change that strategy at all. You can list all the new apps and new usages of memory that you like. The memory makers know more about all of that stuff than you and me. If making memory chips is profitable, they will invest to try in the future to have a higher market share than they have today. That's how you make money, and the larger you are in memory the stronger your position becomes.
The problem is ALL of them will invest to gain future share, and that has to be too much future memory. The only way it cannot be too much future memory is if they underestimate the size of the future memory market. If you guys all know so much about the future size of the memory market, why assume that Samsung and Micron and SK don't know how big it will be?
Why are you so confident all the memory guys aren't going to eventually use their own controllers causing SIMO stock to collapse?
This has always been a concern which, fortunately, never materializes. For SIMO's sweet spot (high volume, low cost) it makes sense that SIMO can make millions of controllers which do the job cheaper than a memory maker. Economies of scale, technical skill set, all that stuff. Memory maker's internal controller groups make the low volume high cost high tech enterprise controllers by the hundreds of thousands, while SIMO cranks out the low tech client controllers by the tens of millions. SIMO seems to believe that as SSDs replace more and more disk drives in PCs, even Samsung will move toward SIMO's cheaper mass market controller rather than use their own internal controller. SIMO can do it more cheaply (sell the same controller model into every PC maker's device).
So we'll see, but this fear for SIMO has never materialized. The trend is actually in the other direction. SIMO is more and more moving from it's very low end role up the controller food chain, and eating away share from the internal controller groups at the NAND makers. SIMO has 2 enterprise controller projects underway in 2018 - selling into some hyperscale cloud project that you guys are so excited about - I gotta assume they are taking that business away from a NAND maker's internal controller division.
If a NAND maker is going to use its internal controller rather than SIMO's external solution it will happen first in eMMC (cell phone controllers). eMMC controllers are 50 cents a pop whereas PC SSD controllers are $5.00. So the cell phone controller is super low tech, if SIMO is going to get the boot, it will be in cell phones before PCs. So far this year SIMO has begun shipping cell phone controllers to MU (new customer for SIMO, so supports the "SIMO is gaining share from internal groups" story). We shall see what SK decides to do in this area by Q2 2018. SK is SIMO's big eMMC customer (about 30% of SIMO's sales). |
