SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

   Technology StocksSilicon Motion Inc. (SIMO)


Previous 10 Next 10 
From: Elroy10/6/2021 11:30:47 AM
   of 2458
 
Here's a new enterprise MU SSD. I wonder who makes the controller? Probably gotta wait a bit for the tech paper reviews....

Micron® 7400 SSD With NVMe™ Delivers PCIe Gen4 Performance for Data Centers

finance.yahoo.com

Share RecommendKeepReplyMark as Last ReadRead Replies (1)


To: Elroy who wrote (2408)10/6/2021 11:31:54 AM
From: Elroy
   of 2458
 
TSMC to expand 7nm and 28nm fab capacities


https://www.digitimes.com/news/a20211006PD210.html?mod=2


Interesting. I think 28nm is where most of SIMO's chips are made.

Share RecommendKeepReplyMark as Last Read


From: Elroy10/7/2021 10:24:56 AM
   of 2458
 
Silicon Motion Announces Preliminary Third Quarter 2021 Revenue and Earnings Conference Call Details

based on its preliminary third quarter financial results, sequential revenue growth is expected to slightly exceed the high-end of its original guidance of 7.5% to 12.5%, which the company issued on July 30, 2021. Gross margin (non-GAAP) is expected to slightly exceed 50.0% and is within the high-end of the company's original 48.5% to 50.5% guidance range.



I was hoping for another blow out, but I guess not.

The model says something like

Revenues = $250m
Gross Margin = $125m
Gross Profit = $125m
Op Ex = $53m
Op Profit = $72m
Tax (20%) = $14m
Net Profit = $58m
EPS (35m shares) = $1.66

All of those numbers can shift in SIMO's favor a bit, but I doubt they get EPS over $1.80.

It represents 100% sales growth over Q3 last year, but is somehow depressing as it indicates more that the hyper growth has perhaps peaked.

Share RecommendKeepReplyMark as Last Read


From: Elroy10/7/2021 1:26:17 PM
   of 2458
 
Well, interesting that in the Q2 2021 call SIMO guided toward about $136m-$149m in Q3 sales, and the shares popped from $70 to $80. Today SIMO says the revenue number is going to be a tad above $149m, and the shares dip to $68.

Ho hum!

Q3 2021 sales are going to be about 100% higher than Q3 2020 sales.

I wonder what it will take for SIMO to get just a reasonable "market multiple" of say .... 18x next year's forecast EPS? Next year they're easily going to make $7.00, and perhaps might make as high as $10 depending on how things turn out. There's lots of moving parts now in revenue growth rates, gross margins, operating expense growth and etc. which make it tough to accurately forecast EPS. But it's clearly going to be above $7.00 unless some cataclysm happens (which is possible, I suppose).

The next bit of info we get is the new dividend rate in about two weeks. I don't know if they base the dividend on trailing EPS or forecast EPS. If trailing, the previous four quarter will be about $1.73 (Q3 2021), $1.50, $1.11 and $0.86, so .......about $5.20. I think they say the dividend is about half of EPS, which means about 55 cents per quarter (probably 50 cents to keep it clean) is reasonable. If the dividend is based on expected EPS and cash flow, then the dividend could be closer to 75 cents, but something tells me 50 cents is more likely.....

Share RecommendKeepReplyMark as Last Read


From: Elroy10/20/2021 2:22:50 PM
   of 2458
 
Ok, we're about due for a week of SIMO news, so lets assess the shares.

News - sometime between tomorrow and next Tuesday SIMO will announce the new dividend amount for the next four quarters. The current dividend is 35 cents per Q. My guess is it goes to 50 cents per Q, but it's really just a guess. The CFO has said their historical plan to return capital to shareholders has been to deliver half of free cash to shareholders in dividends. That doesn't mean the "half" number will be the what they do going forward, and it isn't clear if he meant trailing free cash or expected future free cash. Also, SIMO should be spending $$ to build their new headquarters over the next 6-10 quarters, so it's also not clear how that expenditure affects their calculation of "free cash". Anyways, the dividend should increase from 35 cents per Q, by how much we will find out soon.

Then next Wednesday afternoon we'll get the actual Q3 numbers and Q4 guidance. They've already pre-announced revenues slightly ahead of forecast 12.5% sequential growth, and gross margins between 50.0% and 50.5%. Previously they indicated that Q4 would have higher sales than Q3, and that they expect backlog ($1.5 billion at the end of Q2) to grow throughout the year. They also said they would discuss 2022 in general terms in the Q3 call, and give formal guidance in the Q4 2021 call.

Whew!

SIMO has gained tremendous market share in 2021. Phison is the only other major merchant flash controller maker, and SIMO says they are 4x larger than Phison in SSD controllers, and that there is no merchant competition in eMMC and UFS controllers. It remains to be seen how this significant share gains will affect future profitability, but one would expect gross margins to increase down the road as reduced competition should allow SIMO to raise prices. In addition SIMO expects to rapidly ramp their gen4 SSD flash controller in 2022, and that chip is made at 12nm and will have higher ASPs since it's brand new. So ramping a higher priced chip which may have lower cost of production per unit due to reduced line widths should (one would think) translate into higher corporate gross margins for SIMO in 2022. SIMO says at the end of 2022 they expect that chip to have 50% share of gen4 design slots, and, well, THAT'S A LOT!

The only thing I see going WRONG with SIMO currently is the equity valuation. I think management needs to explain how the 2021 changes in the market (massive share gains by SIMO) perhaps give them a stronger moat going forward, if that is true. The only major major threat to SIMO has always been loss of design slots to internal controller groups inside the NAND flash makers (SK, WDC, MU, etc.). Is that concern increased, diminished, or what as a result of market development of the current year? If that concern is significantly reduced, I think SIMO may experience multiple expansion, but who knows?

Optimistic on the coming big news week, and hoping Lucy does not appear to pull the football away, so lets see what happens....

Share RecommendKeepReplyMark as Last ReadRead Replies (1)


To: Elroy who wrote (2412)10/23/2021 10:16:32 PM
From: Anonymous895
   of 2458
 
Elroy,

SIMO is not actually 4x Phison, as we've spoken about other times. I'm pretty sure that when SIMO says they're 4x the next largest merchant controller, they are specifically not counting any controllers used for internal use or for joint ventures/partnerships. So they're not counting Phison's controllesr for the flash that they themselves sell nor the Phison controllers used in Kingston drives. I think they're not counting the Phison controllers used in Toshiba drives as well. They're considering all this to be "in-house" controllers. But it does of course count as competition to SIMO.

All that said, I do agree that SIMO has gained massive market share this year, and I believe a lot of it will be sticky.

Share RecommendKeepReplyMark as Last ReadRead Replies (1)


To: Anonymous895 who wrote (2413)10/23/2021 11:38:27 PM
From: Elroy
   of 2458
 
SIMO is not actually 4x Phison, as we've spoken about other times. I'm pretty sure that when SIMO says they're 4x the next largest merchant controller, they are specifically not counting any controllers used for internal use or for joint ventures/partnerships. So they're not counting Phison's controllesr for the flash that they themselves sell nor the Phison controllers used in Kingston drives.

Yes, that may be true. I'm not really sure how the controller market share battle between Phison and SIMO and then divided along SSD, UFS and eMMC product lines splits out.

It's a bit tough for me to get a handle on the scope of SIMO's controller business (internal consumption and external sales), and it seems like market share on this topic (flash controllers) isn't easily available from the normal market share guys.

Phison doesn't give broken down revenues by those three product lines (SSD, UFS and eMMC flash controllers), so ..... who knows how they're doing? I don't.

I would be interested to better understand the UFS controller market. SIMO has MU as a large UFS cell phone controller customer. SIMO said something about adding 5 or 6 more UFS OEM customers in the Q1 2021 call, then I think didn't mention it in the Q2 2021 call. I've read speculation that SK may be coming back to SIMO for UFS controllers, but if that's true I'd expect to read/hear more about it. I really don't have a great idea what's going on with UFS controller market in general, and don't know how Phison is doing in that market at all.

SIMO indicated Samsung is exiting the eMMC controller market (because the memory chips for it are too small to pursue), so I assume that sort of leaves SIMO as a monopoly, which seemed attractive. Smart speakers, TVs, etc etc should be an OK market for many years. But is Phison even in the eMMC controller market? Donno.

For SSD controllers I know Phison has products, but it's hard to get their market share. It is this space where I think SIMO means they have 4x the unit volume of Phison, but again I'm not sure.

It would be really helpful if Phison would publish it's annual sales of eMMC, UFS and SSD controllers, then we'd all know the answer!

Share RecommendKeepReplyMark as Last ReadRead Replies (1)


To: Elroy who wrote (2414)10/24/2021 10:12:08 AM
From: Anonymous895
   of 2458
 
My reply was apparently not posted, so I'm trying again.

SIMO said that they're adding one NAND flash manufacturer as a UFS customer (starting small this year and ramping next year) and then another 5 customers next year. They did not say whether these five new customers are NAND flash manufacturers or OEM customers. I expect that some, maybe all, of these five customers are module makers.

Phison definitely makes eMMC controllers. Just google Kingston's eMMC products which go back years and use Phison's controllers. (Until recently, Kingston exclusively used Phison's controllers from the Phison/Kingston joint venture. I'm pretty sure that Silicon Motion considers these "in-house" controllers.) If Phison only made SSD controllers, their ASP would have to be much higher.

When SIMO says that they're basically the only significant merchant supplier in the eMMC market, that may technically be true, but that's not counting any "in-house" controller competition, including Kingston. I don't know what market share Kingston has in the eMMC market, but they are a significant player in the SSD market, and I would not discount Kingston as a player in the eMMC market. Of all the module makers, Kingston, due to their size and importance in the market, is practically an OEM.

Share RecommendKeepReplyMark as Last ReadRead Replies (1)


To: Anonymous895 who wrote (2415)10/24/2021 1:08:21 PM
From: Elroy
   of 2458
 
If Phison only made SSD controllers, their ASP would have to be much higher.

Not sure what you mean by this. Historically SIMO said that eMMC controllers were 50 cents and SSD controllers were $5 on average. I don't think they've updated eMMC ASP info since UFS became widespread, so eMMC may have shifted down somewhat.

Does Phison give controller ASPs?

I haven't got a great handle on Phison's market share in anything because Phison doesn't clearly provide the info. IF Phison were dominant in something, I'd expect them to make it crystal clear to investors.

SIMO says they think they will have 50% of the design slots for gen4 PCIe SSD controllers by the end of 2022. Phison has a comparable gen4 SSD controller. They quote the phenomenal growth of this product year over year in recent press releases, but (if I recall) it is growing from something like $1 million in sales in the quarter a year ago to $2.5 million this year, something like that. High growth, but off a miniscule base.

Do you have any idea of Phison's expected market share in gen4 SSD controllers by the end of 2022? That would give us a fair comparison of the two companies in the leading edge SSD PC controller.

Share RecommendKeepReplyMark as Last ReadRead Replies (2)


To: Elroy who wrote (2416)10/24/2021 9:08:37 PM
From: Anonymous895
   of 2458
 
We've been through this in the past. In 2019, Phison made $143 million USD in revenue from controllers (not counting their own modules of combined flash+controller) and sold approximately 411 million such stand-alone controllers. That comes out to an average ASP of about 35 cents. Which is exactly why I say that they obviously are not selling all that many stand-alone SSD controllers -- their ASP is too low for them to much of it to be coming from SSD controllers. Which is why I think that the majority of these controllers are for lower-end applications, such as eMMC (and, of course, USB keys, SD cards, and the like).

I just looked up the 2020 numbers. In 2020, Phison sold $268 million USD of stand-alone controllers, numbering 355 million units. That's an average ASP of 76 cents, much higher than 2019 but still relatively low.

I agree with you that Silicon Motion is apparently smoking Phison's in SSD controller market share, especially for next year.

There is a lot of information to be gotten from Phison's financials and annual report, freely available (in English) on their website. And you can read the transcripts of their conference calls as well, in not-quite-English.

Share RecommendKeepReplyMark as Last ReadRead Replies (1)
Previous 10 Next 10