UMS Holdings

[KopiKat]

The key is whether will there be new fabs considering the shitty state of the world economy...

IMO, for those who are looking at UMS, the KEY is,

1. Is UMS too dependent on Applied Materials? If so, they'll sink with them when times are bad. For that matter, is Applied Materials able to weather any severe downturn?

2. How robust is their biz relationship with Applied Materials? Since Applied Materials is also a listco, they'll be accountable to their shareholders also. So, I don't see it as a case that they must give UMS priority on any biz. Has to be a matter of cost vs quality.

3. Why isn't UMS succeeding in becoming less dependent on Applied Materials? eg. spread out to more Semicon Eqpt Makers (since this appear to be the core competence) or even to other Industrial biz which requires Precision Engineering services. I know they tried Aerospace, but why no results after so many years??

4. Is the CEO's interest 100% aligned with UMS? If so, why is he selling so many shares and why is he spending time to develop IMT-S and IMT-US which he subsequently sold to UMS (of course after the necessary legal approvals)?

So, ya, we can debate on the health of the economy which affects Electronics->Semicon->Eqpt Makers->UMS. But, ultimately, for UMS shareholders (or those evaluating it), we'd want to know how they are going to be able to continue paying the Div=1ct/Q..

[Boon]

The key is whether will there be new fabs considering the shitty state of the world economy...

IMO, for those who are looking at UMS, the KEY is,

1. Is UMS too dependent on Applied Materials? If so, they'll sink with them when times are bad. For that matter, is Applied Materials able to weather any severe downturn?

2. How robust is their biz relationship with Applied Materials? Since Applied Materials is also a listco, they'll be accountable to their shareholders also. So, I don't see it as a case that they must give UMS priority on any biz. Has to be a matter of cost vs quality.

3. Why isn't UMS succeeding in becoming less dependent on Applied Materials? eg. spread out to more Semicon Eqpt Makers (since this appear to be the core competence) or even to other Industrial biz which requires Precision Engineering services. I know they tried Aerospace, but why no results after so many years??

4. Is the CEO's interest 100% aligned with UMS? If so, why is he selling so many shares and why is he spending time to develop IMT-S and IMT-US which he subsequently sold to UMS (of course after the necessary legal approvals)?

So, ya, we can debate on the health of the economy which affects Electronics->Semicon->Eqpt Makers->UMS. But, ultimately, for UMS shareholders (or those evaluating it), we'd want to know how they are going to be able to continue paying the Div=1ct/Q..

1) UMS is over dependent on Applied Materials in my opinion. They both “sunk” in 2009 but “Sailed” pretty nicely in 2010 and 2011. Please refer to their AR.

Some summary of AM's financial summary in this link
http://www.marketwatch.com/investing/sto...financials

2) This is an interesting “strategic partnership” which I am just as eager to find out more. More than 20% in net profit margin for UMS in 2010 and 2011. I am still looking into this and would offer my opinion later.

3) I think they are trying.

4) These “millions of dollar on the run” issue and also “the selling of IMTs” issue are legitimate concerns that I think only the CEO can address.

Without having the understanding of the cause and effect of (a) the underlying factors affecting demand of semiconductor device; (b) how demand for semiconductor device affect demand for semi-tools;© how demand for semi-tools affect Applied Materials semi-tools business?); and (d) the knock on effect from Applied Material on the bottom line of UMS? How would one evaluate how much UMS is going to make and hence its ability to pay dividend? Any better suggestion?

Again, as Nick has mentioned which I agreed, UMS is in a volatile industry so it should not be confused as a yield play.

[KopiKat]

Without having the understanding of the cause and effect of (a) the underlying factors affecting demand of semiconductor device; (b) how demand for semiconductor device affect demand for semi-tools;© how demand for semi-tools affect Applied Materials semi-tools business?); and (d) the knock on effect from Applied Material on the bottom line of UMS? How would one evaluate how much UMS is going to make and hence its ability to pay dividend? Any better suggestion?

Again, as Nick has mentioned which I agreed, UMS is in a volatile industry so it should not be confused as a yield play.

Yes, you are right. Since this is a highly cyclical industry, the right thing for a long term investor to do, would be to understand the industry and the cycles, as it'll affect how UMS is going to perform.

In my case, I'm lazy and do not want to commit too much of my time on the above (just yet) as I have many concerns (on UMS) which I'd not been able to find answers that I can live with. As such, I'm taking a very short term approach and is focussing only on the recent results and praying that the cycle is not going to nose-dive within the recent Q. If I'm not comfortable, I'd exit immediately. If I like what I see next, perhaps I'll increase my stake and start to put in even more work.

IMO, this industry is just too huge and complex (many sub-industry and segments), with too many different players and market leaders in each sub-segments all over the world. It's going to take a lot of my time and at the moment, the risk-reward is not enticing enough for me to stop being lazy...

PS. Those who are in this industry will have an easier job if they have access to Research Reports from IDC or DataQuest (Gartner)

---

UMS is capitalizing on Applied Materials outsourcing trends towards low cost regions ie Singapore. Applied Materials deals with the manufacturing of semiconductor equipment (not manufacturing semiconductor chips). Hence the key metric is semiconductor equipment spending which was guided by Applied Materials (and its peers) to be weaker in 3Q 12 due to lower demand.

Some summary of AM's financial summary in this link
http://www.marketwatch.com/investing/sto...financials

Extracts from the news from the above,

Applied Materials, Inc., the global leader in providing manufacturing solutions for the semiconductor, display and solar industries, today revised its fiscal year 2012 business outlook due to weaker than expected near-term demand in its semiconductor equipment business, primarily among foundry customers.

For the fiscal year ending October 28, 2012, the company expects net sales to be below the previous outlook of $9.1 billion to $9.5 billion, with non-GAAP earnings per share (EPS) below the previous range of $0.85 to $0.95. The company will provide a new target range during its August 15 earnings call, however it expects that the demand changes could have a $0.15 to $0.20 impact on full-year non-GAAP EPS results.

Applied expects financial results for the third quarter of fiscal 2012 ending July 29, 2012, to remain within the outlook ranges provided during its second quarter earnings call on May 17. Net sales for the third quarter are expected to be at the low end of the previous outlook of flat-to-down-10 percent from $2.54 billion in the second quarter. The company expects non-GAAP EPS for the third quarter to be in the lower half of the previous outlook of $0.21-$0.29.

The company also revised its calendar year 2012 industry forecast for wafer fab equipment spending to $30 to $33 billion, compared to its previous expectation of $32 to $35 billion, in line with the market changes.

[Boon]

“Faster, greener, smarter– reaching beyond the horizon in the world of semiconductors” - A study of chip market trends and the technological and economic drivers behind them – by PWC , January 2012

http://www.pwc.com/en_GX/gx/technology/p...ctions.pdf

Preface
The number of semiconductor components used in our daily lives is expanding constantly. Chips form the essential core of many cutting-edge technological devices such as smart-phones, tablets, flat-screen monitors and television sets, sophisticated cars, new aircraft, and many medical devices. As a consequence, the semiconductor industry has been growing for 40 years. However, almost no other industry swings as dramatically in relation to overall economic developments. So, will the golden era of the semiconductor industry continue? Have semiconductor companies learned and prepared from the last downturn during the financial crisis in 2008–09? How will the market for chips develop in the next few years? Which business model will prove to be robust during the crisis and beyond? Where are the current opportunities? What are the critical factors of success? We will try to look behind the curtain and search for answers to those and other essential questions.
……………………………..

[KopiKat]

“Faster, greener, smarter– reaching beyond the horizon in the world of semiconductors” - A study of chip market trends and the technological and economic drivers behind them – by PWC , January 2012

http://www.pwc.com/en_GX/gx/technology/p...ctions.pdf

Preface
The number of semiconductor components used in our daily lives is expanding constantly. Chips form the essential core of many cutting-edge technological devices such as smart-phones, tablets, flat-screen monitors and television sets, sophisticated cars, new aircraft, and many medical devices. As a consequence, the semiconductor industry has been growing for 40 years. However, almost no other industry swings as dramatically in relation to overall economic developments. So, will the golden era of the semiconductor industry continue? Have semiconductor companies learned and prepared from the last downturn during the financial crisis in 2008–09? How will the market for chips develop in the next few years? Which business model will prove to be robust during the crisis and beyond? Where are the current opportunities? What are the critical factors of success? We will try to look behind the curtain and search for answers to those and other essential questions.
……………………………..

Boon,

Thanks for the link. Very informative, brings back a lot of fond memories and nightmares as I was in this industry for almost my entire working life...
I'd certainly like to hear your comments after you'd read the above report, pertaining to the industry and perhaps, how it relate to UMS.

[Boon]

Boon,

Thanks for the link. Very informative, brings back a lot of fond memories and nightmares as I was in this industry for almost my entire working life...
I'd certainly like to hear your comments after you'd read the above report, pertaining to the industry and perhaps, how it relate to UMS.

KopiKat,

Thanks, there is quite a lot of interesting things to write about in this topic and due to time constraint, I may have to give my comments in parts. But before I do that, here is some questions for all to think about:

1) Who is the No:1 Fabless Semiconductor Company (or IC supplier) in the world ?

Answer: Qualcomm (http://www.eeherald.com/section/news/nws201204152.html)

2) Who is the No: 1 pure-play semiconductor foundry in the world?

Answer : TSMC (http://en.wikipedia.org/wiki/Foundry_model#2011)

Qualcomm is a Fabless Semiconductor Company.
TSMC is the contract manufacturer for Qualcomm.
Unlike Intel, an IDM (Integrated Device Manufacturer) which handles semiconductor manufacturing in-house, Qaulcomm, a Fabless semiconductor company, outsource its production to a third party like TSMC.
TSMC is a client of Applied Material for semi-tools (semiconductor manufacturing equipment).
TSMC accounted for 10% of Applied’s net sales in fiscal 2011 - for products in multiple reportable segments.(this figure can be found in AM’s AR2011).

3) Why is there a shortage of 28nm Snapdragon S4 chip?

4) Why Qualcomm had to plan to launch 4 new lines for 28nm Snapdragon S4 chips?

5) There is an over-capacity issue in the semiconductor industry as claimed by many, but on the other hand, there is a shortage of 28nm Snapdragon S4 chips? Why there exists such a mismatch?

6) Is there a production overcapacity in of chips (IC or Integrated Circuit) that the market does not want?

7) Is there an under-capacity in the production of chips (IC) that the market wants?

8) Is there an under-investment in new technological capabilities (facilities/semi-tools etc) in producing semiconductor devices that the market demands?

------------------------------------------------------------------------------------------------------------
Qualcomm Snapdragon S4 chip shortage extending to 2013
Chip maker won't meet demands on 28nm chips
Qualcomm, whose Snapdragon processors are found in a wide variety of smartphones, will experience chip shortages that will extend into next year.
Supply issues are being traced back to Taiwan Semiconductor Manufacturing Company (TSMC), a contract chip maker that produces Qualcomm's widely used 28nm Snapdragon S4 system on a chip.
TSMC told Focus Taiwan that it expects manufacturing to start catching up in 2012's fourth quarter, with the ability to fully satisfy market demand again in 2013.
Shortage spells delays for new devices
Only TSMC's 28nm chip production will be effected by the shortage, which may also impact Nvidia, who also uses the contract manufacturer for its 28nm chips.
Production of smartphones based on the Snapdragon S4 chip will likely slow down with the shortage, effecting handsets like the HTC One X, HTC Evo 4G LTE, and could lead to a rocky start for the Samsung Galaxy S3.
After Qualcomm warned of chip shortages back in April, rumors started circulating that the Snapdragon S4-based Asus PadPhone will likely see a delay.
Last week, Qualcomm announced plans to launch four new lines of Snapdragon S4 chips, expanding its reach to HDTVs, notebooks and tablets. The firm hasn't commented on how the extended chip shortage will muddle those plans.
Qualcomm chairman Paul Jacobs tried to put up an optimistic front to the news, telling Reuters, "The goal is to get enough supply for everyone."
A lofty goal, to be sure, but with matters seemingly out of Qualcomm's control, the firm will likely have to seek out a new contract chip maker if there is any hope of achieving it.

http://www.techradar.com/news/computing-...13-1084987

[KopiKat]

Boon,

Thanks for the link. Very informative, brings back a lot of fond memories and nightmares as I was in this industry for almost my entire working life...
I'd certainly like to hear your comments after you'd read the above report, pertaining to the industry and perhaps, how it relate to UMS.

KopiKat,

Thanks, there is quite a lot of interesting things to write about in this topic and due to time constraint, I may have to give my comments in parts. But before I do that, here is some questions for all to think about:

1) Who is the No:1 Fabless Semiconductor Company (or IC supplier) in the world ?

Answer: Qualcomm (http://www.eeherald.com/section/news/nws201204152.html)

2) Who is the No: 1 pure-play semiconductor foundry in the world?

Answer : TSMC (http://en.wikipedia.org/wiki/Foundry_model#2011)

Qualcomm is a Fabless Semiconductor Company.
TSMC is the contract manufacturer for Qualcomm.
Unlike Intel, an IDM (Integrated Device Manufacturer) which handles semiconductor manufacturing in-house, Qaulcomm, a Fabless semiconductor company, outsource its production to a third party like TSMC.
TSMC is a client of Applied Material for semi-tools (semiconductor manufacturing equipment).
TSMC accounted for 10% of Applied’s net sales in fiscal 2011 - for products in multiple reportable segments.(this figure can be found in AM’s AR2011).

3) Why is there a shortage of 28nm Snapdragon S4 chip?

4) Why Qualcomm had to plan to launch 4 new lines for 28nm Snapdragon S4 chips?

5) There is an over-capacity issue in the semiconductor industry as claimed by many, but on the other hand, there is a shortage of 28nm Snapdragon S4 chips? Why there exists such a mismatch?

6) Is there a production overcapacity in of chips (IC or Integrated Circuit) that the market does not want?

7) Is there an under-capacity in the production of chips (IC) that the market wants?

8) Is there an under-investment in new technological capabilities (facilities/semi-tools etc) in producing semiconductor devices that the market demands?

------------------------------------------------------------------------------------------------------------
Qualcomm Snapdragon S4 chip shortage extending to 2013
Chip maker won't meet demands on 28nm chips
Qualcomm, whose Snapdragon processors are found in a wide variety of smartphones, will experience chip shortages that will extend into next year.
Supply issues are being traced back to Taiwan Semiconductor Manufacturing Company (TSMC), a contract chip maker that produces Qualcomm's widely used 28nm Snapdragon S4 system on a chip.
TSMC told Focus Taiwan that it expects manufacturing to start catching up in 2012's fourth quarter, with the ability to fully satisfy market demand again in 2013.
Shortage spells delays for new devices
Only TSMC's 28nm chip production will be effected by the shortage, which may also impact Nvidia, who also uses the contract manufacturer for its 28nm chips.
Production of smartphones based on the Snapdragon S4 chip will likely slow down with the shortage, effecting handsets like the HTC One X, HTC Evo 4G LTE, and could lead to a rocky start for the Samsung Galaxy S3.
After Qualcomm warned of chip shortages back in April, rumors started circulating that the Snapdragon S4-based Asus PadPhone will likely see a delay.
Last week, Qualcomm announced plans to launch four new lines of Snapdragon S4 chips, expanding its reach to HDTVs, notebooks and tablets. The firm hasn't commented on how the extended chip shortage will muddle those plans.
Qualcomm chairman Paul Jacobs tried to put up an optimistic front to the news, telling Reuters, "The goal is to get enough supply for everyone."
A lofty goal, to be sure, but with matters seemingly out of Qualcomm's control, the firm will likely have to seek out a new contract chip maker if there is any hope of achieving it.

http://www.techradar.com/news/computing-...13-1084987

My understanding of how the relationship works, between Fabless (Qualcomm) and Fab (TSMC),

1. Fabless has to give Fab a Forecast / Projection - Weekly, Monthly, Annual figures ; they meet regularly to review projections and other technical and sales / logistics issues
2. Fab allocates Capacity to Fabless based on (1) - they may tweak their nos. upwards or downwards based on previous experience. Tendency is Fabless usually gives an overly positive projection to protect themself in case of upsides. If Fab assigns all the capacity projected and it doesn't materialise, there's opportunity cost as it takes times to change their tooling if they were to assign it to another Fabless customer.

My guess is that for Snapdragon S4, Qualcomm may have been caught by an unexpected strong demand (eg. Samsung S3 hp strong demand) from the market. Their projection to TSMC is likely to have been way below market demand and that's why they now have to launch 4 new lines (your news).

I've not been up to date with the latest process. But, they could have been hit by a double-whammy. The 28nm process may be a newer process. During the intial start-up for production, it's very normal for the Fab to have very low yield. The process has to be continuously tweaked to improve the yield and may take weeks or even months.

Take note that for Fab, it's not a very simple matter of a Fabless like Qualcomm bringing their biz to another Fab to produce their Snapdragon S4 to fill their orders. If it's a new cutting edge process, not many other Fabs can handle it as they may not have the capability to develop it (that's why CSM was always lagging behind, playing catch-up). Even if another Fab has the same process eg, 28nm, Qualcomm has to re-design their Snapdragon S4 according to that Fab's process parameters + subsequent qualifications with actual Fab samples. This can take weeks and months...

The smaller Fabless usually do not have their own R&D capabilities to develop their own process. Usually, they license it from the leaders. In such cases, the leaders are unlikely to license out the latest process as they can make higher profit margins (since there's limited supply). So, that's why we continue to see an over-capacity situation for the older process and an under-capacity in the newer process.

This is very unlike a CEM relationship or any other mass production kind of biz where you can easily have your products produced in a much shorter time.

As for Q6 and Q7, Semicon products can be broadly classified into Commodities and non-Commodities.

For Commodities, we have standard products like Memory Devices, Logics, Transistors and also Passive Components like Capacitors and Resistors. The main differentiation for Commodities are PRICING, so ya, this is a very cut-throat market segment, very hard to make money unless there's a market shortage.

As for non-Commodities, this means non-standard products like Snapdragon ie. no one else can provide the same IC except Qualcomm. Other general term products you'll see are SOC (System On Chip), ASIC (Application Specific ICs),... The margins are great here, if you managed to come up with a product that become widely adopted and if you can control the market. That's why companies like Intel, Motorola, TI,.. spent so much time protecting their own turf.

All Qs answered? May not be the correct answers as I'm more than 10 years out of touch with the market..

[mrEngineer]

3) Why is there a shortage of 28nm Snapdragon S4 chip?

4) Why Qualcomm had to plan to launch 4 new lines for 28nm Snapdragon S4 chips?

5) There is an over-capacity issue in the semiconductor industry as claimed by many, but on the other hand, there is a shortage of 28nm Snapdragon S4 chips? Why there exists such a mismatch?

6) Is there a production overcapacity in of chips (IC or Integrated Circuit) that the market does not want?

7) Is there an under-capacity in the production of chips (IC) that the market wants?

8) Is there an under-investment in new technological capabilities (facilities/semi-tools etc) in producing semiconductor devices that the market demands?

Hi Boon,

In order for you to understand a bit better on IC industry. Take a look at how UBS tried to teardown Iphone 4s. (http://9to5mac.com/2011/10/11/iphone-4s-...-analysis/) There are many more components that are required to put together in order to form a single electronic product on top of the processor and the industry needs that to survive.

No doubt the processor (S4 or A5 or whatever) can be found at every electronic product as it serves as the brain of the product, but one should note that one wafer can already give you 500+ A5 chips (depending on yield, http://www.anandtech.com/show/5789/the-i...a5-tested/ ), with typically one lot is 25 wafers. Based on the demand for Samsung Galaxy S2, 10 million units were sold ( http://www.appolicious.com/tech/articles...5-upgrades ), you work it backwards you will get 10mil/500chips = 20,000 wafers for the entire S2 product line. Maximum production capacity of our Global Foundaries Singapore Fab 7 (and there are Fab 1-13) approximately 60,000 wafers per month, (http://www.globalfoundries.com/manufacturing/300mm.aspx ). Let say even if the yield is 20% for advanced technology (which is ridiculous as the fab will confirm make a loss), you will need 20,000/0.2 = 100,000 wfrs. This can all be produced within one fab in few months. Now, think about the number of fabs around the world.

In summary of what I am trying to say here, it is not conclusive to use one single component to justify the entire semicon supply chain is unable to meet production demands and require capex to buy new equipments to do so. The whole semicon supply chain supplies different components for different parts of one electronic product and we should look at it on an overall basis. Furthermore, I have previously explained, it is not so easy for any fab to commission any new process, let alone the equipment for any products. Such commissioning needs alot of justification and time from the management, engineers etc. Another point I want to make is that we should sometimes take news or reports with a pinch of salt.

[Zelphon]

Yup.

I agree with Mr Engineer...

To know whether UMS is experiencing increase revenue or not..
Just pop down to APPLIED MATERIALS MFG plant in Changi on a weekly basis...

Observer the no. of crates there..
More crates observed near the loading/unloading bay = MORE FACTORY ACTIVITY = MORE SALES la..

Can't reveal which company i worked for but that is how we insiders gauge whether how my company is doing...

LOL...

[Boon]

Thank you for your comments, "Mr.Engineer" and "KopiKat". Honestly, I am a layman in the field of semiconductor and therefore have a lot to learn from you both who are expert in this field. For the benefits of myself and other buddies, I look forward to more participation and comments from you.