Intel tops Q3 estimates

[NTMBK]

10nm is already delayed, but Stacy Smith said that the the CAPEX change wasn't due to any downward trend, but more because of reshuffle in equipment orders. CAPEX should increase in 2016.

Makes sense- if 10nm is pushed out, 10nm equipment buys get pushed out too.

 

[Spungo]

Actually net profits aren't an interesting measure to calculate whether a business is faring well or not.

There are a myriad of reasons on why investors DEMAND from companies to release things like EBITDA, expenses and gross margins and just not the net profits and that's because net profits don't say much about the company's business or it's business decisions, it just says a lot on how the company played its balance sheet around.

I agree with Seth Klarman and Warren Buffet on this one - EBITDA is a useless metric.

https://hurricanecapital.wordpress.com/2014/10/27/margin-of-safety-seth-klarman-on-ebitda/

Not thinking of depreciation as an expense is crazy. I can think of a few businesses where one could ignore depreciation charges, but not many. Even with our gas pipelines, depreciation is real — you have to maintain them and eventually they become worthless (though this may be 100 years).

It [depreciation] is reverse float — you lay out money before you get cash. Any management that doesn’t regards depreciation as an expense is living in a dream world, but they’re encouraged to do so by bankers. Many times, this comes close to a flim flam game.

People want to send me books with EBITDA and I say fine, as long as you pay cap ex. There are very few businesses that can spend a lot less than depreciation and maintain the health of the business.

This is nonsense. It couldn’t be worse. But a whole generation of investors have been taught this. It’s not a non-cash expense — it’s a cash expense but you spend it first. It’s a delayed recording of a cash expense.

We at Berkshire are going to spend more this year on cap ex than we depreciate.

Remember what we're dealing with. This is chip manufacturing. It's an extremely capital intense business. The hardware used to make a Pentium 4 cannot be used to make an i7, so depreciation and amortization are real expenses that need to be accounted for.
Excluding depreciation and amortization is literally the same thing as assuming a house has no maintenance costs. The roof never needs to be fixed, windows are never replaced, the floors are never replaced. The cost of owning a house is zero as long as we exclude everything that costs money. The company has zero depreciation expenses as long as we have no intention of upgrading anything and we plan to sell the exact same products for the next 20 years (which is impossible when competitors are selling better and cheaper products every year).

 

[Fjodor2001]

Q3 2015 compared to Q3 2014:
- Notebook platform volumes decreased 14%
- Notebook platform average selling prices increased 4%
- Desktop platform volumes decreased 15%
- Desktop platform average selling prices increased 8%
- Tablet platform volumes of 8 million units decreased 39%
First nine months of 2015 compared to the first nine months of 2014:
- Notebook platform volumes decreased 8%
- Desktop platform volumes decreased 18%
- Desktop platform average selling prices increased 6%
- Tablet platform volumes of 26 million units decreased 7%


2015 is turning to be the worst year for the PC market.

Yup. I think that even though the title of this thead ("Intel tops Q3 estimates") is correct, it is also misleading and does not give a correct overall picture of the result. The specific numbers tell much more about the actual outcome.

More specifically, the title gives you the impression that things are going great, but the actual numbers paint another much more bleak picture.

 

[sm625]

What shortfall? They beat estimates.

Earnings estimates are always in the shape of fishhooks

They start off high, and then are revised lower and lower and lower until right before the earnings are announced. Then they hook upwards. This scam is what allows companies to always magically produce earnings beats. I dont know why it works, but I guess there are a lot of dummies out there.

 

[Spungo]

They start off high, and then are revised lower and lower and lower until right before the earnings are announced. Then they hook upwards. This scam is what allows companies to always magically produce earnings beats. I dont know why it works, but I guess there are a lot of dummies out there.

Also happens in central planner GDP estimates.

Fed in 2012: We'll raise rates next quarter.
Fed in 2013: We'll raise rates next quarter.
Fed in 2014: We'll raise rates next quarter.
Fed in 2015: We'll raise rates next quarter.
(hint: they can't raise rates when the global economy is already in recession)

 

[dealcorn]

This scam is what allows companies to always magically produce earnings beats. I dont know why it works, but I guess there are a lot of dummies out there.

I suspect the fish hook shape curve you see regarding earnings is driven by the ancient saw "The market does not react, it overreacts." Presumably, there is some academic study that shows the market punishes a stock's price less when any earnings shortfall is disclosed prior to release rather than at release. For this reason, the j shape that troubles you is a prudent strategy to deliver news in a way the market finds palatable.

I agree there are a lot of dummies out there, but I lack confidence that you have identified the dummies correctly. Using common sense strategy in communications is typically not viewed as a scam.

 

[teejee]

Teejee, I dont need an education in reading an income statement, but thanks anyway.

well your post still doesn't make sense.
But don't worry, I won't try to educate you anymore

 

[mrmt]

I agree with Seth Klarman and Warren Buffet on this one - EBITDA is a useless metric.

Ebitda is just a starting point for most analysis, a better one than net profits and one less subject to interferences.

 

[witeken]

Makes sense- if 10nm is pushed out, 10nm equipment buys get pushed out too.

So here's the deal.

Intel has this neat 2 year (24-26 months) cadence, which painfully broke dowm recently because III-V is very hard, and even more importantly, because double, triple and at 10nm for Intel maybe even quadruple patterning is very hard. So they're now at something like 2.5 years for 14nm and presumably 10nm as well, even though they learned from 14.

So, obviously 10nm didn't happen 4 years after 22nm. If it had, then Intel would have started 10nm spending in Q4'14 (a year ago). Amazingly, they said at the investor meeting that you would see the front-end of the 10nm startup. Even more amazingly, it actually showed in the quarterly reports of Q4 and I think also Q1 (but not as much as what it had been for 14nm).

But after Q2, they announced the delay. So instead of 4 years after 22nm, 10nm will arrive 5 years after it, averaging 2.5 years for 14 and 10 (actually a bit more than that since IB was released in May).

And this is what we're seeing now. The start-up costs will start in Q4 (minus 1.0 point gross margin reconcillation), and will increase in H1, for (hopefully) production (qualification) in Q4'16 and HVM in H1 for high volume H2 release of 2017.

 

[AtenRa]

If PC market will continue shrinking the next one two quarters, i dont believe we will see 10nm products in retail in 2017.

 

[krumme]

If PC market will continue shrinking the next one two quarters, i dont believe we will see 10nm products in retail in 2017.

Yes we will. The shiny things on mobile is still the driver. Look at apple and ss profit.
In late 2017 we will have loads of 10nm from ss apple and qcom.

http://m.gsmarena.com/qualcomm_snapdragon_830_to_be_made_on_a_10nm_process-news-14476.php

As i wrote 3 years ago Intels venture into mobile would end like it does now (and how much crap did i get from the usual btw). What a waste of ressources, talent and valuable time. Now competing against mediatek, depreciated 28 nm a72, samsung, apple, china and an entire ecosystem with a monopoly x86 solution as alternative. What worse is there to do? Its time to do as nv did. Retire it - and concentrate on the future for eg dcg and more integral solution for business.

 

[mrmt]

Do you know what estimates mean? At least your use of the word doesn't make sense. It is just what a bunch of market analysts predicted shortly before the report. It says nothing about absolute performance.

It gives a good indication of very short term stock development. Nothing else.

Actually the estimates Phynaz was talking about was the estimate Intel laid out on the end of Q2, not what analysts were predicting just before the earning releases. Serious companies usually laid out a few quarters down the road, the most amateur ones give at least single quarter forecasts.

 

[Fjodor2001]

If PC market will continue shrinking the next one two quarters, i dont believe we will see 10nm products in retail in 2017.

Yes we will. The shiny things on mobile is still the driver. Look at apple and ss profit.
In late 2017 we will have loads of 10nm from ss apple and qcom.

This does not compute. If Intel 10 nm will arrive in 2018, it would mean TSMC/GF/Samsung won't have 10 nm until 2022-2023. Just check this:

It seems at 10nm that Intel may be 4-5 years ahead. Assuming that 10nm ever materialize for the foundries.

 

[ShintaiDK]

This does not compute. If Intel 10 nm will arrive in 2018, it would mean TSMC/GF/Samsung won't have 10 nm until 2022-2023. Just check this:

Yes, because 10nm must equal 10nm. Both in sizes and electrical properties. Samsungs 14nm isn't exactly showing anything good with the A9.

And for size.

Maybe you learn one day that the term nm is used...very lighthanded by everyone.

Samsungs 14LPE is 37% bigger than Intels 14nm(dense). TSMCs 16FF is 58% bigger and 16FF+ is 40% bigger. And then there is the materials/electrical properties to discuss.

 

[jpiniero]

As i wrote 3 years ago Intels venture into mobile would end like it does now (and how much crap did i get from the usual btw). What a waste of ressources, talent and valuable time. Now competing against mediatek, depreciated 28 nm a72, samsung, apple, china and an entire ecosystem with a monopoly x86 solution as alternative. What worse is there to do? Its time to do as nv did. Retire it - and concentrate on the future for eg dcg and more integral solution for business.

Servers aren't enough to fuel Intel's capital spending. If PC Sales continue to plummet Intel has big problems even as it wipes out AMD. Wonder if they will seriously go after the Apple A11 foundry deal. Intel's 14 nm would still be very competitive with the fake 10 nm from the foundries.

 

[Arachnotronic]

Servers aren't enough to fuel Intel's capital spending. If PC Sales continue to plummet Intel has big problems even as it wipes out AMD. Wonder if they will seriously go after the Apple A11 foundry deal. Intel's 14 nm would still be very competitive with the fake 10 nm from the foundries.

I think Intel's best shot is to move its modems to the leading edge and win Apple modem business.

 

[ShintaiDK]

I think Intel's best shot is to move its modems to the leading edge and win Apple modem business.

They should get 30% modem allocation with the iPhone 7 or whatever it will be named.

 

[ShintaiDK]

Servers aren't enough to fuel Intel's capital spending. If PC Sales continue to plummet Intel has big problems even as it wipes out AMD. Wonder if they will seriously go after the Apple A11 foundry deal. Intel's 14 nm would still be very competitive with the fake 10 nm from the foundries.

Hopefully they start to accept the current death in smartphones CPU wise. Specially with the race to the bottom for the vast part. And then stay focused on high end tablets using Core parts. That would mean fabs can be used without competition to companies willing to pay in blood. And Apple is that type of company.

Volume is the key. And easier to let Apple pay out the nose than doing it yourself to increase volume. However it would need some kind of long term deal I guess, since Intel is rather limited in the customer area.

 

[Arachnotronic]

They should get 30% modem allocation with the iPhone 7 or whatever it will be named.

Doubt it. Until Intel can successfully integrate CDMA EV-DO into its modems, I don't think Intel has much of a chance of winning Apple biz.

That said they just bought the CDMA assets of Via Telecom, probably because they are dead serious about trying to compete for the Apple biz. It will be interesting if they can get those assets integrated into something that can intercept a 2017 iPhone.

 

[ShintaiDK]

The modem in question is the XMM 7360. And the 70% allocation would come from Qualcomms MDM 9645 if true.
Huawei demonstrated 1.36Gbit 5G with NTT DoCoMo as the first customer as a sidespring.

 

[Arachnotronic]

The modem in question is the XMM 7360. And the 70% allocation would come from Qualcomms MDM 9645.

I would be shocked to see Intel use a 28nm modem w/o CDMA support as well as a 20nm modem w/ CDMA support given that Apple seems to be fairly successfully delivering "global SKUs" of its iPhones.

 

[ShintaiDK]

It is speculation from KGI Research.

XMM 7360 factsheet:
http://download.intel.com/newsroom/kits/mobileworld/2015/pdfs/Intel_7360_factsheet.pdf

5-mode operation, including LTE-FDD, LTE-TDD and TD-SCDMA for world markets

Looks good enough to me.

 

[Arachnotronic]

It is speculation from KGI Research.

XMM 7360 factsheet:
http://download.intel.com/newsroom/kits/mobileworld/2015/pdfs/Intel_7360_factsheet.pdf

5-mode operation, including LTE-FDD, LTE-TDD and TD-SCDMA for world markets

Looks good enough to me.

No CDMA though.

 

[Spungo]

If PC market will continue shrinking the next one two quarters, i dont believe we will see 10nm products in retail in 2017.

What about non-PC things? Is there a demand for 10nm cell phones?
That's a serious question. I noticed calculators haven't improved at all in the last 20 years. The TI-84 still uses a Z80 processor. Cell phones do improve every year, but it seems like there's not much reason to improve. My phone is slow as crap because it lacks memory; it has nothing to do with CPU speed.

 

[Phynaz]

Actually the estimates Phynaz was talking about was the estimate Intel laid out on the end of Q2, not what analysts were predicting just before the earning releases. Serious companies usually laid out a few quarters down the road, the most amateur ones give at least single quarter forecasts.

Thank you.