Question Intel Q4 Results

[jpiniero]

https://d1io3yog0oux5.cloudfront.net/_8020e84ca449012613cb7a51a8b3788c/intel/db/887/8894/earnings_release/Q4+22_EarningsRelease.pdf

https://d1io3yog0oux5.cloudfront.net/_f2b8ef2fed169a734498f0ef90617704/intel/db/887/8894/earnings_presentation/Q4%272022+Earnings+Deck_Final.pdf

Yuck. Revenue down 32% and again they lost money.

CCG revenue down 36%, DCG down 33%... that in Q4 compared to previous year.

Q1 Guidance is even worse... revenue down 40% YoY.

 

[Hitman928]

Ouch. I figured it would be bad, but this is just brutal.

 

[Hitman928]

Foundry services and graphics are technically growing revenue, but they are losing a good chunk of money at the same time so either their costs are exorbitantly higher than their competitors, or they have to offer steep discounts to gain market share (or a mix of both). Mobileye is the only segment growing and doing so profitably but it's still a small piece of the pie. Graphics is the big loser with operating loss being 78.5% higher than revenue. They are losing almost twice as much money running the segment than money coming in.

 

[krumme]

Real men have fabs
Jerry Sanders, AMD

`Real Men Have Fabs'

Odd as it sounds, chipmaker Actel Corp. doesn't make chips. It sells them, all right, and its designs for so-called programmable logic devices are hot. But the nine-year-old company farms out production to true chipmakers with spare capacity in their wafer-fabrication plants. This strategy was a...

www.bloomberg.com

 

[Hitman928]

Q1 outlook is in the basement. If AMD doesn't also have a majorly down outlook on Q1 (based on this, they probably will), then Intel will only be making ~2x the revenue of AMD for the quarter with significantly worse margins. Many analysts are predicting a second half turn around for semiconductors this year. Gelsinger better hope that happens or else I think his seat is going to get very hot, very quickly.

 

[Hitman928]

Unfortunately, I suspect that this is going to lead to many more job cuts to try and reign in costs.

 

[Mopetar]

Anyone with a brain should expect graphics to be a money pit for a while. It's going to take a few more generations before Intel is really able to square off against AMD and NVidia.

Intel should probably focus on getting into the data center market more than gaming. The margins are higher and they can give their driver team time to smooth out the software side of things while they refine their hardware designs.

 

[Markfw]

Yes, not good. But due to the bad server chips, and the cost of Raptor lake (DDR5 setup), I want to see AMD's results before I decide the economy is in trouble. (electronics industry)

 

[Harry_Wild]

Intel (INTC) reported its Q4 2022 earnings after the bell on Thursday, missing analysts' expectations as the chip industry continues to struggle with slowing consumer and enterprise demand. What's more, the company is guiding for an adjusted loss of $0.15 per share in Q1. Wall Street was looking for a profit of $0.25 per share.

Here are the most important numbers from the announcement compared to what analysts anticipated, as compiled by Bloomberg.

• Revenue: $14 billion versus $14.4 billion expected
• Adjusted EPS: $0.10 versus $0.19 expected
• Client Computing: $6.6 billion versus $7.4 billion expected
• Datacenter and AI: $4.3 billion versus $4 billion expected

Shares of Intel were off more than 5% immediately following the announcement.

Intel's Q1 expectations aren't much better than its Q4 performance. The company said it expects revenue of between $10.5 billion and $11.5 billion. The Street was looking for $14 billion. Gross margins are also expected to come in at 39%. Analysts anticipated margins to top 45.5%.

Intel is facing a steep drop in consumer PC sales, as shoppers choose to hold on to the laptops and desktops they purchased at the peak of the pandemic. According to Gartner, Q4 worldwide PC shipments declined a stunning 28.5%, the biggest decline since the firm started following shipments in the mid-1990s.

Intel's Client Computing Group was hammered in the quarter, with revenue declining 36% year-over-year from $10.3 billion to $6.6 billion.

Intel's Datacenter and AI business also took a beating, with revenue declining 33% year-over-year from $6.4 billion to $4.3 billion.
————— —— —————-
Intel stock is holding up pretty well given the terrible numbers, only down 5% or so! I might double down but will wait till 2nd quarter results come out to buy more!

 

[Karnak]

If AMD doesn't also have a majorly down outlook on Q1 (based on this, they probably will), then Intel will only be making ~2x the revenue of AMD for the quarter with significantly worse margins.

According to Intel regarding DCAI revenue:

Lower revenue on TAM contraction and competitive pressure

This will obviously continue throughout Q1 since it's getting worse for Intel. For AMD to deliver such a bad outlook, they literally have to cancel all client products and don't sell a single one. I mean GM of freaking 39% in Q1. The hell is Intel even doing. Pat is an awful CEO. Should probably retire sooner than later.

 

[Vattila]

[AXG is] losing almost twice as much money running the segment than money coming in.

Although AXG losses have shrunk substantially, this is not a pretty picture on which to end Raja Koduri's short-lived business group. As this business is reintegrated into the other groups, ongoing losses will carry over into DCAI (data centre) and CCG (client), which already are operating at very slim margins (9% and 11%, respectively).



PowerPoint Presentation (d1io3yog0oux5.cloudfront.net)

 

[Markfw]

According to Intel regarding DCAI revenue:


This will obviously continue throughout Q1 since it's getting worse for Intel. For AMD to deliver such a bad outlook, they literally have to cancel all client products and don't sell a single one. I mean GM of freaking 39% in Q1. The hell is Intel even doing. Pat is an awful CEO. Should probably retire sooner than later.

You take inferior products, and a "light" market, and that spells disaster. I think that AMD may fare much better, due to its superior server chips that everybody has been waiting for them to deliver, and some pretty competitive desktop and mobile chips. We will see in a week or two.

Edit: see "competitive pressure " note above.

 

[rainy]

We will see in a week or two.

Precisely, in a few days, on 30th January.

 

[sandorski]

With the amount of recent layoffs on the Software side of the industry, I'm tempted to think AMD and others are hurting possibly as much.

 

[exquisitechar]

With the amount of recent layoffs on the Software side of the industry, I'm tempted to think AMD and others are hurting possibly as much.

Nah, not as much. This is a disaster.

 

[scannall]

15 years or more of bad decisions are rolling up on them. It isn't a surprise really. Even if AMD had stayed stagnant and irrelevant it would have come anyway, just later in date. When they started chasing quarterly profits instead of long term engineering they mortgaged their future for some quick pocket cash. The thing is, in an industry with such long lead and development times you can't focus on quarterly's.

To be clear, I am *NOT* suggesting Intel is doomed. They aren't. And they do have a few compelling products. They don't have focus. Or a track. I hope Pat stays in charge with enough strength to re-acquire the engineering talent they need. And to hold off the demands of wallstreet long enough to rebuild the engineering side of the company. The last thing they need is another Otellini.

 

[Saylick]

According to Intel regarding DCAI revenue:


This will obviously continue throughout Q1 since it's getting worse for Intel. For AMD to deliver such a bad outlook, they literally have to cancel all client products and don't sell a single one. I mean GM of freaking 39% in Q1. The hell is Intel even doing. Pat is an awful CEO. Should probably retire sooner than later.

It's not actually 39%. It's actually lower, like 36% because Intel changed up their accounting practices for this particular earnings report to elongate their depreciation schedule on older fabs from 5 years to 8 years, which means they take less of a depreciation hit per year, thus inflating their operating margins.

I really do think Intel is truly at a do-or-die moment now and they no longer have the COVID surge in all things semi to save their ass. It's going to be a really dark tunnel ahead of them and it will be exceptionally painful going through it.

Those fabs, as many have said on these forums over the years, must be feeling real heavy by now, like an albatross around Pat's neck. They said during the call that they plan on keeping a "competitive" dividend, but in my opinion the dividend needs to be cut. If the stock price drops as a result, so be it. We're talking about the survival and long-term health of the company here. But alas, Intel are keeping the dividend and instead cutting employees to reduce operating costs. The issue is that Intel needs engineers more than ever these days, as evidenced by the fact that it was a shortage of pre-silicon validation that tanked their server business. Their decision making is truly short-sighted and it's becoming more obvious that Pat cannot turn the Titanic quick enough. Hell, I'm not even sure he's trying to turn the rudder based on some of his decision making.

 

[Markfw]

It's not actually 39%. It's actually lower, like 36% because Intel elongated their depreciation schedule on older fabs from 5 years to 8 years which means they take less of a depreciation hit per year, thus inflating their operating margins.

I really do think Intel is truly at a do-or-die moment now and they no longer have the COVID surge in all things semi to save their ass. It's going to be a really dark tunnel ahead of them and it will be exceptionally painful going through it.

I totally agree. While Raptor Lake has SOME high points, it just barely competes with Zen 4. And Zen4x3d is comping in the next 30 days ? Intel at that point will have NO compelling products in a horrible declining market (short term IMO). SR for servers can't touch Genoa, laptop is a wash, and desktop productivity is on AMDs side, and now gaming in desktop will be AMD in 30 days ????

As said above, I am not saying Intel is finished, just that YES, the next few quarters will be live or slowly die or become irrelevant.

I hate to say this, but they may trade places with AMD in the next few years. Be the underdog scratching for a place to live, in a minority.

 

[Vattila]

It's not actually 39%. It's actually lower, like 36% because Intel changed up their accounting practices for this particular earnings report to elongate their depreciation schedule on older fabs from 5 years to 8 years, which means they take less of a depreciation hit per year, thus inflating their operating margins.

Yeah, Stacy Rasgon points that out in his immediate reaction:

 

[gdansk]

Many companies will have worse than expected numbers. It's not like Nvidia has bad products right now but they'll still get hammered YoY. Intel with a poor lineup outside of RPL *should* be the worst hit.

I'm also guessing Genoa won't be as successful as it deserves. What timing to have the most competitive server line up in decades during a recession/"correction". Even if they do gain a lot of market share it'll be a smaller market YoY.

 

[Saylick]

Yeah, Stacy Rasgon points that out in his immediate reaction:

Thanks for sharing. Stacy brings up another interesting point, too. Intel isn't even giving full year guidance citing uncertainty in the market... That is NOT confidence inspiring at all.

 

[Saylick]

I'm also guessing Genoa won't be as successful as it deserves. What timing to have the most competitive server line up in decades during a recession/"correction". Even if they do gain a lot of market share it'll be a smaller market YoY.

Yeah, AMD got the short end of the stick twice.

The first time is when Milan was absolutely crushing Cascade Lake but AMD could not capitalize on the COVID demand because there was not enough wafer/packaging supply to fulfil the demand. Meanwhile, Intel was selling out of a worse server product purely because people just needed to expand and they'd buy whatever was available.

The second time is now, like you just explained.

 

[itsmydamnation]

Lots of initial Genoa will go to cloud as well, That will be interesting to see might make the ramp look good but then general availability sales hurt.

I've often wondered how these accounting types plan to cut costs when you have no assets to sweat and everything is an OPEX cost.

 

[mderbarimdiger]

some points of the call

1. Intel mentioned twice they now have higher ASP, but in the same time they mentioned that Xeon pricing was impacted by the competitors products, i am not sure how to make sense of that, maybe they sell at higher prices than Intel's own prior products, but not as high as they could have charged if not for AMD

2. intel said regarding SPR "we are on track to ship 1 million units by midyear"

3. promises made about its foundry business is as good as possible to get from promises, actual foundry revenue also showed a significant increase over last Quarter and over last year, however current revenue is way to low of volume to give us any measure about IDM 2.0 progress.

But those who care about promises should be happy with some big promises, including a promise to disclose big wins for intel 16, 3 and 18a during the year, another promising point is intel's update about 20 and 18A progress, "On Intel 20A and Intel 18A.... internal test chips and those of a major potential foundry customer have taped out with the silicon running in the fab" and he added "we continue to make progress on Intel 18A, and I've already shared the engineering release of PDK0.5 with our lead customers and expect to have the final production release in the next few weeks"

(rumor is saying that Qualcomm is the costumer that Intel is referring to)

4. promised to interduce and ramp emerald rapids and meteor lake in second half of this year.

---
Microsoft already confirmed the obvious, windows sales were considerably down in last Quarter, no Question the desktop market is not in best shape at the moment.

Intel CFO mentioned 4 times in the call underload, which means Intel sits on inventory they can not sell, AMD will not have much better numbers.

I am concerned that AMD will report on lower profit margins regardless of the desktop market conditions, Given that Zen4 is based on TSM advanced nodes, while RPL and ADL are based on a mature and internal node, i suspect that AMD had much higher cost for Zen4 VS intel cost for RPL, I would expect Intel to have a far better cost advantage over AMD.

Initially Zen 4 prices were not competitive with RPL, but they Quickly lowered prices, and we know the reason why.

But the server Business might be the place where AMD will possibly show strength.

I am afraid that AMD is close to reaching the peak, Epyc has a few more years to shine, but ultimately not only will intel better compete than they do currently, but the entire X86 server dominance will at some point fade, All the hyperscale's should be considered as competitors to X86.

not that X86 will somehow collaps very soon, quite the contrary, it will perhaps have its place for a very long time, but its going to be like windows Vs Linux, windows is still a major server OS, but Linux is by now the de-facto server OS for most applications, The same game will playout with with Risc - V, Amazon, Google, Alibaba, etc will not be happy to pay hefty margins for the privilege to use X86, and we already see billions of dollars and great talent being applied to big Risc-V projects, its open sourced and considered a much simpler architecture than any other architecture (i belive I've heard that from Jim Keller), that is besides the growing presence of ARM chips in the server market, this paradigm shift will only accelerate to the benefit of the cloud companies who couldnt care less about AMD or Intel profit margin.

I just think intel will be in a better place Than AMD in 5 years from now, not because Pat is better executing than Lisa, but because of the advantages that foundry will eventually have over designers, i dont expect any other foundry to sprout from nowhere in the next 10 years that will be able to compete with the few big fabs, intel atleast have a chance to improve its fab, whereas AMD have a much smaller chance to take more market share against both challengers, intel's competition on X86, and the Risc-V inevitable bite in servers, + ARMs bigger bites in desktop (in the form of apple or soon Qualcomm and perhaps MediaTek as well).

Intel has little room to fall from here, SPR is definitely inferior to AMDs products, Pat confirmed that several times in the past, but, since everyone is aware about it and its priced in the stock price, this might actually be a good investment opportunity for those who belive that foundry will deliver on its promise, I am in the believer camp but it will take atleast another 18 months to playout.

 

[Tigerick]

Client	Q2 2022	Q3 2022	Q4 2022	Q1 2023
Intel	$7.7 B	$8.1 B	$6.6 B	?
AMD	$2.2 B	$1 B	$0.9 B	?
Server
Intel	$4.6 B	$4.2 B	$4.3 B	?
AMD	$1.48 B	$1.6 B	$1.7 B	?

Above are sequential financial report from Intel and AMD, I will update AMD numbers once they release financial report.

AMD did horribly last quarter compared to Intel gains at the same time which resulted in price cut. I sure hope AMD will do better this quarter but Intel still command large share of corporate and notebook PC market share which AMD have hard time capturing. DIY markets are actually pretty small compared to above mentioned market.

In DC, AMD has better chance to compete, we will see how AMD fare in a few days time. Remember when Lisa mentioned that DC contributed largest revenue in the event, see above turn of revenue between PC and server department, you will understand