Thread regarding Intel Corp. layoffs

you lost me after the first 10,ooo words lol

@egi
Summary

• Have you ever thought of Intel as a bankruptcy candidate? Maybe you should.
• The company is facing challenges in its Data Center business and is heavily reliant on the Client Computing Group for revenue.
• Intel's future success hinges on the launch of new products with its 18A process in 2026, but financial struggles and debt concerns remain.

For quite some time now, Intel (NASDAQ:INTC) management commentary and the reality at Intel were at odds. Intel Q2 earnings report and earnings call were a case of management finally starting to come to grips with the reality at Intel. Investors should not take the Q2 as a “kitchen sink” quarter. As bad as it was, the news is going to get worse in the coming quarters.

Q2 revenues were in-line, but gross margin of 38.7% was well below guidance of 43.5%. A nearly 5% miss! Consequently, EPS came in at $0.02 as opposed to guidance of $0.10.

What was even worse was that Q3 revenue guidance is essentially flat in what should have been a seasonally strong quarter. The gross margin guidance was an abysmal 38% and the company guided for a loss even on a non-GAAP basis.

Segment level data shows that the Data Center business, DCAI, is in a serious decline and only the Client Computing Group, CCG, is the business holding the fort for Intel. Altera and Mobileye are largely irrelevant for operating purposes, as they both are mainly a source of funds for Intel. While Mobileye has already been spun off, it appears that Altera will reach a similar fate by the end of the year. The foundry business is highly inefficient and will remain so until it becomes competitive - something that is unlikely before 2026 at the earliest. Currently, in addition to generating large losses, it also consumes tens of billions of dollars of capex.

Margin Story Is Not Good and Threatens to Get Worse
Management offered several reasons for the lower margins in Q3:

Unfavorable product mix as the Company ramped its Meteor Lake chips faster than expected. This is expected to continue into Q3. What was unsaid was why the mix was so different from the guidance just 2 months back. Beyond The Hype suspects that competition from AMD is forcing Intel to move up the product stack faster than Intel planned.
Intel accelerated its 4nm and 3nm process move from its Oregon 3nm development fab to its Ireland fab, and initial Ireland wafers are more expensive. Intel claims this move saved $1B in capex and will result in better margins in the future. The current headwind is temporary and will disappear over time.
Underutilization penalty (although one would think that this cannot be meaningful, given the revenues came in-line with guidance)
Higher than typical charges related to non-core businesses – Intel did not explain what these were, but they appear to be small.
More competitive price than expected. A different way of looking at the issue is that Intel’s products are less competitive and do not fetch the premium pricing they used to fetch. This is a problem that will get worse as AMD ramps Zen 5.
While that list was long, it appears that the main reason for the shortfall is the product mix. The problem for Intel is that the product mix headwind will remain mostly in place and will get worse as Intel launches its Lunar Lake CPU in late Q3.

Intel feels that Lunar Lake is a strong product and noted that it will power over 80 new Copilot+ PCs across more than 20 OEMs. Note that this compares unfavorably with the “100+ design wins” that Advanced Micro Devices (AMD) claims to have with its Zen 5 laptop CPUs. Never in the PC industry history has Intel gotten fewer design wins for its new product than did AMD. The limited adoption of Lunar Lake likely has to do with the timing of the chip’s release. September availability misses the back-to-school season and was likely a bit late for an aggressive Christmas toll out.

Lunar Lake, which Intel is excited about, is more competitive with AMD Zen 5 but is expensive to make as it uses chiplets manufactured at TSMC (TSM). Furthermore, Lunar Lake integrates a memory chip on package, which Intel has to procure from memory suppliers and sell to OEMs with “0%” mark-up.

Unlike Nvidia (NVDA), which is making 80% margins on the HBM it is integrating into GPUs, Intel is getting no margin for integrating memory! Readers may wonder why this is the case. The answer is that AMD’s Zen 5 is a superior architecture and is able to deliver high performance without needing an on-die memory. A PC OEM buying a mobile Zen 5 CPU from AMD can buy memory directly from memory producers. To compete with AMD, Intel will have to sell Meteor Lake with “0” markup on memory. As such, Intel will have to eat the cost of memory in every package that has to be thrown away. This can be disastrous in the event Intel has yield problems.

With Lunar Lake being a major seller for H2 and much of 2025, even with good yields, margins will continue to be under pressure. As a result, Intel shifted its goal of getting back to 60% margins to 2026.

Near Term Guidance Is Not Inspiring
What makes the margin situation worse is that the Company’s Q3 guidance is weak. Management noted that the second-half trends are more challenging than the Company previously expected.

Management noted that there will be a modest inventory digestion in CCG, with DCAI, NEX, Altera and Mobileye trending below previous forecast. While DCAI is expected to benefit from improved demand for servers, it seems nearly certain that Intel will slip behind AMD in data center revenues in Q3.

It is quite shocking how Intel went from #1 data center player about a year back to a #3 player behind Nvidia and AMD in a few quarters. The Q4 guidance preview is not great either. The Company expects Q4 revenue growth at the high-end of a “seasonal” 0% to 5% range. (the seasonal adjective has little merit in this context).

This weak H2 prognosis is despite the launch of Gaudi 3 in Q3. Management expects Gaudi 3 to be priced at two-thirds the cost of competitive offerings and deliver roughly 3x performance per-dollar compared to H100. While no new guidance was given, previous language suggests that Intel expects $500M in accelerator revenues in 2024, most of which will come from Gaudi 3.

In summary, Management offered several reasons for the weak revenue growth in H2:

PC TAM is only expected to grow about 3% this year.
While CPU server orders are increasing, data center customers are investing mostly in GPUs.
A small amount of inventory build-up which needs to be digested.
Lack of competitive products (this was phrased as the Company becoming more competitive with next-generation products).
The combination of margin decline and revenue pressures means that the Company is now burning cash very rapidly.

Margin Pressure, Low Revenue Growth, & Large Capex Make Cost Cutting Necessary
Given the above reality on margins and revenues, Intel’s EPS has already dipped into negative territory. With the current cost structure, Intel makes no money with the current level of gross margins.

This is alarming because Intel has massive capex needs. Intel’s net capex, after adjusting for what partners pitch in, is still over $10B a year. And the Company has $48B in debt on its balance sheet.

Given these realities, Intel had to take aggressive cost-cutting actions. Otherwise, with negative earnings and large capex spending, it would not take long for the Company to get into debt trouble. Thus, it was not a surprise that Intel announced major cost reduction efforts. These include:

Over 15% headcount reduction with the majority of reduction completed in 2024 (Intel wisely chose "over 15%" as the actual number could be much higher).
Comprehensive reduction in spending. Employees have been informed that many perks and benefits will be cut.
Suspending dividend starting Q4
As a result of these measures, the Company expects to reduce opex and capex significantly. The bad news for Intel investors is that even the cost-cutting is not sufficient. Sooner than later, the Company would need to raise money from the sale of Mobileye and Altera to de-lever its balance sheet.

Intel’s 2026 Hope Rests Entirely On 18A Process
Given the challenges associated with Lunar Lake margins, management’s hopes rest entirely on the new products the Company hopes to launch with its 18A process starting H2 2025 (with high volume in 2026).

One thing to note in this context is that the 18A naming may be a bit misleading as, on paper, the process appears to be comparable to TSMC’s N2 process which is expected to launch in H2 2025. TSMC management is on record saying that 18A may be more comparable to N3 than N2. Time will tell whose claims are more accurate. In Intel does match TSMC’s N2 with 18A, which is doubtful, it would be quite an engineering accomplishment for Intel.

Management claimed that the 18A process is in good health, and Panther Lake and Clearwater Forest products have already powered on with the new process. Management expects Panther Lake to start production in H1, 2025 with customer shipments in H2.

Intel expects much better margins with Panther Lake, as the product will use Intel 18A chiplets instead of chiplets from TSMC N3 or N2. However, given that Panther Lake does not start shipping until H2 2025, high volume production is unlikely until 2026 at the earliest. Any benefits from 18A and the new products Panther Lake and Clearwater Forest will not be seen until 2026.

Prognosis
Intel’s Q2 earnings report and disclosures, while awful, are only the beginning of the Company’s reorganization journey. Intel has yet to disclose the size of the write-downs it will need to take for the actions it announced. The Company has not even disclosed, let alone faced the music, on the recent Raptor Lake failure issue. Massive write-downs are likely to be announced in conjunction with Q3 earnings – if not sooner.

The story on the process front remains shaky, and investors may not know the real status until 18A reaches high volume production – something that may not happen until 2026. A bet on the company at this point is a bet on Panther Lake and Clearwater Forest. Any missteps here could lead to disastrous financial outcomes for Intel. Any delays could lead to very ugly cash flows in 2026 and create financial stress.

The larger problem for Intel is that management has yet to accept the realities of the changing semiconductor industry. At the turn of the century, Microsoft (MSFT) and Intel were the dynamic duo of the PC industry. The world has changed a lot in the ensuing two decades. Microsoft, under the leadership of Satya Nadella, had the vision and guts to move to a cloud future from windows past. Intel’s problem has been that it was unable to let the past go and clung to its fab focus. The world has changed and the value of a fab in the US now is questionable even with massive subsidies. There is now far more value in products – as can be seen from the ascendancy of Nvidia and AMD, but Intel has risked its product roadmap with its fab focus.

Intel has gotten itself to a point where its debt will start becoming a concern. While no one would have thought of debt problems in Intel’s context, they are now clearly on the table.

In the near term, the stock is likely to continue to decline as the recovery has been pushed back once again. There is little reason for growth investors to hold Intel, as the Company seems to have completely missed the GPU train. Dividend funds are likely to exit over the coming weeks as the news of the dividend cut sinks in, putting ongoing pressure on the stock.

The stock has dropped 20%+ post-earnings, but more downside is likely.

Anyone have a non paywall link or summary?

Things will only get worse.