Intel Parts Ways with Arm Holdings, Sells Entire Stake

[Nomad76]

why sell stake in what's gonna be the dominant cpu arch and not sell some of their SaaS holdings ML holdings, sports broadcasting and other junk that aren't doing all that great ??
and this is just the list from 2015 on here's a link to the full list of them pissing away money that should have gone to foundry and their core mission https://en.wikipedia.org/wiki/List_of_mergers_and_acquisitions_by_Intel

Exactly what is puzzling me, why Arm? I guess we have to wait...

 

[trsttte]

Exactly what is puzzling me, why Arm? I guess we have to wait...

Here's a couple guesses:

• They don't need it. They don't need to be a shareholder to license the core designs if it comes to it, there's so many big players they don't really have a meaningful ammount of power (nor does anyone else) and ARM doesn't make that much money, they've made moves to change that but it hasn't yet and they risk antagonizing everyone to the point of starting using alternatives
• Rumours of x86's death (and ARM supremecy) have been greatly exagerated. Though Apple was very successful in their transition from x86 to ARM, Microsoft just proved yet again it can't replicate that and likely never will. Even with Intel struggling to compete, they're still doing it to a point and AMD definitely is doing it, and their combined results against the recent Snapdragon Elite show x86 ain't dead yet.

 

[Daven]

From what I’ve understood, IFS didn’t become a different company by symbolism’s, it’s a bit like Samsung, where things aren’t « free » between the entities. The loss of the CPU side of Intel are gains for IFS. If you work at IFS, you are being paid by IFS, not by Intel the CPU maker. I’m working at a company that operates in a similar way

But if you read the quarterly report, Intel subtracts the IFS revenue out at the end. Why would it matter how workers are internally organized? At the end of the day, Intel is trying to make it seem IFS has a lot of customers when really it has just itself. This is a slimy maneuver.

 

[Neo_Morpheus]

Lovely karma doing its thing.

Die intel die!

And for the hypocrites that will cry competition, none of you have a problem with Ngreedia monopoly, so..shhhh.

 

[64K]

Lovely karma doing its thing.

Die intel die!

And for the hypocrites that will cry competition, none of you have a problem with Ngreedia monopoly, so..shhhh.

People need to stop throwing around the word monopoly. Here's the definition of a monopoly:

the exclusive possession or control of the supply of or trade in a commodity or service.

Note the word exclusive.

 

[Daven]

People need to stop throwing around the word monopoly. Here's the definition of a monopoly:

the exclusive possession or control of the supply of or trade in a commodity or service.

Note the word exclusive.

Exclusive applies to the word ‘control’ too. You can sell a similar product but have no ability to gain market share or be profitable when a monopoly ‘controls’ the market.

Without a doubt, Nvidia exclusively controls the compute GPU market. Anyone who says otherwise should be ignored. Intel use to control the whole CPU market but lost that control years ago.

 

[Random_User]

This is really the intel's business, and I shouldn't have write this, as I'm not a professional in the finances, and it's not me to tell anyone what to do. But this is nevertheless, strange, to say the least. Qualcomm seems have come into intel's room and taking it's dinner. But instead of rivaling, and chewing own piece of a growing ARM business, they give up, leaving themselves with own old inferior tech.
Maybe they have other, superior RISC designs, or they are such confident in their upcoming whatever next uarch, that will carry the new Core Ultra monicker... Or they are reacting impetuously, unwisely, and clearly "cutting corners" in wrong departments.
If that comes so far into the saving money, this is more reasonal, that their marketing, and C-suit are first to get rid off. Just saying.

Wont happen until old staff inc CEO are kicked out who think Intel is too big to fail.

Indeed! There was a lot of big, wealthy and really successful companies, and it took much less mistakes to get them under. Intel was actively burying themselves with an excavator for a good while now. It's a "miracle" they are still afloat, after all their f-ups and unique business sellouts.

Interesting graphic but it is misleading. IFS only made about $100 million or so on third party chip fabbing not $4.3 billion. I'm really disappointed that Intel chose to use accounting tricks to make itself look good. In case you don't know what I'm talking about, just add all the bar graph segments up. It doesn't equal $12.8 billion revenue because Intel is double counting the cost of making it's own chips as if it were a third party customer of IFS.

IFS. After their CPUs got screwed by their own fabs, and the issue, which has been "addressed" on early stages of 13th gen Raptor-Lake manufacturing, slipped into 14th gen... I doubt there would be many really willing clients, even if IFS would be open for other silicon companies, like nVidia, AMD, Apple, Quallcomm, and others.
It honestly, would take intel a trully successful launch if new successful and issue-free priducts and arch, and sell it for a while (at least a year of continous good reviews and reports), in order to recover the lost credibility.
Yes, any company can have manufacturing issues, like nVidia, and AMD, etc. But the problem is, thiae are fabless companies. And intel had years of 14+++++++ "lingering", and troublesome 10nm (intel 7) inception, and now the RPL issues... This all doesn't ecourage, TBH.

 

[trsttte]

Qualcomm seems have come into intel's room and taking it's dinner. But instead of rivaling, and chewing own piece of a growing ARM business, they give up, leaving themselves with own old inferior tech.

They don't need to be shareholders to use any of the core designs. They jumped in the IPO like everyone else in the industry after the nvidia acquisition debacle to help prevent any one competitor company from being able to take control of ARM for themselves but, after that, being a shareholder may or may not be in their interest - I'd argue it's not in any company's interest, it's good to have ARM ownership distributed by many players but it's not really interesting financially for any of them to be shareholders. Intel decided to stop taking that hit because they need to right now.

 

[mechtech]

"suspend dividend payments"

Also suspending all exec bonuses?

 

[SHADOW MAN]

The ever tightening spiral! Someone better get their hand on the wheel or this sucker is going off the road!

 

[Visible Noise]

Is there another tech-company with So Many technological disasters, sells and product / project cancellations?

Here is a Short List:

- Itanium - cancelled / a complete disaster
- XScale ( ARM-based RISC architecture ) - sold to Marvel Technology Group ( mistake! )
- A line of smartphones with Intel CPUs - Stopped manufacturing / Intel could use XScale ( ARM-based RISC architecture instead of Intel processors using CISC architecture )
- Intel Xeon Phi - cancelled / a complete disaster / used Atom-based cores instead of Ivy Bridge-based cores ( mistake! ) / Worked on a project using an Intel Xeon Phi server and exceeded its capabilities in about 4 weeks
- A series of Many Integrated Core processors - cancelled ( these processors were Too expensive compared to NVIDIA GPUs )
- Purchased Altera ( FPGA technology ) to compete with AMD since it purchased Xilinx
- Sold Altera - failed to compete with AMD-Xilinx
- A complete disaster with process technologies smaller than 14 nm / process "chaos", like 14 nm, then 14+ nm, then 14++ nm, then 14+++ nm, then 14++++ nm, and so on
- RISC-V technology - cancelled ( mistake! )
- Inability to scale up and port Almost Perfect Integrated GPU technology to Discrete GPU technology in order to compete with NVIDIA and AMD GPUs for Data Centers
- OneAPI architecture - already over-complicated and will be cancelled in the future

Do you really want to go there?

 

[Dr. Dro]

Lovely karma doing its thing.

Die intel die!

And for the hypocrites that will cry competition, none of you have a problem with Ngreedia monopoly, so..shhhh.

IFS. After their CPUs got screwed by their own fabs, and the issue, which has been "addressed" on early stages of 13th gen Raptor-Lake manufacturing, slipped into 14th gen... I doubt there would be many really willing clients, even if IFS would be open for other silicon companies, like nVidia, AMD, Apple, Quallcomm, and others.
It honestly, would take intel a trully successful launch if new successful and issue-free priducts and arch, and sell it for a while (at least a year of continous good reviews and reports), in order to recover the lost credibility.
Yes, any company can have manufacturing issues, like nVidia, and AMD, etc. But the problem is, thiae are fabless companies. And intel had years of 14+++++++ "lingering", and troublesome 10nm (intel 7) inception, and now the RPL issues... This all doesn't ecourage, TBH.

Via oxidation issue only affected a small batch of early-release 13th Gen CPUs. The vast majority of them are not contaminated. Fab contamination occurs every now and then and generates catastrophic losses, it has happened to TSMC in the past as well

TSMC’s Fab 14B Photoresist Material Incident: $550 Million in Lost Revenue

www.anandtech.com

TSMC said that a batch of photoresist it used included a specific element which was abnormally treated, creating a foreign polymer in the photoresist. The problem was detected late when the wafer yeilds were lower than expected. As it turns out, consequences of the photoresist incident at Fab 14B were more serious than initially calculated by TSMC. There are media reports claiming that between 10,000 and 30,000 wafers were affected and had to be scrapped, but TSMC has never confirmed either of the numbers.

According to media reports, the affected companies include HiSilicon/Huawei, NVIDIA, and MediaTek, but TSMC has not disclosed names of its customers that suffered from the incident. The only thing that TSMC does confirm is that it has already negotiated new delivery scheduled with its customers.

That was a multi-million writeoff then, and even then the industry's faith in TSMC was not "shaken" - mistakes are made every now and then.

Exactly what is puzzling me, why Arm? I guess we have to wait...

My best guess remains that they are going to be launching an offensive in this front sometime. Meteor Lake, for example, advanced significantly in many areas where ARM chips have a clear advantage over the traditional x86 design, including those from AMD. It will be interesting to see

 

[Caring1]

Sounds like they are trimming the fat all around, to pay for the inevitable recall and lawsuits.

 

[LabRat 891]

why sell stake in what's gonna be the dominant cpu arch and not sell some of their SaaS holdings ML holdings, sports broadcasting and other junk that aren't doing all that great ??
and this is just the list from 2015 on here's a link to the full list of them pissing away money that should have gone to foundry and their core mission https://en.wikipedia.org/wiki/List_of_mergers_and_acquisitions_by_Intel
View attachment 358964

I didn't realize Intel ate Centaur Technologies (Cyrix-VIA, C3, C7, etc.)

 

[unwind-protect]

Strange decision, they must anticipate a need for a lot of capital in the near to distant future.

Might also be a fear of government action. It is a direct competitor after all.

 

[Random_User]

Via oxidation issue only affected a small batch of early-release 13th Gen CPUs. The vast majority of them are not contaminated. Fab contamination occurs every now and then and generates catastrophic losses, it has happened to TSMC in the past as well

I'm sorry, but something doesn't add up. Just by simple logic- if the oxidation has been fixed on the early stages of 13th gen production, then how comes, that 14th gen is prone to the all same degradation issues? Where's the chance that most (if not all) high-voltage degradation issues, are indeed being caused by the oxidation alone? Who knows, if the engineering sample was fine with high voltages and clocks long term, but something went off during the manufacturing, the fab scr*wed up,. And the "manufacturer" decided, that instead of uncovering the whole mess, and informing the buyers beforehand about the bad batch, and tell which serial numbers are affected, right when the problem was spotted, to just hide it under the curtain, hoping none gonna find it.
This is unacceptable, even if this would have been the issue of early 13th gen batches alone. Even then, Intel, by their own had to come up to the buyers, supplier and retailer, asking to withdraw the failed SKUs form the shelves, and use. Even then. But the situation went downward spiraling, towards the complete cesspool, when Intel decided to bite their primary and key "hand" that was feeding them, for their lifetime- the OEM/enterprise and bulk buying clients. The ones, who are shafted the most (yeah, yeah, and laptop users as well).

So this brings to the conclusion, and a coulpe of questions:

1. If the 14th gen is unaffected by the oxidation issue, why they being so reluctant, and refuse to honour the RMA replacement, before, the issue went to the web as a wildfire?
2. If the issue is not that significant, as Intel claims, why adjust the warranty term, istad of granting the RMA replacement from the get go? Maybe, because there's not much unaffected/deffect-free chips/SKUs, to swap to?

 

[TumbleGeorge]

I'm sorry, but something doesn't add up. Just by simple logic- if the oxidation has been fixed on the early stages of 13th gen production, then how comes, that 14th gen is prone to the all same degradation issues? Where's the chance that most (if not all) high-voltage degradation issues, are indeed being caused by the oxidation alone? Who knows, if the engineering sa

No, just oxidation was different problem than degradation. The last one isn't consequence from the first.