Semiwiki on LinkedIn Semiwiki on Facebook Semiwiki on Twitter Register / Log in 
Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/former-intel-ceo-craig-barrett-on-saving-intel.23380/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021770
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

Former Intel CEO Craig Barrett on saving Intel

X

Xebec

Well-known member
A couple of teasers from the article:

"6. The current Intel CEO's comments about not investing in new technology (14A) until customers sign up is a joke."

"9. The FFWBMs (four former wise board members) of Intel continue to claim you have to break Intel into two pieces before any customer will invest in Intel. Be serious. There are many company interactions that involve both supply and competition. It is also extremely hard to imagine Intel really competing with the likes of Nvidia, Apple, Meta, Google, Dell, etc in their well established product lines. By all means, if you want to complicate the problem, then take the time to split up Intel and make the FFWBMs happy but if you’re in the business of saving Intel and its core manufacturing strength for the USA then solve the real problem – immediate investment in Intel, committed customers, national security, etc."

fortune.com

Exclusive: Former Intel CEO Craig Barrett outlines rescue plan to save Intel and America's advanced chip manufacturing

Intel needs a cash infusion of roughly $40 billion to remain at the leading edge of chip manufacturing, according to former CEO Craig Barrett. The U.S. government is unlikely to be the main savior—instead cash rich chip customers like Nvidia, Apple, and Google need to step up.
fortune.com fortune.com
 
Daniel Nenni said:
Here is my takeaway:

10. POTUS and DoC can set the stage, the customers can make the necessary investments, the Intel Board can finally do something positive for the company, and we stop writing opinion pieces on the topic.

:ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO:
Click to expand...

Then what are we supposed to do after that?

 
4. Intel is cash poor and can’t afford to invest in the capacity needed in the future to replace TSMC or even a reasonable fraction of TSMC capacity. They probably need a cash infusion of $40B or so to be competitive.

Is $40 billion even enough?
 
Fred Chen said:
4. Intel is cash poor and can’t afford to invest in the capacity needed in the future to replace TSMC or even a reasonable fraction of TSMC capacity. They probably need a cash infusion of $40B or so to be competitive.

Is $40 billion even enough?
Click to expand...

It depends on the goals. It won’t be enough for many of the goals people want Intel to achieve. For example, TSMC spends each Capex dollar solely on its foundry business, whereas Intel splits each Capex dollar between its product division and Intel Foundry.

TSMC’s 2024 Capex is about $30 billion, increasing to about $40 billion in 2025. Intel’s 2024 Capex is $24 billion, with 2025 estimated at $18 billion. From this comparison, even a $40 billion infusion for Intel would be comically insufficient.
 
Last edited:
The (potential) 'customers' are Intel bitter rivals and disgruntled customers from the last 20-30 years, including the period Intel was dominant and used legal and illegal means to stop competition and protect or increase its dominance. They could invest if they outright buy Intel and the parts of Intel that compete with them are divested. But I guess for Craig Barret and many ex-Intel from the glorious days that would hardly be the 'real' Intel.

Intel hasn't been able to prove they have the capacity to deliver a competitive process that can rival TSMC. If anything they seem to be in the same place that Samsung but with less financial backing that Samsung has from their other business lines and a lot of money spent the last decade in buybacks and grandiose plans that never panned out (fabs that will never be used).

The opinion article looks completely out touch.

US can basically force TSMC to 'relocate' into the US for all practical purposes as Taiwan government behaves as an US protectorate and the TSMC board isn't suicidal and wants to keep making good money. If there was any real danger of China liberating Taiwan they would do in weeks, equipment and engineers and scientists relocated to 'safety'.
 
vmoya said:
The (potential) 'customers' are Intel bitter rivals and disgruntled customers from the last 20-30 years, including the period Intel was dominant and used legal and illegal means to stop competition and protect or increase its dominance. They could invest if they outright buy Intel and the parts of Intel that compete with them are divested. But I guess for Craig Barret and many ex-Intel from the glorious days that would hardly be the 'real' Intel.

Intel hasn't been able to prove they have the capacity to deliver a competitive process that can rival TSMC. If anything they seem to be in the same place that Samsung but with less financial backing that Samsung has from their other business lines and a lot of money spent the last decade in buybacks and grandiose plans that never panned out (fabs that will never be used).

The opinion article looks completely out touch.

US can basically force TSMC to 'relocate' into the US for all practical purposes as Taiwan government behaves as an US protectorate and the TSMC board isn't suicidal and wants to keep making good money. If there was any real danger of China liberating Taiwan they would do in weeks, equipment and engineers and scientists relocated to 'safety'.
Click to expand...
Realistically, the capacity equivalent to TSMC (total) cannot be fulfilled within the US, within practical scope of money/time.
 
It is about dual sourcing not fully replacing TSMC so $40 billion should be enough for limited volume and IFS sustenance.

Daniel Nenni said:
Here is my takeaway:

10. POTUS and DoC can set the stage, the customers can make the necessary investments, the Intel Board can finally do something positive for the company, and we stop writing opinion pieces on the topic.

:ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO::ROFLMAO:
Click to expand...
I like this point very much to the point no BS.
 
hist78 said:
It depends on the goals. It won’t be enough for many of the goals people want Intel to achieve. For example, TSMC spends each Capex dollar solely on its foundry business, whereas Intel splits each Capex dollar between its product division and Intel Foundry.

TSMC’s 2024 Capex is about $30 billion, increasing to about $40 billion in 2025. Intel’s 2024 Capex is $24 billion, with 2025 estimated at $18 billion. From this comparison, even a $40 billion infusion for Intel would be comically insufficient.
Click to expand...

Let's poke at this a little bit -- I think you're right, but what are Intel's real Capex requirements? Maybe an extra $40B is enough for 3 years but not longer?

Intel needs more Capex than TSMC:
- Catch up/exceed on process (HighNA, packaging improvements)
- Intel develops silicon products too (i.e. TSMC doesn't design microprocessors)

Intel needs less Capex than TSMC:
- Less variations of nodes (i.e. TSMC due to general Foundry services needs more specialized options for new nodes)
- Less total foundry capacity needed than TSMC (i.e. TSMC needs 4X* the wafer capacity at a leading edge node of Intel)
- Intel isn't expanding into a new country/region (i.e. first time moving into a whole new area/region, the capital requirements will be higher than 'just build another fab in Israel, next to the other fabs')

What other major factors would adjust Intel's Capital needs for the next 3-5 years relative to TSMC?


*I have no idea what the actual # is, but TSMC simply needs more physical fabs at a given leading edge node
 
Majority of points are just "you have money, i don't, so give me your money"... which is kind of dumb. We need better sales pitch. :D
5. The only place the cash can come from is the customers. They are all cash rich and if 8 of them were willing to invest $5B each then Intel would have a chance.
Click to expand...
But fact is that right now investing in Intel is more productive than own capex (honestly mostly AI datacenters) or stock buybacks...
 
Fred Chen said:
Pilot line operations to begin this year, so presumably some expenditures already: https://techhq.com/news/can-rapidus-rebuild-semiconductor-industry-in-japan/

A more practical reference than Rapidus would be TSMC (obviously): https://www.tsmc.com/static/abouttsmcaz/index.htm
$65 billion for the three fabs in AZ covering 5nm down to 2nm. Funding first announced 2020, production at most advanced node in 2028.
Click to expand...
I don't think 2nm will be produced in Arizona before 2029 at least considering H2 28 is 14A HVM
 
Xebec said:
A couple of teasers from the article:

"6. The current Intel CEO's comments about not investing in new technology (14A) until customers sign up is a joke."

"9. The FFWBMs (four former wise board members) of Intel continue to claim you have to break Intel into two pieces before any customer will invest in Intel. Be serious. There are many company interactions that involve both supply and competition. It is also extremely hard to imagine Intel really competing with the likes of Nvidia, Apple, Meta, Google, Dell, etc in their well established product lines. By all means, if you want to complicate the problem, then take the time to split up Intel and make the FFWBMs happy but if you’re in the business of saving Intel and its core manufacturing strength for the USA then solve the real problem – immediate investment in Intel, committed customers, national security, etc."

fortune.com

Exclusive: Former Intel CEO Craig Barrett outlines rescue plan to save Intel and America's advanced chip manufacturing

Intel needs a cash infusion of roughly $40 billion to remain at the leading edge of chip manufacturing, according to former CEO Craig Barrett. The U.S. government is unlikely to be the main savior—instead cash rich chip customers like Nvidia, Apple, and Google need to step up.
fortune.com fortune.com
Click to expand...
It's interesting that AMD was omitted from the list of companies CB thinks Intel can't compete with, but AMD is doing very well.
 
Daniel Nenni said:
semiwiki.com

Should Intel be Split in Half? - Semiwiki

A recent commentary from four former Intel board members argue that Intel should be split into two separate companies with separate CEOs and separate board of directors. Charlene Barshefsky, Reed Hundt, James Plummer, and David Yoffie wrote that Intel shareholders should insist on a split which...
semiwiki.com semiwiki.com
Click to expand...
Been saying this for years, glad some people are finally figuring out what needs to happen.
 
You must log in or register to reply here.
Back
Top