News Jim Keller says a 'great Intel' is worth $1 trillion, company would be sold at fire sale pricing if sold now

[Admin]

Jim Keller would be sad if Intel gets split and its divisions will be run by different companies.

Jim Keller says a 'great Intel' is worth $1 trillion, company would be sold at fire sale pricing if sold now : Read more

 

[bit_user]

if Intel gives up its own manufacturing, it will not only cease to be an integrated device manufacturer (IDM) but also lose one of its strong points: complete control over products and their manufacturing.

Right now, that exists only for their server products. It went out the window for client CPUs with Meteor Lake and they never had it for GPUs.

 

[watzupken]

It may be worth "1 trillion bucks" but this is clearly not the value now. Taking it private sounds like a great plan, but is also a great way to make the individual poorer each day. I am not saying it is not possible, but more around the willingness to pay for something that is a heavy financial burden with no turnaround possible in the mid term given that Intel does not have a healthy list of clients using their fabs.

 

[TommyCC]

Intel cannot make another mistake like hiring the like of Gelsinger, who bragged about 4 process nodes in 5 years. If Intel can find another Lisa Su, it might work.

 

[Pigpig]

Intel cannot make another mistake like hiring the like of Gelsinger, who bragged about 4 process nodes in 5 years. If Intel can find another Lisa Su, it might work.

Pat is doing good actually, intel 18a is nice, but intel board cannot wait anymore…. The problems should be intel ex ceo who refused to buy EUV and keep stuck in 14nm technology.

 

[rluker5]

Intel cannot make another mistake like hiring the like of Gelsinger, who bragged about 4 process nodes in 5 years. If Intel can find another Lisa Su, it might work.

OK they will get just 4 nodes in 4 years since they decided to outsource 2nm to focus on 1.8nm. And how well did that outsourcing work for them?
Apparently the most node progress in the world at this time is a miserable failure to you. What has TSMC done in comparison over the same time span? 3nm, 3nm+, 3nm++?
Meanwhile Lisa Sue can only do what Intel did with 2nm when it comes to manufacturing.

I'm looking forward to seeing how well 18A does later this year, but more so on unlocked desktop chips.

 

[bit_user]

Taking it private sounds like a great plan,

The risk is that private investors stop developing new nodes and just milk the existing ones. That's like page 1 of the private equity playbook. The US government would need to create some incentive to keep them doing new node development. Maybe along the lines of CHIPS, or some kind of equivalent tax breaks + interest-free loans, etc.

given that Intel does not have a healthy list of clients using their fabs.

There's an upside to a spinoff of their fabs, which is that it should be easier for them to attract big clients, the more distance they can put between themselves and the design side of Intel.

 

[bit_user]

Meanwhile Lisa Sue can only do what Intel did with 2nm when it comes to manufacturing.

I'm not really sure what you mean by this, but what's ironic about her leading a fabless company like AMD is that she came up through the fabrication side of Texas Instruments & IBM's semiconductor business and certainly knows a lot more about lithography than about CPU design.

 

[nogaard777]

Intel cannot make another mistake like hiring the like of Gelsinger, who bragged about 4 process nodes in 5 years. If Intel can find another Lisa Su, it might work.

What are you smoking? Intel made MASSIVE gains in it's process nodes under Gelsinger. In those 5 years Intel went from the 14nm+++++++ nightmare and way behind TSMC to actually being ahead again. Hiring Gelsinger put Intel back on track. Firing him is the mistake that tells you it's still the same old Intel run by idiot investors that think changing CEOs more often than toilet paper will bring instant profits.

 

[bit_user]

Intel made MASSIVE gains in it's process nodes under Gelsinger. In those 5 years Intel went from the 14nm+++++++ nightmare and way behind

A lot of that was already baked in the cake, by the time he started. Most of the stuff that Gelsinger deserves credit for is just starting to see the light of day.

I'm not too sure about Intel 3. Maybe @jkflipflop98 can say if he feels Gelsinger had much impact on it.

Hiring Gelsinger put Intel back on track. Firing him is the mistake that tells you it's still the same old Intel run by idiot investors that think changing CEOs more often than toilet paper will bring instant profits.

IMO, we should go back to a higher tax rate on short-term capital gains. That would incentivize investors to be more long-term focused. That influences board composition and their decisions, such as this one.

 

[ekio]

Just like my company.
It’s worth 100k but it could be worth 100 Milions in an hypothetical future where I miraculously suddenly do everything perfect.

Intel is worth not much because their x86 advantage is dying, their GPU are waay behind Nvidia and AMD, their fabs are never going to beat TSMC and they sold their diversified portfolio to stay afloat, and their brand is not anymore synonymous with champion but with falling behind.

I think thais comment is just a courtesy from JK for his friends there who are sinking.

 

[bit_user]

their x86 advantage is dying,

True.

their GPU are waay behind Nvidia and AMD,

Alchemist was more competitive in AI and RT than raster performance. It seems to me that Battlemage closes much of the deficit on Raster, but the main problem is just that it only competes at the lower-end. They need to do a proper mid-range card. Also, it seems launch volumes were not high, because it went out of stock almost immediately.

their fabs are never going to beat TSMC

We don't know that.

and they sold their diversified portfolio to stay afloat, and their brand is not anymore synonymous with champion but with falling behind.

Yes, sadly.

The typical consumer, who hasn't been following this stuff, probably still has a positive impression of the brand.

I think thais comment is just a courtesy from JK for his friends there who are sinking.

I disagree. I do think one way to read it might be as a bid for the CEO's chair, or at least CTO. Then again, Tenstorrent seems to be doing pretty well, so I'd guess he'd really like to see that to fruition.

 

[theboeingman]

This is what happens when you have myopic and greedy shareholders who are incapable of realising that greed does not grow a business. They are to blame for this mess.

 

[Gururu]

This is what happens when you have myopic and greedy shareholders who are incapable of realising that greed does not grow a business. They are to blame for this mess.

This is rooted in Millennials and Gen Z entitlements. There is no longer respect for neighbors, local or even American values anymore. To be fair, it’s not easy to get by let alone secure retirement. So it’s all a money grab, slash and burn. All American companies at risk unfortunately. However, nVidia is on top because of shareholder greed, but they are doing well so we turn a blind eye?

 

[Giroro]

Giving TSMC (or any company) a de facto global monopoly over all of the computers would be, you know, a bad thing.

 

[TerryLaze]

given that Intel does not have a healthy list of clients using their fabs.

Is any of the new FABs even operational yet?!
Of course there would be very little demand for all the old FABs.

Also does intel really need a big list of clients?
Having all of that FAB space available to churn out AI GPUs or whatever makes money would make them be able to sell super cheap and make a big profit.

 

[TerryLaze]

Giving TSMC (or any company) a de facto global monopoly over all of the computers would be, you know, a bad thing.

Ah, but we have ARM, and they do run windows now so from a legal perspective I think that would be enough.

#I'mnotalawyer

 

[bit_user]

This is rooted in <generational conflict>

Oh, please. Let's not go there. Ever since the dawn of human civilization, older generations have been complaining about younger ones. There's a lot of stuff going on in the world and it's unfair to pin it on the younger generations.

To the extent you think it's generational, you're basically giving up any hope of a solution. It would be much more constructive to focus on what's wrong and ways to fix it. Like my point about short term capital gains tax rates (which, by the way, weren't undone by the generations you cited). I think there are other ways we could improve corporate governance standards, but that would be a start.

 

[stuff and nonesense]

Ah, but we have ARM, and they do run windows now so from a legal perspective I think that would be enough.

#I'mnotalawyer

Who produces the silicon circuitry?

 

[Root Canal]

Taiwanese read the situation better than any bean counter on American soil:

Intel's 18A advantage undermines the case for a TSMC takeover​

Amanda Liang, Taipei; Vyra Wu, DIGITIMES Asia​

 

[Gururu]

Oh, please. Let's not go there. Ever since the dawn of human civilization, older generations have been complaining about younger ones. There's a lot of stuff going on in the world and it's unfair to pin it on the younger generations.

To the extent you think it's generational, you're basically giving up any hope of a solution. It would be much more constructive to focus on what's wrong and ways to fix it. Like my point about short term capital gains tax rates (which, by the way, weren't undone by the generations you cited). I think there are other ways we could improve corporate governance standards, but that would be a start.

Hook line and sinker 😛

 

[jkflipflop98]

What are you smoking? Intel made MASSIVE gains in it's process nodes under Gelsinger. In those 5 years Intel went from the 14nm+++++++ nightmare and way behind TSMC to actually being ahead again. Hiring Gelsinger put Intel back on track. Firing him is the mistake that tells you it's still the same old Intel run by idiot investors that think changing CEOs more often than toilet paper will bring instant profits.

I agree with this. Pat's plan was sound and he had everyone in the company fired up and ready to actually pull off the impossible. 5 nodes in 4 years has never been accomplished before in history until now. The money-focused people with no vision are responsible for Pat leaving the way he did. Most people don't understand how long it takes to create a leading-edge node like this. It's a massive endeavor that takes years to do it once.

Pat has always been one of my personal heroes that I look up to, and it's sad to me that someone else is going to swoop in and take credit for "turning the company around" when it was all Gelsinger's vision that made it possible.

 

[systemBuilder_49]

I agree with this. Pat's plan was sound and he had everyone in the company fired up and ready to actually pull off the impossible. 5 nodes in 4 years has never been accomplished before in history until now. The money-focused people with no vision are responsible for Pat leaving the way he did. Most people don't understand how long it takes to create a leading-edge node like this. It's a massive endeavor that takes years to do it once.

Pat has always been one of my personal heroes that I look up to, and it's sad to me that someone else is going to swoop in and take credit for "turning the company around" when it was all Gelsinger's vision that made it possible.

Pat was kind of an idiot. He promised the sun the moon and the wind and the stars to the board, and they were too stupid to see through his lies. Intel just doesn't earn enough money to do 5 nodes in 4 years - and they only FINISHED THREE. So far 18A is a failure - Broadcom rejected it, "This sucks" is basically what they told Intel.

Doing lots of new nodes is not even important! What's important is that Intel has to become a foundry YESTERDAY. The cost of doing the next node is higher than ALL OF INTEL'S BUSINESS COMBINED. The only companies that survive in this environment of spiraling foundry costs are the ones that sign up outside customers because no single market is growing fast enough to pay for the next node - not at Intel, not at TSMC, not anywhere.

 

[jkflipflop98]

Pat was kind of an idiot. He promised the sun the moon and the wind and the stars to the board, and they were too stupid to see through his lies. Intel just doesn't earn enough money to do 5 nodes in 4 years - and they only FINISHED THREE. So far 18A is a failure - Broadcom rejected it, "This sucks" is basically what they told Intel.

Doing lots of new nodes is not even important! What's important is that Intel has to become a foundry YESTERDAY. The cost of doing the next node is higher than ALL OF INTEL'S BUSINESS COMBINED. The only companies that survive in this environment of spiraling foundry costs are the ones that sign up outside customers because no single market is growing fast enough to pay for the next node - not at Intel, not at TSMC, not anywhere.

Absolutely nothing you said here is true. Not a single word. Let the grown-ups talk. Thanks.

 

[sjkpublic]

If Intel is really up for sale, then Management is an abysmal failure. And there has been a switch at Intel upper management from innovation to selling out.