Well, CFO had also mentioned that with the 18A test chips they had lost some customers: https://www.reuters.com/business/in...-chip-manufacturing-tech-cfo-says-2025-05-13/.
Intel 10Q filing - We may cancel 14A if we do not win a large external customer
- Thread starter Xebec
- Start date
This quote was interesting, (article dated May 13
"It's a fair assessment that Lip-Bu isn't thinking about massive changes," Zinsner said during the call.
I guess 25% layoffs, and considering cancelling 14A and all future nodes aren't massive changes?
I think he knows what he is doing. Based on his comments on AI during the call, he understands it (but he did not explain his ideas clearly, a bit hard to decipher what he meant at the first instance):
www.linkedin.com
Lip-Bu Tan - Intel Corporation | LinkedIn
Experienced Chief Executive Officer with a demonstrated history of working in the… · Experience: Intel Corporation · Education: Carnegie Mellon University · Location: San Jose · 500+ connections on LinkedIn. View Lip-Bu Tan’s profile on LinkedIn, a professional community of 1 billion members.
www.linkedin.com
Intel has been hiring mainly good talkers for the past 15 years, so they became very good at talking.
you can say that for a lot of large high tech companies. at least this time with Intel, there is an outsider that can bring something fundamentally different.Intel has been hiring mainly good talkers for the past 15 years, so they became very good at talking.
hist78
Well-known member
As of the end of 2024, Intel had already spent $3.7 billion on constructing new fabs in New Albany, Ohio. I’ve always felt this decision is questionable. Intel could potentially achieve better efficiency and economies of scale by building new fabs at or near its existing sites in Arizona and Oregon.
I found this to be partly driven by politics.
TSMC gross margins have gone from 48% to 58% over the past decade. What do you think they are going to go to when there is no alternative? Whatever the market will bear that won't kill overall demand. Intel is basically saying to fabless leading-edge chipmakers, commit to 14A or your margins will evaporate throughout the 2030s
This is thing for me… TSMC margins are exploding and anytime anyone has margins like that there will always be competition; it’s just too lucrative.
I mean heck this is what happened to Intel Products where everyone came for their historically high margin lunch.
So I understand that leading edge semi manufacturing is one of the most expensive ventures humans have ever done, but when you’re looking at the insane margins of TSMC wouldn’t you — as the player who is closest to competing — go after that at nearly all costs? Like AMD did in going after Intel? It pays off in the long run if you win…
Nothing risked nothing gained. Hope LBT can land a customer and make that risk pay off instead of just hanging it up and making short term shareholders happy.
I don’t believe Intel is going to walk away from 14A. The threat was caveated all over the place and is meant to help spur it along. I think 14A will have highly competitive technology. If customers won’t come over if you have even competitive technology than why do it?
And there aren’t any other options except Samsung I guess that has been executing very badly for a while. It’ll happen.
I can't know what goes on behind the scenes. But it strikes me quite awkward. Intel is not TSMC and to this point it has not been particularly successful in engaging customers, has multiple efforts to become a fab without much success, has a well-documented fiasco with success on recent processes and is now under financial stress in a business that you need truckloads of money. Mentioning that if a customer does not appear to secure 14A future you will abandon it, is it possible to foster a positive culture or expectation for any company to invest its future products and billions of dollars given this Intel situation? I cannot understand who can take those risks. I repeat I do not know what is happening behind the scenes. Maybe Intel has demonstrated in private some process capabilities that are expected to outperform significantly TSMC in such a way that some companies are considering worthy to take the risk. Or maybe there is some political pressure and a few companies half-hearted take the risk, because they know that somehow they are going to be covered (or loose a government contract that maybe does not require such a competitive product - only that it is manufactured in the states).
I am scratching my head, with all the efforts to onshore manufacturing in the States, why Intel is not pursuing a strategy to heavily promote and build business on top of older nodes. The big checks are not there, but an overwhelming and lucrative part of many products in all sorts of categories and volumes, depends on chips that do not require these leading nodes. Maybe I am naive, but a customer would have a much greater confidence in using a proven node from a new foundry rather a new (and uncertain?) node from a new foundry. And although the money gained would be significantly less, those older nodes and fabs are already paid for, so they would constitute pure profit - and even maybe very aggressive pricing to gain customers. A few success stories + the knowledge gained in interacting with external customers and providing support, would be exceptional. After that, and hopefully with an advanced node that goes really well, Intel could scale its business in the domain. The non-leading edge landscape has a lot more competition, but on the other hand, other fabs only have those processes and manufacturing lines to get money. Intel, through its product division, has already paid the costs and made money on those lines. It is a competitive advantage that not many others in the industry enjoy.
I am scratching my head, with all the efforts to onshore manufacturing in the States, why Intel is not pursuing a strategy to heavily promote and build business on top of older nodes. The big checks are not there, but an overwhelming and lucrative part of many products in all sorts of categories and volumes, depends on chips that do not require these leading nodes. Maybe I am naive, but a customer would have a much greater confidence in using a proven node from a new foundry rather a new (and uncertain?) node from a new foundry. And although the money gained would be significantly less, those older nodes and fabs are already paid for, so they would constitute pure profit - and even maybe very aggressive pricing to gain customers. A few success stories + the knowledge gained in interacting with external customers and providing support, would be exceptional. After that, and hopefully with an advanced node that goes really well, Intel could scale its business in the domain. The non-leading edge landscape has a lot more competition, but on the other hand, other fabs only have those processes and manufacturing lines to get money. Intel, through its product division, has already paid the costs and made money on those lines. It is a competitive advantage that not many others in the industry enjoy.
gc_epictetus
New member
Why would Lip Bu mention the risk to 14a in earnings? He’s a smart industry leader. What is the upside to shareholders, customers, or employees by announcing this news?
Could it be he did it to insulate Intel from the liability of a ‘potential’ of a withdrawal on 14a? If you haven’t noticed, Intel has been sued numerous times for defrauding shareholders by concealing internal problems.
Maybe he did it to confuse or distract. Maybe he’s following his own mantra of ‘under promise and over deliver’.
Its hard to believe a nation and an ecosystem could sit idly by and watch one of the only remaining competitors in such a critical industry abandon ship and suffer the long term consequences. There must be more going on behind the scenes with government and industry leadership.
Maybe Lip Bu sees a path the rest of us don’t, which is what drives the value of a visionary leader. I can’t wait to watch the documentary and see how the story ends!
Could it be he did it to insulate Intel from the liability of a ‘potential’ of a withdrawal on 14a? If you haven’t noticed, Intel has been sued numerous times for defrauding shareholders by concealing internal problems.
Maybe he did it to confuse or distract. Maybe he’s following his own mantra of ‘under promise and over deliver’.
Its hard to believe a nation and an ecosystem could sit idly by and watch one of the only remaining competitors in such a critical industry abandon ship and suffer the long term consequences. There must be more going on behind the scenes with government and industry leadership.
Maybe Lip Bu sees a path the rest of us don’t, which is what drives the value of a visionary leader. I can’t wait to watch the documentary and see how the story ends!
Until recently, Intel had always replaced older nodes with newer nodes, so many of those "pure profit" nodes have long since been eliminated. (I think the oldest process still in production is 22 nm(?) while TSMC still has >250 nm nodes.)
You're right. In principle, it makes sense: fully depreciated processes = printing money in perpetuity. Plus, you've got proven products and yields. However, Intel never intended to use their processes and PDKs for foundry services. Otherwise, Intel 10 and 7 would be ideal candidates.
Instead, it's been more worthwhile (financially, ecosystem-wise, customer adoption, etc.) to develop foundry services at the leading edge (18A and now 14A), which is a bit crazy when you think about it.
It is quite difficult to adapt Intel's old nodes for industry use. They have several collaborations going on (with UMC, for example), but it is not as trivial as just reusing the old equipment.
This situation should have been addressed at Intel 3 and/or 18A, but Pat Gelsinger probably didn't realize it until too late.
That being said, catching up to TSMC's leading edge with 18A itself has been a remarkable win for Intel, IMO. Very few people believed this four years ago.
I am still cautiously optimistic about Intel going forward. I am equally bullish on AMD, I have to add.
siliconbruh999
Well-known member
Intel needs to nail the 14A PPA and the TTM should be at least half a year before TSMC A14 considering 14A Risk Production is 2027 also they need a workable PDK not half assed like 18A.
If they had the node gap like they had they would have been fine attracting customer but it's not like that anymore. They need to have some secret sauce otherwise i don't have much hope for 14A.
If they had the node gap like they had they would have been fine attracting customer but it's not like that anymore. They need to have some secret sauce otherwise i don't have much hope for 14A.
Come to think of it, I wonder if Intel is involved in the process design in the UMC 12nm?