Intel's Foundry Plan Remains the Same, Interim Co-CEO David Zinsner Confirms

[AleksandarK]

Intel's Foundry business is the company's current pain point and probably the reason why the company board of directors forced out ex-CEO Pat Gelsinger. However, the new interim co-CEO, David Zinsner, confirmed that the foundry plan would remain the same. At the UBS technology conference, Zinsner indicated that the company's core strategy remains unchanged and reiterated the forecasts shared in October, highlighting optimism about growth in its PC and server segments. This is a healthy sign that Intel will not lose its foundry subsidiary, which, even though difficult to operate, could be Intel's silver lining with growth opportunities ahead.

Yesterday, we covered the choice of Lip-Bu Tan as Intel's next CEO. However, the new co-CEO, Zinsner, stated, "I'm not in the process, but I'm guessing that the CEO will have some capability around foundry as well as on the product side." A new CEO would be left with a lot of work that, apparently, no one so far can finish. There are speculations that Intel's 18A node is yielding 10% of usable silicon, while Intel's head of foundry business, Naga Chandrasekaran, has noted that 18A node is going through evolution phases to improve final yields and remain profitable, noting that "there's nothing fundamentally challenging on this node now. It is about going through the remaining yield challenges, defect density challenges."



View at TechPowerUp Main Site | Source

 

[Wirko]

Well what else can he possibly say right now? Fine print comes later.

 

[TheinsanegamerN]

Well what else can he possibility say right now? Fine print comes later.

Something that's not along the lines of "we're gonna keep doing the thing that is losing us money, nothing to see here".

 

[Taisho]

10%... Is there any hope left?

 

[Daven]

So no sign of a turnaround as Intel digs itself deeper out of arrogance, ignorance and denialism.

 

[trparky]

So no sign of a turnaround as Intel digs itself deeper out of arrogance, ignorance and denialism.

Pretty much.

 

[Chrispy_]

Intel's plan remains the same...

 

[Daven]

Intel stock is down 20% since Monday and still going down. Don't they understand that there is not a single part of the company that can operate in any way shape or form as the past models. They have to adapt to a completely different market from ten years ago or go bankrupt. I just don't get what they are doing over there.

 

[Darmok N Jalad]

They had 10 years to get it right while still making sellable products. Now the competition is all over them and they have no choice but to execute well. Naturally, this is right when the federal government decides it’s time to invest in Intel. It might be Solyndra all over again, only 10x more costly.

 

[trparky]

Now the competition is all over them and they have no choice but to execute well.

At no time in history has Intel faced such competition not only in the x86 space from that of AMD but also from ARM. They were caught completely flatfooted.

You can see that in how they implemented that stupid E-Core and P-Core platform. Like Intel did years ago pushing the "GHz, GHz, GHz!" mantra, this time it's "Cores! Cores! Cores! MOAR CORES!" all while the software landscape was completely incapable of handling the heterogeneous platform. It's pretty bad when you have to use third-party software like Process Lasso to make sure programs such as games are routed to the proper type cores.

 

[phints]

Intel for almost a decade:

• Intel 14nm - use our 14nm+++++ litho it's all good bro
• Intel 10nm - delayed so long we called it Intel 7
• Intel 4 - delayed so long we skipped it
• Intel 3 - who knows
• Intel 20A - yields so bad we called up TSMC 3N
• Intel 18A - damage control has begun

Wtf how are they still in business?

 

[AcE]

Everyone just wants to save their own head, lying about yields. 10% is a disaster. It’s just the repetition of 10nm again, they’re trying to jump nodes to be better than anyone else and that way they’re messing it up royally. I never believed they could beat tsmc and this is the proof for it.

US stop funding this sinking ship, invest the money elsewhere.

 

[Nhonho]

I bet Intel will publish a loss of at least US$ 20 billion next quarter.

So no sign of a turnaround as Intel digs itself deeper out of arrogance, ignorance and denialism.

(2017)
"AMD is launching a new 8-core CPU with high IPC and affordable price..."
Don't worry, we are Intel, nothing is going to happen to us...

(2024)
"Nvidia is going to enter the CPU market for real..."
Don't worry, our boat is only 3/4 full of water. We are Intel, nothing is going to happen to us...

---------------------------

Intel will only avoid bankruptcy if it launches the CPU of the galaxies, with very high IPC, low power consumption, AVX-512 and at an affordable price. And with motherboards that accept CPUs from several generations. Something very unrealistic.

 

[Redwoodz]

speechless

 

[chrcoluk]

At no time in history has Intel faced such competition not only in the x86 space from that of AMD but also from ARM. They were caught completely flatfooted.

You can see that in how they implemented that stupid E-Core and P-Core platform. Like Intel did years ago pushing the "GHz, GHz, GHz!" mantra, this time it's "Cores! Cores! Cores! MOAR CORES!" all while the software landscape was completely incapable of handling the heterogeneous platform. It's pretty bad when you have to use third-party software like Process Lasso to make sure programs such as games are routed to the proper type cores.

To be fair, again engaging the neutral gear here. The cores addiction AMD started, and initially they had issues with that if the wrong cores were used you lost a lot of game performance. There was some bashing from the community, but nowhere near the level we see now, for what I consider obvious reasons. The more cores strategy was successful for AMD so obviously Intel had to react, its not a craze they initiated. Although I think they have gone over board, the amount of e-cores on Alder Lake is what they should have settled on as a maximum.

If Intel had their own way we would probably still have quad core consumer chips.

Also I think the progress made on AMD side has been nothing short of a miracle, these type of gains in CPU sector in such short time is crazy. Its like they making a sandy bridge every generation.

 

[Bwaze]

Intel for almost a decade:

• Intel 14nm - use our 14nm+++++ litho it's all good bro
• Intel 10nm - delayed so long we called it Intel 7
• Intel 4 - delayed so long we skipped it
• Intel 3 - who knows
• Intel 20A - yields so bad we called up TSMC 3N
• Intel 18A - damage control has begun

Wtf how are they still in business?

And yet they're still selling majority of PC processors, people would think they're somehow barely operating...

 

[trparky]

Its like they making a sandy bridge every generation.

That's what happens when you're in the mindset that you have to fight for every customer in the market. For the longest time, Intel has never had to contend with that kind of mentality. To borrow and modify a phrase... "Nobody was ever fired for buying Intel."

 

[chrcoluk]

That's what happens when you're in the mindset that you have to fight for every customer in the market. For the longest time, Intel has never had to contend with that kind of mentality. To borrow and modify a phrase... "Nobody was ever fired for buying Intel."

I think they just have the better engineers right now. Of course company culture has an effect, but I dont think Intel have an answer right now on the overall performance per watt, TDP, core size issues. Even if the CEO was like ok we putting everything into our next CPU, I dont think it would be much better than what is on the table anyway.

Historically the pendulum has swung between AMD and Intel, it was going to swing at some point to AMD, which makes their decisions to drop other projects like Optane silly, as the company would have been in a batter state with more revenue streams.

 

[trparky]

I think they just have the better engineers right now.

That may very well be the case, especially since a little while ago Intel let go of a lot of people and some of them were their top talent. Some went to AMD, some went to Apple.

 

[AcE]

Of course company culture has an effect

I saw videos and news about bad company culture at intel, so I think this effect was rather big and shouldn't be underestimated. People work way better when they're feeling good at their work place. Also better management, this is also a huge factor. They are doing new architectures every generation, something which Intel never did, Intel was famous for "tictoc", which is "new node, new architecture, new node, new arch" and so on, but not new arch new node like AMD sometimes did and then new arch every generation. It's also a mistake that Intel brings new CPUs every year (sometimes multiple refreshes), it costs time and money and is worse than bringing a new arch every 2 years instead, their cadence is like a new arch every 2-3 years at best, with Skylake it was way worse, Skylake was around for 6 years I think.

AMD successfully modernised themselves and something clicked, big time, after Ryzen came. And it's still like that today. AMD is generally a more modern company compared to Intel, Intel has some dinosaury-vibes to me. Maybe i'm exaggerating.

 

[Wirko]

I saw videos and news about bad company culture at intel, so I think this effect was rather big and shouldn't be underestimated. People work way better when they're feeling good at their work place. Also better management, this is also a huge factor.

We need to put things in perspective. A giant company is a giant company. I don't believe any Fortune 500 company is immensely better than any other when it comes to corporate culture, workplace relations, and good and efficient communication between people and departments.

But even if I said no one is better, some still are worse.

 

[pavle]

Intel for almost a decade:

• Intel 14nm - use our 14nm+++++ litho it's all good bro
• Intel 10nm - delayed so long we called it Intel 7
• Intel 4 - delayed so long we skipped it
• Intel 3 - who knows
• Intel 20A - yields so bad we called up TSMC 3N
• Intel 18A - damage control has begun

Wtf how are they still in business?

Strong brand name.

 

[Wirko]

The overly optimistic plan:

5N4Y

What came of it (if two interim officers count as two):

5CEO4Y

(note the colour of Intel vs. the colour of negative money)

 

[chrcoluk]

That may very well be the case, especially since a little while ago Intel let go of a lot of people and some of them were their top talent. Some went to AMD, some went to Apple.

Corporate culture, they tend to over time forget the value of the people that helped propel them up. This of course isnt unique to Intel.

 

[AcE]

I don't believe any Fortune 500 company is immensely better than any other when it comes to corporate culture, workplace relations, and good and efficient communication between people and departments.

That's your opinion, I genuinely believe that there are even big companies that have a healthy and good culture. And I believe Nvidia, AMD, Apple among others are some of them. All the really successful companies.