Holy shit balls is Intel (INTC) Fucked?

[Wiler77 1566]

1 hour ago, Dahobbs said:

A big part of AMD's efficiency and performance gains have been driven by the node improvement rather than by design changes. 

In the server space, the decision to move to chiplets was as important as the node advantage.  They've had a process advantage and design advantage for 5 years.

[Dahobbs 17521]

12 minutes ago, Wiler77 said:

In the server space, the decision to move to chiplets was as important as the node advantage.  They've had a process advantage and design advantage for 5 years.

I know chiplets increased yield and flexibility in terms of product offerings, but I'm unclear what specific advantage they offer in the server space. There is a performance cost to going with chiplets so I'd think monolithic designs would generally fair better all else being equally. Is it just cost you're referring to or is there some other advantage to chiplets in the server space that I'm not aware of?

[Wiler77 1566]

20 minutes ago, Dahobbs said:

I know chiplets increased yield and flexibility in terms of product offerings, but I'm unclear what specific advantage they offer in the server space. There is a performance cost to going with chiplets so I'd think monolithic designs would generally fair better all else being equally. Is it just cost you're referring to or is there some other advantage to chiplets in the server space that I'm not aware of?

Yield is important.  8C die are really easy to yield which makes for a reasonable cost structure as you said.  But the other thing is the ability to scale throughput (i.e. core counts).  AMD has been shipping 64C server CPUs for 5 years.  Chiplets + on-chip fabric made that possible.  With a monolithic design, if yielding a 64C part was even physically possible, it would be in such small volumes to make it unviable to bring to market.

 

 

[Texas Jeff 1942]

BTW, I previously worked for Intel.  They have a horrible internal culture.  I'm not surprised that they are where they are.

[BrazilHorn 6586]

Ancillary to Intel as they used to subsidize them for years in terms of Marketing $ offsets to use Intel exclusively but Dell will be announcing major layoffs soon as well. 

[Wiler77 1566]

Ah, Dell. Another of my favorites.

I guess if Intel suspended dividends, they don't have an extra $1B between the cushions to pay Dell to be their bitch. Sucks that honest,  hardworking people will be the victims in this.

[immamac 42832]

This thread fucks. 

[swraith 345]

15 hours ago, BrazilHorn said:

Ancillary to Intel as they used to subsidize them for years in terms of Marketing $ offsets to use Intel exclusively but Dell will be announcing major layoffs soon as well. 

Dell seems to be in constant reorganization and/or layoffs. My neighbor gets bounced between different buildings in Round Rock, it seems every year. 

[Wally Fairway 5539]

On 8/1/2024 at 5:52 PM, Captainant said:

On the other hand, this goober just yolo'd a 700k inheritance onto INTC this morning so I'm sure it'll all work out lol

That guy makes me look market sage!

[freyguy 4258]

Loving this thread.  So Intel & Samsung suddenly not knowing how to fill up their new fabs....not great Bob.  ASML still controls the world.  TSMC hates the "lazy & entitled" American work ethic.  GF seems to have their shit together for the most part.  TI has been kind of quiet about their new fab in Dallas...I assume that's rolling right along.  NXP & Spansion are selectively investing in their local fabs.  While larger joint ventures in Europe & Vietnam are under way.  Will see if this all induces a worldwide Semiconductor bubble or not.  

[Gravy Train 974]

^TI finished their second Richardson fab in 2022 for 300mm wafer production, following the 2021 acquisition of Micron's fab in Lehi, UT, supporting 65nm and 45nm process:

https://news.ti.com/2021-06-30-TI-to-acquire-Micron-300-mm-semiconductor-factory,-extending-TIs-cost-advantage-and-greater-control-of-supply-chain

https://news.ti.com/tis-new-300-millimeter-wafer-fab-in-richardson-texas-begins-initial-production

Four more fabs are under construction in Sherman, TX, as part of a $30B facility investment:

https://www.ti.com/about-ti/company/ti-at-a-glance/manufacturing/sherman.html

[Gravy Train 974]

Steve / Gamers Nexus completely shreds Intel for their history of gaffes and sour culture, breaking down their timeline responding to the 13th and 14th-gen defects:

Apparently Intel's subreddit moderator has been in damage control mode and moonlights as a reviewer/writer on staff for Tom's Hardware.

[immamac 42832]

2 minutes ago, Gravy Train said:

Steve / Gamers Nexus completely shreds Intel for their history of gaffes and sour culture, breaking down their timeline responding to the 13th and 14th-gen defects:

Apparently Intel's subreddit moderator has been in damage control mode and moonlights as a reviewer/writer on staff for Tom's Hardware.

Watched the first 7 or so minutes. Pretty pretty bad. 

The latest reports are arrow lake lunar lake and meteor lake are all margin dogs because TSMC fab costs are so much higher than internal fab cost. 

So that means they will have an entire generation of lost margins cranking against their cash cow. It's about to get even uglier for Intel unless they can figure out how to get ASPs up. 

[TwiceHorn 31429]

1 hour ago, Gravy Train said:

^TI finished their second Richardson fab in 2022 for 300mm wafer production, following the 2021 acquisition of Micron's fab in Lehi, UT, supporting 65nm and 45nm process:

https://news.ti.com/2021-06-30-TI-to-acquire-Micron-300-mm-semiconductor-factory,-extending-TIs-cost-advantage-and-greater-control-of-supply-chain

https://news.ti.com/tis-new-300-millimeter-wafer-fab-in-richardson-texas-begins-initial-production

Four more fabs are under construction in Sherman, TX, as part of a $30B facility investment:

https://www.ti.com/about-ti/company/ti-at-a-glance/manufacturing/sherman.html

TI seems to have gotten out of the cutting-edge semiconductor business (formerly DRAM for them) and has been a pretty incredibly nimble company over the years.

[Deej 76748]

On 8/1/2024 at 4:56 PM, 0xdeadbeef said:

 Waiting for the axe to swing kills productivity

[texas08 1599]

Hi 

[immamac 42832]

1 hour ago, texas08 said:

Hi 

Why hello there. 

[Horn Under a Bad Sign 20389]

On 8/3/2024 at 12:46 PM, Wally Fairway said:

That guy makes me look market sage!

It strikes me as pretty unusual that a person with that amount of money could die and, two months later, one of her beneficiaries already has the money.  From what I've seen, large distributions like that take much longer to transfer. 

So I need to buy a new Windows laptop that runs Photoshop and the rest of the Adobe suite. What should I get?

Edited August 5 by Horn Under a Bad Sign

[atomheartbevo 89617]

On 8/3/2024 at 1:18 PM, freyguy said:

So Intel & Samsung suddenly not knowing how to fill up their new fabs...

On 8/4/2024 at 11:30 AM, immamac said:

It's about to get even uglier for Intel unless they can figure out how to get ASPs up. 

Everything is under control guys, the intel CEO is going to….pray about it.

 

[freyguy 4258]

Jesus fuck, what the hell?

[Hefeweizen 5308]

12 minutes ago, freyguy said:

Jesus fuck, what the hell?

Appropriate phrasing.  Yeah I’d bail on that one.  Thoughts and prayers INTC 

[PilotsError 2286]

CDC posted some Bible verse right before he hired Schloss, so...I dunno. 

[Blotto 23484]

7 hours ago, atomheartbevo said:

Everything is under control guys, the intel CEO is going to….pray about it.

 

I actually figured Intel would be better off under Gelsinger than the last couple of jabronis that fucked that company up. At least he had an engineering background, and he seemed to do solid work at EMC and VMWare. But holy shit, laying off 15% of your staff and then quoting scripture as the path forward.... fucking joke. 

[52-80 19140]

8 hours ago, freyguy said:

Jesus fuck, what the hell?

brother, you're doing what gelsinger just did

[Dahobbs 17521]

11 hours ago, Blotto said:

I actually figured Intel would be better off under Gelsinger than the last couple of jabronis that fucked that company up. At least he had an engineering background, and he seemed to do solid work at EMC and VMWare. But holy shit, laying off 15% of your staff and then quoting scripture as the path forward.... fucking joke. 

I still think it might be long term. They do seem to be righting the manufacturing ship. Profits will come after that gets working again. 

[Blotto 23484]

Just now, Dahobbs said:

I still think it might be long term. They do seem to be righting the manufacturing ship. Profits will come after that gets working again. 

Back when I was in the industry selling against Intel, they had EVERYONE  beat in process by at least a generation. Their top end shit stood alone, and on the bottom end they simply told their customers, if you use someone else no more "co-marketing" dollars. It was a brutally tough to crack that nut. But it was the tail end of Andy Grove's run (I think he was chairmen, not CEO) and he ran a tight fucking ship.

So part of me enjoys seeing them get rat fucked, but unfortunately I bought some stock a while ago as kind of a hedge against Taiwan fuckery if it occurs, and now I'm down 30%. Guess I'll be a long term holder. Fuck it, maybe AMD will buy em for the  fabs (although the fabless model is working great.....for now). 

[Dahobbs 17521]

36 minutes ago, Blotto said:

Back when I was in the industry selling against Intel, they had EVERYONE  beat in process by at least a generation. Their top end shit stood alone, and on the bottom end they simply told their customers, if you use someone else no more "co-marketing" dollars. It was a brutally tough to crack that nut. But it was the tail end of Andy Grove's run (I think he was chairmen, not CEO) and he ran a tight fucking ship.

So part of me enjoys seeing them get rat fucked, but unfortunately I bought some stock a while ago as kind of a hedge against Taiwan fuckery if it occurs, and now I'm down 30%. Guess I'll be a long term holder. Fuck it, maybe AMD will buy em for the  fabs (although the fabless model is working great.....for now). 

AMD was really mismanaged for a long time and has managed a very quick turn around once some level competence was brought in. For our own domestic manufacturing benefit, I'm hoping Intel can do the same. Plus, someone needs to compete with TSMC. And I don't see anyway AMD jumps back into capex heavy fab ownership anytime soon. 

[immamac 42832]

23 minutes ago, Dahobbs said:

AMD was really mismanaged for a long time and has managed a very quick turn around once some level competence was brought in. For our own domestic manufacturing benefit, I'm hoping Intel can do the same. Plus, someone needs to compete with TSMC. And I don't see anyway AMD jumps back into capex heavy fab ownership anytime soon. 

AMD is still horribly mismanaged. 

[Wiler77 1566]

21 minutes ago, immamac said:

AMD is still horribly mismanaged. 

Curious why you say that

[immamac 42832]

37 minutes ago, Wiler77 said:

Curious why you say that

They've had process dominance for ~4 generations in both datacenter and desktop and have done jack shit to increase market share outside of single digit numbers. They have no concept of how to ecosystem build or what it actually means to be a product leader. The AM5 platform was a complete disaster on launch, to the point where you couldn't recommend building with it for a singificant period of time and the OEM's couldn't ship their shitty ass product because it straight up didn't even work with DIMMs that were fully qualified already.

They went on stage and without anything but a loose ass NDA announced UALink which is kind of fucked on arrival, they still don't even have a damn website up or a trademark. They are pushing UltraEthernet really hard, but not committing anything but their proprietary shit that no one wants to use anyway and then they are wondering why the market isn't swarming around them. They build for AMD first and everyone else second, and that's not how you ecosystem build.

It's also damn near impossible to work with them on anything, back to the ecosystem building part. They just don't get it, everything is about AMD and their shit and you selling their shit for them. That's not how you win, you win by having a strong cohesive ecosystem and set of partners that trust your product set. 

So yeah, they are horribly mismanaged, they can build good enough stuff like they have always done, but their problems haven't changed. Their numbers changed because getting away from Global Foundry was a choice that was rewarded by investors with no real rhyme or reason other than it shed a lot of capital intensity out of their business. 

 

[Wiler77 1566]

9 hours ago, immamac said:

They've had process dominance for ~4 generations in both datacenter and desktop and have done jack shit to increase market share outside of single digit numbers. They have no concept of how to ecosystem build or what it actually means to be a product leader. The AM5 platform was a complete disaster on launch, to the point where you couldn't recommend building with it for a singificant period of time and the OEM's couldn't ship their shitty ass product because it straight up didn't even work with DIMMs that were fully qualified already.

They went on stage and without anything but a loose ass NDA announced UALink which is kind of fucked on arrival, they still don't even have a damn website up or a trademark. They are pushing UltraEthernet really hard, but not committing anything but their proprietary shit that no one wants to use anyway and then they are wondering why the market isn't swarming around them. They build for AMD first and everyone else second, and that's not how you ecosystem build.

It's also damn near impossible to work with them on anything, back to the ecosystem building part. They just don't get it, everything is about AMD and their shit and you selling their shit for them. That's not how you win, you win by having a strong cohesive ecosystem and set of partners that trust your product set. 

So yeah, they are horribly mismanaged, they can build good enough stuff like they have always done, but their problems haven't changed. Their numbers changed because getting away from Global Foundry was a choice that was rewarded by investors with no real rhyme or reason other than it shed a lot of capital intensity out of their business. 

 

This is interesting.  In full disclosure I work at AMD, so I was genuinely interested in your point of view.  I am generally less informed on the consumer side of things, both as an employee and the fact I'm not a gamer or a DIY home PC builder.  So I'll take your POV on AM5 and all things Client at face value.

On the server side, I've seen some shit. I absolutely don't think we've made 100% perfect decisions, but I also absolutely know people underestimate the influence Intel has on customers and the ecosystem, even years after our obvious dominance on the technology side. Our CEO made a major announcement at a event in the last 3 years regarding support from a major enterprise ecosystem partner, one that was not done on a whim and was supported by quotes by the company expressing their support, and the next week that support disappeared. We have enough insiders at this ecosystem partners to know we had met every demand for support and payment they asked for, and after the fact Intel came in with enough zeros on a check to cancel everything. 

I've heard enough stories from former Intel employees now at AMD to know they are fucking dirty, threatening to withold marketing payments companies rely on to stay afloat. Or friends at Dell that were told their intel spiffs werent valid unless they maintained 90% share for Intel. So when you say its about "you selling their shit for them", what I've witnessed is customers saying flat out, "I'm not selling your product unless you pay me to". It has nothing to do with proposing the best solution for customers.  Even with that headwind, we're at 30% market share in server.

Regarding the comment of Global Foundries, that was one of the few good things from the Hector regime. It was a "stop digging and drop the shovel" moment. And it's proven to be the right move, so not sure how that ends up in a list on how AMD has been mismanaged.

In an all-timer, Hector drove his Ferrari to the office on the day he announced laying off 10% of the company. 

[immamac 42832]

9 minutes ago, Wiler77 said:

This is interesting.  In full disclosure I work at AMD, so I was genuinely interested in your point of view.  I am generally less informed on the consumer side of things, both as an employee and the fact I'm not a gamer or a DIY home PC builder.  So I'll take your POV on AM5 and all things Client at face value.

On the server side, I've seen some shit. I absolutely don't think we've made 100% perfect decisions, but I also absolutely know people underestimate the influence Intel has on customers and the ecosystem, even years after our obvious dominance on the technology side. Our CEO made a major announcement at a event in the last 3 years regarding support from a major enterprise ecosystem partner, one that was not done on a whim and was supported by quotes by the company expressing their support, and the next week that support disappeared. We have enough insiders at this ecosystem partners to know we had met every demand for support and payment they asked for, and after the fact Intel came in with enough zeros on a check to cancel everything. 

I've heard enough stories from former Intel employees now at AMD to know they are fucking dirty, threatening to withold marketing payments companies rely on to stay afloat. Or friends at Dell that were told their intel spiffs werent valid unless they maintained 90% share for Intel. So when you say its about "you selling their shit for them", what I've witnessed is customers saying flat out, "I'm not selling your product unless you pay me to". It has nothing to do with proposing the best solution for customers.  Even with that headwind, we're at 30% market share in server.

Regarding the comment of Global Foundries, that was one of the few good things from the Hector regime. It was a "stop digging and drop the shovel" moment. And it's proven to be the right move, so not sure how that ends up in a list on how AMD has been mismanaged.

In an all-timer, Hector drove his Ferrari to the office on the day he announced laying off 10% of the company. 

This embodies why AMD fails at ecosystems. You fund and invest in ecosystems with people who want to build stuff that makes it work and easy to use. You do this for years at a time until poof there's a critical mass and the whole reason it happened was because of AMD. 

Instead AMD pushes stupid shit that's home grown in standards bodies like it's a "collaboration" but it's not, it's just some bullshit to sell more of their own stuff instead of actually fix something. SDXI is a great example of that. So is XGMI. So is Rocm so is etc etc 

The issue isn't that they can't use the ecosystem to their advantage it's that they think Intel is buying the ecosystems goodwill (spoiler: they are not) instead they invest time, effort and code into projects they don't own to make them successful and then eventually they benefit by making the pie bigger overall. 

AMD is always trying to make their slice bigger, but not necessarily make the whole pie bigger. This is their biggest functional misstep from working adjacent and with them for the last 5+ years. It's systemic to the whole apparatus, it's not just like pockets. 

[Captainant 27617]

8 hours ago, immamac said:

This embodies why AMD fails at ecosystems. You fund and invest in ecosystems with people who want to build stuff that makes it work and easy to use. You do this for years at a time until poof there's a critical mass and the whole reason it happened was because of AMD. 

Instead AMD pushes stupid shit that's home grown in standards bodies like it's a "collaboration" but it's not, it's just some bullshit to sell more of their own stuff instead of actually fix something. SDXI is a great example of that. So is XGMI. So is Rocm so is etc etc 

The issue isn't that they can't use the ecosystem to their advantage it's that they think Intel is buying the ecosystems goodwill (spoiler: they are not) instead they invest time, effort and code into projects they don't own to make them successful and then eventually they benefit by making the pie bigger overall. 

AMD is always trying to make their slice bigger, but not necessarily make the whole pie bigger. This is their biggest functional misstep from working adjacent and with them for the last 5+ years. It's systemic to the whole apparatus, it's not just like pockets. 

Maybe you should have made a "is AMD FUCKED" thread instead lol

[immamac 42832]

5 minutes ago, Captainant said:

Maybe you should have made a "is AMD FUCKED" thread instead lol

AMD isn't fucked though. They are who they have always been. 

When you try to compare Intel and AMD as equals or even peers is the issue. They are very different businesses (fully vertically integrated semiconductor vs fabless semiconductor) 

Intel is now putting out puff pieces about how they have onlined some chips on 18A. Which is no doubt an accomplishment, but not one that used to be sunshine pumped. 

[Nivek 16916]

Not a bad rebound today for Intel, up 7.9%    Doesn't make up for the other losses though.

[boilerhorn 1371]

It's hovered around the Aug 8 pricing for the past 2 weeks, giving back another 6% yesterday.

Lip Bu Tan, former executive chair and CEO of Cadence (EDA), resigned from the board yesterday, effective immediately.  Tan joined in 2022 and was part of the board's M&A committee.
https://www.reuters.com/technology/intel-director-lip-bu-tan-is-leaving-chipmakers-board-bloomberg-news-reports-2024-08-22/

[immamac 42832]

The tailspin has to stop with something. If Gaudi 3 lands and the new chiplets work wonders for their datacenter and desktop parts and AI PCs actually start selling it may be just enough to keep them going. That's the unfortunate truth. Their lack of fab customers is wild, especially with the shortage of advanced process node wafer capacity. If they can get yields up and people in it could turn around in a hurry, but it doesn't seem like that's in the cards. 

[boilerhorn 1371]

Agree @immamac.  There is too much inherent value in the company.  It's selling below book - which is wild.  Random, mediocre fabs are priced above 2x book.

Moving material through the fab with external customers will be key for that portion of the business and the recovery.  Production of tip of the spear process technology in high volumes/high yields is non-trivial, which is why there are so few players in that space.

The product side has seen reductions in market share due to competition from AMD and the Arm players - and MSFT has been aggressive in helping elevate the arm ecosystem.  This year, we have QCOM, but next year, expect MediaTek and NVDA to deliver Arm products into the Windows Ecosystem.  The performance is subpar, but the Perf/Watt is better.  None of this was a surprise.

All players have been impacted by NVDA's presence in Data Center / HPC.  They've been VERY aggressive in finding ways to map just about every high-thread-count or compute application into GPU.  They did it with Crypto in ~2015 and that has since migrated to ASIC.  Now it's AI inference and training.  We are seeing some inference-focused development in the market (Groq, others) that *could* reduce NVDA's presence a bit.  Moreover, as model optimization research improves, the overall compute needed to train and infer yesterday's models could go down by orders of magnitude.  Does that result in higher fidelity models or the next "killer app?"  Who knows?  NVDA's main advantage in AI/HPC today: software ecosystem.  AMD and Intel are 1/2-1 generation behind there.

The problem being addressed immediately - and the CEO mentioned it in his quarterly report - is that revenues have dropped ~20% since 2021 and staffing has increased by 15%.  That's not a good combination and is being addressed by EOY, as per his public statements.

[MinerProphet 71]

Read a lot of different source material on Intel this week as it's interesting if nothing else. To paraphrase, from what I can gather, any evaluation of Intel will have to be a long term one. The horrible results from this week should be viewed in the context of decisions made 5~10 years ago, and an appreciation that decisions made more recently won’t pay off for several more years to come. The real measure of Gelsinger’s leadership in particular will come with 16A: that is the node that Intel is betting the company on. That’s when internal chip production should come home, when Intel will finally have a leading or near-leading process operating at volume, and when the question of whether or not the company can actually earn 3rd-party business will be finally be answered.

So, idk, let's see.

[immamac 42832]

4 minutes ago, MinerProphet said:

Read a lot of different source material on Intel this week as it's interesting if nothing else. To paraphrase, from what I can gather, any evaluation of Intel will have to be a long term one. The horrible results from this week should be viewed in the context of decisions made 5~10 years ago, and an appreciation that decisions made more recently won’t pay off for several more years to come. The real measure of Gelsinger’s leadership in particular will come with 16A: that is the node that Intel is betting the company on. That’s when internal chip production should come home, when Intel will finally have a leading or near-leading process operating at volume, and when the question of whether or not the company can actually earn 3rd-party business will be finally be answered.

So, idk, let's see.

18A and yeah that's the bet, Pat has made that clear. He bet the company on 5 nodes in 4 years. It's crazy. The problem for intel is that they don't have any customers or volume. There is also the bet that there is going to be a never ending need for advanced node chips.

[52-80 19140]

On 8/23/2024 at 2:55 PM, boilerhorn said:

Agree @immamac.  There is too much inherent value in the company.  It's selling below book - which is wild.  Random, mediocre fabs are priced above 2x book.

The ~30B gap between book value and market cap happens to be the exact same as the "goodwill" and "intangibles" on their balance sheet.  Maybe investors recognize the worthlessness of those accounting. Other fabs then earn a premium on growth.

[4thgenhorn 38]

So do I buy shares at $20 or what?

[Incredulity 11643]

20 hours ago, 4thgenhorn said:

So do I buy shares at $20 or what?

[boilerhorn 1371]

On 8/23/2024 at 7:21 AM, boilerhorn said:

It's hovered around the Aug 8 pricing for the past 2 weeks, giving back another 6% yesterday.

Lip Bu Tan, former executive chair and CEO of Cadence (EDA), resigned from the board yesterday, effective immediately.  Tan joined in 2022 and was part of the board's M&A committee.
https://www.reuters.com/technology/intel-director-lip-bu-tan-is-leaving-chipmakers-board-bloomberg-news-reports-2024-08-22/

More details on Tan leaving the board are now leaking...
https://www.reuters.com/technology/intel-board-member-quit-after-differences-over-chipmakers-revival-plan-2024-08-27/

The former CEO of chip-software company Cadence Design joined Intel’s board two years ago as part of a plan to restore Intel’s place as the leading global chipmaker. The board expanded Tan’s responsibilities in October 2023, authorizing him to oversee manufacturing operations.

 

Over time, Tan grew frustrated by the company’s large workforce, its approach to contract manufacturing and Intel’s risk-averse and bureaucratic culture, according to the sources, who were not authorized to speak publicly.

 

[52-80 19140]

On 8/26/2024 at 9:58 PM, 4thgenhorn said:

So do I buy shares at $20 or what?

I bottom-fished intc stock once and reeled in a shoe. 

[PilotsError 2286]

 

[Dahobbs 17521]

Teaser on new CPUs:

Intel Core Ultra 200V promises Arm-beating battery life without compatibility issues

https://arstechnica.com/gadgets/2024/09/intel-announces-first-batch-of-second-gen-lunar-lake-core-ultra-laptop-cpus/

[hornbri 1390]

So now they are considering splitting up the business?

https://www.morningstar.com/news/marketwatch/20240830291/intels-stock-soars-on-hope-of-business-split-analyst-says-dont-fall-for-it

[Dahobbs 17521]

14 minutes ago, hornbri said:

So now they are considering splitting up the business?

https://www.morningstar.com/news/marketwatch/20240830291/intels-stock-soars-on-hope-of-business-split-analyst-says-dont-fall-for-it

I think it was among a number of items discussed. I also think that part of the conversation was about 10 minutes. Intel isn't selling off its fabs anytime soon. It bet big on them and foundry and I think has to try and see that through, for better or worse. However, I didn't realize its new laptop chips will not be built on the new Intel processes but rather at TSMC. Allegedly they are focusing their internal production on server chips, but that may suggest there might be yield issues with the newer processes if they are that constrained. Supposedly the next round of laptop chips next year will be back to Intel manufacturing. 

[Captainant 27617]

14 hours ago, Dahobbs said:

I think it was among a number of items discussed. I also think that part of the conversation was about 10 minutes. Intel isn't selling off its fabs anytime soon. It bet big on them and foundry and I think has to try and see that through, for better or worse. However, I didn't realize its new laptop chips will not be built on the new Intel processes but rather at TSMC. Allegedly they are focusing their internal production on server chips, but that may suggest there might be yield issues with the newer processes if they are that constrained. Supposedly the next round of laptop chips next year will be back to Intel manufacturing. 

I've been hearing scuttlebutt that the new Intel mobile chips will have "ARM-like efficiency" so maybe their big bet on their own fabs will pay off yet