Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/intel-reports-fourth-quarter-and-full-year-2024-financial-results.21983/page-3
        )

    [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
)

Intel Reports Fourth-Quarter and Full-Year 2024 Financial Results

nothing outsold raptor lake it must have 200 Million units by now considering we have seen 3 refresh of Raptor lake(13700H,Core 100H Core 200H ) over 3 years
Btw, in my latest video, I asked Intel to start promoting Arrow Lake S for the ML/AI community as I received several questions, regarding what CPUs to use. In addition, in the video, I also suggested Intel that they could contact me to work on some contents regarding that.
 
Last edited:
Please explain how domino circuits are power efficient - I always thought that they burned more power with the need for precharge etc
It’s only for Bitcoin. Power consumption is caused by the toggling of signals. The lowest power consumption for synchronizing signals is achieved by using domino logic, which we implemented in the 32-bit adder.
 
At the very end of the earning call, they are asked how much of Panther lake is manufactured on Intel foundry and they say 70%, but for Nova Lake this percentage will decrease as more more of the tiles are made at TSMC. I wonder if Nova lake will use TSMC N2, or A16
So, on the 18A Panther Lake, I think in the past, I think, Dave, the comment was that you expect to bring roughly 70% of the die in-house. Is that still the plan? And then is it pretty set in stone that you're bringing it back for sure? Or do you have any flexibility whether to bring back more of the die or less of the die if you need to.


Michelle Holthaus -- Interim Co-Chief Executive Officer

Yes. So, we did move Panther Lake inside of 18A design win. But as I stated before, we look at each generation of products based on what's the right product, what's the right process, what's the right market window and what allows our customers to win. So, for Panther Lake, that was 18A.

And as I said, we're very happy with where we are from a performance and yield perspective at this point in the process. So, that will stay on 18A. Then as you look forward, to our next-generation product for client after that, Nova Lake will actually have die both inside and outside for that process. So, you'll actually see compute tiles inside and outside.

Again, it's about optimizing to what allows us to win in the market, what allows us to win with our customers and optimizing the overall product portfolio because at the end of the day, if our customers are successful, we win, that drives more wafers and Intel foundry and that allows us to win, but I'll continue to have a balance. And as I said, we'll be doing the same look across our data center portfolio as well.
 
At the very end of the earning call, they are asked how much of Panther lake is manufactured on Intel foundry and they say 70%, but for Nova Lake this percentage will decrease as more more of the tiles are made at TSMC. I wonder if Nova lake will use TSMC N2, or A16

I don’t think so. Intel has a hefty TSMC N3 agreement that needs to be to be met. Also, I don’t think the supporting chiplets need N2. The word in the trenches is that there is no N2 agreement between Intel and TSMC.
 
I don’t think so. Intel has a hefty TSMC N3 agreement that needs to be to be met. Also, I don’t think the supporting chiplets need N2. The word in the trenches is that there is no N2 agreement between Intel and TSMC.
Well the wafer agreement for N2 was leaked I don't how accurate that is
TSMC-N2-will-be-used-across-many-chips-from-Apple-AMD-Nvidia-and-Intel.jpg
 
I don’t think so. Intel has a hefty TSMC N3 agreement that needs to be to be met. Also, I don’t think the supporting chiplets need N2. The word in the trenches is that there is no N2 agreement between Intel and TSMC.
They are CPU chiplets. Holthaus says that in the earnings call.
 
They are CPU chiplets. Holthaus says that in the earnings call.

Intel said they are using TSMC N2 in the call?

I was told by Intel people that they are being pushed to 18A with BSPD. TSMC N2 does not have BSPD

Where in the call?


Michelle Johnston Holthaus

Thank you, John, and let me add my welcome. It's been roughly two months since Dave and I stepped into our roles as interim co-CEOs. From day one, we have been working closely together alongside the board to drive better execution of our strategy. There are no quick fixes and we are committed to improving our performance and rebuilding our credibility through persistent hard work that delivers tangible results.

As part of this, we are driving more focused investments across the business. We cannot be all things to all people, and we are prioritizing areas where we can drive differentiated value. We are also continuing to simplify our business and become a leaner, more efficient company. And most of all, we are doing a better job of listening to our customers to ensure we meet their needs.

Q4 was a step in the right direction. We delivered revenue, gross margin, and EPS above our guide. Intel Products executed to drive revenue in the quarter even as PC inventory continued to normalize, and Intel Foundry drove incremental operating efficiencies while achieving key grant related milestones which supported solid upside to gross margins.

As co-CEOs, you can expect us to be very straightforward and direct. We only make commitments we are confident we can deliver. We firmly believe that what we say is not nearly as important as what we do and everything we do must be in service of our customers. Innovating to solve their most pressing challenges is the surest path to creating shareholder value.

This is the mindset I have brought to my position as the CEO of Intel Products. This is a great business with great people, partners, and IP to design world class products from edge to cloud. I take nothing for granted but I firmly believe that the core x86 architecture and the ecosystems we have built and invested in over the decades, create a solid foundation for success.

Our customers share this view, but they need us to improve our execution and hit our commitments. I am setting clear priorities and directions in each business to drive better outcomes. I think about Intel products in three buckets.

First, client, edge; second, traditional data center; and third, the AI data center. Let me spend a few moments on each. In client, Intel CPUs power roughly 7 out of every 10 PCs. This is a strong position that gives us advantages in the market that said the market is becoming more competitive, especially as we see new entrants trying to participate in the AI PC category.

Personally, I thrive on competition. It drives a healthy paranoia across everything we do and we're using it as motivation to up our game even more. The success of Core Ultra across Meteor Lake, Arrow Lake, and Lunar Lake has established Intel as the market leader in AI PC CPUs. And we remain on track to ship more than 100 million cumulative systems by the end of 2025. We are innovating at scale unlike any of our competitors.

This was on display earlier this month at CES where we launched the enterprise versions of our AI CPUs with compelling new features to Intel vPRO. This is a testament to the strong ecosystem we have built with IT departments around manageability, security, trust, and brand, and we expect these investments to possession us well as corporations begin their migration to Windows 11.

Alongside our investments in enterprise, our ecosystem reach also positions us well in AI PC consumer markets. We are working with more than 200 ISBs across more than 400 features to optimize their software on our silicon. I'm excited about the new applications I'm seeing in the pipeline that will begin to proliferate over the coming months. Our goal is to innovate partner and fortify our position as the preferred CPU of choice.

Looking ahead to the rest of the year, we will strengthen our client roadmap with the launch of Panther Lake, our lead product on Intel 18A in the second half of 2025. As the first volume customer of Intel 18A, I see the progress that Intel Foundry is making on performance and yield. And I look forward to being in production in the second half as we demonstrate the benefits of our world class design and process technology capabilities.

2026 is even more exciting from a client perspective as Panther Lake achieves meaningful volumes and we introduced our next generation client family code named Nova Lake. Both will provide strong performance across the entire PC stack with significantly better costs and margins for us enhancing our competitive position and reinforcing our value proposition to our partners and customers.

Let me now turn to our traditional data center business. The team has made good progress towards strengthening our offerings and driving better, more predictable execution. This year is all about improving Xeon's competitive position as we fight harder to close the gap to competition.

The ramp of Granite Rapids has been a good first step. We are also making good progress on Clearwater Forest, our first Intel 18A server product, that we plan to launch in the first half of next year. All of this provides a strong foundation on which to build as we execute.

The world's data center workloads still primarily run on Intel silicon. And we have a strong ecosystem especially within enterprise. We are going to leverage these strengths as we work to stabilize our market share in 2025. One of the ways we'll do this is by reengaging the x86 ecosystem.

We have seen a positive response from the x86 Ecosystem Advisory Group we formed last fall, and we are encouraged by the enthusiasm for building both semi-custom and custom products. This is a big area of opportunity for the business, and we look forward to talking more about this as we have news to share.

Turning to the AI data center, I will start by saying that this is an attractive market for us over time, but I am not happy with where we are today.
On the one hand, we have a leading position as the host CPU for AI servers. And we continue to see a significant opportunity for CPU based inference on prem and at the edge as AI infused applications proliferate.

On the other hand, we're not yet participating in the cloud-based AI data center market in a meaningful way. We have learned a lot as we have ramped Gaudi, and we are applying those learnings going forward.

One of the immediate actions I have taken is to simplify our roadmap and concentrate our resources. Many of you heard me temper expectations on Falcon Shores last month. Based on industry feedback, we plan to leverage Falcon Shores as an internal test chip only without bringing it to market. This will support our efforts to develop a system level solution at rack scale with Jaguar Shores to address the AI data center.

More broadly, as I think about our AI opportunity, my focus is on the problems our customers are trying to solve. Most notably bid to lower the cost and increase the efficiency of compute. AI is not a market in the traditional sense. It's an enabling application that needs to span across the compute continuum from data center to the edge as such a one size fits all approach will not work.

And I can see clear opportunities to leverage our core assets in new ways to drive the most compelling total cost of ownership across the continuum. Before I turn the call over to Dave, let me close by speaking of Intel Foundry's largest way for customer.

I have a pretty simple approach. When we are able to combine world class products with world class process technology, we win. As CEO of Intel Products, I will always make process technology decisions based on what is best for my customers. And Intel Foundry will need to earn my business every day just as I need to earn the business of my customers.

Having said that I'm confident in Intel Foundry team ability to support my current and future product roadmap. And I'm excited to do more business with them as their process technology continues to advance. A stronger Intel Products combined with a more competitive Intel Foundry is a recipe for success for Intel overall.

Dave over to you.
 
No, she said: "So, you'll actually see compute tiles inside and outside." That's at the very end of the call

Srinivas Pajjuri
Yes. A quick one. So on the 18A Panther Lake, I think in the past, I think, Dave, the comment was that you expect to bring roughly 70% of the die in-house. Is that still the plan? And then is it pretty set in stone that you're bringing it back for sure? Or do you have any flexibility whether to bring back more of the die or less of the die if you need to. So just trying to understand.

David Zinsner
I'm going to let Michelle answer that because it really is her decision on how she builds her products.
Yes. So we did move Panther Lake inside of 18A design win. But as I stated before, we look at each generation of products based on what's the right product, what's the right process, what's the right market window and what allows our customers to win. So for Panther Lake, that was 18A. And as I said, we're very happy with where we are from a performance and yield perspective at this point in the process. So that will stay on 18A. Then as you look forward, to our next-generation product for client after that, Nova Lake will actually have die both inside and outside for that process. So you'll actually see compute tiles inside and outside. Again, it's about optimizing to what allows us to win in the market, what allows us to win with our customers and optimizing the overall product portfolio because at the end of the day, if our customers are successful, we win, that drives more wafers and Intel Foundry and that allows us to win, but I'll continue to have a balance. And as I said, we'll be doing the same look across our data center portfolio as well.


From what I have heard Pat G was not successful with renegotiating the massive N3 contract so Intel is on the hook for some serious N3 wafers. I certainly hope Intel will use the latest and greatest TSMC nodes but relations between the companies are strained to say the least. BSPD is also an issue. Maybe a new Intel CEO will come in and change the relationship. Hock Tan could certainly do it.

Michelle should go into politics......... :ROFLMAO: Answering a question while not answering a question is a fine skill.
 
Supposed TSMC N2 client list (image source: Jukanlosreve)

I can confirm 4 of those tape-outs but Intel is not one of them, I will check into it.
I think Intel is doing small volumes on N2 she herself said Nova Lake is Dual sourced
People are misinterpreting that they will only do small volume on 18A and big volume on TSMC if that's the case their margin will never recover
 
but she did answer the question

"I think, Dave, the comment was that you expect to bring roughly 70% of the die in-house. Is that still the plan?"

As President she should know what % of wafers will be Intel and TSMC. You have to pick a foundry when you start a design and that is 2-3 years before manufacturing starts so she should have a clear vision of production 3-5 years out. So either she does not know or she does not want to say because Wall Street might punish her. Either way it is a non answer, my opinion.

When you sign a wafer agreement that is a legally binding contract. Intel did a huge prepay. Will Intel use 18A and eat the TSMC penalty? Or will Intel honor their N3 contract and not use 18A with BSPD?

Intel BSPD offers a significant advantage over AMD. For Intel to skip BSPD in favor of TSMC N2 it would be a huge vote of no confidence in Intel innovation. Again, just my opinion.
 
I think Intel is doing small volumes on N2 she herself said Nova Lake is Dual sourced
People are misinterpreting that they will only do small volume on 18A and big volume on TSMC if that's the case their margin will never recover
It doesn't make any sense why Panther lake is using 18A and then they switch back to TSMC for Nova Lake. Maybe yields are too low for the big dies required for Nova Lake, or maybe N2 performs better.
Intel's fabs will be mostly idle in the next few years - once Arrow lake takes over from Raptor lake. Even Panther lake should really be 100% Intel foundry. 70% is not good enough
 
Absolutely. Nor any of the major cloud leaders is holding their breath while Intel is struggling to deliver. I'm afraid that train has already left the station for Intel.
I don't think that way. There are always trains in a station. Currently, the train is Deepseek. For any company, they need to react to information.
 
It doesn't make any sense why Panther lake is using 18A and then they switch back to TSMC for Nova Lake. Maybe yields are too low for the big dies required for Nova Lake, or maybe N2 performs better.
Intel's fabs will be mostly idle in the next few years - once Arrow lake takes over from Raptor lake. Even Panther lake should really be 100% Intel foundry. 70% is not good enough
You don't know ? Arrow Lake volume is not much they are still serving market with Raptor Lake and they still sell Intel 7 silicon to keep the fab running there is a Panther lake SKUs with 90% silicon on Intel 18A and Intel 3 combined they have outsourced PCH or the die that contains the IO/PCH to N6 which makes sense cause N6 is cheaper than I7
Panther Lake has 3 SKUU
4+8+4/ 4+4 CPU+ IMC + Display Logic + NPU on 18 A for all 3 and PCH on N6 for all 3 the GPU for 1 config is N3E and rest 2 is Intel 3 all in Panther lake is mostly internal from the rumors.
I am hearing there are SKU with complete Intel Nodes+Packing but also TSMC Node+Intel Node + Intel packing for NVL
 
"I think, Dave, the comment was that you expect to bring roughly 70% of the die in-house. Is that still the plan?"

As President she should know what % of wafers will be Intel and TSMC. You have to pick a foundry when you start a design and that is 2-3 years before manufacturing starts so she should have a clear vision of production 3-5 years out. So either she does not know or she does not want to say because Wall Street might punish her. Either way it is a non answer, my opinion.

When you sign a wafer agreement that is a legally binding contract. Intel did a huge prepay. Will Intel use 18A and eat the TSMC penalty? Or will Intel honor their N3 contract and not use 18A with BSPD?

Intel BSPD offers a significant advantage over AMD. For Intel so skip BSPD in favor of TSMC N2 it would be a huge vote of no confidence in Intel innovation. Again, just my opinion.
Might there be risks for higher power designs (like desktop) where the rumored hot spots that are a problem in BSPDN could be problematic in a desktop design, but not a laptop design?

Still, seems like a metric butt ton of work to design a compute tile for 2 different processes (and process types!) Hard to fathom Intel would go this path.
 
Might there be risks for higher power designs (like desktop) where the rumored hot spots that are a problem in BSPDN could be problematic in a desktop design, but not a laptop design?

Still, seems like a metric butt ton of work to design a compute tile for 2 different processes (and process types!) Hard to fathom Intel would go this path.
It is a lot of work but at least it's not as much as previous designs which were custom Intel EDA Tools now all the design are industry standard EDA Tools.

I bet they are doing something else to make this process more efficient that no one else can at this level
 
Last edited:
Back
Top