Semiwiki on LinkedIn Semiwiki on Facebook Semiwiki on Twitter Register / Log in 
Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/intel-employees-very-optimistic.22950/page-8
        )

    [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 Employees "Very Optimistic"

dkr1986 said:
Intel's avg 18A wafer D0 is probably higher (may be 0.2+/sq-cm if you consider Pat's 0.4 D0 comment back in Q3'24), it is their Best Wafers that are at HVM D0 now (I assumed that they are talking about 0.1/sq-cm here as HVM level D0). TSMC N2 D0 is unknown at this time. Only information we can infer is from that SRAM yield they have shared which may be not accurate (the calculations).



I also just noticed that M5 (Macbook chips) is also marked as "wafer in" at End'25. So that could be first product out on N2 before AMD too (The last two gen of Macs launched in March). We will have to wait to see.
Click to expand...
Tsmc N2 d0 is very good, comparable to their best nose in recent history. https://www.tomshardware.com/tech-i...ower-than-n3-at-the-same-stage-of-development

dkr1986 said:
Vivek Arya asked this question to Michelle on the recent BofA conference and this is how the interaction went,

Vivek Arya - You mentioned Panther Lake is up and running. You're happy with that success. So if that progress is going per your expectations, why go back to TSMC for Nova Lake?

Michelle - Yes. So maybe just to baseline everybody on Panther Lake. So Panther Lake is a product that's going to launch in the second half of this year. And it is all built on Intel 18A. And really -- so Panther Lake is an all-mobile stack. When you get to the next-generation Nova Lake, it is both a mobile stack and a desktop stack. And so one of the things about the desktop market, which is a place that we have lost market Sigma share, it is a very elastic market. The best product at the time of graphics card launch is really how you kind of take advantage of that TAM. And so being able to land on a node that is already ramped is at very high performance plus yield is very important. So you can imagine, I'm looking at how much yield and product can I get in a very short amount of time. And so when you look at that, you might actually pick maybe not the latest dot of a node at TSMC, but you know you can get a lot of wafers and a lot of product in a really short amount of time. And so you'd put that SKU on TSMC. And so when I say I'm pragmatic, I literally look at it by SKU and where it makes the most sense. And so I like personally a portfolio where I use both boundaries because at times, I want to be cost. At times, I want to be about volume and at times, I want to be about performance. And depending on which is most important for the customer and the segment, that's what I pick. And so it isn't anything about Intel doing something wrong. It's more about optimizing the mix to be able to deliver best on behalf of my customers and hit a particular market window with a particular amount of volume.

One can take away that Michelle is saying N2 is higher performance or higher yielding node at that point of time compared to 18A-P (rumored to be the node for NVL-S, it could be just vanilla 18A too, not sure yet) or Intel is balancing between Internal and external wafer volume with constrained fab capacity. It is up to interpretation. 🤷‍♂️
Click to expand...
I read Michelle's comment as that N2 not necessarilly having best performace, but can ramp up high volume in a short peroid of time. Basically when the high end GPU starts to sell, high end gaming CPUs will sell well, too. So time to ramp high volume is absolutely important, and N2 is the best node to deliver that.
 
kevin01 said:
I read Michelle's comment as that N2 not necessarily having best performance, but can ramp up high volume in a short period of time. Basically when the high end GPU starts to sell, high end gaming CPUs will sell well, too. So time to ramp high volume is absolutely important, and N2 is the best node to deliver that.
Click to expand...

I read her comment and thought spin spin spin spin spin. This is why they don't let me ask questions on investor calls. No way would I let that one slip by.
 
siliconbruh999 said:
Exactly reminds me when Intel presented 18A SRAM some random analyst asked what is 18A in 18A 🤣🤣 banking media should be banned from asking question.
Click to expand...
Be careful what you wish for, Darwin is watching you.......as he has been at Intel's Board since 2009, when the US-financial crisis hit the global world (and perhaps Intel also?), he must have learned the node nomenclature by now in 2025......just kidding...

https://semiwiki.com/forum/threads/...tire-in-latest-shakeup-amid-turnaround.22422/
 
Daniel Nenni said:
I read her comment and thought spin spin spin spin spin. This is why they don't let me ask questions on investor calls. No way would I let that one slip by.
Click to expand...
TSMC is more flexible on capacity and ramping. That is why Intel Products choose TSMC. If Intel process was better than TSMC, they would choose Intel .... but they did not... so... ??
 
MKWVentures said:
TSMC is more flexible on capacity and ramping. That is why Intel Products choose TSMC. If Intel process was better than TSMC, they would choose Intel .... but they did not... so... ??
Click to expand...

Another word for process ramping is improving yield. TSMC is a very high yield foundry so it will be tough to beat them at 3nm and now 2nm.
 
MKWVentures said:
TSMC is more flexible on capacity and ramping. That is why Intel Products choose TSMC. If Intel process was better than TSMC, they would choose Intel .... but they did not... so... ??
Click to expand...
But it is confirmed that >80% of Nova Lake will be on Intel Foundry nodes. Rumor is this one halo NVL-S SKU\tile (desktop SKU) will be on TSMC N2 and this SKU is likely low volume compared to other SKUs of Nova Lake. It is also rumored that this one Halo SKU is bLLC SKU (big large level cache?) to compete with AMD's x3D chips. This is one area where Intel has lost a lot of their market share.

So, like @kevin01 said, it could be like Michelle said they want right volume of this SKU ready for Nvidia's next gen RTX 6xxx GPU launch to capitalize on that. Also, if it is really 18A-P, they are targeting for this product, that node maybe not mature enough to yield during this launch window to do large cache chips. We can speculate but will not know for sure why Intel made this decision (until these products hit the shelves and reviewers get their hands on it).
 
Last edited:
MKWVentures said:
@siliconbruh999 @nghanayem @dkr1986

I'm doing IFS P&L modeling for 2026...

Will Intel ship more than 10M Panther Lake Chips to customers in Q1 2026? 5M? 15M?

They will have been running risk production for 9 Months when Q1 2026 begins... thats a lot of chips
Click to expand...
Intel shipped ~40 Million Unit of AI PC in entirety of 2024 this include some 1.5M LNL some ARL less than LNL cause it didn't launch let's say maybe (1/3 of LNL for 500K ) so ~2 Million unit out of 40 Million we get roughly ~38 Million.

Now since the supply would have increased gradually per quarter let's say 7/9/10/11 per quarter also in Q1 24 Intel were running hot lots as well so the initial supply must be low basing on my calculations. I would say they would likely fell short for 10Million units and would be roughly 8-9 Million Units.

This is entirely my assumption though based on the public data available.
 
MKWVentures said:
Will Intel ship more than 10M Panther Lake Chips to customers in Q1 2026? 5M? 15M?
Click to expand...
Intel management indicates that Panther Lake will follow similar launch and volume profile as Meteor Lake. Intel launched Meteor Lake notebooks in December 2023. Per Intel's earnings calls and various events that Intel participated (that I followed online). This is how Meteor Lake shipment were reported.

Q1'24 earnings press release - 5Mu since December launch. Mentioned 40Mu of AI PCs targeted to be shipped for CY2024 (note - Meteor Lake is the only AI PC with an NPU among Intel's products)
Q2'24 earnings press release - >15Mu since December launch. So >10 Mu shipped since Q2 alone.
Q3'24 earnings press release - no mention of no of units shipped. But AI PC target changed to 100Mu of AI PCs to be shipped for CY2024 & CY2025 combined.
October 2024 - On Lenovo event, Pat's keynote presentation - >20Mu AI PC shipped. (by this time Lunar Lake was launched @September 2024 as well but I doubt they sold any of this at that point of time).
Q4'24 earnings press release - no mention of no of units shipped. But AI PC target of 100Mu of AI to be shipped for CY2024 & CY2025 combined is reiterated.
On January 2025 - CES 2025- Michelle's Keynote - 1.5Mu of Lunar Lake was shipped in 2024 (this means 1.5Mu in Q4'24 alone as it launched in September 2024 which is not bad since this is targeted at thin/light premium niche segment - CFO also said Intel now expects to ship 3x original volume planned on LNL in 2025 due to positive reception).

Considering the Q4 is usually strongest quarter for CCG, I am guessing Intel shipped around roughly ~30Mu of Meteor Lake CPUs in 2024.

Now 2025 & 2026 is supposed to be stronger year for PC refresh (barring any macro issues like recession or tariffs) due to Windows 10 End of Life & PC refresh cycle after Post Covid Boom & Bust (5 years after initial buy in 2020/2021 for WFH - warranty expired in most cases).

Per Intel's recent commentary, Intel said Panther Lake / 18A ramp (yield) is going well ahead compared to Meteor lake at this point of time. So depending on when they launch the Panther Lake SKU in Q4'25, I would expect 6-8 Mu of Panther Lake CPU shipments in Q1'26 because of the tailwinds I mentioned above.

So considering similar ramp profile for Panther Lake with a little bonus for tailwinds, I expect them to ship 40-60Mu panther lake CPUs in 2026. That represents about 26% to 40% of all notebook CPUs sold by Intel in 2024 (Intel notebook CPU volume roughly 150Mu).

Assuming 400 good dies of 114sq-mm compute tile (very approximate, accounting for process yield & parametric yield) of Panther Lake CPU per 300mm wafer, total wafer volume required is 100k to 150k wafers per year in 2026 to support this shipment volume. i.e 8.33kwspm to 12.5kwspm.
Assuming a $22.5k/wafer pricing and another 1.5x to account for base tile & advanced packaging costs, I get about $3.375B to $5.06B in revenue for IFS from Panther Lake alone in 2026. This accounts for 19% to 29% of total IFS revenue of $17.5B for 2024.

This is a very rough estimate by a non semi industry guy, so I would take it with a boat load of salt and apply a big margin of safety for +/- error of my ignorance :LOL:. I did some sanity check at the end to make sure the Cost of Goods Sold for whole Intel is not too egregious for 50% margin.

1749697180887.png


Now in the mean time, Intel is expected to launch Xeon 7 (Clearwater Forrest & Diamond Rapids) and Nova Lake desktop CPUs too but unsure what the volume will be for 2026.
 
You must log in or register to reply here.
Back
Top