Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/index.php?threads/tsmc-4th-qtr-2020-earnings-discussion.13609/page-2
        )

    [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] => 2021370
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

TSMC 4th QTR 2020 Earnings Discussion

Iam not sure, if intel is honest they haven't decided if they outsource yet (also they want to wait on new CEO).
I think its more likely that AMD made a big order and also Apple increased quite a bit...
Intel is honest, for sure. They will "announce" their decision this Thursday or postpone the announcement to a later day. But as big as Intel is, there's no way they haven't made the firm decision yet. They are trying to give the public an impression that whatever decision they already made is agreed and supported by the old and new CEO.
 
I think that author misunderstood what described in the original Anandtech article:

"“This is the result of a demand focused environment, rather than manufacturing issues,” Su said."
"There is tightness in the supply chain due to demand, and that invariably puts pressure on our consumer, PC, and gaming product lines. As it relates to our semiconductor production, we’re putting in additional capacity to meet this unexpected demand. It will take time to catch up, but that’s what we’re seeing."
 
This would explain the CAPEX bump:


I wonder if TrendForce will print a retraction when they are proven wrong?

I think this plan, which we've heard rumors about before, makes sense on the surface. I don't know if it's true, but moving i3 over this year with higher end PC chips over next year makes sense to me. It frees up 10nm capacity that Intel can use for datacenter chips, assuming yields are improving.
 
This would explain the CAPEX bump:


I wonder if XXXXXX will print a retraction when they are proven wrong?
Bloomberg notes that there are clues indicating that Intel has decided to outsource.

Key points:
1. ASML said it is shifting orders for some of its most advanced machines from one customer to others (such as TSMC and Samsung). It didn't say who but Bloomberg believes the customer is Intel.
2. Intel has talked with TSMC and Samsung about making some of its best chips
3. TSMC and Samsung have production technology that's now more advanced than Intel
4. Intel may not deliver its final answer today (Thursday). Swan had said he would announce whether to outsource production & how much during the first quarter but the incoming CEO Gelsinger may need more time to develop his own strategy.

One more note....if TrendForce's Jan 13 prediction fails to materialize, I imagine there will be significant reputational damage to their brand. However, if their predictions pan out then many people will be paying more attention to their press releases going forward. For reference, XXXX is predicting that:
1. Intel will kick off mass production of Core i3 CPUs at TSMC’s 5nm node in 2H21
2. Intel’s mid-range and high-end CPUs are projected to enter mass production using TSMC’s 3nm node in 2H22
 
Last edited by a moderator:
This would explain the CAPEX bump:


I wonder if TrendForce will print a retraction when they are proven wrong?

  • On the Q4 earnings call, incoming Intel CEO Pat Geisinger says he believes "a majority" of the company's 2023 products will still be handled internally, though it's "likely" outside foundries will be used more than in the past.
Ooops......
 
  • On the Q4 earnings call, incoming Intel CEO Pat Geisinger says he believes "a majority" of the company's 2023 products will still be handled internally, though it's "likely" outside foundries will be used more than in the past.
Ooops......
Hm strange why the huge capex expansion @TSMC??
 
Hm strange why the huge capex expansion @TSMC??
Apple (smartphones and Macs), Qualcomm, AMD, Nvidia, MediaTek, Intel, and Microsoft.

Microsoft can be an interesting one. I think Microsoft is very worry about if Intel can't regain technology and manufacturing leadership, it can cause serious problem for Microsoft to compete against Apple.
 
In the Oct 14, 2020 ASML earnings call, Peter Wennink (ASML President & CEO) was asked about his thoughts on the possibility of the number of customers producing leading edge devices going down. He stated:
Now, on the number of customers in the leading edge, as one thing is absolutely certain, things are not getting easier. The next nodes will increase complexity and I think only the very large customers can deal with that. We've seen over the last 20 years a significant reduction in number of leading edge customers and I think it's going to continue with less than a handful. And I'm going to not go to speculate who those are going to be because they are all dear to my heart. And I think they'll have to figure it out who has the best designs, the best production technology, who are the most efficient and the lowest cost. But in the end, it doesn't matter, because in the end what we always talk about is the number of wafers and device and chips that are needed in the digital transformation that need to be made somewhere on the planet.

And that means that I do believe that our large customers are going to be larger and they are going to be dominant in areas of chip production, which is only going to grow. That is why I said at the end of my prepared remarks, our confidence in our 2025 outlook has only grown. As always, ultimately, you could argue also bit as a result of what we've seen as a part of the COVID crisis how important this digital infrastructure is. So, in that sense, yes, I think there will be fewer and fewer customers, but it will be much bigger than they are today.

Given that TSMC, Samsung and Intel are ASML's only three leading edge customers at the time....it would appear Peter didn't believe that all three would be able to stay on the leading edge. Now take what he said about his leading-edge customers two days ago:
Yes. We're not going to be specific on any customer as you can imagine because of the fact that we only have very few. So the issue is that when we look at 2021, two things are impacting our shipment scheduled to our leading-edge customers is, one is what you referred to is effectively -- has there been a transition from tools that we already planned for customer A potentially to customer B and C? I think, yes, that has happened. But on top of that I think there is the increased demand for advanced nodes. So it's the combination of the two that actually caters for maximizing the shipments out of our shipment capability. And it's the redistribution that has happened. So, yes, I think we are -- that's been sold out for this year. But perhaps with one or two upsides referring to the previous question. But I think no impact on this foreseen in our 2021 numbers.

The big question is who is "customer A" that Peter referred to? Given TSMC's $25-$28 billion capex for 2021, they would probably be "customer B" or "customer C". Samsung would also seem NOT to be "customer A" given that Samsung Electronics Vice Chairman Lee Jae-young asked for more EUV equipment from ASML in late 2020. So Intel would seem to be "customer A" given that their current 10nm process doesn't use EUV. Given that Intel has yet to release a product using EUV manufacturing and Peter's comments, it certainly casts some doubt on Intel's ability to leapfrog TSMC.

Note that incoming CEO Pat Gelsinger didn't dispute an analyst statement stating that when "Intel gets to 7nm by 2023, the competition (TSMC) will be one effective node ahead". But he stated Intel is committed to delivering leadership products with the goal of re-establishing its lead in process node technology.
Majority of our products in 2023 will be on Intel’s 7nm, but we will have increasing use of external foundry. Intel is committed to building a competitive product with leadership in the marketplace. This means packaging, software, external and internal fabs, and we are confident in delivering a product leadership across all categories

Now yesterday's Intel earnings call didn't appear to me to out-right reject what was stated in a Taiwanese research report Jan 13 press release. Gelsinger stated that the "[m]ajority" of Intel's products will be using Intel 7 nm. But given that Intel processors include Xeon Platinum, Xeon D, Xeon W, Xeon E, X-series, i9, i7, i5, i3 and m3, Pentium Gold, Pentium Silver, Celeron, Atom, etc....TSMC producing Core i3 CPUs for 2022 & Core i5 (maybe Core i7?) for Intel in 2023 can "fit" with what he stated. Gelsinger was deliberately vague here.
 
Last edited by a moderator:
The big question is who is "customer A" that Peter referred to? Given TSMC's $25-$28 billion capex for 2021, they would probably be "customer B" or "customer C". Samsung would also seem NOT to be "customer A" given that Samsung Electronics Vice Chairman Lee Jae-young asked for more EUV equipment from ASML in late 2020. So Intel would seem to be "customer A" given that their current 10nm process doesn't use EUV. Given that Intel has yet to release a product using EUV manufacturing and Peter's comments, it certainly casts some doubt on Intel's ability to leapfrog TSMC.
It is particularly interesting, since IIRC Intel was the first with a big order of 15 tools, but this was before the NXE3400 came out.

I wonder if they (1) had already bought 15 tools, which are no longer up to date and therefore not competitive or (2) they bought a few, then had credit for future ones, which they may be holding out for high NA.
 
The number is too big for AMD and Apple increasing their orders. We are talking about a 50% jump in TSMC capex over 2020, and a 25-40% jump from expected capex just a few months ago. I'm sure some of it is Apple asking for more wafers for M1, and AMD trying to get more capacity to fill demand, but I think it's likely part (if not most) of it is Intel. Don't forget Intel already oursources a lot to TSMC, so it's not a matter of "to outsource or not" it's a matter of "which products and how much".
I think it is a good strategy. TSMC will grow very quickly cause Intel will have to outsource some of the production. Intel will eventually go into full IDM mode again, but at that time AMD will be able to fill the gap pretty easily.
 
Back
Top