Discussion Intel leading customer for TSMC 3nm?

[Failnaught]

There is news that Intel is the co-leading customer on the TSMC 3nm node, with Apple. Did they manage to outflank AMD on process technology? What are they going to make on this node?

https://asia.nikkei.com/Business/Te...become-first-to-adopt-TSMC-s-latest-chip-tech

 

[LikeLinus]

One thing everyone seems to have missed...

"Intel, on the other hand, will be using TSMC to produce some of its consumer chips for the first time, something it has typically used its own foundries to do despite using TSMC for some of its other product lines.

"Intel has said these will be fabricated on its own 7 nm production lines."

Why does everyone think Intel is producing all it's chips with TSMC. Most likely they will continue 10nm/7nm production for a vast majority of their products and use 3nm for the high performance stuff. Even then, their production on 3nm will most likely surpass AMD and I'm baffled why everything things TSMC wouldn't accept more money for their product or a new customer to make more money? redacted people, do any of you actually work for a company that looks to raise profits???? Adding a customer or more production is huge for them.

There is nothing to say they don't have the compacity to produce Apple and Intel 3nm at the same time depending on supply. It's all jumping to conclusions, at this point.

We have a zero tolerance policy for profanity in the tech forums.
You need to stop posting profanity.

Iron Woode
Super Moderator

 

[beginner99]

How is it that Intel gained marketshare in 1Q 2021 over AMD? You know, cause people for the past two years have been saying Intel was dead and they can't survive. Makes me curious?

Because of the chip shortage. The one who produces the most sells the most, regardless of the performance or performance/watt figures. If you can sell everything, then the only thign that matters for market share is your production speed.

The pandemic was a huge blessing for intel, especially in the server space.

 

[coercitiv]

redacted people, do any of you actually work for a company that looks to raise profits???? Adding a customer or more production is huge for them.

Does the company you work for provide their best services and products to their main competitor just to raise profits? Read the thread, there's plenty of insights on what's actually happening, including information about the political game both Intel and TSMC are playing to gain favors from the US government.

Profits from selling 3nm capacity pale in comparison to what's actually at stake here.

 

[DrMrLordX]

While you say that I don't have any faith in Intel delivering 10nm SuperFIN

10SF already exists. It's Tiger Lake.

It wasn't a matter of "choosing". If you come to TSMC with cash in hand they'll sell you future wafers. Intel bought up future capacity before AMD did, that's why they have it.

See earlier in thread: this is potentially stupid for TSMC so I'll believe it when I see it.

If you want to blame someone, blame AMD's management for deciding to spend precious cash on stock buybacks instead of future wafer allocation.

The irony of someone upbraiding a company for stock buybacks in an Intel thread.

mindwhatever (some German retail site)

lol okay

Intel still gained marketshare in the Laptop/desktop segment.

Go look at their DCG segment. Woof.

Why does everyone think Intel is producing all it's chips with TSMC.

Zero confidence in 7nm production + yields.

 

[Hitman928]

I think @Doug S is right. AMD is tossing $4B to investors to pump up their stock prices (and make AMD executive rich). It's BS. What if they waved that cash in front of TSMC? That's allot to pass up. AMD is going from being the smaller scrappier company to another typical Wall Street playa. It's a sad turn that could jeopardize their medium term success.

If this news is true, it won't be because AMD couldn't pay TSMC for the wafers.

 

[moinmoin]

The fact that this "news" is not instantly laughed out of court by everybody shows how Intel was capable of meandering on pretending everything's fine for so long despite all the ongoing publicly visible issues. In the end this kind of unthinking behavior is good for nobody. It's gross that this kind of "reporting" is directly affecting stock markets.

 

[jpiniero]

The fact that this "news" is not instantly laughed out of court by everybody shows how Intel was capable of meandering on pretending everything's fine for so long despite all the ongoing publicly visible issues. In the end this kind of unthinking behavior is good for nobody. It's gross that this kind of "reporting" is directly affecting stock markets.

What you don't think it's true? Sounds better than ending up with 10+++++++++ when they inevitably delay 7 nm again.

 

[LightningZ71]

At this point, I believe that there will be a "7nm" node from Intel next year. It will be an improvement from 10sf in most every way. However, I also believe that, just like 10nm has been, there will be several "revisions" of it over time that get it to where it was originally intended to be. Why? Past performance. If they were that confident about it, they would be beating a drum for it. They wouldn't be nosing around TSMC and giving them money that they could use to keep ahead of Intel.

 

[Doug S]

At this point, I believe that there will be a "7nm" node from Intel next year. It will be an improvement from 10sf in most every way. However, I also believe that, just like 10nm has been, there will be several "revisions" of it over time that get it to where it was originally intended to be. Why? Past performance. If they were that confident about it, they would be beating a drum for it. They wouldn't be nosing around TSMC and giving them money that they could use to keep ahead of Intel.

Intel sold about 100,000 10nm CPUs in 2018 in a single SKU, just to keep promises about their 10nm timeline.

Even today you can't claim they are fully on board with 10nm since it hasn't rolled out across their whole product line like every past process has done. So the date when they begin shipping 7nm is largely irrelevant, since they have shown they will paper launch a process for the sole reason of fooling Wall Street.

What matters is when it makes up a significant part of their revenue mix, which won't happen next year and I'm not convinced will even happen in 2023.

 

[Ajay]

What matters is when it makes up a significant part of their revenue mix, which won't happen next year and I'm not convinced will even happen in 2023.

The main issue, I think, will be the number of EUV lithographs they have. It's possible 7nm (+, ++) will be a bust for Intel. Hopefully, they have gotten back in line for EUV systems for the next node (5nm, or are they going to try to skip a node - IBM 2nm?).

 

[moinmoin]

What you don't think it's true? Sounds better than ending up with 10+++++++++ when they inevitably delay 7 nm again.

It's not up to Intel what TSMC is going to do.

 

[jpiniero]

It's not up to Intel what TSMC is going to do.

Correct. But that doesn't mean TSMC won't take Intel's money, especially if it's a very large number.

 

[moinmoin]

Correct. But that doesn't mean TSMC won't take Intel's money, especially if it's a very large number.

Correct. TSMC will gladly take Intel's money, Apple's money, AMD's money etc. pp.

The lot is kidding themselves though if they think TSMC is voluntarily backing itself into a corner and taking away capacity of its other customers just because Intel is waving a check.

 

[jpiniero]

The lot is kidding themselves though if they think TSMC is voluntarily backing itself into a corner and taking away capacity of its other customers just because Intel is waving a check.

They might be able to get some Biden Bux out of it too if TSMC plays their cards right. A free or discounted 3 nm fab in the US might be nice.

Just because it might be bad for AMD doesn't mean TSMC won't do it.

 

[moinmoin]

They might be able to get some Biden Bux out of it too if TSMC plays their cards right. A free 3 nm fab in the US might be nice.

Might be. Doesn't make the "news" with its many red flags this thread is about any more plausible though.

 

[jpiniero]

Might be. Doesn't make the "news" with its many red flags this thread is about any more plausible though.

Article says Intel confirmed they are releasing products made at TSMC in 2023.

 

[moinmoin]

Just because it might be bad for AMD doesn't mean TSMC won't do it.

Article says Intel is getting capacity at the cost of Apple. You don't bat an eye at that?

Article says Intel confirmed they are releasing products made at TSMC in 2023.

Intel is releasing products made at TSMC for ages already, that's not news.

 

[jpiniero]

Article says Intel is getting capacity at the cost of Apple. You don't bat an eye at that?

Apple can't use it for the 2022 iPhone since it won't be ready in time for them. That was explained earlier in the thread.

 

[moinmoin]

Apple can't use it for the 2022 iPhone since it won't be ready in time for them. That was explained earlier in the thread.

The article says so. Other places such as AT say N3 is for H2 2022, not "end of 2022 at the earliest" as the article says.

 

[jpiniero]

The article says so. Other places such as AT say N3 is for H2 2022, not "end of 2022" as the article says.

Even early H2 is still far too late. Apple wants production started in February or early March for a Sept/October release.

 

[Doug S]

The main issue, I think, will be the number of EUV lithographs they have. It's possible 7nm (+, ++) will be a bust for Intel. Hopefully, they have gotten back in line for EUV systems for the next node (5nm, or are they going to try to skip a node - IBM 2nm?).

Every time I've heard the idea of "skipping a node" to catch up suggested, either in the recent past or 20 years ago, process engineers have said that such a thing is almost impossible. A new node isn't a single change, it is dozens of changes that build on what came before. In order to skip a node you still have to develop all those incremental steps, because it isn't the same things that are changing between nodes.

If skipping a node was feasible we would have seen that happen before, but I can't think of a single case where anyone has ever skipped a node - let alone doing so anywhere near leading edge. Perhaps some tier two foundry could skip from 45nm to 28nm, but that's not the same thing.

And IBM's 2nm was just a technology demonstration. I'm sure TSMC and Samsung have 2nm wafers in their development fabs too. That doesn't mean they're ready to roll them out anytime soon. Even if IBM handed all their work over to Intel it isn't like they can skip a couple nodes and start shipping 2nm in 2024 to leapfrog TSMC. That's just not possible.

 

[Doug S]

Correct. TSMC will gladly take Intel's money, Apple's money, AMD's money etc. pp.

The lot is kidding themselves though if they think TSMC is voluntarily backing itself into a corner and taking away capacity of its other customers just because Intel is waving a check.

Who says they are "taking capacity away" from anyone. What is so hard to understand about first come first served? If you come to TSMC tomorrow with a check for a couple billion dollars, you might be first in line for N2 wafers. Since Apple is probably already first in line for N2, more likely you'd be second. If you wait until N2 is around the corner, like N3 is now, you won't be first, or second, or maybe even third.

 

[leoneazzurro]

I don't understand how big the issue is. Capacities are booked early, true, but they are discussed by quarters and having a early booking on N3 or N2 does not mean you will have access to that capacity for the whole cycle. AMD's leaked roadmaps have hinted that the 5nm product for AMD is ZEN4, and projected date for it is not early 2022. So in late 2022 and large part of 2023 at least AMD would have used anyway the 5nm process node. AMD wil llikely use the 3nm node when they are ready to.

 

[Abwx]

They might be able to get some Biden Bux out of it too if TSMC plays their cards right. A free or discounted 3 nm fab in the US might be nice.

Just because it might be bad for AMD doesn't mean TSMC won't do it.

It will be bad for AMD (and Apple) today and then bad for TSMC once Intel get their processes working as they will get back all this procuction in their own fabs, in the meantime they will stop marketshare bleeding thanks to a complacent and short sighted TSMC...

 

[jpiniero]

It will be bad for AMD (and Apple) today and then bad for TSMC once Intel get their processes working as they will get back all this procuction in their own fabs, in the meantime they will stop marketshare bleeding thanks to a complacent and short sighted TSMC...

If taking Intel's order pressures AMD to go to 3 nm sooner than they would like and/or accept way higher wafer prices, that's not exactly a problem for TSMC.

Plus TSMC has plenty of other customers.