Discussion Leading Edge Foundry Node advances (TSMC, Samsung Foundry, Intel) - [2020 - 2025]

[DisEnchantment]

TSMC's N7 EUV is now in its second year of production and N5 is contributing to revenue for TSMC this quarter. N3 is scheduled for 2022 and I believe they have a good chance to reach that target.


N7 performance is more or less understood.


This year and next year TSMC is mainly increasing capacity to meet demands.

AnandTech Forums: Technology, Hardware, Software, and Deals

Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

www.anandtech.com

For Samsung the nodes are basically the same from 7LPP to 4 LPE, they just add incremental scaling boosters while the bulk of the tech is the same.

Samsung is already shipping 7LPP and will ship 6LPP in H2. Hopefully they fix any issues if at all.
They have two more intermediate nodes in between before going to 3GAE, most likely 5LPE will ship next year but for 4LPE it will probably be back to back with 3GAA since 3GAA is a parallel development with 7LPP enhancements.

AnandTech Forums: Technology, Hardware, Software, and Deals

Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

www.anandtech.com

Samsung's 3GAA will go for HVM in 2022 most likely, similar timeframe to TSMC's N3.
There are major differences in how the transistor will be fabricated due to the GAA but density for sure Samsung will be behind N3.
But there might be advantages for Samsung with regards to power and performance, so it may be better suited for some applications.
But for now we don't know how much of this is true and we can only rely on the marketing material.

This year there should be a lot more available wafers due to lack of demand from Smartphone vendors and increased capacity from TSMC and Samsung.
Lots of SoCs which dont need to be top end will be fabbed with N7 or 7LPP/6LPP instead of N5, so there will be lots of wafers around.

Most of the current 7nm designs are far from the advertized density from TSMC and Samsung. There is still potential for density increase compared to currently shipping products.
N5 is going to be the leading foundry node for the next couple of years.

For a lot of fabless companies out there, the processes and capacity available are quite good.

---------------------------------------------------------------------------------------------------------------------------------------------------


FEEL FREE TO CREATE A NEW THREAD FOR 2025+ OUTLOOK, I WILL LINK IT HERE

 

[coercitiv]

The original Gelsinger promise / pitch was that 18A would leapfrog TSMC.

That's ok, they'll leapfrog with 14A. /s

Remember, they always talk about leadership. They never talk about meeting deadlines, predictable outcomes, good mix of performance/volume/cost.

 

[igor_kavinski]

The original Gelsinger promise / pitch was that 18A would leapfrog TSMC.

Probably why he got fired coz his frog failed to leap. Something that's actually expected from a frog.

I think LBT may be trying to prevent retooling existing fabs to 18A and go for 14A instead. That would save money in the end but they will need to make do with limited 18A capacity in the meanwhile and rely on TSMC for the bulk of their maufacturing needs. LBT may also be in a better position to negotiate cheaper prices with TSMC because he doesn't have a tendency to run his mouth faster than his brain.

My suggestion for the 18A capacity. Fab a limited edition run of one million Tejas CPUs and call it Pentium 4 Anniversary Edition. They can simply shrink the die with AI. 6 months tops.

 

[Io Magnesso]

That's ok, they'll leapfrog with 14A. /s

Remember, they always talk about leadership. They never talk about meeting deadlines, predictable outcomes, good mix of performance/volume/cost.

Well, they didn't say anything
However, as long as we do the FOUNDRY business of providing it to the outside,
From now on, I can't lie
Because external customers' products come out around 2026-2027 It would be nice if I could get a contract with a customer who has already signed a contract...

 

[511]

It says Intel is contemplating to no longer offer 18A to external customers, which could lead to a partial write off of development costs of the node - ranging from 100s of millions to billions.

So basically they are going to use it internally and not for extenal customers except where they have contractual obligations.

Probably why he got fired coz his frog failed to leap. Something that's actually expected from a frog.

I think LBT may be trying to prevent retooling existing fabs to 18A and go for 14A instead. That would save money in the end but they will need to make do with limited 18A capacity in the meanwhile and rely on TSMC for the bulk of their maufacturing needs. LBT may also be in a better position to negotiate cheaper prices with TSMC because he doesn't have a tendency to run his mouth faster than his brain.

Well if that happens than they are even more bad position cause they don't have anything to sell how i see it is that winding down any Fab Capacity/Equipment that is supposed plan for external not internal use they will ramp it for Internal otherwise No NVL/DMR/WCL/Celestial/PTL/CLW-F.

My suggestion for the 18A capacity. Fab a limited edition run of one million Tejas CPUs and call it Pentium 4 Anniversary Edition.

LOL.

 

[Joe NYC]

Probably why he got fired coz his frog failed to leap. Something that's actually expected from a frog.

I think LBT may be trying to prevent retooling existing fabs to 18A and go for 14A instead. That would save money in the end but they will need to make do with limited 18A capacity in the meanwhile and rely on TSMC for the bulk of their maufacturing needs. LBT may also be in a better position to negotiate cheaper prices with TSMC because he doesn't have a tendency to run his mouth faster than his brain.

My suggestion for the 18A capacity. Fab a limited edition run of one million Tejas CPUs and call it Pentium 4 Anniversary Edition.

Or, maybe more precisely, Intel will have to buy less 18A ready equipment if it is no longer offered to outside customers, if it is only serving intel.

That would defer some of the equipment purchases by a couple of years.

My "thesis" about Intel's problem has been that under Gelsinger's plan, Intel would simply run out of cash, and that's the reason Gelsinger was fired.

By reducing and deferring this major cash expenditure + mass layoffs, LBT has a better shot at saving Intel from running out of cash.

 

[igor_kavinski]

Well if that happens than they are even more bad position cause they don't have anything to sell how i see it is that winding down any Fab Capacity/Equipment that is supposed plan for external not internal use they will ramp it for Internal otherwise No NVL/DMR/WCL/Celestial/PTL/CLW-F.

I think LBT is thinking of putting the fabs in hibernation mode for a while until they can get their act together. He's probably going to try to run Intel like AMD with outsourced manufacturing. That seems logical because Intel's 2024 losses are mainly fab related.

 

[Io Magnesso]

The question is that 18A has already completed the process itself. After that, I think the main work is to increase the yield...
Why would you want to stop providing it?
There is talk that 18AP will be canceled, but if you use it for your company's products, it's an excuse. I don't think it would cost that much to develop with a 18A variation node.
It's a matter of course to focus on 14A,
Isn't it possible to wait and see instead of not providing 18A?
Is it a Reuters in the first place because it's fuzzy information?

 

[Joe NYC]

The question is that 18A has already completed the process itself. After that, I think the main work is to increase the yield...
Why would you want to stop providing it?
There is talk that 18AP will be canceled, but if you use it for your company's products, it's an excuse. I don't think it would cost that much to develop with a 18A variation node.
It's a matter of course to focus on 14A,
Isn't it possible to wait and see instead of not providing 18A?
Is it a Reuters in the first place because it's fuzzy information?

There is another reason why external customers rejected 18A, according to Daniel Nenni: Intel's PDK definition of the 18A process was delayed, and when it arrived, it was still quite inferior to TSMC. And by now, all the customers that matter have moved to TSMC N2.

 

[Io Magnesso]

There is another reason why external customers rejected 18A, according to Daniel Nenni: Intel's PDK definition of the 18A process was delayed, and when it arrived, it was still quite inferior to TSMC. And by now, all the customers that matter have moved to TSMC N2.

That's right, I agree
I don't think it's a to stop providing it to external customers because the PDK is inferior.
The experience gained from 14A may lead to the enhancement of the PDK of 18A.

 

[Io Magnesso]

If you want to improve, I think it is necessary to gain experience now rather than procrastination.

 

[igor_kavinski]

it was still quite inferior to TSMC.

Intel seriously needs to swallow its pride and poach some top TSMC talent. There's no shame in it.

 

[511]

I think LBT is thinking of putting the fabs in hibernation mode for a while until they can get their act together. He's probably going to try to run Intel like AMD with outsourced manufacturing. That seems logical because Intel's 2024 losses are mainly fab related.

It's not possible he is putting external customer to Hibernation not Intel Internal volume

 

[511]

Intel seriously needs to swallow its pride and poach some top TSMC talent. There's no shame in it.

Funnily their SVP is Ex Intel Kevin Zhang.

 

[poke01]

So is PTL still on track?? I don’t get why no external customer would not 18A. Isn’t it cheaper than N3E/P?

 

[511]

So is PTL still on track?? I don’t get why no external customer would not 18A. Isn’t it cheaper than N3E/P?

Plans are decided 2-3 years in advanced and PDK came out last year and was horrible for external they missed timelines also big client's like AMD/NVIDIA won't trust Intel even if the process is fine everyone is skeptical and combined that with poor PDK even if the node is fine.

also PTL is on track but it's still ES haven't seen a single leak pointing to QS

 

[poke01]

plot thickens

 

[maddie]

Remember, they always talk about leadership.

That is a much bigger psychological problem than most assume, and generally leads to poor outcomes.

 

[511]

https://twitter.com/x/status/1940367537168060848

 

[Io Magnesso]

In this turmoil... Raichu said...
Intel 18A has a 4th variation...
(intel 18A/1st 18AP/2 18A-PT/3 18A-？/4？)

 

[Josh128]

Intel Considers Abandoning 18A Node for 14A Chipmaking Process

Intel's CEO Lip-Bu Tan is looking at a big change to how the company makes chips for others, Reuters reports citing people who know about it. The new plan might mean Intel stops offering its 18A process technology to other companies. This is different from the strategy of the former CEO, Pat...

www.techpowerup.com

Uhh...what the hell? 20A scrapped for better looking 18A. 18A used for internal, external put aside for 14A. Not good.

 

[marees]

Intel Considers Abandoning 18A Node for 14A Chipmaking Process

Intel's CEO Lip-Bu Tan is looking at a big change to how the company makes chips for others, Reuters reports citing people who know about it. The new plan might mean Intel stops offering its 18A process technology to other companies. This is different from the strategy of the former CEO, Pat...

www.techpowerup.com

Uhh...what the hell? 20A scrapped for better looking 18A. 18A used for internal, put aside for 14A. Not good.

Keep moving all the eggs to a new basket

 

[511]

Keep moving all the eggs to a new basket

That basket is going to be full soon.

 

[511]

 

[Win2012R2]

As expected this is what Pat was fired for - "18A on track, yields great blah blah blah", quiet clearly renaming 20A to 18A does not conjure premium customers out of thin air the moment PDK is "ready". The -P version looks like what 18A should have been from start - this year.

It seems to me also that their gamble on High-NA is a bust too, TSMC made the right decision not to rush it.

It's over.

 

[Josh128]

As expected this is what Pat was fired for - "18A on track, yields great blah blah blah", quiet clearly renaming 20A to 18A does not conjure premium customers out of thin air the moment PDK is "ready".

It seems to me also that their gamble on High-NA is a bust too, TSMC made the right decision not to rush it.

It's over.

I REALLY hope they can recover and begin to compete again, but its looking pretty bleak at this time.