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

 

[Geddagod]

TSMC 2nm delayed into 2026?

https://twitter.com/x/status/1703748422723883349

Good news for Intel 18A if true.

TSMC denied those rumors, and also TSMC 2nm HVM was announced to be in 2025, but very likely it would have been very late 2025. I can see TSMC "stretching the truth" about entering HVM manufacturing for 2nm late 2025 if the delay (if the delay even exists, which TSMC denied) is only like a quarter or maybe two.
I'm giving TSMC the benefit of the doubt, esp since I've given Intel's foundry benefit of the doubt on their own fabs as well lol

 

[Dayman1225]

Intel claims Intel 4 has the best yields at product launch that they’ve seen in the last decade.

 

[Thibsie]

View attachment 85959
Intel claims Intel 4 has the best yields at product launch that they’ve seen in the last decade.

Is Intel's 14nm considered as this or previous one?
'cos if they are comparing to their 10nm woes, that very simple to do lol

 

[Ajay]

Is Intel's 14nm considered as this or previous one?
'cos if they are comparing to their 10nm woes, that very simple to do lol

11 years would go back the introduction of 22nm FinFET based products. So I have no idea what Intel means by "the last decade".

 

[Dayman1225]

11 years would go back the introduction of 22nm FinFET based products. So I have no idea what Intel means by "the last decade".

22nm was Intel best ever yielding node so I’m assuming that’s why it’s not on the graph.

Is Intel's 14nm considered as this or previous one?
'cos if they are comparing to their 10nm woes, that very simple to do lol

I’m confused, you see that 14nm is on the photo, correct?

 

[jpiniero]

Intel claims Intel 4 has the best yields at product launch that they’ve seen in the last decade.

Gonna call shens unless this is them validating their yieldable die strategy.

Now will we see 1+0+2 Meteor Lake?

 

[DrMrLordX]

Is Intel's 14nm considered as this or previous one?
'cos if they are comparing to their 10nm woes, that very simple to do lol

Early 14nm wasn't that great either, unless memory fails.

 

[jpiniero]

Early 14nm wasn't that great either, unless memory fails.

Broadwell Core M was pretty papery... but it was also the full die.

 

[Ajay]

There was also a six month delay in getting 14nm production ready.

 

[lightisgood]

iPhone 15 Pro Max Suffers From Severe Thermal Throttling; A17 Pro Only Achieves High Single, Multi-Core Scores Using An External Cooler

Despite featuring a cutting-edge 3nm A17 Pro, the iPhone 15 Pro Max still overheats and needs an external cooler to maintain decent benchmark scores

wccftech.com

I'd like to give a wry smile to TSMC node guys.

 

[eek2121]

Gonna call shens unless this is them validating their yieldable die strategy.

Now will we see 1+0+2 Meteor Lake?

It is believable as Intel 4 is using EUV. IIRC TSMC reported something similar with N7+/N6 and also N5.

That is why I am surprised Intel 4/3 isn’t longer lived. Seems like they aren’t using it long enough to break even on development costs. Also, yes, I am aware of IFS plans. Still…

 

[DrMrLordX]

It is believable as Intel 4 is using EUV. IIRC TSMC reported something similar with N7+/N6 and also N5.

That is why I am surprised Intel 4/3 isn’t longer lived. Seems like they aren’t using it long enough to break even on development costs. Also, yes, I am aware of IFS plans. Still…

Either that or Intel is just playing a shell game with that statement, and yields/performance characteristics on Intel 4 are bad enough that they're gonna sweep it under the rug and pretend like it never happened. It's easier for them to bugfix it, rename it Intel 3, and move on to that (and even Intel 3 won't be showing up in any known Intel consumer products).

 

[Dayman1225]

Either that or Intel is just playing a shell game with that statement, and yields/performance characteristics on Intel 4 are bad enough that they're gonna sweep it under the rug and pretend like it never happened. It's easier for them to bugfix it, rename it Intel 3, and move on to that (and even Intel 3 won't be showing up in any known Intel consumer products).

Yields on Intel4 is pretty good, Intel 3 is an extension of Intel4 with additional libraries and will be a long life node for Intel and foundry customers

 

[DrMrLordX]

Yields on Intel4 is pretty good

We'll never know since Intel is mothballing it immediately after Meteor Lake.

, Intel 3 is an extension of Intel4 with additional libraries and will be a long life node for Intel and foundry customers

How many Intel products will there be on Intel 3 besides Granite Rapids and Sierra Forest? Intel may schlep it to IFS2.0 customers, sure, but it sure looks like they're ready to move past that one pretty quickly as well for their own stuff.

 

[Dayman1225]

We'll never know since Intel is mothballing it immediately after Meteor Lake.



How many Intel products will there be on Intel 3 besides Granite Rapids and Sierra Forest? Intel may schlep it to IFS2.0 customers, sure, but it sure looks like they're ready to move past that one pretty quickly as well for their own stuff.

It will be used in future tiles on future products. Just because Intel hasn’t announced other products using it doesn’t mean it won’t be used. Not much more I can say

 

[DrMrLordX]

It will be used in future tiles on future products. Just because Intel hasn’t announced other products using it doesn’t mean it won’t be used. Not much more I can say

Guess we'll find out, but for now it looks like the critical tiles for consumer products are moving from Intel 4 (Meteor Lake) to 20a (Arrow Lake mobile).

 

[JoeRambo]

Guess we'll find out, but for now it looks like the critical tiles for consumer products are moving from Intel 4 (Meteor Lake) to 20a (Arrow Lake mobile).

Let's not loose context here: this is their first node using EUV machines. EUV machines that will carry them till EUV NA machines will step in to be used on 'Intel next ar whatever they call it" process.
Since they don't have infinite number of EUV machines, "4" looking abandoned, having limited production, might not have anything to do with process and everything to do with their plans for 20A and 18A ?
Intel is still catching up now, would not make much sense to bet the farm on TSMC N5 like process in 2024 ?

And surely yields with EUV are great, we already know that from TSMC 5nm that is using it heavily. I think things will only go down once they need multipatterning with EUV.

 

[DrMrLordX]

Since they don't have infinite number of EUV machines, "4" looking abandoned, having limited production, might not have anything to do with process and everything to do with their plans for 20A and 18A ?

If that's the case, that doesn't necessarily bode well for any of their EUV nodes unless they're taking delivery of a lot more equipment soon. Assuming the Mizuho report was right (and things may have changed since then), Intel might not be taking significant deliveries of EUV equipment until 2025. And some of that was supposed to be High NA EUV equipment.

Plus unless ASML has significantly increased their capacity to produce EUV machines, all the extra N3 capacity TSMC built out for Intel at one of their research fabs had to have eaten into the available supply of EUV equipment.

Intel is still catching up now, would not make much sense to bet the farm on TSMC N5 like process in 2024 ?

I think the question is, are they actually catching up? Seems like people are bending over backwards to make excuses for them despite them having been a recent market leader in silicon manufacturing and how they're absorbing massive amounts of . . . shall we say, outside investment to prop up their foundry business.

And surely yields with EUV are great

N3B's yields are what,55 - 60%? Not amazing. EUV isn't a magic bullet that automatically makes yields better.

 

[moinmoin]

And surely yields with EUV are great, we already know that from TSMC 5nm that is using it heavily.

TSMC's competitive advantage was for a long time making good use of EUV at good yield. Intel's major reason for stagnating node process for so long was that they didn't manage to make EUV work well for them. So Intel finally using EUV won't equal to great yield, the latter is still to be seen (but also about time).

 

[lightisgood]

View attachment 85959
Intel claims Intel 4 has the best yields at product launch that they’ve seen in the last decade.

I can't understand that some people are arguing about MTL yeild now.
MTL is exceeding Skylake's yield.
This means MTL has attained roughly perfect yield.
This is why Intel will be able to reach PRQ such big tile products as GNR/SRF soon.

 

[DrMrLordX]

MTL is exceeding Skylake's yield.

No, it exceed's early Broadwell yield. Not exactly the same thing.

 

[João Bortolace]

No, it exceed's early Broadwell yield. Not exactly the same thing.

the slide clearly shows the meteor above the skylake

 

[JoeRambo]

If that's the case, that doesn't necessarily bode well for any of their EUV nodes unless they're taking delivery of a lot more equipment soon. Assuming the Mizuho report was right (and things may have changed since then), Intel might not be taking significant deliveries of EUV equipment until 2025. And some of that was supposed to be High NA EUV equipment.

Plus unless ASML has significantly increased their capacity to produce EUV machines, all the extra N3 capacity TSMC built out for Intel at one of their research fabs had to have eaten into the available supply of EUV equipment.

Let us rest assured -> this is commercial secret, IF Intel is coming up with products and releasing them, things much be getting better for them. But obviuosly they are moving on to 18/20A that also require same machines. So i will reiterate my claim -> product volume might not be due to process being unhealthy and more about R&D and manufacturing concerns.

I think the question is, are they actually catching up? Seems like people are bending over backwards to make excuses for them despite them having been a recent market leader in silicon manufacturing and how they're absorbing massive amounts of . . . shall we say, outside investment to prop up their foundry business.

Impossible to say when all products on market are 7nm, next year will be crucial. But surely if those products they've announced are not wapoware they will be just fine.

N3B's yields are what,55 - 60%? Not amazing. EUV isn't a magic bullet that automatically makes yields better.

What N3B has to do with Intel's 4? Despite all forum warrior blablering, Intel's 4 is more like TSMC N5 and Intel's 3 is N4? Give or take some nm's.
So 1st generation EUV products and not something where TSMC pushed things to run into trouble.

 

[jpiniero]

I can't understand that some people are arguing about MTL yeild now.
MTL is exceeding Skylake's yield.
This means MTL has attained roughly perfect yield.
This is why Intel will be able to reach PRQ such big tile products as GNR/SRF soon.

See I think the slide is misleading. With the Meteor tile being just cores, they could in theory be able to salvage nearly all chips (down to 1+0 if need be) even with crap defect rates. Now will they sell 1+0+2 Meteor, we shall see, but it wouldn't surprise me that the reason Meteor was 'delayed' was because they needed more time to get enough fully enabled dies to even do a paper launch.

The other products mentioned on the slide have a lot of the die... that if there was a defect there, they would have to toss it in the trash can.

 

[Geddagod]

Intel's 4 is more like TSMC N5 and Intel's 3 is N4?

Intel 4 HP density is the same as TSMC 3nm. If Intel 3 adds HD libs, then I expect iso fin count for Intel 3 to be closer to TSMC's 3nm density than TSMC 4nm. Closer to a Pseudo 3nm node rather than a better 4nm node.
SRAM density might get closer to N5, if they improve that. If they don't, it's marginally worse than N5.
Both not having HD or lower SRAM density aren't that big deals for Intel due to design choices, but could be an issue for external customers.