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

 

[adroc_thurston]

Pat's attempt at saving Intel failed, but it's in no way over yet. Everything is on fire, but the company can survive for years in this state, stumbling, hopefully towards salvation. Maybe Tan can do better than Pat.

Unified Core better be hail mary or they're toast.

 

[poke01]

Unified Core better be hail mary or they're toast.

what makes unified core good?

 

[DrMrLordX]

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

Nah, external customers will probably be looking at 18AP

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 certainly does look like 18a is, at best, a half-node step from 20a, and is still b0rked enough that nobody outside of Intel wants it.

I think Intel 18A/18AP and Intel 20A are different things.
Or rather, 18A is a 20A half node or a finished product...

Yeah it's unlikely that 18a is to 20a as Intel 7 is to 10ESF. Still it doesn't give much hope for Panther Lake.

 

[adroc_thurston]

what makes unified core good?

Well idk if it's gonna be good, but it's Atom guys heralding the new microarch stack for Intel.

 

[DZero]

what makes unified core good?

A Conroe Jump... again

 

[DavidC1]

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.

Ok, this explains why Clearwater Forest was delayed. It may not have been just packaging.

Another thing is this is in line with Exist50's very gloomy projections. He said they won't have leadership even with 14A.

what makes unified core good?

Stephen Robinson, the Chief architect of the E cores since at least Tremont is taking over the next gen core project.

I'm expecting the dead end P cores that started since Sandy Bridge to be abandoned and go with the E core route, which is: no uop cache, clustered decode, no SMT, no thoughtless expansions.

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

Intel did, and would wish to.

They are unappealing to work, that's why they leave Intel. The solution is to "fix" the mess they are having. Poaching at this point just wastes money.

If you look at Intel 3 vs 4 there are clear difference in metals and pitches and RC characteristics would be correct to call two different things as one?

Do you not remember a slide from Intel saying Intel 3 and 4 is the same thing?

I said when they downgraded 18A performance from 15% + 10% to 15% that it's basically rebranded 20A. If you claim X offers 10% extra performance, but suddenly you take that away, what else is there?

 

[DavidC1]

So they claimed "5 nodes in 4 years", which turned out to be:
Intel 7, 4, 3, 20A, and 18A.

Claims-
Intel 4: 18% gain with significant area
Intel 3: 18% gain with more libraries, HD cell availability and improvements in density for high performance cells
Intel 20A: RibbonFET and PowerVia with 15% perf
Intel 18A: 10% more perf

Reality-
Intel 4 did gain significant density, but broke off from the traditional 2x gains. It was more like 60%.
Intel 3 performed better, but density wise, the hinted at HP gains weren't there. And even if you went high density, it was only 10%
Intel 20A was "cancelled", which got replaced with 18A being 15% better. Density gains are only 30% with PowerVia.

So you can't believe what Intel says anymore, even at IEDM and Hot Chips. They either lie or completely miss their targets.

 

[Io Magnesso]

Ok, this explains why Clearwater Forest was delayed. It may not have been just packaging.

Another thing is this is in line with Exist50's very gloomy projections. He said they won't have leadership even with 14A.

Stephen Robinson, the Chief architect of the E cores since at least Tremont is taking over the next gen core project.

I'm expecting the dead end P cores that started since Sandy Bridge to be abandoned and go with the E core route, which is: no uop cache, clustered decode, no SMT, no thoughtless expansions.

Intel did, and would wish to.

They are unappealing to work, that's why they leave Intel. The solution is to "fix" the mess they are having. Poaching at this point just wastes money.

Do you not remember a slide from Intel saying Intel 3 and 4 is the same thing?

I said when they downgraded 18A performance from 15% + 10% to 15% that it's basically rebranded 20A. If you claim X offers 10% extra performance, but suddenly you take that away, what else is there?

Is EXIST50 the person who often said that ARC is dead?
Well, this is not the gloomy projections of EXIST50, Let's just say it's your gloomy projections

 

[Io Magnesso]

TSMC is God TSMC is an obligation

 

[Io Magnesso]

All companies other than TSMC will go bankrupt from now on.

 

[Io Magnesso]

Intel is a group of people with intellectual disabilities, so they are liars.

 

[LightningZ71]

Rapidus has VERY deep pockets backing it, the Japanese government with a vested interest, IBM deeply involved in the lithography and supposedly some notable faces from the industry involved. Japan views the success of Rapidus as a national security level initiative.

 

[Io Magnesso]

Rapidus has VERY deep pockets backing it, the Japanese government with a vested interest, IBM deeply involved in the lithography and supposedly some notable faces from the industry involved. Japan views the success of Rapidus as a national security level initiative.

Well, it will fail
If it fails, you can donate the manufacturing equipment to TSMC. You may be unexpectedly smart

 

[Hulk]

Goodbye Intel 18A. It was nice not knowing you.

So much thunder from Intel and as always it signifies absolutely nothing.

Somebody please put Intel out of its misery. This is a slow, agonizing, painful death. Just hit the bottom already.

Intel might axe the 18A process node for foundry customers, essentially leaving TSMC with no rival — Intel reportedly to focus on 14A (Updated)

TSMC's N2 and A16 may have no rivals

www.tomshardware.com

 

[511]

So they claimed "5 nodes in 4 years", which turned out to be:
Intel 7, 4, 3, 20A, and 18A.

Claims-
Intel 4: 18% gain with significant area
Intel 3: 18% gain with more libraries, HD cell availability and improvements in density for high performance cells
Intel 20A: RibbonFET and PowerVia with 15% perf
Intel 18A: 10% more perf

Reality-
Intel 4 did gain significant density, but broke off from the traditional 2x gains. It was more like 60%.
Intel 3 performed better, but density wise, the hinted at HP gains weren't there. And even if you went high density, it was only 10%
Intel 20A was "cancelled", which got replaced with 18A being 15% better. Density gains are only 30% with PowerVia.

So you can't believe what Intel says anymore, even at IEDM and Hot Chips. They either lie or completely miss their targets.

Well if you looks correctly they said 20A - 10-15% better perf than Intel 3 and 18A 5-10% they only achived their target at 1.1V.

It's 4N4Y btw not 5N4Y

Rapidus has VERY deep pockets backing it, the Japanese government with a vested interest, IBM deeply involved in the lithography and supposedly some notable faces from the industry involved. Japan views the success of Rapidus as a national security level initiative.

It's IBM so I don't have high hopes for them better would be to give TSMC/Intel the money.

 

[511]

Goodbye Intel 18A. It was nice not knowing you.

So much thunder from Intel and as always it signifies absolutely nothing.

Somebody please put Intel out of its misery. This is a slow, agonizing, painful death. Just hit the bottom already.

Intel might axe the 18A process node for foundry customers, essentially leaving TSMC with no rival — Intel reportedly to focus on 14A (Updated)

TSMC's N2 and A16 may have no rivals

www.tomshardware.com

This is just FUD being spread

https://twitter.com/x/status/1940484301671616923

 

[Joe NYC]

Goodbye Intel 18A. It was nice not knowing you.

So much thunder from Intel and as always it signifies absolutely nothing.

Somebody please put Intel out of its misery. This is a slow, agonizing, painful death. Just hit the bottom already.

Intel might axe the 18A process node for foundry customers, essentially leaving TSMC with no rival — Intel reportedly to focus on 14A (Updated)

TSMC's N2 and A16 may have no rivals

www.tomshardware.com

"for foundry customers". Intel will still use it for its own processors.

We will see how aggressively Intel can shift from Meteor Lake, Lunar Lake to Panther Lake. Panther Lake has a late start...

 

[Io Magnesso]

This is just FUD being spread

https://twitter.com/x/status/1940484301671616923

You already have a contract, so you can't just ignore it and quit 18A.
I think that no matter how small the number is, it will be manufactured with Intel 18A.

 

[Io Magnesso]

Goodbye Intel 18A. It was nice not knowing you.

So much thunder from Intel and as always it signifies absolutely nothing.

Somebody please put Intel out of its misery. This is a slow, agonizing, painful death. Just hit the bottom already.

Intel might axe the 18A process node for foundry customers, essentially leaving TSMC with no rival — Intel reportedly to focus on 14A (Updated)

TSMC's N2 and A16 may have no rivals

www.tomshardware.com

Intel please die

 

[poke01]

Do we know who these customers are that Dr Ian mentioned for 18A-P?

Is it Amazon, Qualcomm, Microsoft, Nvidia or Google?

 

[Joe NYC]

Do we know who these customers are that Dr Ian mentioned for 18A-P?

Is it Amazon, Qualcomm, Microsoft, Nvidia or Google?

Apparently, some trivial orders from Microsoft and Amazon (AWS).

 

[511]

Apparently, some trivial orders from Microsoft and Amazon (AWS).

Yeah the volume is not great but it's what it is who is going to sign up with Intel Foundry these are their decade old partners especially MSFT.

I won't be surprised if these customers turns out to be partners.

 

[Saylick]

View attachment 126517

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

Intel 18A+++ lol which I guess is really more like 20A++++

 

[DZero]

Well, it will fail
If it fails, you can donate the manufacturing equipment to TSMC. You may be unexpectedly smart

And what if is succesful?

Now seeing this...

Goodbye Intel 18A. It was nice not knowing you.

So much thunder from Intel and as always it signifies absolutely nothing.

Somebody please put Intel out of its misery. This is a slow, agonizing, painful death. Just hit the bottom already.

Intel might axe the 18A process node for foundry customers, essentially leaving TSMC with no rival — Intel reportedly to focus on 14A (Updated)

TSMC's N2 and A16 may have no rivals

www.tomshardware.com

Seems that Intel is totally effed up.

Before this new I tought that Samsung and SMIC are really bad, but this makes them looks into a better light

 

[511]

And what if is succesful?

Now seeing this...


Seems that Intel is totally effed up.

Before this new I tought that Samsung and SMIC are really bad, but this makes them looks into a better light

Oh no you don't actually know how worse Samsung is in the node game rn 🤣