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

Page 228 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DisEnchantment

Golden Member
Mar 3, 2017
1,777
6,791
136
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.

1587737990547.png
N7 performance is more or less understood.
1587739093721.png

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

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.


1587739615344.png

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
 
Last edited:

511

Diamond Member
Jul 12, 2024
4,525
4,146
106
It's pretty close.
5.72 million GNR+SRF units in, 4.35 million Turin units (Q2 2025). AMD ships more 4/3nm using parts than Intel does in server.
Well it's not 5.72 Million lol that's total Intel server Chips it's around 572K vs 435K for GNR/SRF vs Turin.
If you are counting Genoa as well count SPR/EMR as well. It's been over 2-3 years for their launch.
 
  • Like
Reactions: DKR

Geddagod

Golden Member
Dec 28, 2021
1,524
1,620
106
Well it's not 5.72 Million lol that's total Intel server Chips it's around 572K vs 435K for GNR/SRF vs Turin.
If you are counting Genoa as well count SPR/EMR as well. It's been over 2-3 years for their launch.
Lol that's my bad. I genuinely thought they shipped way more than that.
Counting SPR and EMR doesn't make sense when they don't use 4nm/5nm nodes though. Intel still isn't shipping all that advanced node volume.
 

511

Diamond Member
Jul 12, 2024
4,525
4,146
106
Counting SPR and EMR doesn't make sense when they don't use 4nm/5nm nodes though. Intel still isn't shipping all that advanced node volume.
You know that Intel 3 has been in Hvm For only like 1.5 year it will take some time to reach more Volume on I3. Anyway my comparison was with AMD Turin cause both of them launched at nearly the same time.
 
  • Like
Reactions: DKR

511

Diamond Member
Jul 12, 2024
4,525
4,146
106
Back in the day, Nodes were 2 years basically.
Lol that was when Intel execution was unmatched everyone was f**ing around and Intel was here is how you do it. You can't generate ROI on modern Nodes in 2 years now even TSMC can't.
 

Joe NYC

Diamond Member
Jun 26, 2021
3,650
5,189
136
It's not that difficult tbh Intel ships way more CPU than AMD just the ASP is less and most of them are SPR/EMR/ICL.

It's only the SFR + GNR numbers that I am doubting. Intel management went out of their way to downplay interest in SFR the speed of their ramp of GNR.
 

511

Diamond Member
Jul 12, 2024
4,525
4,146
106
It's only the SFR + GNR numbers that I am doubting. Intel management went out of their way to downplay interest in SFR the speed of their ramp of GNR.
There is a difference from point of view they must have expected more than what we are seeing.
For you it's more and for them it's less.
If Intel only ships 1 Million Client chips on particular arch it might be Lower than their expectations.
 
Jul 27, 2020
28,034
19,132
146
Why would they target HPC when the volume play for any foundry is almost always mobile? Were they thinking margins or . . . ?
Maybe they wanted to win some super lucrative supercomputer contract? Maybe the "brains" at Intel think that a win like that would impress the world that Intel is back?
 

Joe NYC

Diamond Member
Jun 26, 2021
3,650
5,189
136
Maybe they wanted to win some super lucrative supercomputer contract? Maybe the "brains" at Intel think that a win like that would impress the world that Intel is back?
Intel officially exited supercomputer business after the Aurora fiasco.
 

Joe NYC

Diamond Member
Jun 26, 2021
3,650
5,189
136
I don't think they did cause there are few Supercomputer announced after Aurora.

I think the exit was for Intel as a general contractor (of the whole project). In any subsequent ones, Intel was likely only a sub-contractor, parts supplier.
 

fastandfurious6

Senior member
Jun 1, 2024
762
957
96
There is not a single Zen 4/5 CPU reaching 5.3 GHz on Mobile except the ones based on Desktop die or the halo as well.

are we in 2005 still measuring computer quality by Ghz? Why is it even relevant anymore lol

Halo trumps everything
 

511

Diamond Member
Jul 12, 2024
4,525
4,146
106

fastandfurious6

Senior member
Jun 1, 2024
762
957
96
@fastandfurious6 remember this this aged like milk


flash-rise based on viral/FOMOeconomics and wrong assumptions, won't last long

non-tech market gives it a positive spin "nvidia sees something in intel, partnership, good products etc"

but in reality nvidia is just signing a profitable deal with a very weakened intel. intel needs money/investors/nvidia, not the other way
 

511

Diamond Member
Jul 12, 2024
4,525
4,146
106
flash-rise based on viral/FOMOeconomics and wrong assumptions, won't last long

non-tech market gives it a positive spin "nvidia sees something in intel, partnership, good products etc"

but in reality nvidia is just signing a profitable deal with a very weakened intel. intel needs money/investors/nvidia, not the other way
Nice way to cover up you are wrong
 

DKR

Junior Member
Nov 19, 2024
19
32
51
It's not that difficult tbh Intel ships way more CPU than AMD just the ASP is less and most of them are SPR/EMR/ICL.
Those are not estimates from Morgan Stanley. The data is from UBS research report which references Mercury Research data. Mercury Research tracks the CPU and GPU unit shipments.
 
  • Like
Reactions: 511