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

Page 203 - 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
6,787
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:

marees

Golden Member
Apr 28, 2024
1,347
1,954
96

TSMC terminated nine employees for leaking 2-nanometer chip data to Tokyo Electron staff​

- Three were trial production staff; six involved in R&D, using mobile phones to transmit sensitive info.

- The breach highlights risks in semiconductor IP protection amid fierce global competition.

- TSMC’s swift action and enhanced security aim to safeguard its tech leadership in advanced chips.

https://www.ainvest.com/news/tsmc-fires-employees-2-nanometer-chip-data-leak-2508/
 

Josh128

Golden Member
Oct 14, 2022
1,106
1,671
106
I have to say it's quite surreal reading @SiliconFly's endless praises to Intel's processes on X/Twitter and then reading threads like this on Reddit:

Quiet quit for a PhD?? Sounds like bro needs to quit whining, get his a$$ to work, and actually do some work for the undoubtedly high salary Intel is most likely paying him. Does he want to help turn the company around or help drag it down is what he should be asking himself. If he hasnt been cut yet, his position has likely been deemed an important one.
 

511

Diamond Member
Jul 12, 2024
3,427
3,312
106
Quiet quit for a PhD?? Sounds like bro needs to quit whining, get his a$$ to work, and actually do some work for the undoubtedly high salary Intel is most likely paying him. Does he want to help turn the company around or help drag it down is what he should be asking himself. If he hasnt been cut yet, his position has likely been deemed an important one.
He is working for free so they can't cut him but he nothing compared to Meng this dude is insane he makes MLID Feel legit
 

Josh128

Golden Member
Oct 14, 2022
1,106
1,671
106
He is working for free so they can't cut him but he nothing compared to Meng this dude is insane he makes MLID Feel legit
I never read anything about him working for free in his post or the thread. Who is this Meng dude? Can you share some fun links?
 

jpiniero

Lifer
Oct 1, 2010
16,589
7,077
136
EDIT: Never Mind it's a parody account :D

Not wrong though.

In theory, having a Foundry that doesn't do leading edge but only semi-sorta edge could work. You could do Things That Don't Matter (ie: chipsets, IGP, etc) on that node while doing The Stuff That Does (ie: CPU tiles) at TSMC.

I don't think it would make enough to pay the bills tho. And Intel is definitely not the company to be able to pull that off.
 

Doug S

Diamond Member
Feb 8, 2020
3,380
5,960
136
TSMC has A16 with similar properties, right?

A16 has BSPDN but does it relax M0 pitch?

I've been wondering about that because yes you can relax M0 thanks to BSPDN but supposedly A16 is basically "N2 with BSPDN". I'll defer to the guys with actual semi/fab experience, but I'd think changing something as basic as M0 pitch (and the pitches of M1/M2 as well no doubt) would make calling it the same process ridiculous. It would have different design rules, different PDKs. I mean yes that's kinda obvious since the addition of BSPDN changes everything, but why tell people that A16 is the same N2 process with BSPDN added?

Since TSMC is apparently going to continue to dual track at least for the next couple generations offering the "same" process both with and without BSPDN can they relax M0? Or would trying to maintain the "same" process mean maintaining the same metal stack for the base process both with and without BSPDN?
 
  • Like
Reactions: coercitiv

LightningZ71

Platinum Member
Mar 10, 2017
2,386
3,033
136
Lest we forget, even if 18A was this spectacular node, Intel doesn't have anywhere near the capacity to feed the market, and would likely struggle to handle their own internal needs on top of any customer of significant size with even perfect yields on the process for a long time. They also don't have the mountain of third party foundry experience that TSMC has to accelerate the design and layout process.

Lest we forget, Intel is deliberately choosing to make chunks of some of their leading edge products on the "inferior" N2, despite supposedly sitting on possibly the "best foundry process to ever be released on the market."

Until Intel actually releases product on this node, and it proves to be competitive in the ways that actually matter, and it has viable enough yields to at least break even on costs, the best they are doing is Intel 3.

The sad part is that I WANT Intel to succeed! There NEEDS to be competition to TSMC besides Rapidus that will be volume constrained for a long time and whatever shadow of Samsung hangs around after their smattering of sudden layoffs this week.
 

511

Diamond Member
Jul 12, 2024
3,427
3,312
106
lol Pet Gelsinger. It got me too for about 5 seconds while I was thinking "Why tf would Pat be replying to this shill on Xitter?", then I saw it.
You are not the only one on Twitter 😂. There have been funnier incident regarding the parafoy account.
 

OneEng2

Senior member
Sep 19, 2022
742
994
106
Lest we forget, even if 18A was this spectacular node, Intel doesn't have anywhere near the capacity to feed the market, and would likely struggle to handle their own internal needs on top of any customer of significant size with even perfect yields on the process for a long time.
Great point, and a sad reality for Intel.
The sad part is that I WANT Intel to succeed! There NEEDS to be competition to TSMC besides Rapidus that will be volume constrained for a long time and whatever shadow of Samsung hangs around after their smattering of sudden layoffs this week.
Me too. I also don't want the US to be under threat of the loss of this technology should something go badly in Thailand.
 

RTX

Member
Nov 5, 2020
179
131
116
TSMC too stronk pls nerf
was 2nm actually leaked to tokyo/rapidus/ibm?

Only Core Ultra 3s at launch
 
  • Like
Reactions: techjunkie123