As announced last week by TSMC, later this year the company is set to start high-volume manufacturing on its N3P fabrication process, and this will be the company's most advanced node for a while. Next year things will get a bit more interesting as TSMC will have two process technologies that could actually compete against each other when they enter high-volume manufacturing (HVM) in the second half of 2025.

Advertised PPA Improvements of New Process Technologies
Data announced during conference calls, events, press briefings and press releases
Compiled
by
AnandTech
TSMC
N3
vs
N5
N3E
vs
N5
N3P
vs
N3E
N3X
vs
N3P
N2
vs
N3E
N2P
vs
N3E
N2P
vs
N2
A16
vs
N2P
Power -25%
-30%
-34% -5%
-10%
-7%*** -25%
-30%
-30%
-40%
-5%
-10%
-15%
-20%
Performance +10%
+15%
+18% +5% +5%
Fmax @1.2V**
+10%
+15%
+15%
+20%
+5
+10%
+8%
+10%
Density* ? 1.3x 1.04x 1.10x*** 1.15x 1.15x ? 1.07x
1.10x
HVM Q4
2022
Q4
2023
H2
2024
H2
2025
H2
2025
H2
2026
H2
2026
H2
2026

*Chip density published by TSMC reflects 'mixed' chip density consisting of 50% logic, 30% SRAM, and 20% analog.
**At the same area. 
***At the same speed.

The production nodes are N3X (3nm-class, extreme performance-focused) as well as N2 (2nm-class). TSMC says that when compared to N3P, chips made on N3X can either lower power consumption by 7% at the same frequency by lowering Vdd from 1.0V to 0.9V, increase performance by 5% at the same area, or increase transistor density by around 10% at the same frequency. Meanwhile, the key advantage of N3X compared to predecessors is its maximum voltage of 1.2V, which is important for ultra-high-performance applications, such as desktop or datacenter GPUs.

TSMC's N2 will be TSMC's first production node to use gate-all-around (GAA) nanosheet transistors and this will significantly enhance its performance, power, and area (PPA) characteristics. When compared to N3E, semiconductors produced on N3 can cut their power consumption by 25% - 30% (at the same transistor count and frequency), increase their performance by 10% - 15% (at the same transistor count and power), and increase transistor density by 15% (at the same speed and power). 

While N2 will certainly be TSMC's undisputed champ when it comes to power consumption and transistor density, N3X could possibly challenge it when it comes to performance, especially at high voltages. For many customers N3X will also have a benefit of using proven FinFET transistors, so N2 will not be automatically the best of TSMC's nodes in the second half of 2025.

2026: N2P and A16

In the following year TSMC will again offer two nodes that are set to target generally similar smartphone and high-performance computing applications: N2P (performance-enhanced 2nm-class) and A16 (1.6nm-class with backside power delivery).

N2P is expected to deliver a 5% - 10% lower power (at the same speed and transistor count) or a 5% - 10% higher performance (at the same power and transistor count) compared to the original N2. Meanwhile, A16 is set to offer an up to 20% lower power (at the same speed and transistors), up to 10% higher performance (at the same power and transistors), and up to 10% higher transistor density compared to N2P. 

Keeping in mind that A16 features enhanced backside power delivery network, it will likely be the node of choice for performance-minded chip designers. But of course, it will be more expensive to use A16 because of the backside power delivery, which requires additional process steps.

Source: TSMC

Comments Locked

47 Comments

View All Comments

  • Terry_Craig - Wednesday, May 22, 2024 - link

    Intel and Samsung are so done. *Just like consumers with no competition for TSMC.
  • meacupla - Wednesday, May 22, 2024 - link

    Samsung still has DRAM, NAND, and HBM they can count on.
    There are also rumors that AMD will use Samsung for their low end Ryzen and Radeon chips. TSMC has limited capacity being bought by Apple and Nvidia, so it would make sense to buy from cheaper Samsung. Kind of like what Nvidia did with RTX 3000 series.

    Intel on tthe otoher hand... all they have is brand name recognition momentum carrying them.
  • shabby - Thursday, May 23, 2024 - link

    Intel has 10nm...
  • Turbofrog - Thursday, May 23, 2024 - link

    This is a super weird take given that Intel will be delivering 2nm to OEMs at the end of 2024, potentially 6 months before TSMC.
  • Terry_Craig - Thursday, May 23, 2024 - link

    Point 1, "2nm" according to who? How does this compare to the competition in scale, price, and yield?
    Point 2, when did Intel say something in the last few years and deliver?
  • mattbe - Thursday, May 23, 2024 - link

    Actually they have been getting much closer in delivering what they promised. Intel 2 Lunar Lake will be here in the second half. Considering we are already in May it is very unlikely that gets delayed. Do you even follow the news or do you only resort to add uninformed snarky remarks?
  • Dante Verizon - Friday, May 24, 2024 - link

    News = Useless announcements/ads from the company that continually lies and fails to deliver for half a decade, if there's no product on the market, there's nothing to talk about*

    do you know that intel can't scale Meteor Lake volume because of yields and packing problems ? Intel has all the problems in the world falling on its head.
  • my_wing - Thursday, May 23, 2024 - link

    TSMC is done, Intel King of the world in 2024 onward, TSMC clearly on the tail.

    Not competition for TSMC for the next 10 years ha ha.

    TSMC Die in water
  • Terry_Craig - Monday, June 10, 2024 - link

    Here's a cookie, Pat's lackey.
  • sharath.naik - Friday, May 24, 2024 - link

    You are mistaken, TMSC is simply a factory floor, the machines are built by ASML. Intel buys from ASML too, for what ever reason they delayed buying the EUV machines from them. So Intel will catch up in a couple of years. So it comes down to if intel's chip packaging can make the difference.
    In my opinion starting with lunar lake, Intel will retake the lead from AMD on mobile processors.

Log in

Don't have an account? Sign up now