Intel’s tGPU for Meteor Lake CPUs using TSMC’s 3nm node delayed mass production until late 2023

In a report published by TrendForce, we get our second confirmation that Intel’s tGPU for the 14th Gen Meteor Lake CPUs, which is expected to use TSMC’s 3nm process node, has been delayed for mass production until the end. of 2023.

Intel’s 3nm tGPU mass production delay hinders TSMC, 14th generation Meteor Lake postponed to end of 2023?

It was previously rumored that Intel could push its 14th generation Meteor Lake CPUs until the end of 2023, and it looks like we’re getting similar reports from Trendforce sources close to Taiwanese chip maker TSMC. It is reported that Intel initially planned to have mass production of the tGPU (tiled GPU) on the 14th generation Meteor Lake CPUs starting in the second half of 2022. It was later postponed to the first half of 2023 due to the design and process. product verification problems, but now we may see further delay.

According to TrendForce research, Intel plans to outsource the tGPU chipset in Meteor Lake to TSMC for manufacturing. Mass production of this product was initially planned for 2H22, but was later postponed to 1H23 due to product design and process verification issues. Recently, the mass production schedule of the product was again postponed to the end of 2023 for some reason, almost completely canceling the 3nm production capacity originally booked in 2023 with only a marginal amount of wafer input remaining for verification. engineering.

However, the development status of Intel’s Intel 4 process and its outsourcing situation are still important potential growth drivers for TSMC. If Intel 4 fails to mass produce as planned, Intel could outsource its processing panels to TSMC, driving growth strongly in 2024. However, if the Intel process develops smoothly, the possibility remains that the company will choose to produce related products and cancel TSMC’s orders.

via TrendForce

According to the latest data, TSMC may not begin mass production on Intel’s 3nm tGPU until late 2023. A report from Digitimes last month had similar things to say. Intel CEO Pat Gelsinger was rumored to personally go to TSMC to correct its outsourcing plan with the Taiwanese chip maker. As Intel was a major 3nm customer, a delay could potentially damage and slow down TSMC. While there are other major 3nm vendors like Apple, AMD, and Qualcomm, their products aren’t expected to start production until 2024.

It is also highlighted that Intel’s “Intel 4” process node may have some problems and if it fails to produce any viable volume for Meteor Lake CPUs, Chipzilla may reconsider outsourcing other IPs such as Compte Tile to TSMC as well. This could help TSMC further boost revenue, but based on Intel’s recent claims, it looks like progress has been made on the internal process node and is showing good results, but we’ll see when we enter 2023.

We are building on Alder Lake’s leadership with Raptor Lake in the second half of this year and Meteor Lake in 2023, exemplifying how our innovative design decisions can drive leadership performance even before reestablishing the best transistor technology.

And in 2023, we will deliver our first Intel 4-based unbundled CPU, Meteor Lake, which is showing good health in our and our customer’s labs.

Yes. Maybe as we quickly go through the five nodes, Intel 7 done, volume shipments, we said five nodes, four years, Intel 7, 35 million units, you can tear one apart. I’m sure our competitors have taken it apart, done. Intel 4, right, we said is, hey, Meteor Lake, at this point it looks good.

Intel CEO Pat Gelsinger (Q2 2022 Earnings Call)

The tGPU will be one of the core components of the 14th Generation Meteor Lake CPUs, with Arc graphics architecture to bring Intel on par with AMD and Apple’s integrated GPU solutions This is just a rumor for now, but the details of the platform have leaked recently for Meteor Lake CPUs confirm that the launch on the consumer market is expected in the second half of 2023.

Intel Mobility CPU Range:

CPU family Meteor lake Rapace Lake Alder lake
Process node Intel 4 ‘7nm EUV’ Intel 7 ’10nm ESF’ Intel 7 ’10nm ESF’
CPU architecture Hybrid (triple core) Hybrid (Dual Core) Hybrid (Dual Core)
P-Core architecture Redwood Bay Bay of Raptors Golden Bay
E-Core Architecture Crestmont Gracemont Gracemont
Superior configuration 6 + 8 (Series H) 6 + 8 (Series H) 6 + 8 (Series H)
Maximum number of cores / threads 14/20 14/20 14/20
Scheduled training H / P / U Series H / P / U Series H / P / U Series
GPU architecture Xe2 ‘Xe-GPL’ Warrior Mage Iris Xe (Jan 12) Iris Xe (Jan 12)
GPU execution unit 128 UE (1024 cores) 96 UE (768 cores) 96 UE (768 cores)
Memory support DDR5-5600
LPDDR5-7400
LPDDR5X – 7400+
DDR5-5200
LPDDR5-5200
LPDDR5-6400
DDR5-4800
LPDDR5-5200
LPDDR5X-4267
Memory capacity (maximum) 96 GB 64 GB 64 GB
Thunderbolt 4 ports 4 2 2
Wi-Fi capability Wi-Fi 6E Wi-Fi 6E Wi-Fi 6E
TDP 15-45W 15-45W 15-45W
Launch 2H 2023 1H 2023 1H 2022

Leave a Reply

%d bloggers like this: