In a report published by TrendForcewe’re getting our second confirmation that Intel’s tGPUs for 14th generation Meteor Lake processors, which are expected to use TSMC’s 3nm process node, have been delayed until 2023. the end
Intel’s 3nm tGPU mass production delays hamper TSMC, 14th-gen Meteor Lake delayed until 2023 the end?
Intel was previously rumored to be pushing its 14th generation Meteor Lake processors to 2023. end, and it looks like we’re getting similar reports from Trendforce’s sources close to Taiwanese chipmaker TSMC. Intel reportedly originally planned to mass-produce tGPUs (tileto GPUs) in its 14th-generation Meteor Lake processors starting in the second half of 2022. side It was later pushed back to 2023. first half due to product design and process. verification issues, but we can now consider an even longer delay.
According to TrendForce research, Intel plans to outsource its tGPU chipset to Meteor Lake for TSMC production. Mass production of this product was initially planned for 2H22, but was later delayed to 1H23 due to product design and process verification issues. Recently, the mass production schedule of the product has been pushed back again to 2023 due to some reasons. by the end of the year, almost completely eliminating the 3nm production capacity originally ordered in 2023, leaving only a small amount of wafers for engineering verification.
However, the development status of Intel’s own Intel 4 process and the related outsourcing situation are still important potential growth factors for TSMC. If Intel 4 doesn’t go into mass production as planned, Intel could hand over its computing chips to TSMC, which would be a big boost to growth in 2024. However, if Intel’s process goes smoothly, the company may decide to manufacture related products on its own and go out of business. TSMC orders.
TSMC may not begin mass production of Intel’s 3nm tGPUs until 2023, according to recent data. the end A Digitimes report last month made similar points. it was said that Intel CEO Pat Gelsinger would visiting TSMC himself fix its outsourcing plan with a Taiwanese chipmaker. Since Intel has been a major 3nm customer, a delay could hurt and slow down TSMC. While there are other major 3nm vendors such as Apple, AMD and Qualcomm, their products are not expected to enter production until 2024.
It also points out that Intel’s own processor node, Intel 4, may have some issues, and if it is unable to create a decent volume of Meteor Lake processors, Chipzilla may also consider other IPs, such as Compte Tile , transfer to TSMC. This could help TSMC boost revenue even more, but based on Intel’s latest claims, their in-house process node seems to be making progress and showing good results, but we’ll see as we enter 2023.
Building on Alder Lake’s leadership position in the second half of this year with Raptor Lake and Meteor Lake in 2023, we are demonstrating how our innovative design solutions can drive leadership performance even before we recreate best-in-class transistor technology.
And in 2023 we will present our first forked processor based on Intel 4, Meteor Lake, which is showing good health both in our labs and in our customers’ labs.
Yes. Maybe because we’re just quickly running through five nodes, Intel 7 finished, the supply, we said five nodes, four years, Intel 7, 35 million units, you can break one out. I’m sure our competitors ruined it, they did. Intel 4, that’s right, we said Meteor Lake looks good right now.
Intel CEO Pat Gelsinger (Q2 2022 Earnings Call)
The tGPU will be one of the key components of the 14th generation Meteor Lake processors with Arc graphics architecture to bring Intel on par with AMD and Apple’s GPU solutions. It’s just a rumor for now, though recently leaked platform information Meteor Lake processors are expected to ship to users in the second half of 2023, to confirm. side
Intel Mobility Processor Series:
| CPU family | Meteor Lake | Raptor Lake | Alksni lake |
|---|---|---|---|
| A process node | Intel 4 ‘7nm EUV’ | Intel 7 ’10nm ESF’ | Intel 7 ’10nm ESF’ |
| CPU architecture | Hybrid (Tri-core) | Hybrid (Dual Core) | Hybrid (Dual Core) |
| P-Core architecture | Redwood Cove | Raptor Cove | Golden Bay |
| E-Core architecture | Crestmont | Gracemont | Gracemont |
| Top configuration | 6+8 (H series) | 6+8 (H series) | 6+8 (H series) |
| Max Core/Threads | 14/20 | 14/20 | 14/20 |
| The composition is planned | H/P/U series | H/P/U series | H/P/U series |
| GPU architecture | Xe2 Battlemage Xe-LPG | Iris Xe (12th generation) | Iris Xe (12th generation) |
| GPU execution units | 128 EU (1024 colors) | 96 EU (768 colors) | 96 EU (768 colors) |
| Memory support | DDR5-5600 LPDDR5-7400 LPDDR5X – 7400+ |
DDR5-5200 LPDDR5-5200 LPDDR5-6400 |
DDR5-4800 LPDDR5-5200 LPDDR5X-4267 |
| Memory capacity (max.) | 96 GB | 64 GB | 64 GB |
| Thunderbolt 4 ports | 4 | 2 | 2 |
| WiFi available | Wi-Fi 6E | Wi-Fi 6E | Wi-Fi 6E |
| TDP | 15-45W | 15-45W | 15-45W |
| Run | in 2023 2 h | in 2023 1 H | in 2022 1 H |