AMD has unveiled its consideration of chiplet designs for its Ryzen APU family for laptops, but it faces significant challenges in terms of cost and power efficiency. The concept of chiplet-based configurations is gaining traction in the industry. A chiplet is essentially a combination of different chips integrated into a single package, interconnected to facilitate the concept of “process shrinking.” Multiple chiplets with the same or different core IPs can be mixed and matched to deliver optimal performance for a product segment.
These chiplet designs are challenging the traditional monolithic configuration that has been dominant in the industry for years. While monolithic designs are gradually shrinking, especially in the high-end segment, AMD still sees value in using monolithic chips for its mainstream laptop segment, particularly within its Ryzen APU series.
AMD has acknowledged its potential interest in adopting a chiplet-based approach for mainstream Ryzen APUs. However, the company remains cautious due to the limitations associated with this design. While chiplet configurations offer various advantages, such as the ability to shrink individual nodes, target specific workloads, and reduce costs, they struggle to maintain power efficiency. This concern was emphasized by David Afee, Corporate VP and General Manager of the Client Channel Business at AMD, during a Q&A session in South Korea. He believes that, at present, transitioning to chiplet designs for power-efficient chips is not the most viable option.
When asked about the absence of chiplet architecture in the laptop market, Afee explained that AMD considers both monolithic and chiplet structures for both desktops and laptops. However, introducing chiplets to laptops is challenging due to power consumption constraints. There is a power penalty associated with chiplet designs, and it will only be considered when it’s deemed worthwhile.
Furthermore, cost-effectiveness is another critical factor in the decision-making process. For entry-level and mainstream applications, chiplets may not provide the same cost-effectiveness as monolithic designs. Nevertheless, AMD has adopted chiplets for its high-end Dragon Range CPUs in enthusiast-grade laptops, where power and cost considerations are less of a concern.
It’s anticipated that AMD will employ chiplet-based designs for its next-gen Strix Point (Halo) chips launching in the coming year. However, the full realization of chiplets in the mainstream laptop “Ryzen APU” segment is not expected until 2026-2027. Intel is also embracing the chiplet approach with its Meteor Lake and future CPUs, which feature a fully disaggregated chip design with multiple tiles hosting various core IPs and IO capabilities.
In conclusion, the implementation of power-efficient “chiplet-based” designs for Ryzen APUs is a work in progress. AMD is planning for it, but it will take some time before it becomes a standard feature in the mainstream laptop segment. AMD’s pioneering work with chiplets in both desktop and laptop platforms suggests exciting developments on the horizon for Ryzen APUs.
AMD Ryzen Mobility CPUs:
CPU FAMILY NAME | AMD KRACKAN POINT | AMD FIRE RANGE | AMD STRIX POINT HALO | AMD STRIX POINT | AMD HAWK POINT | AMD DRAGON RANGE | AMD PHOENIX | AMD REMBRANDT | AMD CEZANNE | AMD RENOIR | AMD PICASSO | AMD RAVEN RIDGE |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Family Branding | AMD Ryzen 9040 (H/U-Series) | AMD Ryzen 8055 (HX-Series) | AMD Ryzen 8050 (H-Series) | AMD Ryzen 8050 (H/U-Series) | AMD Ryzen 8040 (H/U-Series) | AMD Ryzen 7045 (HX-Series) | AMD Ryzen 7040 (H/U-Series) | AMD Ryzen 6000 AMD Ryzen 7035 | AMD Ryzen 5000 (H/U-Series) | AMD Ryzen 4000 (H/U-Series) | AMD Ryzen 3000 (H/U-Series) | AMD Ryzen 2000 (H/U-Series) |
Process Node | 4nm | 5nm | 4nm | 4nm | 4nm | 5nm | 4nm | 6nm | 7nm | 7nm | 12nm | 14nm |
CPU Core Architecture | Zen 5 | Zen 5D | Zen 5C | Zen 5D + Zen 5C | Zen 4 | Zen 4 | Zen 4 | Zen 3+ | Zen 3 | Zen 2 | Zen + | Zen 1 |
CPU Cores/Threads (Max) | TBD | 16/32 | 16/32 | 12/24 | 8/16 | 16/32 | 8/16 | 8/16 | 8/16 | 8/16 | 4/8 | 4/8 |
L2 Cache (Max) | TBD | TBD | TBD | TBD | 4 MB | 16 MB | 4 MB | 4 MB | 4 MB | 4 MB | 2 MB | 2 MB |
L3 Cache (Max) | TBD | TBD | 64 MB | 32 MB | 16 MB | 32 MB | 16 MB | 16 MB | 16 MB | 8 MB | 4 MB | 4 MB |
Max CPU Clocks | TBD | TBD | TBD | TBD | TBD | 5.4 GHz | 5.2 GHz | 5.0 GHz (Ryzen 9 6980HX) | 4.80 GHz (Ryzen 9 5980HX) | 4.3 GHz (Ryzen 9 4900HS) | 4.0 GHz (Ryzen 7 3750H) | 3.8 GHz (Ryzen 7 2800H) |
GPU Core Architecture | TBD | RDNA 3+ 4nm iGPU | RDNA 3+ 4nm iGPU | RDNA 3+ 4nm iGPU | RDNA 3 4nm iGPU | RDNA 2 6nm iGPU | RDNA 3 4nm iGPU | RDNA 2 6nm iGPU | Vega Enhanced 7nm | Vega Enhanced 7nm | Vega 14nm | Vega 14nm |
Max GPU Cores | TBD | 2 CUs (128 cores) | 40 CUs (2560 Cores) | 16 CUs (1024 Cores) | 12 CUs (786 cores) | 2 CUs (128 cores) | 12 CUs (786 cores) | 12 CUs (786 cores) | 8 CUs (512 cores) | 8 CUs (512 cores) | 10 CUs (640 Cores) | 11 CUs (704 cores) |
Max GPU Clocks | TBD | TBD | TBD | TBD | TBD | 2200 MHz | 2800 MHz | 2400 MHz | 2100 MHz | 1750 MHz | 1400 MHz | 1300 MHz |
TDP (cTDP Down/Up) | 15W-45W (65W cTDP) | 55W-75W (65W cTDP) | 25-1250W | 15W-45W (65W cTDP) | 15W-45W (65W cTDP) | 55W-75W (65W cTDP) | 15W-45W (65W cTDP) | 15W-55W (65W cTDP) | 15W -54W(54W cTDP) | 15W-45W (65W cTDP) | 12-35W (35W cTDP) | 35W-45W (65W cTDP) |
Launch | 2025? | 2H 2024? | 2H 2024? | 2H 2024? | Q1 2024? | Q1 2023 | Q2 2023 | Q1 2022 | Q1 2021 | Q2 2020 | Q1 2019 | Q4 2018 |