Newly released photos and specifications of Nvidia’s upcoming AD106 and AD107 graphics processors have been revealed. Designed for the company’s upcoming GeForce RTX 40-series solutions for desktops and laptops, it’s small but has the potential to deliver powerful performance.
this week Moore’s Law is dead (Youtube) has captured images of Nvidia’s upcoming AD106 and AD107 GPUs. Tech Power Up publish detailed specifications such as die size and transistor count. Nvidia’s AD106 and AD107 chips essentially flesh out the company’s Ada Lovelace family and complement his known AD102, AD103, and AD104 graphics processors.
| GPUs | RTX 4080 Laptop | RTX 4070 laptop | RTX 4060 laptop | RTX 4050 laptop |
|---|---|---|---|---|
| architecture | AD104 | AD106 | AD107 | AD107 |
| process technology | TSMC 4N | TSMC 4N | TSMC 4N | TSMC 4N |
| Transistor (billion) | 35.8 | ? | ? | ? |
| Die size (mm^2) | 294.5 | 190 | 146 | 146 |
| streaming multiprocessor | 60 | 36 | twenty four | 20 |
| GPU core (shader) | 7680 | 4608 | 3072 | 2560 |
| Tensor cores | 240 | 144 | 96 | 80 |
| Ray Tracing Core | 60 | 36 | twenty four | 20 |
| TMU | 240 | 144 | 96 | 80 |
| ROP | 80 | 48 | 32 | 32 |
| L2 Cache (MB) | 48 | 32 | 32 | 12 |
| memory interface (bits) | 192 | 128 | 128 | 96 |
| Memory speed (GT/s) | twenty one | 16 | 16 | 16 |
In terms of die size, all Ada Lovelace GPUs are smaller than their Ampere counterparts. Nvidia’s latest graphics processors are made with TSMC’s 4N (4nm-class) manufacturing technology, while previous generation chips used Samsung’s 8LPP (8nm-class node derived from 10nm-class technology). This is expected because So, on the one hand, a smaller die size reduces costs, which is essential. At the same time, however, it is becoming increasingly difficult for chip designers to narrow down all the necessary interfaces (such as memory and display output) while meeting the required performance goals.
TSMC charges more for chips made with 4N technology than Samsung charges for GPUs made using the 8LPP node (which is in line with industry trends). ), we can only speculate whether the smaller Ada Lovelace GPUs are cheaper to manufacture than the larger Ampere GPUs. Still, as yields rise, GeForce RTX 40-series processors will inevitably cost less.
When Nvidia announced its GeForce RTX 40-series laptop graphics processors earlier this year, the entry-level AD107 and mid-range AD106 mobile Ada Lovelace GPUs were only available in laptops starting at $999 and $1,500 respectively. said to be deaf. A $999 gaming notebook with his discrete GeForce RTX 40 GPU sounds tempting, but the question is what else such a machine offers and how prevalent such laptops will be. is.
