NVIDIA GeForce RTX 40 Series Reportedly Has Node Advantage over AMD RDNA 3 GPUs with TSMC 4N

Tsing

The FPS Review
Staff member
Joined
May 6, 2019
Messages
12,595
Points
113
NVIDIA's next generation of GeForce GPUs is reportedly being manufactured on a more advanced process than what AMD is using for its RNDA 3 products. This is according to new rumors shared by Moore's Law Is Dead, who has claimed in a new tweet that green team's Lovelace-based GPUs are being built on TSMC's 4N, an optimized version of the foundry's 5 nm process node custom-designed for NVIDIA with optimizations for power efficiency, performance, and more.

Go to post
 
Well, this would be a first. Can't remember the last time nvidia had the manufacturing process advantage... if ever.
 
Can't remember a time Nvidia cared about power efficiency?
 
Stoly said:
Well, this would be a first. Can't remember the last time nvidia had the manufacturing process advantage... if ever.
Click to expand...
I'm not sure that's true. However, NVIDIA has had more power efficient architectures even when AMD had a process advantage. The 20 series was way more efficient at 12nm than AMD's were at the time.
 
Dan_D said:
I'm not sure that's true. However, NVIDIA has had more power efficient architectures even when AMD had a process advantage. The 20 series was way more efficient at 12nm than AMD's were at the time.
Click to expand...
I think it is true, at least since Kepler

From what I can remember, AMD always pushed for the newer/smaller process. Had it given an efficiency edge is another issue; case in point Pascal.
 
You must log in or register to reply here.
Become a Patron!
Back
Top