NVIDIA Rubin GPU vs AMD Radeon 8040S

GPU Specs Compare

G

NVIDIA Rubin GPU

NVIDIA 2026-01
G

AMD Radeon 8040S

AMD 2025-01

Graphics Card

Release Date
Jan 2026
Jan 2025
Generation
Server Rubin
Navi Mobile
Type
Desktop
Integrated
Bus Interface
PCIe 6.0 x16
PCIe 5.0 x16

Clock Speeds

Base Clock
700 MHz
1295 MHz
Boost Clock
2267 MHz
2800 MHz
Memory Clock
2695 MHz
System Shared

Memory

Memory Size
288GB
System Shared
Memory Type
HBM4
System Shared
Memory Bus
16384bit
System Shared
Bandwidth
22.1TB/s
System Dependent

Render Config

SM Count
224
-
Shading Units
28672
1024
TMUs
896
64
ROPs
24
32
Tensor Cores
896
-
L1 Cache
256 KB (per SM)
-
L2 Cache
128 MB
8 MB
Compute Units
-
16
RT Cores
-
16
L3 Cache
-
64 MB

Theoretical Performance

Pixel Rate
54.41 GPixel/s
89.60 GPixel/s
Texture Rate
2031.2 GTexel/s
179.2 GTexel/s
FP16 (half)
260.0 TFLOPS
11.47 TFLOPS
FP32 (float)
130.0 TFLOPS
5.734 TFLOPS
FP64 (double)
32.50 TFLOPS
179.2 GFLOPS

Graphics Processor

GPU Name
GR100
Strix Halo
Architecture
Rubin
RDNA 3.5
Foundry
TSMC
TSMC
Process Size
3 nm
4 nm
Transistors
336 billion
Unknown
Die Size
1456 mm²
308 mm²

Board Design

TDP
2300W
55W
Suggested PSU
2700 W
-
Outputs
No outputs
Portable Device Dependent
Power Connectors
-
None

Graphics Features

DirectX
N/A
12 Ultimate (12_2)
OpenGL
N/A
4.6
OpenCL
3.0
2.1
Vulkan
N/A
1.3
CUDA
10.7
-
Shader Model
N/A
6.8