GeForce RTX 4070 SUPER and GeForce RTX 5090 - video card comparison
GeForce RTX 4070 SUPER
GeForce RTX 5090
Architecture
Ada Lovelace
Blackwell 2.0
GPU code name
AD104
GB202
Market segment
Desktop
Desktop
Release date
January 2024
2025 (recently)
Interface
PCIe 4.0 x16
PCIe 5.0 x16
Supplementary power connectors
1x 16-pin
1x 16-pin
Boost clock speed
2475 MHz
2520 MHz
Core clock speed
1980 MHz
-
Manufacturing process technology
5 nm
0 nm
Number of transistors
35,800 million
-
Pipelines / CUDA cores
7168
20480
Power consumption (TDP)
220 Watt
500 Watt
Texture fill rate
584.6
1,613
Memory type
GDDR6X
GDDR7
Maximum RAM amount
12 GB
28 GB
Memory bus width
192 Bit
448 Bit
Memory clock speed
21000 MHz
5 GB/s
Memory bandwidth
504.2 GB/s
280.0 GB/s
Shared memory
-
-
Display Connectors
1x HDMI 2.1, 3x DisplayPort 1.4a
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
+
+
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
CUDA
8.9
9.1
OpenCL
3.0
3.0
OpenGL
4.6
4.6
Shader Model
6.7
6.7
Vulkan
1.3
1.3