GeForce RTX 3090 and GeForce RTX 4070 AD103 - video card comparison
GeForce RTX 3090
GeForce RTX 4070 AD103
Architecture
Ampere (2020−2022)
Ada Lovelace
GPU code name
Ampere GA102
AD103
Market segment
Desktop
Desktop
Release date
September 2020
March 2024 (recently)
Interface
PCIe 4.0 x16
PCIe 4.0 x16
Supplementary power connectors
1x 12-pin
1x 16-pin
Boost clock speed
1700 MHz
2475 MHz
Core clock speed
1400 MHz
-
Manufacturing process technology
8 nm
5 nm
Number of transistors
28,300 million
45,900 million
Pipelines / CUDA cores
10496
5888
Power consumption (TDP)
350 Watt
200 Watt
Texture fill rate
556.0
455.4
Memory type
GDDR6X
GDDR6X
Maximum RAM amount
24 GB
12 GB
Memory bus width
384 Bit
192 Bit
Memory clock speed
19500 MHz
21 GB/s
Memory bandwidth
936.2 GB/s
504.2 GB/s
Shared memory
-
-
Display Connectors
1x HDMI, 3x DisplayPort
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
+
+
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
CUDA
8.5
8.9
OpenCL
2.0
3.0
OpenGL
4.6
4.6
Shader Model
6.5
6.7
Vulkan
1.2
1.3