NVIDIA GeForce RTX 4070 GDDR6 vs NVIDIA GeForce RTX 5090
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 4070 GDDR6 and 32GB VRAM GeForce RTX 5090 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4070 GDDR6 's Advantages
Boost Clock has increased by 3% (2475MHz vs 2407MHz)
Lower TDP (200W vs 575W)
NVIDIA GeForce RTX 5090 's Advantages
More VRAM (32GB vs 12GB)
Larger VRAM bandwidth (1.79TB/s vs 480.0GB/s)
15872 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce RTX 4070 GDDR6
29.15 TFLOPS
GeForce RTX 5090
+259%
104.8 TFLOPS
Graphics Card
Aug 2024
Release Date
Jan 2025
GeForce 40
Generation
GeForce 50
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
1920 MHz
Base Clock
2017 MHz
2475 MHz
Boost Clock
2407 MHz
2500 MHz
Memory Clock
1750 MHz
Memory
12GB
Memory Size
32GB
GDDR6
Memory Type
GDDR7
192bit
Memory Bus
512bit
480.0GB/s
Bandwidth
1.79TB/s
Render Config
-
-
-
46
SM Count
170
5888
Shading Units
21760
184
TMUs
680
64
ROPs
176
184
Tensor Cores
680
46
RT Cores
170
128 KB (per SM)
L1 Cache
128 KB (per SM)
36 MB
L2 Cache
96 MB
-
-
-
Theoretical Performance
158.4 GPixel/s
Pixel Rate
423.6 GPixel/s
455.4 GTexel/s
Texture Rate
1637 GTexel/s
29.15 TFLOPS
FP16 (half)
104.8 TFLOPS
29.15 TFLOPS
FP32 (float)
104.8 TFLOPS
455.4 GFLOPS
FP64 (double)
1.637 TFLOPS
Board Design
200W
TDP
575W
550 W
Suggested PSU
950 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
1x HDMI 2.1b
3x DisplayPort 2.1b
1x 16-pin
Power Connectors
1x 16-pin
Graphics Processor
AD104
GPU Name
GB202
AD104-2XX-A1
GPU Variant
GB202-300-A1
Ada Lovelace
Architecture
Blackwell 2.0
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
35.8 billion
Transistors
92.2 billion
294 mm²
Die Size
750 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.4
8.9
CUDA
12.0
6.8
Shader Model
6.8
