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