GeForce RTX 4080 vs GeForce RTX 5070
We compared two Desktop platform GPUs: 16GB VRAM GeForce RTX 4080 and 12GB VRAM GeForce RTX 5070 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GeForce RTX 4080 's Advantages
More VRAM (16GB vs 12GB)
Larger VRAM bandwidth (716.8GB/s vs 672.0GB/s)
3584 additional rendering cores
GeForce RTX 5070 's Advantages
Released 2 years and 6 months late
Boost Clock has increased by 0% (2512MHz vs 2505MHz)
Lower TDP (250W vs 320W)
Score
Benchmark
FP32 (float)
GeForce RTX 4080
+57%
48740
GeForce RTX 5070
30870
3DMark Time Spy
GeForce RTX 4080
+24%
28173
GeForce RTX 5070
22544
3DMark Time Spy Extreme
GeForce RTX 4080
+29%
14002
GeForce RTX 5070
10789
3DMark Speed Way
GeForce RTX 4080
+23%
7254
GeForce RTX 5070
5896
Graphics Card
Sep 2022
Release Date
Mar 2025
GeForce 40
Generation
GeForce 50
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
2205 MHz
Base Clock
2325 MHz
2505 MHz
Boost Clock
2512 MHz
1400 MHz
Memory Clock
1750 MHz
Memory
16GB
Memory Size
12GB
GDDR6X
Memory Type
GDDR7
256bit
Memory Bus
192bit
716.8GB/s
Bandwidth
672.0GB/s
Render Config
76
SM Count
48
9728
Shading Units
6144
304
TMUs
192
112
ROPs
80
304
Tensor Cores
192
76
RT Cores
48
128 KB (per SM)
L1 Cache
128 KB (per SM)
64 MB
L2 Cache
48 MB
Theoretical Performance
280.6 GPixel/s
Pixel Rate
201.0 GPixel/s
761.5 GTexel/s
Texture Rate
482.3 GTexel/s
48.74 TFLOPS
FP16 (half)
30.87 TFLOPS
48.74 TFLOPS
FP32 (float)
30.87 TFLOPS
761.5 GFLOPS
FP64 (double)
482.3 GFLOPS
Board Design
320W
TDP
250W
700 W
Suggested PSU
600 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
AD103
GPU Name
GB205
AD103-300-A1
GPU Variant
GB205-300-A1
Ada Lovelace
Architecture
Blackwell 2.0
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
45.9 billion
Transistors
31.1 billion
379 mm²
Die Size
263 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
10.1
6.7
Shader Model
6.8
