GeForce RTX 4070 Mobile vs GeForce RTX 2070 Mobile Refresh
We compared two Mobile platform GPUs: 8GB VRAM GeForce RTX 4070 Mobile and 8GB VRAM GeForce RTX 2070 Mobile Refresh to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GeForce RTX 4070 Mobile 's Advantages
Released 2 years and 10 months late
Boost Clock has increased by 16% (1695MHz vs 1455MHz)
2304 additional rendering cores
GeForce RTX 2070 Mobile Refresh 's Advantages
Larger VRAM bandwidth (352.0GB/s vs 256.0GB/s)
Score
Benchmark
FP32 (float)
GeForce RTX 4070 Mobile
+132%
15620
GeForce RTX 2070 Mobile Refresh
6705
3DMark Time Spy
GeForce RTX 4070 Mobile
+62%
12501
GeForce RTX 2070 Mobile Refresh
7705
3DMark Time Spy Extreme
GeForce RTX 4070 Mobile
+56%
5699
GeForce RTX 2070 Mobile Refresh
3634
3DMark Speed Way
GeForce RTX 4070 Mobile
+85%
2924
GeForce RTX 2070 Mobile Refresh
1574
Graphics Card
Jan 2023
Release Date
Mar 2020
GeForce 40 Mobile
Generation
GeForce 20 Mobile
Mobile
Type
Mobile
PCIe 4.0 x8
Bus Interface
PCIe 3.0 x16
Clock Speeds
1395 MHz
Base Clock
1260 MHz
1695 MHz
Boost Clock
1455 MHz
2000 MHz
Memory Clock
1375 MHz
Memory
8GB
Memory Size
8GB
GDDR6
Memory Type
GDDR6
128bit
Memory Bus
256bit
256.0GB/s
Bandwidth
352.0GB/s
Render Config
36
SM Count
36
4608
Shading Units
2304
144
TMUs
144
48
ROPs
64
144
Tensor Cores
288
36
RT Cores
36
128 KB (per SM)
L1 Cache
64 KB (per SM)
32 MB
L2 Cache
4 MB
Theoretical Performance
81.36 GPixel/s
Pixel Rate
93.12 GPixel/s
244.1 GTexel/s
Texture Rate
209.5 GTexel/s
15.62 TFLOPS
FP16 (half)
13.41 TFLOPS
15.62 TFLOPS
FP32 (float)
6.705 TFLOPS
244.1 GFLOPS
FP64 (double)
209.5 GFLOPS
Board Design
115W
TDP
115W
Portable Device Dependent
Outputs
No outputs
None
Power Connectors
None
Graphics Processor
AD106
GPU Name
TU106B
GN21-X6
GPU Variant
-
Ada Lovelace
Architecture
Turing
TSMC
Foundry
TSMC
5 nm
Process Size
12 nm
22.9 billion
Transistors
10.8 billion
188 mm²
Die Size
445 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
1.3
8.9
CUDA
7.5
6.8
Shader Model
6.6
