NVIDIA GeForce RTX 4090 Mobile vs NVIDIA GeForce RTX 3050 Ti Mobile
We compared two Mobile platform GPUs: 16GB VRAM GeForce RTX 4090 Mobile and 4GB VRAM GeForce RTX 3050 Ti Mobile to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4090 Mobile 's Advantages
Released 1 years and 8 months late
Boost Clock has increased by 64% (1695MHz vs 1035MHz)
More VRAM (16GB vs 4GB)
Larger VRAM bandwidth (576.0GB/s vs 192.0GB/s)
7168 additional rendering cores
NVIDIA GeForce RTX 3050 Ti Mobile 's Advantages
Lower TDP (75W vs 120W)
Score
Benchmark
FP32 (float)
GeForce RTX 4090 Mobile
+522%
32.98 TFLOPS
GeForce RTX 3050 Ti Mobile
5.299 TFLOPS
3DMark Time Spy
GeForce RTX 4090 Mobile
+300%
21498
GeForce RTX 3050 Ti Mobile
5369
3DMark Time Spy Extreme
GeForce RTX 4090 Mobile
+327%
10528
GeForce RTX 3050 Ti Mobile
2464
3DMark Speed Way
GeForce RTX 4090 Mobile
+1162%
5604
GeForce RTX 3050 Ti Mobile
444
Blender
GeForce RTX 4090 Mobile
+430%
7692
GeForce RTX 3050 Ti Mobile
1450
OctaneBench
GeForce RTX 4090 Mobile
+340%
714
GeForce RTX 3050 Ti Mobile
162
Graphics Card
Jan 2023
Release Date
May 2021
GeForce 40 Mobile
Generation
GeForce 30 Mobile
Mobile
Type
Mobile
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1335 MHz
Base Clock
735 MHz
1695 MHz
Boost Clock
1035 MHz
2250 MHz
Memory Clock
1500 MHz
Memory
16GB
Memory Size
4GB
GDDR6
Memory Type
GDDR6
256bit
Memory Bus
128bit
576.0GB/s
Bandwidth
192.0GB/s
Render Config
-
-
-
76
SM Count
20
9728
Shading Units
2560
304
TMUs
80
112
ROPs
32
304
Tensor Cores
80
76
RT Cores
20
128 KB (per SM)
L1 Cache
128 KB (per SM)
64 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
189.8 GPixel/s
Pixel Rate
33.12 GPixel/s
515.3 GTexel/s
Texture Rate
82.80 GTexel/s
32.98 TFLOPS
FP16 (half)
5.299 TFLOPS
32.98 TFLOPS
FP32 (float)
5.299 TFLOPS
515.3 GFLOPS
FP64 (double)
82.80 GFLOPS
Board Design
120W
TDP
75W
-
-
-
Portable Device Dependent
Outputs
Portable Device Dependent
None
Power Connectors
None
Graphics Processor
AD103
GPU Name
GA106
GN21-X11
GPU Variant
-
Ada Lovelace
Architecture
Ampere
TSMC
Foundry
Samsung
5 nm
Process Size
8 nm
45.9 billion
Transistors
12 billion
379 mm²
Die Size
276 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
8.6
6.8
Shader Model
6.7
