GeForce RTX 3060 vs GeForce RTX 5090
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 3060 and 32GB VRAM GeForce RTX 5090 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GeForce RTX 3060 's Advantages
Lower TDP (170W vs 575W)
GeForce RTX 5090 's Advantages
Released 4 years late
Boost Clock has increased by 35% (2407MHz vs 1777MHz)
More VRAM (32GB vs 12GB)
Larger VRAM bandwidth (1.79TB/s vs 360.0GB/s)
18176 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce RTX 3060
12740
GeForce RTX 5090
+722%
104800
3DMark Time Spy
GeForce RTX 3060
8747
GeForce RTX 5090
+443%
47510
3DMark Time Spy Extreme
GeForce RTX 3060
4097
GeForce RTX 5090
+530%
25825
3DMark Speed Way
GeForce RTX 3060
2181
GeForce RTX 5090
+570%
14621
Graphics Card
Jan 2021
Release Date
Jan 2025
GeForce 30
Generation
GeForce 50
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
1320 MHz
Base Clock
2017 MHz
1777 MHz
Boost Clock
2407 MHz
1875 MHz
Memory Clock
1750 MHz
Memory
12GB
Memory Size
32GB
GDDR6
Memory Type
GDDR7
192bit
Memory Bus
512bit
360.0GB/s
Bandwidth
1.79TB/s
Render Config
28
SM Count
170
3584
Shading Units
21760
112
TMUs
680
48
ROPs
176
112
Tensor Cores
680
28
RT Cores
170
128 KB (per SM)
L1 Cache
128 KB (per SM)
3 MB
L2 Cache
96 MB
Theoretical Performance
85.30 GPixel/s
Pixel Rate
423.6 GPixel/s
199.0 GTexel/s
Texture Rate
1637 GTexel/s
12.74 TFLOPS
FP16 (half)
104.8 TFLOPS
12.74 TFLOPS
FP32 (float)
104.8 TFLOPS
199.0 GFLOPS
FP64 (double)
1.637 TFLOPS
Board Design
170W
TDP
575W
450 W
Suggested PSU
950 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
1x HDMI 2.1b
3x DisplayPort 2.1b
1x 12-pin
Power Connectors
1x 16-pin
Graphics Processor
GA106
GPU Name
GB202
GA106-300-A1
GPU Variant
GB202-300-A1
Ampere
Architecture
Blackwell 2.0
Samsung
Foundry
TSMC
8 nm
Process Size
5 nm
12 billion
Transistors
92.2 billion
276 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
8.6
CUDA
12.0
6.6
Shader Model
6.8
