NVIDIA GeForce RTX 3060 vs NVIDIA GeForce RTX 3070 graphics comparison: performance score, cores and clock speeds, VRAM, TGP, cache, and RT and AI units. The better value in each row is highlighted in green.
Summary
NVIDIA GeForce RTX 3060 offers a higher boost clock (1703 MHz vs 1620 MHz). NVIDIA GeForce RTX 3070 offers more compute units (CUDA cores: 5120 vs 3840), more VRAM (8 GB vs 6 GB), a wider memory bus (256 bit vs 192 bit), higher memory bandwidth (448 GB/s vs 336 GB/s), and more L2 cache (4 MB vs 3 MB).
Performance score
0
Coming soon
0
Coming soon
Key specs
CUDA cores3840
CUDA cores5120
Boost1703 MHz
Boost1620 MHz
VRAM6 GB
VRAM8 GB
Memory bus192 bit
Memory bus256 bit
TGP60–130 W
TGP80–140 W
MemoryGDDR6
MemoryGDDR6
General
SeriesRTX 30 Mobile
SeriesRTX 30 Mobile
GPU architectureAmpere
GPU architectureAmpere
GPU chipGA106
GPU chipGA104
Chip variantGN20-E3
Chip variantGN20-E5
FoundrySamsung
FoundrySamsung
Process node8 nm
Process node8 nm
Die size276 mm²
Die size392 mm²
Transistors12 billion
Transistors17.4 billion
InterfacePCIe 4.0 x16
InterfacePCIe 4.0 x16
Release dateJanuary 12, 2021
Release dateJanuary 12, 2021
Cores and clocks
CUDA cores3840
CUDA cores5120
30× SM
40× SM
Boost clock (max)1703 MHz
Boost clock (max)1620 MHz
Boost clock (min)1283 MHz
Boost clock (min)1290 MHz
Memory
VRAM6 GB
VRAM8 GB
MemoryGDDR6
MemoryGDDR6
Memory bus192 bit
Memory bus256 bit
Memory speed14 Gb/s
Memory speed14 Gb/s
Bandwidth336 GB/s
Bandwidth448 GB/s
TGP
TGP range60 - 115 W
TGP range80 - 125 W
Dynamic Boost15 W
Dynamic Boost15 W
Cache
L2 cache3 MB
L2 cache4 MB
L1 cache128 KB
L1 cache128 KB
Units
TMUs120
TMUs160
ROPs48
ROPs80
RT cores30
RT cores40
Tensor cores120
Tensor cores160
SM blocks30
SM blocks40
AI
AI performanceN/A
AI performanceN/A