NVIDIA GeForce RTX 4070 vs NVIDIA GeForce RTX 4080 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 4080 offers more compute units (CUDA cores: 7424 vs 4608), a higher boost clock (2280 MHz vs 2175 MHz), more VRAM (12 GB vs 8 GB), a wider memory bus (192 bit vs 128 bit), and higher memory bandwidth (432 GB/s vs 256 GB/s).
Performance score
0
Coming soon
0
Coming soon
Key specs
CUDA cores4608
CUDA cores7424
Boost2175 MHz
Boost2280 MHz
VRAM8 GB
VRAM12 GB
Memory bus128 bit
Memory bus192 bit
TGP35–140 W
TGP60–175 W
MemoryGDDR6
MemoryGDDR6
General
SeriesRTX 40 Mobile
SeriesRTX 40 Mobile
GPU architectureAda Lovelace
GPU architectureAda Lovelace
GPU chipAD106
GPU chipAD104
Chip variantGN21-X6
Chip variantGN21-X9
FoundryTSMC
FoundryTSMC
Process node5 nm
Process node5 nm
Die size190 mm²
Die size295 mm²
Transistors22.9 billion
Transistors35.8 billion
InterfacePCIe 4.0 x8
InterfacePCIe 4.0 x16
Release dateJanuary 3, 2023
Release dateJanuary 3, 2023
Cores and clocks
CUDA cores4608
CUDA cores7424
36× SM
58× SM
Boost clock (max)2175 MHz
Boost clock (max)2280 MHz
Boost clock (min)1230 MHz
Boost clock (min)1350 MHz
Memory
VRAM8 GB
VRAM12 GB
MemoryGDDR6
MemoryGDDR6
Memory bus128 bit
Memory bus192 bit
Memory speed16 Gb/s
Memory speed18 Gb/s
Bandwidth256 GB/s
Bandwidth432 GB/s
TGP
TGP range35 - 115 W
TGP range60 - 150 W
Dynamic Boost25 W
Dynamic Boost25 W
Cache
L2 cache32 MB
L2 cache48 MB
L1 cache128 KB
L1 cache128 KB
Units
TMUs144
TMUs232
ROPs48
ROPs80
RT cores36
RT cores58
Tensor cores144
Tensor cores232
SM blocks36
SM blocks58
AI
AI performance321 TOPS
AI performance542 TOPS