NVIDIA RTX 3500 vs NVIDIA RTX 4000 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 RTX 4000 offers more compute units (CUDA cores: 7424 vs 5120), a higher boost clock (1665 MHz vs 1545 MHz), more RT units (58 vs 40), and higher AI performance (538 TOPS vs 369 TOPS).
Performance score
0
Coming soon
0
Coming soon
Key specs
CUDA cores5120
CUDA cores7424
Boost1545 MHz
Boost1665 MHz
VRAM12 GB
VRAM12 GB
Memory bus192 bit
Memory bus192 bit
TGP60–140 W
TGP60–175 W
MemoryGDDR6
MemoryGDDR6
General
SeriesRTX Ada Mobile
SeriesRTX Ada Mobile
GPU architectureAda Lovelace
GPU architectureAda Lovelace
GPU chipAD104
GPU chipAD104
FoundryTSMC
FoundryTSMC
Process node5 nm
Process node5 nm
Die size295 mm²
Die size295 mm²
Transistors35.8 billion
Transistors35.8 billion
InterfacePCIe 4.0 x16
InterfacePCIe 4.0 x16
Release dateMarch 21, 2023
Release dateMarch 21, 2023
Cores and clocks
CUDA cores5120
CUDA cores7424
40× SM
58× SM
Boost clock1545 MHz
Boost clock1665 MHz
Memory
VRAM12 GB
VRAM12 GB
MemoryGDDR6
MemoryGDDR6
Memory bus192 bit
Memory bus192 bit
Memory speed18 Gb/s
Memory speed18 Gb/s
Bandwidth432 GB/s
Bandwidth432 GB/s
TGP
TGP range60 - 140 W
TGP range60 - 175 W
Cache
L2 cache48 MB
L2 cache48 MB
L1 cache128 KB
L1 cache128 KB
Units
TMUs160
TMUs232
ROPs64
ROPs80
RT cores40
RT cores58
Tensor cores160
Tensor cores232
SM blocks40
SM blocks58
AI
AI performance369 TOPS
AI performance538 TOPS