GPU Wiki

GeForce GTX 1070 vs AMD Radeon RX 580

Swap Clear comparison
Edit GPUs
Add a third or fourth GPU

Key Differences

  • No game was measured on every one of these cards on one shared test system, so there is no speed comparison to make.
  • GTX 1070 has the most memory bandwidth: 256.3 GB/s against 256.0 GB/s.
  • GTX 1070 draws the least power: 150 W against 185 W.
  • Core counts are not comparable across these architectures.
Spec GeForce GTX 1070 AMD Radeon RX 580 AMD
Overview
Product name NVIDIA GeForce GTX 1070 AMD Radeon RX 580
Architecture Pascal GCN 4 (Polaris)
Generation / family GeForce GTX 10 Series Radeon RX 500 Series
GPU codename GP104 Polaris 20 XT
Launch date 2016-06-10 2017-04-18
Launch MSRP Unknown 229
Market segment Desktop in GeForce GTX 10 Series Desktop, Mainstream
Board type / entry type Reference GPU Reference GPU
GPU
Process node TSMC 16 nm 14 nm
Die size 314 mm² 232 mm²
Transistor count 7.2 billion 5700 million
GPU chip GP104 Polaris 20
SM / multiprocessor count 15 36
Shader processors Raw counts do not scale linearly across architectures or generations. 1920 CUDA Cores 2304 Stream Processors
ROPs 64 32
TMUs 120 144
Clocks
Base / Core clock 1,506 MHz 1257 MHz
Boost clock 1,683 MHz 1340 MHz
Memory clock 2,002 MHz 2000 MHz
Memory
Memory size 8 GB 8 GB
Memory type GDDR5 GDDR5
Memory bus width 256-bit 256-bit
Memory bandwidth 256.3 GB/s 256.0 GB/s
Bus / Interface
Bus interface PCIe 3.0 x16 PCIe 3.0 x16
AGP / PCI / PCIe distinction PCIe PCIe
Power
TDP / TBP / board power 150 W 185 W
Recommended PSU 500 W 450 W
Power connectors 1x PCIe 8-pin 2x 8-pin
Slot width Dual-slot, 267 mm length, 40 mm width Dual-slot
Display / Media
Display outputs 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a 1x HDMI 2.0b, 3x DisplayPort 1.4a
Features
DirectX support DirectX 12.1 DirectX 12
OpenGL support OpenGL 4.6 OpenGL 4.6
Vulkan support Vulkan 1.4 Vulkan 1.2
CUDA support 6.1 Not applicable
Shader Model 6.8 6.7
Performance / Benchmarks
FP32 throughput 6.46 TFLOPS 6.17 TFLOPS
FP16 throughput 0.1 TFLOPS 6.17 TFLOPS
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.

Frame rates compare runs of the same game at the same quality preset on one test system. DLAA counts as native resolution, because it applies anti-aliasing at full render resolution rather than upscaling; DLSS and FSR upscaling are never counted as native. Where a capture named ray tracing without a readable on or off value, it is read as on.