NVIDIA Tesla K40s vs AMD Radeon Pro 580

GPU · Specs Compare

Products
GPU NVIDIA Tesla K40s NVIDIA · 2013-11
GPU AMD Radeon Pro 580 AMD · 2017-06
Graphics Card
Release Date
Nov 2013
Jun 2017
Generation
Tesla Kepler
Radeon Pro Mac
Type
Professional
Mobile
Bus Interface
PCIe 3.0 x16
PCIe 3.0 x16
Clock Speeds
Base Clock
706 MHz
1100 MHz
Boost Clock
706 MHz
1200 MHz
Memory Clock
1502 MHz
1695 MHz
Memory
Memory Size
12GB
8GB
Memory Type
GDDR5
GDDR5
Memory Bus
384bit
256bit
Bandwidth
288.4GB/s
217.0GB/s
Render Config
Shading Units
2880
2304
TMUs
240
144
ROPs
48
32
L1 Cache
16 KB (per SMX)
16 KB (per CU)
L2 Cache
1536 KB
2 MB
Compute Units
—
36
Theoretical Performance
Pixel Rate
42.36 GPixel/s
38.40 GPixel/s
Texture Rate
169.4 GTexel/s
172.8 GTexel/s
FP32 (float)
4.067 TFLOPS
5.530 TFLOPS
FP64 (double)
1355.5 GFLOPS
345.6 GFLOPS
FP16 (half)
—
5.530 TFLOPS
Graphics Processor
GPU Name
GK110B
Ellesmere
Architecture
Kepler
GCN 4.0
Foundry
TSMC
GlobalFoundries
Process Size
28 nm
14 nm
Transistors
7.08 billion
5.7 billion
Die Size
561 mm²
232 mm²
GPU Variant
—
Ellesmere XTA(216-0886220)
Board Design
TDP
245W
185W
Suggested PSU
550 W
—
Outputs
No outputs
No outputs
Power Connectors
—
None
Graphics Features
DirectX
12 (11_1)
12 (12_0)
OpenGL
4.6
4.6
OpenCL
3.0
2.1
Vulkan
1.2.175
1.2
CUDA
3.5
—
Shader Model
6.5 (5.1)
6.4
Benchmark
3DMark Time Spy
—
3,791
3DMark Time Spy Extreme
—
1,684
▲ Specs Compare Data sourced from official specifications and benchmark tests.