Home GPU Comparison NVIDIA Tesla K40st vs NVIDIA Tesla S870

NVIDIA Tesla K40st vs NVIDIA Tesla S870

We compared two Professional market GPUs: 12GB VRAM Tesla K40st and 1536MB VRAM Tesla S870 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA Tesla K40st 's Advantages
Released 6 years and 6 months late
More VRAM (12GB vs 1GB)
Larger VRAM bandwidth (288.4GB/s vs 76.80GB/s)
2752 additional rendering cores
Lower TDP (245W vs 800W)

Score

Benchmark

FP32 (float)
Tesla K40st +857%
3.312 TFLOPS
Tesla S870
0.346 TFLOPS
VS

Graphics Card

Nov 2013
Release Date
May 2007
Tesla
Generation
Tesla
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 1.0 x16

Clock Speeds

-
Base Clock
-
-
Boost Clock
-
1502 MHz
Memory Clock
800 MHz

Memory

12GB
Memory Size
1536MB
GDDR5
Memory Type
GDDR3
384bit
Memory Bus
384bit
288.4GB/s
Bandwidth
76.80GB/s

Render Config

-
SM Count
16
-
Compute Units
-
2880
Shading Units
128
240
TMUs
32
48
ROPs
24
-
Tensor Cores
-
-
RT Cores
-
16 KB (per SMX)
L1 Cache
-
1536 KB
L2 Cache
96 KB

Theoretical Performance

34.50 GPixel/s
Pixel Rate
14.40 GPixel/s
138.0 GTexel/s
Texture Rate
38.40 GTexel/s
-
FP16 (half)
-
3.312 TFLOPS
FP32 (float)
345.6 GFLOPS
1104 GFLOPS
FP64 (double)
-

Graphics Processor

GK110B
GPU Name
G80
-
GPU Variant
-
Kepler
Architecture
Tesla
TSMC
Foundry
TSMC
28 nm
Process Size
90 nm
7.08 billion
Transistors
0.681 billion
561 mm²
Die Size
484 mm²

Board Design

245W
TDP
800W
550 W
Suggested PSU
1200 W
No outputs
Outputs
No outputs
-
Power Connectors
-

Graphics Features

12 (11_1)
DirectX
11.1 (10_0)
4.6
OpenGL
3.3
3.0
OpenCL
1.1 (1.0)
1.1
Vulkan
N/A
3.5
CUDA
1.0
5.1
Shader Model
4.0

Related GPU Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy