Tesla D870 vs Tesla K20c
We compared two Professional market GPUs: 1536MB VRAM Tesla D870 and 5GB VRAM Tesla K20c to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Tesla K20c 's Advantages
Released 5 years and 6 months late
More VRAM (5GB vs 1536GB)
Larger VRAM bandwidth (208.0GB/s vs 76.80GB/s)
2368 additional rendering cores
Lower TDP (225W vs 520W)
Score
Benchmark
FP32 (float)
Tesla D870
346
Tesla K20c
+918%
3524
Graphics Card
May 2007
Release Date
Nov 2012
Tesla
Generation
Tesla
Professional
Type
Professional
PCIe 1.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
800 MHz
Memory Clock
1300 MHz
Memory
1536MB
Memory Size
5GB
GDDR3
Memory Type
GDDR5
384bit
Memory Bus
320bit
76.80GB/s
Bandwidth
208.0GB/s
Render Config
16
SM Count
-
128
Shading Units
2496
32
TMUs
208
24
ROPs
40
-
L1 Cache
16 KB (per SMX)
96 KB
L2 Cache
1280 KB
Theoretical Performance
14.40 GPixel/s
Pixel Rate
36.71 GPixel/s
38.40 GTexel/s
Texture Rate
146.8 GTexel/s
345.6 GFLOPS
FP32 (float)
3.524 TFLOPS
-
FP64 (double)
1175 GFLOPS
Board Design
520W
TDP
225W
900 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
-
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
G80
GPU Name
GK110
-
GPU Variant
GK110-885-KA-A1
Tesla
Architecture
Kepler
TSMC
Foundry
TSMC
90 nm
Process Size
28 nm
0.681 billion
Transistors
7.08 billion
484 mm²
Die Size
561 mm²
Graphics Features
11.1 (10_0)
DirectX
12 (11_0)
3.3
OpenGL
4.6
1.1 (1.0)
OpenCL
3.0
N/A
Vulkan
1.1
1.0
CUDA
3.5
4.0
Shader Model
5.1
