AMD Radeon Instinct MI8 vs NVIDIA Tesla M2070 Q
We compared two Professional market GPUs: 4GB VRAM Radeon Instinct MI8 and 6GB VRAM Tesla M2070 Q to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD Radeon Instinct MI8 's Advantages
Released 5 years and 5 months late
Larger VRAM bandwidth (512.0GB/s vs 150.3GB/s)
3648 additional rendering cores
Lower TDP (175W vs 225W)
NVIDIA Tesla M2070 Q 's Advantages
More VRAM (6GB vs 4GB)
Score
Benchmark
FP32 (float)
Radeon Instinct MI8
+696%
8.192 TFLOPS
Tesla M2070 Q
1.028 TFLOPS
Graphics Card
Dec 2016
Release Date
Jul 2011
Radeon Instinct
Generation
Tesla
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
-
-
-
-
-
-
500 MHz
Memory Clock
783 MHz
Memory
4GB
Memory Size
6GB
HBM
Memory Type
GDDR5
4096bit
Memory Bus
384bit
512.0GB/s
Bandwidth
150.3GB/s
Render Config
64
Compute Units
-
-
SM Count
14
4096
Shading Units
448
256
TMUs
56
64
ROPs
48
-
-
-
-
-
-
16 KB (per CU)
L1 Cache
64 KB (per SM)
2 MB
L2 Cache
768 KB
-
-
-
Theoretical Performance
64.00 GPixel/s
Pixel Rate
16.07 GPixel/s
256.0 GTexel/s
Texture Rate
32.14 GTexel/s
8.192 TFLOPS
FP16 (half)
-
8.192 TFLOPS
FP32 (float)
1028 GFLOPS
512.0 GFLOPS
FP64 (double)
513.9 GFLOPS
Board Design
175W
TDP
225W
450 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
1x 8-pin
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
Fiji
GPU Name
GF100
Fiji XT CA
(215-0862120)
GPU Variant
GF100-876-A1
GCN 3.0
Architecture
Fermi
TSMC
Foundry
TSMC
28 nm
Process Size
40 nm
8.9 billion
Transistors
3.1 billion
596 mm²
Die Size
529 mm²
Graphics Features
12 (12_0)
DirectX
12 (11_0)
4.6
OpenGL
4.6
2.1
OpenCL
1.1
1.2.170
Vulkan
N/A
-
CUDA
2.0
6.5
Shader Model
5.1
