NVIDIA GRID M10 8Q vs AMD Radeon Instinct MI8
We compared two Professional market GPUs: 8GB VRAM GRID M10 8Q and 4GB VRAM Radeon Instinct MI8 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GRID M10 8Q 's Advantages
Boost Clock1306MHz
More VRAM (8GB vs 4GB)
AMD Radeon Instinct MI8 's Advantages
Released 7 months late
Larger VRAM bandwidth (512.0GB/s vs 83.20GB/s)
3456 additional rendering cores
Lower TDP (175W vs 225W)
Score
Benchmark
FP32 (float)
GRID M10 8Q
1.672 TFLOPS
Radeon Instinct MI8
+389%
8.192 TFLOPS
Graphics Card
May 2016
Release Date
Dec 2016
GRID
Generation
Radeon Instinct
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1033 MHz
Base Clock
-
1306 MHz
Boost Clock
-
1300 MHz
Memory Clock
500 MHz
Memory
8GB
Memory Size
4GB
GDDR5
Memory Type
HBM
128bit
Memory Bus
4096bit
83.20GB/s
Bandwidth
512.0GB/s
Render Config
-
Compute Units
64
-
-
-
640
Shading Units
4096
40
TMUs
256
16
ROPs
64
-
-
-
-
-
-
64 KB (per SMM)
L1 Cache
16 KB (per CU)
2 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
20.90 GPixel/s
Pixel Rate
64.00 GPixel/s
52.24 GTexel/s
Texture Rate
256.0 GTexel/s
-
FP16 (half)
8.192 TFLOPS
1.672 TFLOPS
FP32 (float)
8.192 TFLOPS
52.24 GFLOPS
FP64 (double)
512.0 GFLOPS
Board Design
225W
TDP
175W
550 W
Suggested PSU
450 W
No outputs
Outputs
No outputs
1x 8-pin
Power Connectors
1x 8-pin
Graphics Processor
GM107
GPU Name
Fiji
GM107-570-A2
GPU Variant
Fiji XT CA
(215-0862120)
Maxwell
Architecture
GCN 3.0
TSMC
Foundry
TSMC
28 nm
Process Size
28 nm
1.87 billion
Transistors
8.9 billion
148 mm²
Die Size
596 mm²
Graphics Features
12 (11_0)
DirectX
12 (12_0)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.3
Vulkan
1.2.170
5.0
CUDA
-
5.1
Shader Model
6.5
