AMD Radeon Instinct MI8 vs NVIDIA A10 PCIe
We compared two Professional market GPUs: 4GB VRAM Radeon Instinct MI8 and 24GB VRAM A10 PCIe to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA A10 PCIe 's Advantages
Released 4 years and 4 months late
Boost Clock1695MHz
More VRAM (24GB vs 4GB)
Larger VRAM bandwidth (600.2GB/s vs 512.0GB/s)
5120 additional rendering cores
Lower TDP (150W vs 175W)
Score
Benchmark
FP32 (float)
Radeon Instinct MI8
8.192 TFLOPS
A10 PCIe
+281%
31.24 TFLOPS
Graphics Card
Dec 2016
Release Date
Apr 2021
Radeon Instinct
Generation
Tesla
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
-
Base Clock
885 MHz
-
Boost Clock
1695 MHz
500 MHz
Memory Clock
1563 MHz
Memory
4GB
Memory Size
24GB
HBM
Memory Type
GDDR6
4096bit
Memory Bus
384bit
512.0GB/s
Bandwidth
600.2GB/s
Render Config
64
Compute Units
-
-
SM Count
72
4096
Shading Units
9216
256
TMUs
288
64
ROPs
96
-
Tensor Cores
288
-
RT Cores
72
16 KB (per CU)
L1 Cache
128 KB (per SM)
2 MB
L2 Cache
6 MB
-
-
-
Theoretical Performance
64.00 GPixel/s
Pixel Rate
162.7 GPixel/s
256.0 GTexel/s
Texture Rate
488.2 GTexel/s
8.192 TFLOPS
FP16 (half)
31.24 TFLOPS
8.192 TFLOPS
FP32 (float)
31.24 TFLOPS
512.0 GFLOPS
FP64 (double)
976.3 GFLOPS
Board Design
175W
TDP
150W
450 W
Suggested PSU
450 W
No outputs
Outputs
No outputs
1x 8-pin
Power Connectors
1x 8-pin
Graphics Processor
Fiji
GPU Name
GA102
Fiji XT CA
(215-0862120)
GPU Variant
GA102-890-A1
GCN 3.0
Architecture
Ampere
TSMC
Foundry
Samsung
28 nm
Process Size
8 nm
8.9 billion
Transistors
28.3 billion
596 mm²
Die Size
628 mm²
Graphics Features
12 (12_0)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
2.1
OpenCL
3.0
1.2.170
Vulkan
1.3
-
CUDA
8.6
6.5
Shader Model
6.6
