NVIDIA Tesla P100 PCIe 12 GB vs AMD Radeon Instinct MI8
We compared two Professional market GPUs: 12GB VRAM Tesla P100 PCIe 12 GB and 4GB VRAM Radeon Instinct MI8 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA Tesla P100 PCIe 12 GB 's Advantages
Boost Clock1329MHz
More VRAM (12GB vs 4GB)
Larger VRAM bandwidth (549.1GB/s vs 512.0GB/s)
AMD Radeon Instinct MI8 's Advantages
Released 6 months late
512 additional rendering cores
Lower TDP (175W vs 250W)
Score
Benchmark
FP32 (float)
Tesla P100 PCIe 12 GB
+16%
9.526 TFLOPS
Radeon Instinct MI8
8.192 TFLOPS
Graphics Card
Jun 2016
Release Date
Dec 2016
Tesla
Generation
Radeon Instinct
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1190 MHz
Base Clock
-
1329 MHz
Boost Clock
-
715 MHz
Memory Clock
500 MHz
Memory
12GB
Memory Size
4GB
HBM2
Memory Type
HBM
3072bit
Memory Bus
4096bit
549.1GB/s
Bandwidth
512.0GB/s
Render Config
-
Compute Units
64
56
SM Count
-
3584
Shading Units
4096
224
TMUs
256
96
ROPs
64
-
-
-
-
-
-
24 KB (per SM)
L1 Cache
16 KB (per CU)
3 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
127.6 GPixel/s
Pixel Rate
64.00 GPixel/s
297.7 GTexel/s
Texture Rate
256.0 GTexel/s
19.05 TFLOPS
FP16 (half)
8.192 TFLOPS
9.526 TFLOPS
FP32 (float)
8.192 TFLOPS
4.763 TFLOPS
FP64 (double)
512.0 GFLOPS
Board Design
250W
TDP
175W
600 W
Suggested PSU
450 W
No outputs
Outputs
No outputs
1x 8-pin
Power Connectors
1x 8-pin
Graphics Processor
GP100
GPU Name
Fiji
-
GPU Variant
Fiji XT CA
(215-0862120)
Pascal
Architecture
GCN 3.0
TSMC
Foundry
TSMC
16 nm
Process Size
28 nm
15.3 billion
Transistors
8.9 billion
610 mm²
Die Size
596 mm²
Graphics Features
12 (12_1)
DirectX
12 (12_0)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.3
Vulkan
1.2.170
6.0
CUDA
-
6.4
Shader Model
6.5
