Tesla M40 vs Radeon Instinct MI60
We compared two Professional market GPUs: 12GB VRAM Tesla M40 and 32GB VRAM Radeon Instinct MI60 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Tesla M40 's Advantages
Lower TDP (250W vs 300W)
Radeon Instinct MI60 's Advantages
Released 3 years late
Boost Clock has increased by 62% (1800MHz vs 1112MHz)
More VRAM (32GB vs 12GB)
Larger VRAM bandwidth (1024GB/s vs 288.4GB/s)
1024 additional rendering cores
Score
Benchmark
FP32 (float)
Tesla M40
6832
Radeon Instinct MI60
+115%
14750
Graphics Card
Nov 2015
Release Date
Nov 2018
Tesla Maxwell
Generation
Radeon Instinct
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
948 MHz
Base Clock
1200 MHz
1112 MHz
Boost Clock
1800 MHz
1502 MHz
Memory Clock
1000 MHz
Memory
12GB
Memory Size
32GB
GDDR5
Memory Type
HBM2
384bit
Memory Bus
4096bit
288.4GB/s
Bandwidth
1024GB/s
Render Config
-
Compute Units
64
3072
Shading Units
4096
192
TMUs
256
96
ROPs
64
48 KB (per SMM)
L1 Cache
16 KB (per CU)
3 MB
L2 Cache
4 MB
Theoretical Performance
106.8 GPixel/s
Pixel Rate
115.2 GPixel/s
213.5 GTexel/s
Texture Rate
460.8 GTexel/s
-
FP16 (half)
29.49 TFLOPS
6.832 TFLOPS
FP32 (float)
14.75 TFLOPS
213.5 GFLOPS
FP64 (double)
7.373 TFLOPS
Board Design
250W
TDP
300W
600 W
Suggested PSU
700 W
No outputs
Outputs
1x mini-DisplayPort 1.4a
8-pin EPS
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
GM200
GPU Name
Vega 20
GM200-895-A1
GPU Variant
Vega 20 GL
Maxwell 2.0
Architecture
GCN 5.1
TSMC
Foundry
TSMC
28 nm
Process Size
7 nm
8 billion
Transistors
13.23 billion
601 mm²
Die Size
331 mm²
Graphics Features
12 (12_1)
DirectX
12 (12_1)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.3
Vulkan
1.2
5.2
CUDA
-
6.7
Shader Model
6.4
