Home
NVIDIA Tesla M40
NVIDIA Tesla M40
NVIDIA Tesla M40 24 GB
NVIDIA Tesla M40
This is a GPU manufactured with TSMC 28nm process, based on Nvidia Maxwell 2.0 architecture and released on Nov 2015. It features 8 billion transistors, 3072 CUDA cores and 12GB GDDR5 memory, with 3MB L2 cache, theoretical performance of 6.832TFLOPS, with total power consumption of 250W.
Graphics Card
[Report Issues]
Release Date
Nov 2015
Generation
Tesla Maxwell
Type
Professional
Bus Interface
PCIe 3.0 x16
Clock Speeds
[Report Issues]
Base Clock
948 MHz
Boost Clock
1112 MHz
Memory Clock
1502 MHz
Memory
[Report Issues]
Memory Size
12GB
Memory Type
GDDR5
Memory Bus
384bit
Bandwidth
288.4GB/s
Render Config
[Report Issues]
Shading Units
3072
TMUs
192
ROPs
96
L1 Cache
48 KB (per SMM)
L2 Cache
3 MB
Theoretical Performance
[Report Issues]
Pixel Rate
106.8 GPixel/s
Texture Rate
213.5 GTexel/s
FP32 (float)
6.832 TFLOPS
FP64 (double)
213.5 GFLOPS
Graphics Processor
[Report Issues]
GPU Name
GM200
GPU Variant
GM200-895-A1
Architecture
Maxwell 2.0
Foundry
TSMC
Process Size
28 nm
Transistors
8 billion
Die Size
601 mm²
Board Design
[Report Issues]
TDP
250W
Suggested PSU
600 W
Outputs
No outputs
Power Connectors
8-pin EPS
Graphics Features
[Report Issues]
DirectX
12 (12_1)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
5.2
Shader Model
6.7
Rankings
[Report Issues]
FP32 (float)
3DMark Time Spy
3DMark Time Spy Extreme
Blender
OctaneBench