Tesla M40 24 GB and GeForce GTX 680 - video card comparison
Passmark - benchmark score Tesla M40 24 GB vs GeForce GTX 680
Tesla M40 24 GB
10730
GeForce GTX 680
5721
Tesla M40 24 GB
GeForce GTX 680
Architecture
Maxwell 2.0 (2015−2019)
Kepler (2012−2018)
GPU code name
GM200
GK104
Market segment
Workstation
Desktop
Release date
November 2015
March 2012
Bus support
-
PCI Express 3.0
Interface
PCIe 3.0 x16
PCIe 3.0 x16
Supplementary power connectors
8-pin EPS
Two 6-pin
SLI options
-
+
Boost clock speed
1112 MHz
1058 MHz
Core clock speed
-
1006 MHz
CUDA cores
-
1536
Floating-point performance
-
3,090.4 gflops
Manufacturing process technology
28 nm
28 nm
Number of transistors
8,000 million
3,540 million
Pipelines / CUDA cores
3072
1536
Power consumption (TDP)
250 Watt
195 Watt
Texture fill rate
213.5
128.8 billion/sec
Memory type
GDDR5
GDDR5
Maximum RAM amount
24 GB
2048 MB
Memory bus width
384 Bit
256-bit GDDR5
Memory clock speed
6 GB/s
6000 MHz
Memory bandwidth
288.4 GB/s
192.2 GB/s
Shared memory
-
-
Display Connectors
No outputs
One Dual Link DVI-I, One Dual Link DVI-D, One HDMI, One DisplayPort
Multi monitor support
-
4 displays
HDMI
-
+
HDCP
-
+
Maximum VGA resolution
-
2048x1536
Audio input for HDMI
-
Internal
DirectX
12 (12_1)
12 (11_0)
CUDA
5.2
+
OpenCL
3.0
1.2
OpenGL
4.6
4.2
Shader Model
6.7
5.1
Vulkan
1.3
1.1.126