GeForce GTX 680MX and Radeon Pro 575 - video card comparison
GeForce GTX 680MX
Radeon Pro 575
Architecture
Kepler (2012−2018)
GCN 4.0 (2016−2020)
GPU code name
-
Polaris 20
Market segment
Laptop
Mobile workstation
Release date
October 2012
June 2017
Bus support
PCI Express 3.0
-
Interface
-
PCIe 3.0 x16
Supplementary power connectors
-
None
SLI options
+
-
Core clock speed
720 MHz
1096 MHz
CUDA cores
1536
-
Floating-point performance
-
4,506 gflops
Manufacturing process technology
28 nm
14 nm
Number of transistors
3540 Million
5,700 million
Pipelines / CUDA cores
1536
2048
Power consumption (TDP)
122 Watt
120 Watt
Texture fill rate
92.2 billion/sec
140.3
Memory type
GDDR5
GDDR5
Maximum RAM amount
2 GB
4 GB
Memory bus width
256 Bit
256 Bit
Memory clock speed
2500 MHz
6780 MHz
Memory bandwidth
160 GB/s
217.0 GB/s
Shared memory
-
-
Display Connectors
-
No outputs
DirectX
12 API
12 (12_0)
CUDA
+
-
OpenCL
1.1
2.0
OpenGL
4.5
4.6
Shader Model
-
6.4
Vulkan
-
1.2.131
3D Vision
+
-
Optimus
+
-