Radeon Pro Vega 48 and GeForce GTX 1660 Super - video card comparison
Passmark - benchmark score Radeon Pro Vega 48 vs GeForce GTX 1660 Super
Radeon Pro Vega 48
11638
GeForce GTX 1660 Super
13154
Radeon Pro Vega 48
GeForce GTX 1660 Super
Architecture
GCN 5.0 (2017−2020)
Turing (2018−2021)
GPU code name
Vega 10
Turing TU116
Market segment
Mobile workstation
Desktop
Release date
March 2019
October 2019
Interface
PCIe 3.0 x16
PCIe 3.0 x16
Supplementary power connectors
None
1x 8-pin
Boost clock speed
1300 MHz
1785 MHz
Core clock speed
1200 MHz
1530 MHz
Manufacturing process technology
14 nm
12 nm
Number of transistors
12,500 million
6,600 million
Pipelines / CUDA cores
3072
1408
Power consumption (TDP)
-
125 Watt
Texture fill rate
249.6
157.1
Memory type
HBM2
GDDR6
Maximum RAM amount
8 GB
6 GB
Memory bus width
2048 Bit
192 Bit
Memory clock speed
1572 MHz
14000 MHz
Memory bandwidth
402.4 GB/s
336.0 GB/s
Shared memory
-
-
Display Connectors
No outputs
1x DVI, 1x HDMI, 1x DisplayPort
HDMI
-
+
HDCP
-
+
G-SYNC support
-
+
DirectX
12 (12_1)
12 (12_1)
CUDA
-
7.5
OpenCL
2.0
1.2
OpenGL
4.6
4.6
Shader Model
6.4
6.5
Vulkan
1.1.125
1.2.131
Ansel
-
+
NVENC
-
+