Radeon Pro Duo and Tesla P40 - video card comparison
Passmark - benchmark score Radeon Pro Duo vs Tesla P40
Radeon Pro Duo
8409
Tesla P40
12873
Radeon Pro Duo
Tesla P40
Architecture
GCN 3.0 (2014−2017)
Pascal (2016−2021)
GPU code name
Capsaicin
GP102
Market segment
Workstation
Workstation
Release date
April 2016
September 2016
Bus support
PCIe 3.0
-
Interface
PCIe 3.0 x16
PCIe 3.0 x16
Supplementary power connectors
3x 8-pin
8-pin EPS
Boost clock speed
1000 MHz
1531 MHz
Compute units
128
-
Core clock speed
-
1303 MHz
Floating-point performance
2x 8,192 gflops
11,758 gflops
Manufacturing process technology
28 nm
16 nm
Number of transistors
8,900 million
11,800 million
Pipelines / CUDA cores
4096
3840
Power consumption (TDP)
350 Watt
250 Watt
Texture fill rate
256.0
367.4
Memory type
High Bandwidth Memory (HBM)
GDDR5
Maximum RAM amount
8 GB
24 GB
Memory bus width
4096 Bit
384 Bit
Memory clock speed
500 MHz
7.2 GB/s
Memory bandwidth
512 GB/s
347.1 GB/s
Display Connectors
1x HDMI, 3x DisplayPort
No outputs
Eyefinity
1
-
Number of Eyefinity displays
6
-
HDMI
+
-
DisplayPort support
+
-
DirectX
DirectX® 12
12 (12_1)
CUDA
-
6.1
Mantle
+
-
OpenCL
2.0
3.0
OpenGL
4.5
4.6
Shader Model
6.0
6.7
Vulkan
+
1.3
AppAcceleration
+
-
CrossFire
1
-
Enduro
+
-
FreeSync
1
-
FRTC
1
-
HD3D
+
-
LiquidVR
1
-
PowerTune
+
-
TressFX
1
-
TrueAudio
+
-
UVD
+
-
VCE
+
-
ZeroCore
+
-