P104 100 vs Radeon Pro Duo

If you are going to buy a new graphics card and are choosing between P104 100 and Radeon Pro Duo, there are a couple of things to consider. Cards with more VRAM in general perform better and allow you to play on higher graphics settings. Size also makes a difference. A model with a large heatsink can occupy up to three expansion slots on a motherboard. Be sure you have enough room in your PC case. When comparing GPUs with different architectures, more processing cores and even higher TFLOPS will not always translate to better performance. To help you decide which GPU you need, we have measured frame rates in a number of popular games. For more on how the P104 100 stacks up against Radeon Pro Duo, check out specs charts below.

Main Specs

  P104 100 Radeon Pro Duo
Power consumption (TDP) 350 Watt
Interface PCIe 3.0 x16 PCIe 3.0 x16
Supplementary power connectors 1x 8-pin 3x 8-pin
Memory type GDDR5X High Bandwidth Memory (HBM)
Maximum RAM amount 4 GB 8 GB
Display Connectors No outputs 1x HDMI, 3x DisplayPort
  Check Price
  • Both video cards are using PCIe 3.0 x16 interface connection to a motherboard.
  • Radeon Pro Duo has 4 GB more memory, than P104 100.
  • Both cards are used in Desktops.
  • P104 100 is build with Pascal architecture, and Radeon Pro Duo - with GCN 3.0.
  • P104 100 is manufactured by 16 nm process technology, and Radeon Pro Duo - by 28 nm process technology.
  • Radeon Pro Duo is 10 mm longer, than P104 100.
  • Memory clock speed of P104 100 is 9508 MHz higher, than Radeon Pro Duo.

Game benchmarks

high / 1080p 35−40 50−55
ultra / 1080p 21−24 30−35
QHD / 1440p 16−18 27−30
4K / 2160p 10−11 14−16
low / 720p 60−65 75−80
medium / 1080p 40−45 60−65
The average gaming FPS of Radeon Pro Duo in Assassin's Creed Odyssey is 37% more, than P104 100.

Full Specs

  P104 100 Radeon Pro Duo
Architecture Pascal GCN 3.0
Code name GP104 Capsaicin
Type Desktop Workstation
Release date 12 December 2017 26 April 2016
Pipelines 1920 4096
Core clock speed 1607 MHz
Boost Clock 1733 MHz 1000 MHz
Transistor count 7,200 million 8,900 million
Manufacturing process technology 16 nm 28 nm
Texture fill rate 208.0 256.0
Floating-point performance 8,873 gflops 2x 8,192 gflops
Length 267 mm 277 mm
Memory bus width 256 Bit 4096 Bit
Memory clock speed 10008 MHz 500 MHz
Memory bandwidth 320.3 GB/s 512 GB/s
DirectX 12 (12_1)
Shader Model 6.4 6.0
OpenGL 4.6 4.5
OpenCL 1.2 2.0
Vulkan 1.2.131 +
CUDA 6.1
FreeSync +
Bus support PCIe 3.0
HDMI +
Bitcoin / BTC (SHA256) 809 Mh/s
Eyefinity +
HD3D +
PowerTune +
TrueAudio +
ZeroCore +
Mantle +
Design reference
Number of Eyefinity displays 6
DisplayPort support +
CrossFire +
VCE +
Compute units 128
AppAcceleration +
LiquidVR +
TressFX +
UVD +
Enduro +
FRTC +
  Check Price