P104 100 vs Radeon HD 7670A

When choosing between P104 100 and Radeon HD 7670A, it is worth examining the specifications of the models in detail. Do they meet the recommended requirements of modern games and software? Storage capacity, form factor, TDP, available ports, warranty and manufacturer support are all important. For example, the size of a PC case can limit the maximum thickness and length of the card. Often, instead of the factory overclocked card and RGB backlight, it is better to choose a reference model with a more efficient GPU. And make sure that your current power supply unit has the correct connection pins (using adapters is not recommended). This GPUs compare tool is meant to help you to choose the best graphics card for your build. Let's find out the difference between P104 100 and Radeon HD 7670A.
P104 100
Radeon HD 7670A

Main Specs

  P104 100 Radeon HD 7670A
Power consumption (TDP) 45 Watt
Interface PCIe 3.0 x16 MXM-A (3.0)
Supplementary power connectors 1x 8-pin
Memory type GDDR5X GDDR5
Maximum RAM amount 4 GB 1 GB
Display Connectors No outputs No outputs
 
  • P104 100 is connected by PCIe 3.0 x16, and Radeon HD 7670A uses MXM-A (3.0) interface.
  • P104 100 has 3 GB more memory, than Radeon HD 7670A.
  • Both cards are used in Desktops.
  • P104 100 is build with Pascal architecture, and Radeon HD 7670A - with TeraScale 2.
  • Core clock speed of P104 100 is 1007 MHz higher, than Radeon HD 7670A.
  • P104 100 is manufactured by 16 nm process technology, and Radeon HD 7670A - by 40 nm process technology.
  • Memory clock speed of P104 100 is 6008 MHz higher, than Radeon HD 7670A.

Game benchmarks

high / 1080p 35−40 1−2
ultra / 1080p 21−24 0−1
QHD / 1440p 16−18 0−1
4K / 2160p 10−11
low / 720p 60−65 12−14
medium / 1080p 40−45 3−4
The average gaming FPS of P104 100 in Assassin's Creed Odyssey is 683% more, than Radeon HD 7670A.

Full Specs

  P104 100 Radeon HD 7670A
Architecture Pascal TeraScale 2
Code name GP104 Turks
Type Desktop Desktop
Release date 12 December 2017 5 January 2012
Pipelines 1920 480
Core clock speed 1607 MHz 600 MHz
Boost Clock 1733 MHz
Transistor count 7,200 million 716 million
Manufacturing process technology 16 nm 40 nm
Texture fill rate 208.0 14.40
Floating-point performance 8,873 gflops 576.0 gflops
Length 267 mm
Memory bus width 256 Bit 128 Bit
Memory clock speed 10008 MHz 4000 MHz
Memory bandwidth 320.3 GB/s 64 GB/s
DirectX 12 (12_1) 11.2 (11_0)
Shader Model 6.4 5.0
OpenGL 4.6 4.4
OpenCL 1.2 1.2
Vulkan 1.2.131 N/A
CUDA 6.1