Template:AMD graphics API support

From HandWiki

The following table shows the graphics and compute APIs support across AMD GPU microarchitectures. Note that a branding series might include older generation chips.

[ VisualEditor ] [view · talk · edit]

Chip series Micro-architecture Fab Supported APIs AMD support Year introduced Introduced with
Rendering Computing
Vulkan[1] OpenGL[2] Direct3D HSA OpenCL
Wonder Fixed-pipeline[lower-alpha 1] 1000nm
800nm
N/A N/A N/A N/A N/A Ended 1986
Mach 800nm
600nm
1991
3D Rage 500nm 5.0 1996 3D Rage
Rage Pro 350nm 1.1 6.0 1997 Rage Pro
Rage 128 250nm 1.2 1998 Rage 128GL/VR
R100 180 nm
150 nm
1.3 7.0 2000 Original "ATI Radeon", as well as Radeon DDR, 7000, 7500, VE, and LE models
R200 Programmable
pixel & vertex
pipelines
150 nm 8.1 2001 8500, 9000, 9200 and 9250
R300 150 nm
130 nm
110 nm
2.0[lower-alpha 2] 9.0
11 (FL 9_2)
2002 9500–9800, X300–X600, X1050
R420 130 nm
110 nm
9.0b
11 (FL 9_2)
2004 X700–X850
R520 90 nm
80 nm
9.0c
11 (FL 9_3)
2005 X1300–X1950
R600 TeraScale 1 80 nm
65 nm
3.3 10.0
11 (FL 10_0)
ATI Stream 2007 HD 2000 series, HD 3410
RV670 55 nm 10.1
11 (FL 10_1)
ATI Stream APP[3] 2007 HD 3450–3870, Mobility HD 2000 and 3000 series
RV770 55 nm
40 nm
1.0 2008 HD 4000 series
Evergreen TeraScale 2 40 nm 4.5
(Linux 4.2)
[4][5][6][lower-alpha 3]
11 (FL 11_0) 1.2 2009 HD 5000 series
Northern Islands TeraScale 2
TeraScale 3
2010 HD 6000 series, and IGP 7000 series
Southern Islands GCN 1st gen 28 nm 1.0 4.6
(Mesa 4.5)
11 (FL 11_1)
12 (FL11_1)
Yes 1.2
2.0 possible
Current 2012 HD 7000 series
Sea Islands GCN 2nd gen 1.1 11 (FL 12_0)
12 (FL 12_0)
2.0
(1.2 in MacOS, Linux)
2.1 Beta in Linux ROCm
2.2 possible
2013 Radeon 200 series
Volcanic Islands GCN 3rd gen 2014 Radeon 300 series
Arctic Islands GCN 4th gen 14 nm 2016 Radeon 400 series
Vega GCN 5th gen 14 nm
7 nm
11 (FL 12_1)
12 (FL 12_1)
2017 Radeon Vega series
Navi RDNA 1st gen 7 nm 2019 Radeon RX 5000 series
  1. Radeon 7000 Series has programmable pixel shaders, but do not fully comply with DirectX 8 or Pixel Shader 1.0. See article on R100's pixel shaders.
  2. These series do not fully comply with OpenGL 2+ as the hardware does not support all types of non-power-of-two (NPOT) textures.
  3. OpenGL 4+ compliance requires supporting FP64 shaders and these are emulated on some TeraScale chips using 32-bit hardware.

[7][8][9]

References