De Wikipedia, la enciclopedia libre
Saltar a navegación Saltar a búsqueda
AMD Radeon Pro
Radeon Pro Logo
Firma de diseñoMicro dispositivos avanzados
TipoEstaciones de trabajo profesionales

Radeon Pro es la marca de AMD de GPU de orientación profesional . Reemplazó a la marca FirePro de AMD en 2016. En comparación con la marca Radeon para productos de consumo / jugadores convencionales, la marca Radeon Pro está diseñada para su uso en estaciones de trabajo y la ejecución de diseño asistido por computadora (CAD), imágenes generadas por computadora (CGI), creación de contenido digital (DCC), computación de alto rendimiento / aplicaciones GPGPU y creación y ejecución de programas y juegos de realidad virtual . [1]

La línea de productos Radeon Pro compite directamente con Nvidia , es decir, su línea Quadro (ya descontinuada) de tarjetas profesionales para estaciones de trabajo. [2]

Productos [ editar ]

Artículos principales: Lista de GPU de estaciones de trabajo AMD y Lista de GPU de estaciones de trabajo móviles AMD

Radeon Pro Duo 2016 [ editar ]

La primera tarjeta que se lanzó con el nombre de Radeon Pro fue la GPU Radeon Pro Duo dual en abril de 2016. La tarjeta cuenta con 2 núcleos R9 Nano refrigerados por líquido y se comercializó fuertemente tanto para la ejecución como para la creación de contenido de realidad virtual con el lema "For Gamers Who Create y Creators Who Game ". [3] [4] La estética y el marketing del Pro Duo siguen a los del resto de los productos Fury de la serie 300 .

Radeon Pro SSG [ editar ]

Fiji Radeon Pro SSG [ editar ]

Utilizando la arquitectura GCN 1.2 de AMD Radeon , la Radeon Pro SSG se presentó en julio de 2016. SSG significa S olid S tate G raphics, y la tarjeta acoplará el núcleo Fiji de AMD con almacenamiento de estado sólido para aumentar el búfer de fotogramas para el renderizado. Por lo tanto, esta expansión del almacenamiento de acceso rápido aliviará el problema de la latencia que ocurre cuando una GPU tiene que recuperar información de un dispositivo de almacenamiento masivo a través de la CPU cuando la VRAM limitada de una tarjeta está al máximo en cargas de trabajo pesadas. [5] Los usuarios podrán agregar hasta 1 TB de memoria flash PCIe M.2 NANDpara mejorar los tiempos de renderizado y fregado. [6] AMD demostró un aumento de 5,3 veces en el rendimiento en la depuración de video 8K . [7] Este espacio de almacenamiento SSD puede estar disponible para el sistema operativo o controlado por completo por la GPU. [8] La tarjeta Radeon Pro SSG basada en Fiji estaba disponible como programa beta. [9] [10]

Vega Radeon Pro SSG [ editar ]

En julio de 2017, AMD lanzó la Radeon Pro SSG basada en Vega . [11] La tarjeta utiliza 16 GB de memoria de alto ancho de banda ECC de segunda generación (HBM2), una actualización de los 4 GB de memoria HBM de primera generación de la tarjeta basada en Fiji. La tarjeta Vega también aumentó el almacenamiento de estado sólido integrado a 2TB.

Serie Radeon Pro WX [ editar ]

La serie Radeon Pro WX son tarjetas gráficas diseñadas específicamente para aplicaciones profesionales utilizadas en ingeniería, diseño, creación de contenido y ciencia. Las primeras tarjetas Radeon Pro con el prefijo WX que se anunciaron fueron la WX 7100 , la WX 5100 y la WX 4100 en julio de 2016. [2] Estas tarjetas basadas en Polaris están dirigidas una vez más al mercado profesional tradicional y están preparadas para reemplazar las Serie FirePro Wx100 y serie FirePro Wx300 . Estas tarjetas, junto con el Pro SSG, utilizarán el nuevo YInMn Blue , no tóxico y energéticamente eficiente , descubierto por Mas Subramanian.. Esta estética única de la línea Radeon Pro distinguirá los productos profesionales de la serie Radeon de consumo . [12]

La tarjeta más pequeña, la WX 4100 de media altura , se comercializa para su uso en estaciones de trabajo de factor de forma pequeño . [13] Diseñado para motores de contenido en tiempo real y fabricación CAD y CAM, el WX 5100 encaja entre el WX 4100 y el WX 7100 en términos de rendimiento, y este último se comercializa una vez más con énfasis en la aplicación de realidad virtual y otros medios. creación, mientras afirma ser "La solución de estación de trabajo más asequible". [1]

En junio de 2017, AMD anunció la incorporación de las tarjetas WX 2100 y WX 3100 de menor potencia a la serie Radeon Pro WX. [14] Ambas tarjetas se basan en la GPU Polaris y tienen una potencia nominal de 1,25 TFLOPS . El WX 2100 tiene 2 GB de GDDR5 SDRAM , mientras que el WX 3100 tiene 4 GB de memoria GDDR5.

En septiembre de 2017, AMD lanzó el WX 9100 basado en la arquitectura Vega. La tarjeta cuenta con 16 GB de memoria ECC HBM2 y tiene una capacidad nominal de 12,29 TFLOPS. [15] Como nuevo buque insignia de la línea WX, supera con creces el rendimiento del antiguo WX 7100, que tiene una potencia de 5,73 TFLOPS. [16] El WX 9100 tiene controladores certificados por ISV (proveedor de software independiente) para aplicaciones profesionales que incluyen Siemens NX, PTC Creo, Dassault Systèmes CATIA y 3DExperience Platform, Dassault Systèmes SOLIDWORKS y Autodesk® Revit®. [15] El WX 9100 es particularmente adecuado para cargas de trabajo de misión crítica y modelado científico complejo porque la memoria ECC ayuda a corregir "errores de bit simple o doble como resultado de la radiación de fondo que ocurre naturalmente". [15]

Serie Radeon Pro 400 [ editar ]

Las piezas de Mobile Radeon Pro se revelaron por primera vez con el lanzamiento de la actualización de 2016 de la MacBook Pro de Apple de 15 " . [17] Parecen ser piezas derivadas de Polaris 11 con unidades de cómputo GCN de cuarta generación 10-16 , que proporcionan entre 1 y 1,86 TFLOPS de desempeño. [18] [19]

Radeon Pro Duo 2017 [ editar ]

In April 2017 AMD announced a new version of the Radeon Pro Duo for release the following month.[20] The newer version of the Pro Duo utilizes dual GPUs from the Polaris architecture, using the same GPUs as in the WX7100. While this results a smaller number of compute units and lower theoretical performance, it allows for the inclusion of 32GB GDDR5 SDRAM and a lower board power.

Radeon Vega Frontier Edition[edit]

AMD announced in May 2017 the Radeon Vega Frontier Edition, for release in June of that year.[21][22] While not branded as a Pro product, the card is marketed within the Radeon Pro series.[23] The Radeon Vega Frontier Edition uses the new "Next-Gen Compute Unit" and 16GB of HBM2 memory for an expected 13.1 TFLOPs of single precision and 26.2 TFLOPs of half precision performance. Ultimately, two Frontier Edition products were released with either air or liquid cooling.[24] The liquid cooling part supported a higher TDP, and was able to reach and sustain higher clock speeds,[25] but otherwise the two products have similar hardware specifications.

Radeon Pro 500 series[edit]

Released in conjunction with the 2017 Apple iMac refresh, the Radeon Pro 500 series serve as GPUs for the 4K and 5K Retina Display iMacs.[26] The 500 series ranges supports 2 to 8 GB of graphics RAM with performance from 1.3 to 5.5 TFLOPS.

Radeon Pro Vega[edit]

The Radeon Pro Vega product line of GPUs were first announced in 2017 as a part of Apple's iMac Pro. The two models, Radeon Pro Vega 56 and 64, support 8 and 16 GB of HBM2 memory, respectively.[27] On October 30, 2018, Apple added graphics upgrade options for their 15-inch MacBook Pro lineup consisting Radeon Pro Vega 16 and 20. Derived from Vega 12 GPU that was only used on Apple laptops, both GPU features a 4GB HBM2 memory stack and performance up to 3.3 TFLOPS. [28]

The second-generation, 7 nm Radeon Pro Vega II was announced in 2019 as part of Apple's third-generation Mac Pro desktop computer. The Pro Vega II supports 32 GB of HBM2 memory, while the Pro Vega II Duo combines two Vega GPUs and supports 64 GB of HBM2 memory. The Mac Pro supports up to two Pro Vega II or Pro Vega II Duo graphics cards, allowing up to four Vega GPUs and 128 GB of HBM2 memory to be used in a system.[29]

The Radeon Pro VII was announced on May 2020, as a professional variant of the Radeon VII.[30]

Radeon Pro 5000M series[edit]

Released in conjunction with the 2019 Apple 16 inch MacBook Pro.[31] Two models were announced, the 5300M and the 5500M. Both feature GDDR6 memory interfaces, with 192 GB/s bandwidth. The 5500M supports up to 8 GB of GDDR6 and 4.0 TFLOPS.[32] In June 2020, a new 5600M GPU model with 8 GB of HBM2 memory was quietly released.

Radeon Pro W5000 series[edit]

The Radeon Pro W5700, which is based on RDNA Architecture for desktop workstations, was officially released on November 19, 2019.[33] The smaller model Radeon Pro W5500 was released in February 2020.[34]

Software[edit]

Project Loom[edit]

At an AMD event in 2016 Project Loom was announced as a collaboration between AMD and Radiant Images.[35] The real-time GPU accelerated photo and video stitching program will complement AMD's virtual reality development platform. While traditional photo stitching is not that much of a complex task, Project Loom aims to improve render times when tasked with the heavy workload of stitching together multiple high resolution angles to form a 360 degree VR experience, either to headsets or mobile devices.[36] Using AMD's Direct GMA protocol, the software allows Radeon Pro graphics cards to work directly with video capture hardware to stitch together a 30 fps, 360 degree 4k resolution video from 24, 1080p cameras at 60 fps.[37]

The software is to be competitive with Nvidia's VRWorks 360 Video SDK, and is reportedly set to be made open-source through GPUOpen.[38]

ProRender[edit]

The successor to FireRender, Radeon ProRender works with high-end graphics programs as an OpenCL photorealistic offline 3D renderer and raytracing engine.[39] ProRender aims to compete with programs such as NVIDIA's Iray and other expensive, proprietary solutions. However, AMD is making ProRender free, open source and available for all graphics hardware.[36] ProRender was released by AMD in June 2016 with support for Blender, 3D Studio Max, SolidWorks, and Maya.[40]

Driver[edit]

API OpenGL 4.5 is supported and 4.6 is in development. API Vulkan 1.0 is supported for all with GCN Architecture. Vulkan 1.1 (GCN 2. Gen. or 1.2 and higher) will be supported with actual drivers in 2018.[41]

As with other GPU architectures, the floating-point performance is dependent on the precision and the GCN generation:

  • In 4th Gen GCN, FP64 is 1/16 of FP32. Newer gaming cards have better ratios, which should be reflected on newer derivative "Pro" versions:
    • The gaming card Radeon R9 295X2 has it bumped up to 1/8 FP32.
    • The gaming card Radeon VII has it bumped up to 1/4 FP32.
    • The Radeon Pro Vega 20 has the ratio bumped up to 1/2 FP32.
  • In 5th Gen GCN, FP16 is double of FP32. In 1st Gen to 4th it was equal to FP32.

For those requiring higher FP64 performance, a form of FP64 distinct from the IEEE double-precision can be emulated with the much faster FP32 operations. The cost is around a ~1/3 performance compared to FP32, much better than what the native support could provide.[42]

Chipset table[edit]

Workstation[edit]

Radeon Pro WX x100, SSG, Duo and V series[edit]

  • v
  • t
  • e
Model
(Codename)		Release Date
& Price		Architecture
& Fab		Transistors
& Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTBP (W)Bus interfaceGraphic output
ports
Config[e]Clock[a] (MHz)Texture
(GT/s)		Pixel
(GP/s)		HalfSingleDouble[f]Bus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro WX 2100
(Lexa)[43][44][45]		June 2017
$149 USD		GCN 4th gen
14 nm		2200×106
103 mm2		512:32:16
8 CU		121939.019.5SP1,25078GDDR5
64-bit		2600048<35PCIe 3.0 ×81× DP 1.4
2× mini-DP 1.4
Radeon Pro WX 3100[43][46]June 2017
$200 USD		512:32:16
8 CU		121939.019.5SP1,25078GDDR5
128-bit		4600096<50PCIe 3.0 ×81× DP 1.4
2× mini-DP 1.4
Radeon Pro WX 4100
(Polaris 11)[47][48][49][50][51][52]		November 2016
$399 USD		3000×106
123 mm2		1024:64:16
16 CU		925
1170		59.2
74.9		14.8
18.7		SP1,890
2,400		118
150		GDDR5
128-bit		4700011250PCIe 3.0 ×84× mini-DP 1.4
Radeon Pro WX 5100
(Polaris 10)[48][49][50][53][52]		November 2016
$499 USD		5700×106
232 mm2		1792:112:32
28 CU		926
1090		103.7
122.1		29.6
34.9		SP3,320
3,910		208
244		GDDR5
256-bit		86600211.275PCIe 3.0 ×164× Display
Port 1.4
Radeon Pro WX 7100
(Polaris 10)[48][49][50][54][55][52]		November 2016
$799 USD		5700×106
232 mm2		2304:144:32
36 CU		900
1240		115.2
158.7		28.8
39.7		SP4,150
5,710		259
357		GDDR5
256-bit		88000256130PCIe 3.0 ×164× Display
Port 1.4
Radeon Pro WX 9100
(Vega)[56][57][58][59][60][61][62]		September 2017
$2,199 USD		GCN 5th gen
14 nm		12500×106
484mm2		4096:256:64[63]
64 CU		1500384.096.024,576
(2× SP)		12,288768
(1⁄16× SP)		HBM2
2048-bit		161890483.84<250PCIe 3.0 ×166× mini-DP 1.4
Radeon Pro SSG
(Fiji)[64]		July 2016
prototype only		GCN 3rd gen
28 nm		8900×106
596 mm2		4096:256:64
64 CU		1000?256?64.0?8,192?8,192?512?HBM + SSG
4096-bit		4
+ 1 TB SSD		1000512200?PCIe 3.0 ×16Un­known
Radeon Pro SSG
(Vega)[65][58][60][61]		September 2017
$6,999 USD		GCN 5th gen
14 nm		12500×106
484 mm2		4096:256:64[63]
64 CU		1500384.096.024,60012,290768HBM2 + SSG
2048-bit		16
+ 2 TB SSD		1890483.84<300PCIe 3.0 ×166× mini-DP 1.4
Radeon Pro Duo
(Polaris 10)[48][49][50][54][55][52][66]		April 2017
$999 USD		GCN 4th gen
14 nm
2×5700×106
232 mm2
2×2304:128:32
36 CU

1243		2× 179.02× 39.78SP2× 5,728
(1⁄16× SP)		GDDR5
256-bit		2×167000224<250PCIe 3.0 ×163× DP 1.4
1× HDMI 2.0
Radeon Pro V340
(Vega)[67]		Q4 2018
TBA		GCN 5th gen
14 nm		12500×106
484 mm2		3584:224:64
56 CU		852
1000		2246414,336
(2× SP)		7,168448
(1⁄16× SP)		HBM2
2048-bit		162000512230PCIe 3.0 ×16no
outputs
Radeon Pro V340 MxGPU
(Vega)[68][69][70]		Q4 2018
TBA
2×12500×106
484 mm2
2×3584:224:64
56 CU
852
1000		2× 2242× 642× 14,336
(2× SP)		2× 7,1682× 448
(1⁄16× SP)		HBM2
2048-bit		2× 162000512300PCIe 3.0 ×16no
outputs
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)
  6. ^ To calculate driver enabled FP64 processing power see driver.


Radeon Vega Frontier Edition series[edit]

  • view
  • talk
  • edit
Model
(Codename)		LaunchArchitecture
(Fab)		Transistors
Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTBPBus interfaceRelease Price (USD)
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)HalfSingleDoubleBus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Vega Frontier Edition
(Air Cooled)[71][72]		27 June 2017GCN 5th gen
(14 nm)		12.5×109
484 mm2		4096:256:641382409.6102.42264311321707.6HBM2
2048-bit		161890483.8300 WPCIe 3.0 ×16$999
Radeon Vega Frontier Edition
(Liquid Cooled)[71][72]		16002621413107819.2350 W$1499
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units


Radeon Pro WX x200[edit]

  • view
  • talk
  • edit
Model
(Codename)		LaunchArchitecture
(Fab)		Transistors
Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(TFLOPS)		MemoryTBP (W)Bus interfaceGraphic output PortRelease Price (USD)
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)HalfSingleDouble[f]Bus type &
width (bit)		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro
WX 3200
(Polaris)[73]		July 2019GCN 4th gen
(14 nm)		640:40:?
(10)		SP1.660.104GDDR5
128		460009650PCIe 3.0 ×164x mini-DP 1.4$200
Radeon Pro
WX 8200
(Vega) [74]		August 2018GCN 5th gen
(14 nm)		12500×106
484mm2		3584:224:64
(56)		1200
1530		342.797.922x SP10.9671/16 SPHBM2
2048		82000512<230PCIe 3.0 ×164x mini-DP 1.4$999
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and (Compute Units)
  6. ^ To calculate driver enabled FP64 processing power see driver.



Radeon Pro Vega (for Apple Mac Pro)[edit]

Model
(Codename)		Release Date
& Price		Architecture
& Fab		Transistors
& Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTDPBus interfaceGraphic output
ports
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)SingleDoubleBus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro Vega II
(Vega 20)[75]		November 2019$2,800 USDGCN 5th gen
7FF		0.0×109
? mm2		4096:256:64
64 CU		170043510914200880HBM2
4096-bit		3220001000UnknownPCIe 3.0 ×164× Thunderbolt 3

1× HDMI 2.0

Radeon Pro Vega II Duo
(Vega 20)[75]		November 2019$5,600 USD
2×4096:256:64
64 CU
17002× 4352× 1092× 142002× 880HBM2
4096-bit		2× 3220001000
  • v
  • t
  • e
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)


Mobile Workstation[edit]

Radeon Pro 400 series[edit]

Model
(Codename)		Release DateArchitecture
& Fab		Transistors
& Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTDPBus interface
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)SingleDoubleBus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro 450 [76][77][78][79][80]October 2016GCN 4th gen
14 nm		3.0×109
123 mm2		640:40:16
10 CU		725
775		29
31		11.6
12.4		928
992		1/16 SPGDDR5
128-bit		250008035 WPCIe 3.0 ×16
Radeon Pro 455 [76][77][78][81]October 2016768:48:16
12 CU		85040.813.61305.61/16 SPGDDR5
128-bit		250008035 W
Radeon Pro 460 [76][77][78][82]October 20161024:64:16
16 CU		850
910		54.4
58.2		13.6
14.56		1740.8
1863.7		1/16 SPGDDR5
128-bit		450008035 W
  • v
  • t
  • e
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)


Radeon Pro 500 series (for Apple iMac & MacBook Pro)[edit]

Model
(Codename)		Release DateArchitecture
& Fab		Transistors
& Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTDPBus interface
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)SingleDoubleBus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro
555
(Polaris 21) [83][84][85][86]		June 2017GCN 4th gen
14 nm		3.0×109
123 mm2		768:48:16
12 CU		85040.813.61306
1/16 SPGDDR5
128-bit		2510081.6050 WPCIe 3.0 ×16
Radeon Pro
555X
(Polaris 21) [83][84][85][87]		July 2018768:48:16
12 CU		90743.5414.511393
1/16 SPGDDR5
128-bit		4510081.6050 W
Radeon Pro
560
(Polaris 21) [83][84][85][88]		June 20171024:64:16
16 CU		90758.0514.511858
1/16 SPGDDR5
128-bit		4508081.2850 W
Radeon Pro
560X
(Polaris 21) [83][84][85][89]		July 20181024:64:16
16 CU		100464.2616.062056
1/16 SPGDDR5
128-bit		4508081.2850 W
Radeon Pro
570
(Polaris 20) [83][84][85][90]		June 20175.7×109
232 mm2		1792:112:32
28 CU		1000112323584
1/16 SPGDDR5
256-bit		46800217.6120 W
Radeon Pro
575
(Polaris 20) [83][84][85][91]		June 20172048:128:32
32 CU		1100140.835.24506
1/16 SPGDDR5
256-bit		46800217.6120 W
Radeon Pro
580
(Polaris 20) [83][84][85][92]		June 20172304:144:32
36 CU		1100
1200
172.8
38.4		5530
1/16 SPGDDR5
256-bit		86800217.0150 W
  • v
  • t
  • e
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)


Radeon Pro 5000M series (for Apple MacBook Pro)[edit]

Model
(Codename)		Release DateArchitecture
& Fab		Transistors
& Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(TFLOPS)		MemoryTDPBus interface
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)SingleDoubleBus type
& width		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro
5300M
(Navi 14)[93][94]		November 2019RDNA
7 nm		6.4×109
155 mm2		1280:80:32
20 CU		1000
1250		100403.2
1/16 SPGDDR6
128-bit		412000192.050 WPCIe 4.0 ×16
Radeon Pro
5500M
(Navi 14)[93][95]		November 20191536:96:32
24 CU		1000
1300		124.841.64.0
1/16 SP4 or 8
Radeon Pro
5600M
(Navi 12)[93]		June 2020RDNA
7 nm		10.3×109
251 mm2		2560:160:64
40 CU		1000
1030		164.865.95.3
1/16 SPHBM2
2048-bit		81540394.050 WPCIe 4.0 ×16
  • v
  • t
  • e
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)


Radeon Pro WX mobile series[edit]

  • Half Precision Power (FP16) is equal single precision power (FP32) in 4th GCN Generation (in 5th Gen: Half Precision (FP16) = 2x SP (FP32))
Model
(Codename)		LaunchArchitecture
(Fab)		Transistors
Die Size		CoreFillrate[a][b][c]Processing power[a][d]
(GFLOPS)		MemoryTDP (W)Bus interfaceRelease Price (USD)
Config[e]Clock[a] (MHz)Texture (GT/s)Pixel (GP/s)SingleDoubleBus type
& width (bit)		Size (GiB)Clock (MT/s)Band-
width (GB/s)
Radeon Pro
WX 2100 (Mobile)
(Polaris 12)[96]		March 2017GCN 4th gen
(14 nm)[97]		2.2×109
101 mm2		512:32:16:8?
??1250
78
1/16 SP		GDDR5
64-bit		260004835PCIe 3.0 ×16$149
Radeon Pro
WX 3100 (Mobile)
(Polaris 12)[98][99]		March 2017512:32:16:8?
??1250
78
1/16 SP		GDDR5
128-bit		460009650$199
Radeon Pro
WX 4130 Mobile
(Polaris 11)[100]		March 20173.0×109
123 mm2		640:40:16:101000
1050		16.8542.121348
84.24
1/16 SP		GDDR5
128-bit		460009650Un­known
Radeon Pro
WX 4150 Mobile
(Polaris 11)[101]		March 2017896:56:16:141000
1050		16.8558.971887
118
1/16 SP		GDDR5
128-bit		460009650Un­known
Radeon Pro
WX 4170 Mobile
(Polaris 11)[102]		March 20171024:64:16:161000
1050		16.8567.392157
135
1/16 SP		GDDR5
128-bit		4700011250$399 See WX4100
Radeon Pro
WX 7100 Mobile
(Polaris 10)[103]		March 20175.7×109
232 mm2		2304:144:32:36118839.78179.05728
358
1/16 SP		GDDR5
256-bit		85000160130$799 See WX7100
  1. ^ a b c Boost values (if available) are stated below the base value in italic.
  2. ^ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
  3. ^ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
  4. ^ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
  5. ^ Unified Shaders : Texture Mapping Units : Render Output Units : Compute Units

See also[edit]

  • AMD FirePro – AMD's predecessor to Radeon Pro
  • Nvidia Quadro – Nvidia's competing workstation graphics solution
  • Nvidia Tesla – Nvidia's competing GPGPU solution
  • List of AMD graphics processing units

References[edit]

  1. ^ a b Amjad, Talha (9 August 2016). "AMD Radeon Rro WX Series GPUs: VR Content Creation And More". Tech Frag. Retrieved 19 August 2016.
  2. ^ a b Ung, Gordon Mah (25 July 2016). "AMD introduces a new Radeon Pro WX series to replace FirePro". PC World. Retrieved 19 August 2016.
  3. ^ "Radeon Pro Duo". AMD. Retrieved 19 August 2016.
  4. ^ Ung, Gordon Mah (14 March 2016). "AMD's $150 dual-GPU Radeon Pro Duo graphics card is built for virtual reality". PC World. Retrieved 19 August 2016.
  5. ^ Evangelho, Jason (26 July 2016). "AMD's Radeon Pro SSG Could Be A Game Changer For Developers And Content Creators". Forbes. Retrieved 19 August 2016.
  6. ^ Walton, Mark (26 July 2016). "AMD unveils Radeon Pro SSG graphics card with up to 1TB of M.2 flash memory". Ars Technica. Retrieved 19 August 2016.
  7. ^ Chiappetta, Marco (25 July 2016). "AMD Unveils Radeon Solid State Storage Architecture And 1TB Radeon Pro SSG For Massive Pro Graphics Datasets". Hot Hardware. Retrieved 26 July 2016.
  8. ^ Alcorn, Paul (8 August 2016). "Examining AMD Radeon Pro SSG: How NAND Changes The GPU Game". Tom's Hardware. Retrieved 8 August 2016.
  9. ^ Smith, Ryan (25 July 2016). "AMD Announces Radeon-pro SSG :Polaris with m.2 SSDs Onboard". Anandtech. Retrieved 26 July 2016.
  10. ^ https://www.amd.com/Documents/Radeon-Pro-SSG-Technical-Brief.pdf
  11. ^ Manion, Wayne (31 July 2017). "Vega goes pro on the Radeon Pro WX 9100 and Radeon Pro SSG". Tech Report. Retrieved 2 August 2017.
  12. ^ "Radeon Pro WX Series and YInMn Blue". YouTube. AMD. 15 August 2016. Retrieved 22 August 2016.
  13. ^ "Radeon Pro Graphics". AMD. Retrieved 19 August 2016.
  14. ^ Killian, Zak (1 June 2017). "AMD Radeon Pro WX 2100 and 3100 fit any workstation". Tech Report. Retrieved 1 June 2017.
  15. ^ a b c "Radeon Pro WX 9100". AMD. Retrieved 8 February 2018.
  16. ^ "Radeon Pro WX 7100". AMD. Retrieved 8 February 2018.
  17. ^ Cunningham, Andrew (27 October 2016). "Apple introduces brand-new 13- and 15-inch MacBook Pros for $1,799 and $2,399". Ars Technica. Retrieved 27 October 2016.
  18. ^ "Radeon Pro". Radeon.com. AMD. Retrieved 27 October 2016.
  19. ^ Kampman, Jeff (27 October 2016). "Radeon Pro specs hint at a full-fat Polaris 11 GPU in MacBook Pros". Tech Report. Retrieved 27 October 2016.
  20. ^ Cunningham, Andrew (25 April 2017). "AMD puts two GPUs and 32GB of RAM on its latest Radeon Pro Duo graphics card". Ars Technica. Retrieved 25 April 2017.
  21. ^ Kampman, Jeff (17 May 2017). "Spitballing the performance of AMD's Radeon Vega Frontier Edition graphics card". Tech Report. Retrieved 24 May 2017.
  22. ^ Smith, Ryan (17 May 2017). "AMD Unveils the Radeon Vega Frontier Edition". Anandtech. Retrieved 24 May 2017.
  23. ^ "Radeon Vega Frontier Edition". AMD. Retrieved 24 May 2017.
  24. ^ Forrest, Derek (27 June 2017). "AMD Vega Frontier Edition Now Available For Pre-Order With $999 Price Tag". Tom's Hardware. Retrieved 27 June 2017.
  25. ^ Shrout, Ryan (17 July 2017). "The AMD Radeon Vega Frontier Edition 16GB Liquid-Cooled Review". PC Perspective. Retrieved 17 July 2017.
  26. ^ Kampman, Jeff (5 June 2017). "iMacs and MacBook Pros take a dip in Kaby Lake". Tech Report. Retrieved 6 June 2017.
  27. ^ Manion, Wayne (12 December 2017). "iMac Pro details and release date come into focus". Tech Report. Retrieved 12 December 2017.
  28. ^ "Remarkable Graphics Performance for Premium Notebooks AMD Radeon Pro Vega 20 and Radeon Pro Vega 16 Graphics". PCWorld. Retrieved 2020-01-11.
  29. ^ Hill, Brandon (June 4, 2019). "AMD Announces Radeon Pro Vega II And Pro Vega II Duo: Up To 28.2 TFLOPS And 64GB HBM2". HotHardware. Retrieved 2019-06-05.
  30. ^ Smith, Ryan. Anandtech https://www.anandtech.com/show/15777/amd-reveals-radeon-pro-vii. Retrieved 11 April 2021. Missing or empty |title= (help)
  31. ^ Axon, Samuel (14 November 2019). "Apple introduces a redesigned, thicker MacBook Pro". Ars Technica. Retrieved 27 November 2019.
  32. ^ "Radeon Pro 5000M Series". AMD. Retrieved 27 November 2019.
  33. ^ https://www.amd.com/en/support/professional-graphics/radeon-pro/radeon-pro-w5000-series/radeon-pro-w5700
  34. ^ https://www.techpowerup.com/gpu-specs/radeon-pro-w5500.c3479
  35. ^ "Radiant Images and AMD Collaborate on Project Loom, a Multi-Cam Real-Time 360 Stitching Platform". radiantimages.com. 21 August 2016. Retrieved 23 September 2016.
  36. ^ a b Demerjian, Charlie (July 25, 2016). "AMD unveils Loom and ProRender software". semiaccurate.com. Retrieved 23 September 2016.
  37. ^ Lang, Ben (July 26, 2016). "AMD Announces Radeon Pro WX 7100 GPU Focused on Professional VR Film Editing". roadtovr.com. Retrieved 23 September 2016.
  38. ^ Kampman, Jeff (July 26, 2016). "Nvidia and AMD ease 360-degree video production with new APIs". techreport.com. Retrieved 23 September 2016.
  39. ^ Killian, Zak (July 26, 2016). "AMD FireRender is now the open-source Radeon ProRender". techreport.com. Retrieved 23 September 2016.
  40. ^ Michaud, Scott (28 June 2017). "AMD Releases Radeon ProRender for Blender and SolidWorks". PC Perspective. Retrieved 30 June 2017.
  41. ^ "The Khronos Group". The Khronos Group. June 4, 2019.
  42. ^ Thall, Andrew (March 2007). Extended-Precision Floating-Point Numbers for GPU Computation (PDF) (Report). Tech. Rep. CIM-007-01.
  43. ^ a b Killian, Zak (1 June 2017). "AMD Radeon Pro WX 2100 and 3100 fit any workstation". Tech Report. Retrieved 1 June 2017.
  44. ^ "Radeon Pro WX 2100". Radeon Technologies. AMD. Retrieved 1 June 2017.
  45. ^ "AMD Radeon Pro WX 2100 Specs". TechPowerUp.
  46. ^ "Radeon Pro WX 3100". Radeon Technologies. AMD. Retrieved 1 June 2017.
  47. ^ "Radeon™ Pro WX 4100 Graphics". Advanced Micro Devices. Retrieved 29 June 2018.
  48. ^ a b c d Mujtaba, Hassan (July 2016). "AMD Announces Radeon Pro WX Series With Polaris GPUs – Radeon Pro WX 7100 Leads The Pack With Polaris 10". wccftech.com. Retrieved 17 September 2016.
  49. ^ a b c d Smith, Ryan (26 July 2016). "AMD Announces Radeon Pro WX Series: WX 4100, WX 5100, & WX 7100 Bring Polaris to Pros". Anandtech. Retrieved 18 September 2016.
  50. ^ a b c d Farrell, Nick (27 July 2016). "AMD releases Radeon Pro WX 4100". techeye.net. Retrieved 18 September 2016.
  51. ^ "AMD Radeon Pro WX 4100". techpowerup.com. Retrieved 4 November 2016.
  52. ^ a b c d Smith, Ryan (7 November 2016). "Now Shipping: AMD Radeon Pro WX Series". Anandtech. Retrieved 7 November 2016.
  53. ^ "AMD Radeon Pro WX 5100". techpowerup.com. Retrieved 4 November 2016.
  54. ^ a b Lang, Ben (26 July 2016). "AMD Announces Radeon Pro WX 7100 GPU Focused on Professional VR Film Editing". Roadtovr.com. Retrieved 18 September 2016.
  55. ^ a b "AMD Radeon Pro WX 7100". techpowerup.com. Retrieved 4 November 2016.
  56. ^ "Radeon™ Pro WX 9100 Graphics". Advanced Micro Devices. Retrieved 29 June 2018.
  57. ^ "Introducing the Radeon Pro WX 9100". pro.radeon.com. Radeon Technologies Group. Retrieved 31 July 2017.
  58. ^ a b "Radeon™ Pro WX 9100 and Radeon™ Pro SSG: Defying Convention to Forge the Future of Professional Content Creation". globenewswire.com. GlobeNewswire, Inc. 30 July 2017. Retrieved 31 July 2017.
  59. ^ Wan, Samuel. "AMD Readying Radeon Pro WX 9100 with Vega GPU". eteknix.com. eteknix. Retrieved 31 July 2017.
  60. ^ a b Manion, Wayne (31 July 2017). "Vega goes pro on the Radeon Pro WX 9100 and Radeon Pro SSG". Tech Report. Retrieved 31 July 2017.
  61. ^ a b Lily, Paul. "AMD put 2TB of memory in a new $7000 graphics card". pcgamer.com. PC Gamer. Retrieved 2 August 2017.
  62. ^ "AMD Radeon Pro WX 9100 Specs". TechPowerUp.
  63. ^ a b Oh, Nate (26 October 2017). "Now Shipping: Radeon Pro WX 9100 and SSG, with New Vega Pro Drivers". Anandtech. Retrieved 26 October 2017.
  64. ^ Walton, Mark (26 July 2016). "AMD unveils Radeon Pro SSG graphics card with up to 1TB of M.2 flash memory". Ars Technica. Retrieved 2 August 2017.
  65. ^ "Introducing the Radeon Pro SSG". pro.radeon.com. Radeon Technologies Group. Retrieved 31 July 2017.
  66. ^ "AMD Radeon Pro Duo Polaris Specs". TechPowerUp.
  67. ^ "AMD Radeon Pro V340 Specs". TechPowerUp.
  68. ^ "V340. Built for Builders". Amd.com. Retrieved 27 August 2018.
  69. ^ "Professional Graphics Specifications". Amd.com. Retrieved 27 August 2018.
  70. ^ Demerjian, Charlie (26 August 2018). "AMD V-Series virtualization GPUs launch with the V340". semiaccurate.com. Retrieved 27 August 2018.
  71. ^ a b "Radeon™ Pro Graphics - Workstation Solutions - AMD". www.amd.com.
  72. ^ a b Shrout, Ryan (17 July 2017). "The AMD Radeon Vega Frontier Edition 16GB Liquid-Cooled Review". PC Perspective. Retrieved 26 July 2017.
  73. ^ "Introducing AMD Radeon™ Pro WX 3200 -- Most cost-effective, ISV-certified AMD graphics card under $200 for small form factor CAD workstations". AMD. Retrieved 3 July 2019.
  74. ^ "AMD Radeon Pro WX 8200 Specs". TechPowerUp.
  75. ^ a b "AMD Radeon Pro Vega II". amd.com. Retrieved 18 Nov 2019.
  76. ^ a b c Cunningham, Andrew (27 October 2016). "Apple introduces brand-new 13- and 15-inch MacBook Pros for $1,799 and $2,399". Ars Technica. Retrieved 27 October 2016.
  77. ^ a b c "Radeon Pro". Radeon.com. AMD. Archived from the original on 28 October 2016. Retrieved 27 October 2016.
  78. ^ a b c Kampman, Jeff (27 October 2016). "Radeon Pro specs hint at a full-fat Polaris 11 GPU in MacBook Pros". Tech Report. Retrieved 27 October 2016.
  79. ^ Smith, Ryan (27 October 2016). "Apple Announces 4th Generation MacBook Pro Family: Thinner, Lighter, with Thunderbolt 3 & "Touchbar"". Anandtech. Retrieved 27 October 2016.
  80. ^ "AMD Radeon Pro 450". notebookcheck.com. Retrieved 22 May 2017.
  81. ^ "AMD Radeon Pro 455". notebookcheck.com. Retrieved 22 May 2017.
  82. ^ "AMD Radeon Pro 460". notebookcheck.com. Retrieved 22 May 2017.
  83. ^ a b c d e f g "Introducing Radeon Pro". radeon.com. Archived from the original on 10 June 2017. Retrieved 6 June 2017.
  84. ^ a b c d e f g Kampman, Jeff (5 June 2017). "iMacs and MacBook Pros take a dip in Kaby Lake". Tech Report. Retrieved 6 June 2017.
  85. ^ a b c d e f g Manion, Wayne (6 June 2017). "AMD spills the beans on Radeon Pro chips inside the new Macs". Tech Report. Retrieved 6 June 2017.
  86. ^ "AMD Radeon Pro 555 Specs". TechPowerUp.
  87. ^ "AMD Radeon Pro 555X Specs". TechPowerUp.
  88. ^ "AMD Radeon Pro 560 Specs". TechPowerUp.
  89. ^ "AMD Radeon Pro 560X Specs". TechPowerUp.
  90. ^ "AMD Radeon Pro 570 Specs". TechPowerUp.
  91. ^ "AMD Radeon Pro 575 Specs". TechPowerUp.
  92. ^ "AMD Radeon Pro 580 Specs". TechPowerUp.
  93. ^ a b c "Radeon Pro 5000M Series". amd.com. Retrieved 27 November 2019.
  94. ^ "AMD Radeon Pro 5300M Specs". TechPowerUp.
  95. ^ "AMD Radeon Pro 5500M Specs". TechPowerUp.
  96. ^ https://www.amd.com/en/products/professional-graphics/radeon-pro-wx-2100-mobile
  97. ^ Hinum, Klaus. "AMD's Polaris Architecture". Notebookcheck.
  98. ^ https://www.amd.com/en/products/professional-graphics/radeon-pro-wx-3100-mobile
  99. ^ https://www.notebookcheck.net/AMD-Radeon-Pro-WX-3100-Mobile-GPU.283050.0.html
  100. ^ "AMD Radeon Pro WX 4130 Mobile Specs". TechPowerUp.
  101. ^ "AMD Radeon Pro WX 4150 Mobile Specs". TechPowerUp.
  102. ^ "AMD Radeon Pro WX 4170 Mobile Specs". TechPowerUp.
  103. ^ "AMD Radeon Pro WX 7100 Mobile Specs". TechPowerUp.

External links[edit]

  • AMD Radeon Pro Graphics
  • Radeon Pro
  • Radeon ProRender Technology