Report an Error

AMD Radeon Pro 575

Graphics Processor
Polaris 20
Cores
2048
TMUs
128
ROPs
32
Memory Size
4 GB
Memory Type
GDDR5
Bus Width
256 bit
The Radeon Pro 575 is a professional graphics card by AMD, launched in June 2017. Built on the 14 nm process, and based on the Polaris 20 graphics processor, in its Polaris 20 XL variant, the card supports DirectX 12.0. The Polaris 20 graphics processor is an average sized chip with a die area of 232 mm² and 5,700 million transistors. Unlike the fully unlocked Radeon RX 580X, which uses the same GPU but has all 2304 shaders enabled, AMD has disabled some shading units on the Radeon Pro 575 to reach the product's target shader count. It features 2048 shading units, 128 texture mapping units and 32 ROPs. AMD has placed 4,096 MB GDDR5 memory on the card, which are connected using a 256-bit memory interface. The GPU is operating at a frequency of 1096 MHz, memory is running at 1695 MHz.
Being a igp card, the AMD Radeon Pro 575 does not require any additional power connector, its power draw is rated at 120 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Radeon Pro 575 is connected to the rest of the system using a PCI-Express 3.0 x16 interface.

Graphics Processor

GPU Name
Polaris 20
GPU Variant
Polaris 20 XL
(215-0910052)
Architecture
GCN 4.0
Foundry
GlobalFoundries
Process Size
14 nm
Transistors
5,700 million
Die Size
232 mm²

Graphics Card

Release Date
Jun 5th, 2017
Generation
Radeon Pro
(500)
Production
Active
Bus Interface
PCIe 3.0 x16

Relative Performance

2%
GeForce 210
3%
GeForce 9400 GT
4%
Radeon HD 4550
5%
Radeon HD 5450
6%
Radeon HD 6450
6%
GeForce GT 520
8%
GeForce GT 220
10%
GeForce GT 430
10%
Radeon HD 5570
11%
Radeon HD 4670
12%
GeForce GT 440
12%
GeForce GT 240
14%
GeForce 9600 GT
15%
Radeon HD 5670
16%
GeForce GT 640
16%
GeForce 9800 GT
17%
Radeon HD 6670
17%
Radeon HD 4830
18%
Radeon HD 4770
19%
Radeon HD 4850
20%
GeForce GTS 250
20%
GeForce GTS 450
21%
Radeon HD 5750
24%
Radeon HD 7750
24%
Radeon HD 5770
25%
GeForce GTX 260
25%
Radeon HD 4870
25%
Radeon Vega 8
25%
GeForce GTX 550 Ti
26%
GeForce GTX 650
27%
Radeon HD 5830
29%
Radeon HD 6790
29%
Radeon HD 4890
30%
GeForce GTX 460
30%
Radeon RX Vega 11
31%
GeForce GT 1030
31%
GeForce GTX 275
31%
Radeon HD 7770 GHz Edition
31%
GeForce GTX 280
32%
GeForce GTX 465
33%
Radeon HD 6850
34%
GeForce GTX 285
36%
Radeon HD 5850
37%
GeForce GTX 650 Ti
37%
Radeon HD 7790
39%
Radeon RX 550
39%
GeForce GTX 470
40%
Radeon HD 6870
43%
Radeon HD 5870
44%
GeForce GTX 560 Ti
44%
Radeon HD 4870 X2
45%
GeForce GTX 750 Ti
45%
Radeon HD 6950
47%
GeForce GTX 295
47%
GeForce GTX 650 Ti Boost
48%
Radeon HD 7850
49%
GeForce GTX 480
51%
Radeon HD 6970
52%
Radeon R7 265
53%
Radeon R7 370
53%
Radeon RX 560
53%
GeForce GTX 660
53%
GeForce GTX 570
57%
GeForce GTX 950
58%
Radeon HD 7870 GHz Edition
59%
Radeon R9 270X
60%
GeForce GTX 660 Ti
60%
GeForce GTX 580
61%
GeForce GTX 1050
61%
Radeon HD 7950
62%
Radeon RX 460
62%
Radeon HD 5970
64%
GeForce GTX 760
69%
GeForce GTX 670
70%
GeForce GTX 960
72%
Radeon R9 380
73%
Radeon R9 285
73%
Radeon HD 7970
74%
GeForce GTX 680
76%
GeForce GTX 1050 Ti
78%
GeForce GTX 770
80%
Radeon HD 6990
81%
Radeon R9 280X
83%
GeForce GTX 590
83%
Radeon HD 7970 GHz Edition
92%
GeForce GTX 780
100%
Radeon Pro 575
101%
Radeon RX 470
102%
Radeon R9 290
109%
Radeon R9 390
110%
Radeon R9 290X
111%
GeForce GTX 970
113%
GeForce GTX TITAN
114%
Radeon RX 570
114%
GeForce GTX 780 Ti
116%
Radeon R9 390X
116%
Radeon HD 7990
117%
Radeon RX 480
118%
GeForce GTX 690
124%
Radeon RX 580
127%
GeForce GTX 980
128%
GeForce GTX 1060 6 GB
128%
Radeon R9 FURY
129%
Radeon RX 590
141%
Radeon R9 295X2
144%
Radeon R9 FURY X
146%
GeForce GTX 980 Ti
157%
GeForce GTX TITAN X
165%
GeForce GTX 1660 Ti
165%
GeForce GTX 1070
173%
Radeon RX Vega 56
187%
GeForce GTX 1070 Ti
192%
Radeon RX Vega 64
192%
GeForce RTX 2060
197%
GeForce GTX 1080
226%
Radeon VII
226%
GeForce RTX 2070
237%
TITAN X Pascal
243%
GeForce GTX 1080 Ti
263%
GeForce RTX 2080
304%
GeForce RTX 2080 Ti
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
1096 MHz
Memory Clock
1695 MHz
6780 MHz effective

Memory

Memory Size
4 GB
Memory Type
GDDR5
Memory Bus
256 bit
Bandwidth
217.0 GB/s

Render Config

Shading Units
2048
TMUs
128
ROPs
32
Compute Units
32
L1 Cache
16 KB (per CU)
L2 Cache
2 MB

Theoretical Performance

Pixel Rate
35.07 GPixel/s
Texture Rate
140.3 GTexel/s
FP16 (half) performance
4.489 TFLOPS (1:1)
FP32 (float) performance
4.489 TFLOPS
FP64 (double) performance
280.6 GFLOPS (1:16)

Board Design

Slot Width
IGP
TDP
120 W
Outputs
No outputs
Power Connectors
None

Graphics Features

DirectX
12.0 (12_0)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.1.101
Shader Model
6.0

Polaris 20 GPU Notes

Architecture Codename: Arctic Islands
Codename: Ellesmere
CLRX Version: GCN 1.2
Graphics/Compute: GFX8 (gfx803)
Display Core Engine: 11.2
Unified Video Decoder: 6.3
Video Compression Engine: 3.4