Report an Error

AMD Radeon Pro 580X

Graphics Processor
Ellesmere
Cores
2304
TMUs
144
ROPs
32
Memory Size
8 GB
Memory Type
GDDR5
Bus Width
256 bit
The Radeon Pro 580X is a professional mobile graphics chip by AMD, launched in March 2019. Built on the 14 nm process, and based on the Ellesmere graphics processor, in its Ellesmere XTA variant, the chip supports DirectX 12. The Ellesmere graphics processor is an average sized chip with a die area of 232 mm² and 5,700 million transistors. It features 2304 shading units, 144 texture mapping units, and 32 ROPs. AMD has paired 8 GB GDDR5 memory with the Radeon Pro 580X, which are connected using a 256-bit memory interface. The GPU is operating at a frequency of 1100 MHz, which can be boosted up to 1200 MHz, memory is running at 1710 MHz (6.8 Gbps effective).
Its power draw is rated at 185 W maximum. Display outputs include: 2x HDMI. Radeon Pro 580X is connected to the rest of the system using a PCI-Express 3.0 x16 interface.

Graphics Processor

GPU Name
Ellesmere
GPU Variant
Ellesmere XTA
(216-0886220)
Architecture
GCN 4.0
Foundry
GlobalFoundries
Process Size
14 nm
Transistors
5,700 million
Die Size
232 mm²

Mobile Graphics

Release Date
Mar 18th, 2019
Generation
Radeon Pro Mac
(500X Series)
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
5%
Radeon HD 6450
6%
GeForce GT 520
7%
GeForce GT 220
9%
GeForce GT 430
9%
Radeon HD 5570
10%
Radeon HD 4670
10%
GeForce GT 440
11%
GeForce GT 240
12%
GeForce 9600 GT
13%
Radeon HD 5670
14%
GeForce GT 640
14%
GeForce 9800 GT
15%
Radeon HD 6670
16%
Radeon HD 4830
17%
Radeon HD 4770
17%
Radeon HD 4850
18%
GeForce GTS 250
18%
GeForce GTS 450
19%
Radeon HD 5750
22%
Radeon HD 7750
22%
Radeon HD 5770
22%
GeForce GTX 260
23%
Radeon HD 4870
23%
Radeon Vega 8
23%
GeForce GTX 550 Ti
23%
GeForce GTX 650
24%
Radeon HD 5830
26%
Radeon HD 6790
26%
Radeon HD 4890
27%
GeForce GTX 460
27%
Radeon RX Vega 11
28%
GeForce GT 1030
28%
GeForce GTX 275
28%
Radeon HD 7770 GHz Edition
28%
GeForce GTX 280
28%
GeForce GTX 465
30%
Radeon HD 6850
30%
GeForce GTX 285
33%
Radeon HD 5850
33%
GeForce GTX 650 Ti
33%
Radeon HD 7790
35%
Radeon RX 550
35%
GeForce GTX 470
36%
Radeon HD 6870
38%
Radeon HD 5870
40%
GeForce GTX 560 Ti
40%
Radeon HD 4870 X2
41%
GeForce GTX 750 Ti
41%
Radeon HD 6950
42%
GeForce GTX 295
43%
GeForce GTX 650 Ti Boost
43%
Radeon HD 7850
44%
GeForce GTX 480
46%
Radeon HD 6970
47%
Radeon R7 265
47%
Radeon RX 460
47%
Radeon R7 370
48%
GeForce GTX 660
48%
GeForce GTX 570
49%
Radeon RX 560
52%
GeForce GTX 950
52%
Radeon HD 7870 GHz Edition
53%
Radeon R9 270X
54%
GeForce GTX 660 Ti
54%
GeForce GTX 580
55%
GeForce GTX 1050
55%
Radeon HD 7950
56%
Radeon HD 5970
57%
GeForce GTX 760
62%
GeForce GTX 670
63%
GeForce GTX 960
65%
Radeon R9 380
65%
Radeon R9 285
66%
Radeon HD 7970
67%
GeForce GTX 680
68%
GeForce GTX 1050 Ti
70%
GeForce GTX 770
72%
Radeon HD 6990
73%
Radeon R9 280X
74%
GeForce GTX 590
74%
Radeon HD 7970 GHz Edition
82%
GeForce GTX 780
85%
GeForce GTX 1650
91%
Radeon RX 470
92%
Radeon R9 290
97%
Radeon RX 570
98%
Radeon R9 390
99%
Radeon R9 290X
100%
GeForce GTX 970
100%
Radeon Pro 580X
101%
GeForce GTX TITAN
103%
GeForce GTX 780 Ti
104%
Radeon R9 390X
105%
Radeon HD 7990
105%
Radeon RX 480
106%
GeForce GTX 690
109%
GeForce GTX 1060 6 GB
110%
Radeon RX 5500 OEM
112%
Radeon RX 580
113%
Radeon RX 5500 XT
114%
GeForce GTX 980
115%
GeForce GTX 1650 SUPER
115%
Radeon R9 FURY
121%
Radeon RX 590
127%
Radeon R9 295X2
128%
GeForce GTX 1660
129%
Radeon R9 FURY X
131%
GeForce GTX 980 Ti
135%
GeForce GTX TITAN X
143%
GeForce GTX 1660 SUPER
147%
GeForce GTX 1660 Ti
147%
GeForce GTX 1070
156%
Radeon RX Vega 56
166%
GeForce GTX 1070 Ti
170%
Radeon RX Vega 64
173%
GeForce RTX 2060
173%
GeForce GTX 1080
180%
Radeon RX 5700
194%
GeForce RTX 2060 SUPER
201%
GeForce RTX 2070
206%
Radeon RX 5700 XT
208%
Radeon VII
213%
TITAN X Pascal
220%
GeForce GTX 1080 Ti
222%
GeForce RTX 2070 SUPER
236%
GeForce RTX 2080
251%
GeForce RTX 2080 SUPER
277%
GeForce RTX 2080 Ti
279%
GeForce RTX 3070
365%
GeForce RTX 3080
402%
GeForce RTX 3090
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for 2080 Ti and faster.
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

Base Clock
1100 MHz
Boost Clock
1200 MHz
Memory Clock
1710 MHz
6.8 Gbps effective

Memory

Memory Size
8 GB
Memory Type
GDDR5
Memory Bus
256 bit
Bandwidth
218.9 GB/s

Render Config

Shading Units
2304
TMUs
144
ROPs
32
Compute Units
36
L1 Cache
16 KB (per CU)
L2 Cache
2 MB

Theoretical Performance

Pixel Rate
38.40 GPixel/s
Texture Rate
172.8 GTexel/s
FP16 (half) performance
5.530 TFLOPS (1:1)
FP32 (float) performance
5.530 TFLOPS
FP64 (double) performance
345.6 GFLOPS (1:16)

Board Design

Slot Width
IGP
TDP
185 W
Outputs
2x HDMI
Power Connectors
None

Graphics Features

DirectX
12 (12_0)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2
Shader Model
6.4

Ellesmere GPU Notes

Architecture Codename: Arctic Islands
Codename: Polaris 10
Old Codename: Cozumel / Hawaii Refresh
CLRX Version: GCN 1.2
Graphics/Compute: GFX8 (gfx803)
Display Core Engine: 11.2
Unified Video Decoder: 6.3
Video Compression Engine: 3.4