Report an Error

AMD Radeon Pro Duo

Capsaicin x2
Graphics Processor
4096 x2
Cores
256 x2
TMUs
64 x2
ROPs
4096 MB x2
Memory Size
HBM
Memory Type
4096 bit x2
Bus Width
AMD Radeon Pro Duo Photo AMD Capsaicin Photo
The Radeon Pro Duo is a enthusiast-class professional graphics card by AMD, launched in April 2016. Built on the 28 nm process, and based on the Capsaicin graphics processor, in its Capsaicin XT variant, the card supports DirectX 12.0. The Capsaicin graphics processor is a large chip with a die area of 596 mm² and 8,900 million transistors. Radeon Pro Duo combines two graphics processors to increase performance. It features 4096 shading units, 256 texture mapping units and 64 ROPs per GPU. AMD has placed 8,192 MB HBM memory on the card, which are connected using a 4096-bit memory interface per GPU (each GPU manages 4,096 MB). The GPU is operating at a frequency of 1000 MHz, memory is running at 500 MHz.
Being a dual-slot card, the AMD Radeon Pro Duo draws power from 3x 8-pin power connectors, with power draw rated at 350 W maximum. Display outputs include: 1x HDMI, 3x DisplayPort. Radeon Pro Duo is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card measures 277 mm in length, and features a dual-slot cooling solution. Its price at launch was 1499 US Dollars.

Graphics Processor

GPU Name
Capsaicin
GPU Variant
Capsaicin XT
Architecture
GCN 3.0
Process Size
28 nm
Transistors
8,900 million
Die Size
596 mm²

Graphics Card

Release Date
Apr 26th, 2016
Generation
Radeon Pro
(Fiji)
Production
Active
Launch Price
1,499 USD
Bus Interface
PCIe 3.0 x16

Relative Performance

2%
2%
3%
4%
4%
5%
6%
7%
7%
8%
9%
9%
10%
11%
11%
12%
12%
13%
13%
14%
15%
15%
15%
18%
18%
18%
18%
18%
19%
19%
20%
21%
21%
22%
22%
22%
23%
23%
23%
23%
24%
25%
27%
27%
27%
28%
28%
29%
31%
32%
32%
33%
33%
34%
35%
35%
36%
37%
38%
39%
39%
39%
39%
42%
42%
43%
44%
44%
44%
44%
45%
45%
47%
50%
51%
53%
53%
54%
54%
55%
57%
58%
60%
60%
60%
67%
74%
75%
80%
80%
81%
82%
83%
83%
85%
85%
85%
86%
91%
93%
94%
94%
100%
Radeon Pro Duo
103%
105%
107%
115%
126%
131%
141%
141%
148%
174%
178%
194%
221%
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
1000 MHz
Memory Clock
500 MHz
1000 MHz effective

Memory

Memory Size
4096 MB
Memory Type
HBM
Memory Bus
4096 bit
Bandwidth
512.0 GB/s

Render Config

Shading Units
4096
TMUs
256
ROPs
64
Compute Units
64

Theoretical Performance

Pixel Rate
64.00 GPixel/s
Texture Rate
256.0 GTexel/s
FP16 (half) performance
8,192 GFLOPS (1:1)
FP32 (float) performance
8,192 GFLOPS
FP64 (double) performance
512.0 GFLOPS (1:16)

Board Design

Slot Width
Dual-slot
Length
10.9 inches
277 mm
TDP
350 W
Outputs
1x HDMI
3x DisplayPort
Power Connectors
3x 8-pin
Board Number
C888

Graphics Features

DirectX
12.0 (12_0)
OpenGL
4.5
OpenCL
2.0
Vulkan
1.1.82
Shader Model
6.0

Capsaicin GPU Notes

Chip Variant: Fiji
Graphics/Compute: GFX8 (gfx803)
Display Core Engine: 10.0
Unified Video Decoder: 6.0
Video Compression Engine: 3.0