Report an Error

AMD Radeon Pro Vega 16

Vega 12
Graphics Processor
1024
Cores
64
TMUs
32
ROPs
4 GB
Memory Size
HBM2
Memory Type
1024 bit
Bus Width
AMD Radeon Pro Vega 16 Photo
The Radeon Pro Vega 16 is a professional graphics card by AMD, launched in November 2018. Built on the 14 nm process, and based on the Vega 12 graphics processor, the card supports DirectX 12.0. Unlike the fully unlocked Radeon Pro Vega 20, which uses the same GPU but has all 1280 shaders enabled, AMD has disabled some shading units on the Radeon Pro Vega 16 to reach the product's target shader count. It features 1024 shading units, 64 texture mapping units and 32 ROPs. AMD has placed 4,096 MB HBM2 memory on the card, which are connected using a 1024-bit memory interface. The GPU is operating at a frequency of 300 MHz, which can be boosted up to 1190 MHz, memory is running at 1200 MHz.
Being a igp card, its power draw is rated at 75 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Radeon Pro Vega 16 is connected to the rest of the system using a PCI-Express 3.0 x16 interface.

Graphics Processor

GPU Name
Vega 12
Architecture
GCN 5.0
Foundry
TSMC
Process Size
14 nm
Transistors
unknown
Die Size
unknown

Graphics Card

Release Date
Nov 14th, 2018
Generation
Radeon Pro
(Pro Vega)
Production
Active
Bus Interface
PCIe 3.0 x16

Relative Performance

6%
8%
9%
12%
14%
15%
18%
24%
24%
26%
28%
28%
32%
34%
37%
38%
40%
41%
44%
46%
47%
48%
50%
57%
58%
59%
59%
59%
60%
61%
64%
67%
69%
71%
72%
72%
73%
73%
73%
75%
78%
80%
86%
87%
87%
91%
91%
95%
100%
Radeon Pro Vega 16
101%
104%
105%
106%
107%
111%
112%
113%
115%
120%
123%
124%
125%
125%
125%
135%
136%
140%
141%
142%
143%
144%
146%
146%
151%
162%
166%
170%
171%
173%
175%
179%
184%
188%
193%
195%
195%
216%
238%
241%
258%
259%
262%
266%
268%
270%
273%
275%
276%
278%
293%
299%
303%
303%
333%
339%
345%
371%
407%
422%
454%
454%
479%
535%
560%
575%
627%
713%
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
300 MHz
Boost Clock
1190 MHz
Memory Clock
1200 MHz
2400 MHz effective

Memory

Memory Size
4 GB
Memory Type
HBM2
Memory Bus
1024 bit
Bandwidth
307.2 GB/s

Render Config

Shading Units
1024
TMUs
64
ROPs
32
Compute Units
16

Theoretical Performance

Pixel Rate
38.08 GPixel/s
Texture Rate
76.16 GTexel/s
FP16 (half) performance
4,874 GFLOPS (2:1)
FP32 (float) performance
2,437 GFLOPS
FP64 (double) performance
152.3 GFLOPS (1:16)

Board Design

Slot Width
IGP
TDP
75 W
Outputs
No outputs

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.1.82
Shader Model
6.1

Vega 12 GPU Notes

Graphics/Compute: GFX9 (gfx904)
Display Core Engine: 12.0
Unified Video Decoder: 7.0
Video Compression Engine: 4.0
ROCm Support