AMD’s Raja Koduri Celebrates Vega 10 GPU Development Milestone – Next-Gen, HBM2 Powered Chips For Radeon 2017 Family

Author Photo
Jun 23, 2016
32Shares
Submit

According to a tweet by Raja Koduri (SVP and Chief Architect of Radeon Technologies Group, AMD has achieved a development milestone for their upcoming Vega 10 GPU series.  The tweet by Raja confirms that AMD is on track with the development of their next-generation graphics products which are expected to launch next year in the Radeon family.

AMD Celebrates Vega 10 GPU Development Milestone – Powering High-End Radeon Cards in 2017

The tweet confirms that Raja Koduri is in Shanghai this week along with the GPU design team who developed the upcoming Polaris GPUs and the next-generation Vega GPUs. The design team is already celebrating the first Vega 10 GPU milestone which shows that the development process is going smooth and steady but Raja also states  that it will be a long way to go before you see it. We already know from AMD’s official slides that Vega GPUs are planned for launch next year as Polaris will be taking over the mainstream market in 2016.

Following is a picture of the GPU design team over at Radeon Technologies Group who are responsible for the development of Polaris and Vega GPUs. AMD has most of their GPU design team concentrated in Shanghai in mainland China and that’s where most of the development and engineering process is handled.

AMD Vega 10 GPU Features HBM2 – High-Performance Graphics Cards To Replace Fury Series Next Year

AMD’s goal with Vega 10 would be to replace their Fiji based Fury series with high-performance offerings that feature the second generation High-Bandwidth memory solution. The Vega series of GPUs are based on brand new technology and architecture designs. You may call it a step better than Polaris which itself is a 2.8x leap on AMD’s older 28nm products. Vega is going to deliver better performance per watt than Polaris due to a efficient memory architecture and a fine tuned 14nm FinFET process.

AMD’s latest GPU architecture roadmap confirms Vega in 2017 and Navi in 2018.

We have little details regarding Vega 10 at the moment but previous leaks suggest that the will feature 4096 stream processors. These are not the current generation stream processors but utilize the advancements made in the IP v9.0 generation of graphics SOCs under development by AMD. It is also noted that this chip is the “Leading Chip” of the first graphic IP v9.0 chip generation. We can expect to see a liquid cooled design and a new dual GPU offering under the Vega family when it launches in 2017.

AMD Radeon Graphics Architectures

WCCFTech Year Consumer
Product
GPU Process Transistors In Billions Performance Memory Bandwidth
Southern Islands 2012 HD 7970 Ghz Tahiti 28nm 4.3 4.1 TFLOPS 3GB GDDR5 264GB/s
Volcanic Islands 2013 R9 290X Hawaii 28nm 6.2 5.6 TFLOPS 4GB GDDR5 320GB/s
Caribbean Islands 2015 R9 Fury X Fiji 28nm 8.9 8.6 TFLOPS 4GB HBM1 512GB/s
POLARIS 2016 RX 480 Polaris 10 14nm 5.7 5.2 TFLOPS 8GB GDDR5 256 GB/s
VEGA 2017 RX Vega 64 Vega 10 14nm 12.5 13 TFLOPS 8GB HBM2 480GB/s
VEGA Refresh
VEGA 2
2018

2018
TBA

TBA
TBA

Vega 20
12nm

7nm
12.5

TBA
TBA

TBA
TBA

32GB HBM2
TBA

1 TB/s
NAVI 2019 TBA Navi 10 7nm TBA TBA Nextgen Memory TBA

AMD’s Vega is also expected to shake things up in the FirePro and HPC market. Currently, AMD do not see a big success as their counterparts in the professional field but with high-performance product input, things can change. The Vega 10 GPU is expected to be featured in multiple designs ranging from consumer products and all the way up to multi-TFLOPs FirePro graphics cards and even HPC APUs which could feature the full Vega 10 ASIC on the same piece of silicon.

AMD expects to grow the VR ecosystem  with great software and hardware capabilities.

AMD also wants to deploy a new interconnect fabric during the same time as Opteron Zen and Vega FirePro products are introduced to keep up in the HPC space against the likes of Intel and NVIDIA who are quite dominant in this sector.

The GPUs arriving in 2017 will also be the first to deploy HBM2 memory which means that current GPUs will have to stick with either GDDR5. We know that AMD is using GDDR5 memory on their fastest Polaris 10 chip hence it is possible that AMD is reserving HBM2 for their compute oriented cards and makes sense since their collaborative partner,SK Hynix, also plans to begin HBM2 memory production in Q3 2016.

GPU Family AMD Vega AMD Navi NVIDIA Pascal NVIDIA Volta
Flagship GPU Vega 10 Navi 10 NVIDIA GP100 NVIDIA GV100
GPU Process 14nm FinFET 7nm FinFET TSMC 16nm FinFET TSMC 12nm FinFET
GPU Transistors 15-18 Billion TBC 15.3 Billion 21.1 Billion
GPU Cores (Max) 4096 SPs TBC 3840 CUDA Cores 5376 CUDA Cores
Peak FP32 Compute 13.0 TFLOPs TBC 12.0 TFLOPs >15.0 TFLOPs (Full Die)
Peak FP16 Compute 25.0 TFLOPs TBC 24.0 TFLOPs 120 Tensor TFLOPs
VRAM 16 GB HBM2 TBC 16 GB HBM2 16 GB HBM2
Memory (Consumer Cards) HBM2 HBM3 GDDR5X GDDR6
Memory (Dual-Chip Professional/ HPC) HBM2 HBM3 HBM2 HBM2
HBM2 Bandwidth 484 GB/s (Frontier Edition) >1 TB/s? 732 GB/s (Peak) 900 GB/s
Graphics Architecture Next Compute Unit (Vega) Next Compute Unit (Navi) 5th Gen Pascal CUDA 6th Gen Volta CUDA
Successor of (GPU) Radeon RX 500 Series Radeon RX 600 Series GM200 (Maxwell) GP100 (Pascal)
Launch 2017 2019 2016 2017
Submit