AMD has just launched their first graphics cards based on the Polaris architecture and we will be looking at even more cards in the coming weeks. While Polaris is great for the mainstream side of things, AMD is expected to launch their next-generation Vega 10 GPU in the first half of 2017 which will be featured on very powerful cards aimed at the high-end and enthusiast sector of the PC gaming community.
AMD Vega 10 GPUs Rumored For Launch in First Half of 2017 - HBM2 For The High-End Radeons
The latest rumor regarding Vega 10 comes from Fudzilla who are reporting that the first graphics card based on this GPU could launch in first half of 2017. We have known since AMD's Capsaicin 2016 event that Vega would be arriving in 2017 and would succeed the Polaris 10 and Polaris 11 GPUs which are available in the retail GPU market. The Vega architecture had already reached a crucial design milestone suggesting that the development of the chip is underway and we could see it in market as early as Q1 2017.
Vega 10 is the successor to Polaris and if all goes well you will see the world’s first HBM 2.0 based GPU from the house of AMD – RTG in the first part of 1H 2016, possibly even in Q1 2017. via Fudzilla

The official AMD Graphics Roadmap shows Vega launching in 2017 with HBM2.
Some main features of the Vega 10 GPU would include an improved GCN design which will lead to better GPU efficiency compared to Polaris and use of HBM2 standard. It's not known whether Vega will consist of an entire stack of HBM2 GPUs or just the top-end stack. The way AMD shows the product positioning of Vega in the road map shows that the former could be true with both Vega 10 and Vega 11 utilizing the latest HBM2 standard that delivers insane amount of bandwidth, higher VRAM capacity and lower power consumption compared to HBM1 and GDDR5.
| 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 |









