Over the last weeks we have already seen some details and various leaks about AMD RX Vega and today there is a news about the upcoming versions, which are supposed to become available. According to latest rumors, AMD RX Vega will available with 4GB and 8GB HBM2 versions and this also explains why there have been several different device IDs visible in the latest Linux GPU drivers.


According to Scott Herkelman from AMD, the RX Vega launch is just around the corner and while the AMD Fury X was limited to 4GB of HBM, Vega will be available with 4GB or 8GB of HBM2 memory. According to Herkelman, Vega will use HBM2 memory that has different capacity stacks. In other words, AIB partners will have to choose between single stack or double stack HBM2 memory. RX Vega might be available with one or two stacks and this fact might also affect the memory bandwidth, since with more stacks the memory bandwidth is higher. Furthermore if HBM2 can be produced in 2GB stacks, we might also see 4GB GPUs with two full stacks of memory as well.

Aside from learning that Vega is “just around the corner” we also know that Vega would in fact be available in gaming notebooks. This way AMD will be competing with Nvidia in the high-end desktop and laptop GPU space.

However, before we’re going to see Vega, we will first have to deal with the Radeon RX 500 series, which - unfortunately - is nothing more than a refresh of the current Polaris based RX 400 series. It looks like the only differences between the Radeon RX 580 and the Radeon RX 570 compared to their predecessors RX 480 and RX 470 are going to be the clock speeds and improved energy efficiency. Back in the days, with the Radeon R9 300 series, AMD did the exact same trick.






Source: Videocardz