Samsung's new HBM2E enables up to 64GB VRAM at 1.64TB/sec

Samsung's latest version of HBM is HBM2E, offers 16Gb per die, could lead to 64GB HBM2E at 1.64TB/sec bandwidth.

Published
Updated
57 seconds read time

Samsung had some interesting things to show at NVIDIA's GTC 2019 this year, with the latest version of High Bandwidth Memory in something called Flashpoint, which offers a 33% increase in memory performance over HBM2.

Samsung's new HBM2E enables up to 64GB VRAM at 1.64TB/sec | TweakTown.com

We're looking at 3.2Gbps of bandwidth per pin, with double the density to 16Gb per die in stacks of 8. This means we're looking at a single package that can pack 16GB of HBM2E with up to 410GB/sec of memory bandwidth, but in a 4 stacks of HBM2E we're looking at an insane 64GB at 1.64TB/sec of memory bandwidth. The current 4x 4-Hi stacks of HBM2 on the AMD Radeon VII has 16GB with 1TB/sec of memory bandwidth, so this is quite the increase in bandwidth.

Jinman Han, Senior Vice President of Memory Product Planning and Application Engineering Team at Samsung Electronics explains: "Flashbolt's industry-leading performance will enable enhanced solutions for next-generation datacenters, artificial intelligence, machine learning, and graphics applications. We will continue to expand our premium DRAM offering, and improve our 'high-performance, high capacity, and low power' memory segment to meet market demand".

We don't know when Samsung's new HBM2E will be in mass production and ready for professional graphics cards, but don't expect this on Navi as we'll most likely see AMD make its first use of GDDR6 when those cards are announced and released in the coming months.

Anthony joined the TweakTown team in 2010 and has since reviewed 100s of graphics cards. Anthony is a long time PC enthusiast with a passion of hate for games built around consoles. FPS gaming since the pre-Quake days, where you were insulted if you used a mouse to aim, he has been addicted to gaming and hardware ever since. Working in IT retail for 10 years gave him great experience with custom-built PCs. His addiction to GPU tech is unwavering and has recently taken a keen interest in artificial intelligence (AI) hardware.

Newsletter Subscription

Related Tags