Categories: Graphic cards

SK Hynix presents HBM2E with almost doubled bandwidth

When AMD equipped the graphics cards in the Radeon R9 Fury series with HBM memory, it added a substantial increase in bandwidth and energy efficiency, while the stacked memory circuits took up significantly less circuit space compared to the corresponding amount of graphics memory of GDDR5 variant. With the Radeon RX Vega series and then the Radeon VII, the company took the step further to HBM2 memory.

However, HBM memory is significantly more expensive to manufacture, and with the Radeon RX 5700 series, AMD chose to equip the graphics cards with not as fast but significantly cheaper GDDR6 memory. SK Hynix now presents a developed variant of the memory technology called HBM2E which further consolidates the technology’s role as regent regarding bandwidth.

HBM2E builds on HBM2 but increases the bandwidth per memory capsule from 256 GB / s to 460 GB / s. Each individual capsule can now also hold a larger amount of memory of 16 GB per capsule, up from 8 GB per capsule in regular HBM2 memory. By comparison, today’s fastest graphics card boasts 11 GB of GDDR6 memory and a bandwidth of approximately 616 GB / s.

Nvidia’s Geforce RTX 2080 Ti boasts the market’s fastest bandwidth for GDDR6 memory.

However, the HBM2 memory in Radeon VII boasts the highest memory bandwidth in the graphics card market.

GDDR6 memory gets its total bandwidth from the combined bandwidth of the different memory circuits. To achieve high bandwidth, the graphics cards need to be equipped with a larger amount of memory circuits. The amount of GDDR6 memory together with the memory bus width and the memory speed determine the bandwidth of the graphics circuit.

The market’s graphics card with the highest bandwidth is currently AMD Radeon VII, whose 16 GB HBM2 memory has a total bandwidth of 1,024 GB / s. If the same graphics card were equipped with HBM2E memory, the bandwidth would increase to 1,840 GB / s, an increase of close to 80 percent.

However, SK Hynix’s press release does not mention whether HBM2E memory will be more cost-effective compared to HBM2, which will determine whether it will be possible in terms of cost to equip next year’s graphics card with the technology or whether GDDR6 will continue to be trusted.

Mass production of HBM2E memory will begin in 2020 and the company expects to sell the technology to customers in the graphics card market as well as acceleration of machine learning and AI calculations.

Read more about HBM memory:

Miners Hashrate

Recent Posts

Mining RTX 3070 at NiceHash: Overclocking, tuning, profitability, consumption

Mining on RTX 3070. Overclocking, tuning, profitability, consumption: If you are interested in finding more…

6 months ago

Mining GTX 1660, 1660 Ti, 1660 Super: Overclocking, settings, consumption

Mining with GTX 1660, 1660 Ti, 1660 Super. Overclocking, settings, consumption, profitability, comparisons - If…

6 months ago

Mining RTX 2070 and 2070 Super: Overclocking, profitability, consumption

Mining with RTX 2070 and 2070 Super. Overclocking, profitability, consumption, comparison What the RTX 2070…

6 months ago

Mining with RTX 3060, 3060 Ti. Limitations, overclocking, settings, consumption

Mining with RTX 3060, 3060 Ti. Limitations, overclocking, settings, consumption, profitability, comparison Let's look at…

6 months ago

Alphacool Eisblock Aurora Acryl GPX-A Sapphire – test: 2.8 GHz++ are not an issue

Alphacool Eisblock Aurora Acryl GPX-A (2022) with Sapphire Radeon RX 6950 XT Nitro+ Pure in…

6 months ago

Corporate Crypto Strategies 4.0: Leading with Bitcoin Expertise

In the ever-evolving landscape of business strategy, Bitcoin has emerged as a pivotal asset. With…

6 months ago

This website uses cookies.


Notice: ob_end_flush(): failed to send buffer of zlib output compression (1) in /home/gamefeve/bitcoinminershashrate.com/wp-includes/functions.php on line 5420

Notice: ob_end_flush(): failed to send buffer of zlib output compression (1) in /home/gamefeve/bitcoinminershashrate.com/wp-includes/functions.php on line 5420