Samsung has just confirmed that they have begun mass production of the GDDR6 memory for fast graphics products arriving this year. The new GDDR6 memory standard will not only be faster than GDDR5, but will deliver higher bandwidth and VRAM capacities in a lower power package. This will address a range of issues that current graphics cards face such as lower available bandwidth, high power consumption and higher manufacturing costs on high-bandwidth solutions that are faced in using HBM standard.
Samsung Officially Begins Mass Production of GDDR6 Memory - Reaching Speeds of Up To 18 Gbps and 16Gb Density To Produce High-End Graphics Cards With Up To 24 GB VRAM, 864 GB/s Bandwidth
According to Samsung Electronics, the company has begun mass production of the industry's first 16 Gb GDDR6 memory. The application of this memory will be in advanced graphics processing for gaming devices and graphics cards which will power automotive, network, artificial intelligence and deep neural AI markets.
“Beginning with this early production of the industry’s first 16Gb GDDR6, we will offer a comprehensive graphics DRAM line-up, with the highest performance and densities, in a very timely manner,” said Jinman Han, senior vice president, Memory Product Planning & Application Engineering at Samsung Electronics. “By introducing next-generation GDDR6 products, we will strengthen our presence in the gaming and graphics card markets and accommodate the growing need for advanced graphics memory in automotive and network systems.” via Samsung
The production is right on time as previously, Samsung had confirmed that they will initiate mass production of GDDR6 memory in early 2018. Micron and SK Hynix also have plans to initiate mass production of their GDDR6 memory in the first half of 2018 but here, Sammy has taken the lead when it comes to mass producing the memory. There are a range of graphics products expected this year that can leverage from GDDR6 memory.
NVIDIA's upcoming GeForce products are going to feature support for GDDR6 memory allowing them to deliver faster speeds than conventional G5 and G5X solutions which have been around for some time. AMD can opt for the new memory solution if they plan on launching a budget tier Radeon 600 series lineup but that so far hasn't been confirmed.
The Samsung GDDR6 Memory Specifications
Coming to the specifications in-detail, the Samsung 16 Gb GDDR6 memory die will be built on a the 10nm process node which Samsung is calling as the most advanced memory node to date. It will double the density of their GDDR5 solution which was composed of a 20nm 8 Gb die. According to Samsung, their solution will be operating at 18 Gbps against a previous standard speed of 16 Gbps and that is a big deal here. Each die will be able to deliver a data transfer rate of 72 Gbps and hold a capacity of 2 GB VRAM. The solution will be able to do all of this with 35% lower power input at just 1.35V compared to 1.55V.
Built on Samsung’s advanced 10-nanomter (nm) class* process technology, the new GDDR6 memory comes in a 16Gb density, which doubles that of the company’s 20-nanometer 8Gb GDDR5 memory. The new solution performs at an 18-gigabits-per-second (Gbps) pin speed with data transfers of 72 gigabytes per second (GBps), which represents a more than two-fold increase over 8Gb GDDR5 with its 8Gbps pin speed.
Using an innovative, low-power circuit design, the new GDDR6 operates at 1.35V to lower energy consumption approximately 35 percent over the widely used GDDR5 at 1.55V. The 10nm-class 16Gb GDDR6 also brings about a 30 percent manufacturing productivity gain compared to the 20nm 8Gb GDDR5. via Samsung
This means that a solution based on a 384-bit interface and surrounded by 12 DRAM dies could feature up to 24 GB of VRAM while a 256-bit solution can house up to 16 GB of VRAM. That's twice the VRAM capacity as current generation cards. While VRAM is one thing, the maximum bandwidth output on a 384-bit card can reach a blistering fast 864 GB/s while the 256-bit solution can reach a stunning 576 GB/s transfer rate.
GPU Memory Technology Updates
|Graphics Card Name||Memory Technology||Memory Speed||Memory Bus||Memory Bandwidth||Release|
|AMD Radeon R9 Fury X||HBM1||1.0 Gbps||4096-bit||512 GB/s||2015|
|NVIDIA GTX 1080||GDDR5X||10.0 Gbps||256-bit||320 GB/s||2016|
|NVIDIA Tesla P100||HBM2||1.4 Gbps||4096-bit||720 GB/s||2016|
|NVIDIA Titan Xp||GDDR5X||11.4 Gbps||384-bit||547 GB/s||2017|
|AMD RX Vega 64||HBM2||1.9 Gbps||2048-bit||483 GB/s||2017|
|NVIDIA Titan V||HBM2||1.7 Gbps||3072-bit||652 GB/s||2017|
|NVIDIA Tesla V100||HBM2||1.7 Gbps||4096-bit||901 GB/s||2017|
|NVIDIA RTX 2080 Ti||GDDR6||14.0 Gbps||384-bit||672 GB/s||2018|
|AMD Instinct MI100||HBM2||2.4 Gbps||4096-bit||1229 GB/s||2020|
|NVIDIA A100 80 GB||HBM2e||3.2 Gbps||5120-bit||2039 GB/s||2020|
|NVIDIA RTX 3090||GDDR6X||19.5 Gbps||384-bit||936.2 GB/s||2020|
|AMD Instinct MI200||HBM2e||3.2 Gbps||8192-bit||3200 GB/s||2021|
|NVIDIA RTX 3090 Ti||GDDR6X||21.0 Gbps||384-bit||1008 GB/s||2022|
This means that the 256-bit solution will be faster than any 384-bit GDDR5 solution available to date while the 384-bit solution will be able to reach speeds currently only possible with the fastest HBM2 solutions like the one on Volta V100 which reaches slightly beyond 900 GB/s bandwidth. While Samsung has produced faster HBM2 solutions, they are very expensive and out of reach for consumers so implementing them on gaming graphics cards will not turn out to be a good deal. With such fast speeds and high video ram capacities, we can expect to see a new generation of products to drive high-performance graphics tasks including 4K/8K gaming.