Samsung is well known in the semi-conductor world and is an established leader when it comes to memory. The company has been working on 16 gigabit GDDR6 memory modules since last year, the chips even went into mass production back in January. Now, we are seeing the proper debut, as Samsung’s graphics memory is being utilised in Nvidia’s new Quadro RTX series of graphics cards.
Late last night, Nvidia officially launched the Turing architecture, which is featured in the new Quadro RTX series. The RTX 8000, RTX 6000 and RTX 5000 all feature Samsung’s new 16Gb GDDR6 memory, which brings along improved performance and efficiency. Samsung is first to the mark here too, as it is the first company to offer 16Gb GDDR6.
According to Samsung, this new GDDR6 memory uses 35 percent less power draw than required by GDDR5. This is achieve by a low-power circuit design, which allows Samsung’s GDDR6 modules to operate at 1.35V rather than 1.55V. We also see performance jump to 14Gbps pin speed with data transfers running at 56GB/s. This is around a 75 percent performance increase over 8Gb GDDR5 with 8Gbps pin speed.
As you would expect, Samsung is suitably pleased with the situation. Corporate Senior VP of Samsung Semiconductor, Jim Elliott, stated the following: “It’s a tremendous privilege to have been selected by NVIDIA to launch Samsung’s 16Gb GDDR6, and to have enjoyed the full confidence of their design team in making our key contribution to the NVIDIA Quadro RTX GPUs”.
If you are interested in reading more about Turing and the Quadro RTX series, then you can find those details, HERE.
KitGuru Says: We’ve known about Samsung’s 16Gb GDDR6 modules for a long time now, so it is nice to finally see them being put to good use. Are any of you excited about the Turing launch? Are you planning on picking up a new GPU?