Micron is rushing to release the latest GDDR6 graphics memory by the end of the year, and eSports is a major driver behind the plans.
The memory company is speeding up the release of GDDR6 to cope with faster PC and console upgrades. GDDR6 will be significantly faster than its predecessor, GDDR5X, which is still reaching GPUs. Micron had originally planned to release GDDR6 next year.
Virtual reality is also driving PC upgrades, but not at the same pace as gaming and eSports, which are forcing faster development of new graphics memory technologies, said Tom Eby, vice president of the computing and networking business unit at Micron.
A projected 500 million people will be eSports fans by the end of the decade. Gaming PCs are now being upgraded every three years, which is faster than the previous five-year cycle.
"There will be more people watching major eSports championships than any other North American professional championship with the exception of the Super Bowl," Eby said.
Over time, GDDR6 will transfer data at 16Gbps (bits per second), making it two times faster than GDDR5, which is used in most GPUs today. It will also be faster than GDDR5X, which had an initial target speed of 12Gbps.
Users will want higher bandwidth and low-latency memory to get better graphics on their PCs and gaming consoles, Eby said.
GDDR memory is targeted at mainstream computing, Eby said. However, it's not yet known when GPUs with GDDR6 will appear.
Micron will make GDDR6 chips and is already making GDDR5X memory. The GDDR5X memory will continue to be used as a low-end graphics memory replacement to GDDR5.
While GDDR is in most GPUs, there are also graphics memory technologies like HBM (high-bandwidth memory), used in AMD's Radeon R9 Fury X and its successor HBM2, which is in Nvidia's Tesla P100. HBM2 is being manufactured in volume by Samsung and SK Hynix but is more expensive than GDDR memory.
GDDR6 could be used in network switches and routers, which in some instances are equipped with GDDR5 memory. The memory technology will also find a place in high-performance computers for machine learning.