This site may earn affiliate commissions from the links on this folio. Terms of employ.

GDDR5X-GPU

This annunciation might seem a bit anticlimactic at present that Nvidia has already stated that the GeForce GTX 1080 volition utilize GDDR5X RAM, but Micron has appear that GDDR5X is now in mainstream production, months ahead of its originally expected schedule.

Ane interesting thing almost the GTX 1080 is that the RAM it uses is at the lower-cease of the memory Micron said information technology could ship. GDDR5X has previously been touted as a solution capable of providing up to 12.0 Gbps of bandwidth in quad-information rate (QDR) manner. Nvidia, in contrast, has stated that the GTX 1080 uses just 10Gbps memory — towards the low-end of the scale. One potential advantage of starting at the lower end, however, is that the company has plenty of headroom to calibration in the futurity.

Absolute bandwidth on the GTX 1080 is 320GB/south, which most matches the GTX 980 Ti while retaining the GTX 980's 256-bit retention bus. Keeping the retentiveness motorcoach modest is one way to control die size and lower product costs; large band buses eat upwardly die space and consume a significant corporeality of power.

Brand-new standard or GDDR4?

We doubtable that GDDR5X will be similar to GDDR4. GDDR4 was introduced past AMD in 2006 as a way to improve memory bandwidth and GPU performance. Nvidia never switched to GDDR4, preferring instead to retain GDDR3 until GDDR5 became available. AMD used GDDR4 in diverse products from 2006 – 2009, at which point it moved to GDDR5.

AMD hasn't said whether or not it will use GDDR5X in upcoming products, but the fact that Nvidia is only using GDDR5X for the GTX 1080 suggests that the retentivity is still rather expensive; the GTX 1070 uses GDDR5. AMD'south Polaris is expected to target the mainstream PC marketplace, which ways it'll likely utilize GDDR5 too. Meanwhile, both AMD and Nvidia are planning late-yr launches for next-generation products, both of which are confirmed to utilize HBM2.

micron-gddr5x-chip-3

The implication here is that GDDR5X'southward marketplace position long-term will depend on whether or not AMD and Nvidia come across a need for the memory to offset higher costs in HBM2. If HBM2 costs come down relatively apace, at that place may not be much room left for GDDR5X in the market, given that GDDR5 is produced past multiple companies and is relatively cheap (Micron, thus far, is the only company building GDDR5X). If, on the other hand, interposer costs stay high, we may still see all 3 standards in-market for some period of fourth dimension.

What'due south highly unlikely is that we'll see any single GPU supporting both HBM2 and GDDR5X. Sources we've spoken to have indicated that supporting both memory standards simultaneously in the same silicon would be extremely hard for minimal benefits. Supporting GDDR5 and 5X in the same silicon is much simpler than an HBM2 + GDDR5X configuration.