facebook rss twitter

AMD said to have secured priority access to SK Hynix's HBM2 chips

by Mark Tyson on 14 July 2015, 13:11


Quick Link: HEXUS.net/qacsvw

Add to My Vault: x

One of the key technologies behind AMD's Radeon Fury range of graphics cards is the new type of memory known as High Bandwidth Memory, or HBM for short. The HEXUS editor wrote a full feature article about HBM back in May. First generation HBM chips run at a much lower clock speed to rival GDDR5 technology but offer a huge 1024-bit bus, with dense stacked RAM chips very close to the GPU, to provide much more bandwidth to the graphics system.

HBM single GPU graphics cards are restricted to a maximum of 4GB of video memory. With 4K gaming and beyond said to be 'where it's at' in the future this could be a stumbling block faced by adoptees in the coming months/years. However HBM designer and manufacturer SK Hynix already has already shown off plans for HBM2.

At the GPU Technology Conference in March this year HBM2 wafers were shown and its properties and capabilities discussed. With its greater densities the HBM2 memory architecture will allow graphics cards to be designed with up to 32GB of RAM on board.

A little earlier that month Nvidia unveiled the Nvidia GeForce GTX Titan X and sketched out details of its next gen Pascal GPU. According to Nvidia, sometime in 2016 we will see Pascal based graphics products featuring, among other things, 3D memory (with SK Hynix's HBM2 earmarked). Of course AMD isn't standing still and intends to make use of the next gen HBM2 in its Radeon Fury follow up flagship products. Then comes the crunch…

Apparently the first gen HBM has been in rather short supply. With the red and green graphics card goliaths hoping to facilitate mass production of competitive graphics cards based on HBM2 - it looks like there will be some kind of collision. However AMD might have the upper hand, if the sources speaking to WCCF Tech are correct. The tech site says that thanks to AMD's support for HBM and close, long standing working relationship with SK Hynix, it will get "Priority Access To HBM2 Production Capacity". Furthermore it will strive to capture "as much of the initial production capacity as possible," in a move that could stifle Nvidia's chances beyond a paper launch.

Whatever happens with the HBM2 allocation, the above scenario sounds like it will be a scrap worth watching.

HEXUS Forums :: 39 Comments

Login with Forum Account

Don't have an account? Register today!
This could be quite a coup for AMD. Combined with the experience their engineers will already have with working on HBM1 it could give them quite an advantage over Nvidia in the next gen of cards.

One can hope at least.
Sound like a double edged sword, does AMD hoover up as much HBM2 as they can in the hope that they can shift plenty of GPUs and hopefully limit Nvidia's supply, running the risk of having unsold stock.

Or is something like memory production not done on a pre-order basis?
They probably only have to take enough stock that nvidia couldn't launch a new gpu line using it, since they would never go to the expense of having 1 specific card using it and all others on the old ram system.
given that HBM is in limited supply - holding up the Fury range - contracting to buy everything from1 supplier until they dont need anymore would be the Apple way of doing it.
Could be AMDs final bad move if they try and leverage that to stifle competition. Do they have the funds to buy it all up and let it sit in a warehouse gathering dust?

After some of the comments about nVidias business practices recently, not one of them come close to trying to prevent a competitor from making a new product like this seems to imply.

I guess this could get interesting and may force nVidia to throw it's weight around a bit more.