Long story short, HBM trades the high clock rates of GDDR5 for a super wide bandwith. Most graphic cards have a 256 bit connection, HBM has 4096, allowing to clock the VRAM much slower than in GDDR5. This simplifies the chips and helps them consume much less energy.
The tradeoffs are that both the HBM and the Graphics Chip must be posed together on an "connection chip" for all those lines to be connected with as PCBs (Printed Circuit Boards) can't achieve this density, increasing costs. Also, first generation HBM can only adress up to 4GiB of Memory, which is why Fury is lacking in that department. Second Generation will be able to go up to 16 GiB, which will be more approbpriate for high-end cards in the future.
The Nano does intrigue me, seems like a perfect card for most gamers: compact, yet powerful with low energy consumption and thus low TDP, and here the smaller VRAM shouldn't be as much of a concern compared to the bigger Fury cards. Now if we had a pricetag for that one...
Existing User Log In
New User Registration
Register for a free account to gain full access to the VGChartz Network and join our thriving community.
