Side-by-Side Comparison
Graphics Cards
GDDR6 vs. HBM3
Comparing 2 standards across 5 specifications. 5 key differences identified.
Comparing:
GDDR6
HBM3
GDDR6, standardized by JEDEC in July 2017 and reaching mass production in early 2018, doubled data rates over GDDR5 to as much as 16-18 Gbit/s per pin -- the current mainstream graphics memory standard for most non-flagship cards.
HBM3, announced by JEDEC in January 2022, doubled channel count to 16 within the same 1024-bit bus width and reaches roughly 819 GB/s of bandwidth per stack. It's used predominantly in AI and datacenter accelerators like Nvidia's H100, rather than consumer graphics cards.
Specification Comparison Table
5 Parameters| Parameter / Spec | ||
|---|---|---|
|
JEDEC Standard
|
July 2017 | January 2022 |
|
Max Bandwidth / Data Rate
|
Up to 16 Gbit/s per pin (18 Gbit/s in mass-production variants) | ~819 GB/s per stack |
|
Mass Production
|
January 2018 | — |
|
Channels
|
— | 16 x 64-bit (1024-bit total bus, same as HBM2) |
|
Primary Use
|
— | AI/datacenter accelerators |