HBM2 vs. GDDR6: New Video Compares, Contrasts Memory Types

HBM2 vs. GDDR6: New Video Compares, Contrasts Memory Types

As we’ve talked about various aspects of GPU and memory technology, one point that readers have raised are how HBM2 and GDDR6 compare against each other, and which one is really seen as the better option for various types of applications.

A new video by SemiconductorEngineering dives into the topic, with a comparison by Steven Woo, a Rambus fellow.

One point Woo makes is that the differences between HBM2 and GDDR6 are less about raw capability and more about what kind of design tradeoffs the engineering team wants to make. With GDDR5 versus HBM, there was a clear and absolute bandwidth advantage that HBM could offer. This is less the case with GDDR6, thanks to that memory’s higher bandwidth capabilities, but there are still use-cases where HBM2 has an advantage.

GDDR6 vs HBM2
GDDR6 vs HBM2

This slide, from 11:44 in the talk, illustrates some of the differences between the two memory types. Data rates on GDDR6 are much higher per-pin, but there are far fewer pins overall. The amount of area dedicated to the PHY (the circuitry required to implement the actual physical connection) and the power costs required. The PHY area is 1.5x – 1.75x larger, while the power cost can be 3.5x – 4.5x higher.

This echoes statements AMD made back when it explained why it used HBM for its Fury X GPUs. AMD’s explanation was that, above certain capacities, the energy and area requirements for GDDR5 made it a substantially worse choice than HBM. Woo’s comments indicate that this remains the case today, with HBM2 still offering better performance/watt and better overall power consumption.

That’s our expectation as well. Navi, at least, is anticipated to be a GDDR6 chip and all of Nvidia’s refreshed Turing cards have relied on the RAM type. It’s generally expected that AMD will also move away from HBM2 once it refreshes its high-end GPUs.

Continue reading

Samsung Now Producing 16Gbit GDDR6 for Upcoming GPUs
Samsung Now Producing 16Gbit GDDR6 for Upcoming GPUs

Samsung is ramping up production of GDDR6 with an 18Gbit clock and a new process node.

With GDDR6 Memory Production Scaling Up, New GPUs Won’t Be Far Behind
With GDDR6 Memory Production Scaling Up, New GPUs Won’t Be Far Behind

SK Hynix expects to ramp up GDDR6 volume production in the near future — and they ain't ramping up production without customers to sell it to.

Micron Announces Mass Production of GDDR6
Micron Announces Mass Production of GDDR6

Micron announced it's begun mass production of GDDR6, the next-generation memory standard for video cards that's expected to take over for GDDR5 in the larger market.

Micron: Full GDDR6 Production Underway for Nvidia’s RTX Launch
Micron: Full GDDR6 Production Underway for Nvidia’s RTX Launch

Micron has announced full production of GDDR6 as part of the run-up to the Nvidia RTX launch — but it's done so with some frankly peculiar language.