Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps

NVIDIA's next-gen GeForce RTX 3090: GDDR6X memory at 21Gbps will offer memory bandwidth of up to 1TB/sec on new Ampere GPU.

@anthony256

Published Aug 14, 2020 7:01 AM CDT
Updated Nov 3, 2020 12:42 PM CST

13 minutes & 32 seconds read time

It seems today is another huge day for NVIDIA leaks on the purported new Ampere-based flagship GeForce RTX 3090, with Micron confirming the new card will use its next-gen GDDR6X memory.

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 11

VIEW GALLERY - 13 IMAGES

Our good friends at VideoCardz have spotted a new technology brief, Micron confirmed it had worked with NVIDIA on the new GeForce RTX 3090 graphics card -- outing it right there in the documents itself. The company says it will have 12GB of GDDR6X on a 384-bit memory interface on the GeForce RTX 3090.

Micron explains: "Micron GDDR6X is the world's fastest graphics memory, delivering legendary performance to help driveray tracing, shadow mapping and silky-smooth animation for an immersive PC gaming experience".

This has the GeForce RTX 3090 packing over 1TB/sec of memory bandwidth, all thanks to that huge GDDR6X memory pumping away at 21Gbps.

This all feeds into that huge '21' push that NVIDIA is doing right now, something I covered in this article -- but here are some tidbits on the numerology use by NVIDIA and the number 21:

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 12

Now add to that list, the new GeForce RTX 3090 has 12GB (21 reversed) of GDDR6X memory at 21Gbps (21). You can't make this up, it's right here in our face now -- 21 is a big deal for NVIDIA in so many ways it's uncanny.

The company continues: "In Summer of 2020, Micron announced the next evolution of Ultra-Bandwidth Solutions in GDDR6X. Working closely with NVIDIA on their Ampere generation of graphics cards, Micron's 8Gb GDDR6X will deliver up to 21Gb/s (data rate per pin) in 2020. At 21Gb/s, a graphic card with 12pcs of GDDR6X will be able to break the 1TB/s of system bandwidth barrier!"

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 04

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 05

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 06

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 07

Micron adds: "Micron's roadmap also highlights the potential for a 16Gb GDDR6X in 2021 with the ability to reach up to 24Gb/s. GDDR6X is powered by a revolutionary new PAM4 modulation technology for Ultra-Bandwidth Solutions. PAM4 has the potential to drive even more improvements in data rate".

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 08

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 09

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 10

The rest of the Micron documents confirming GDDR6X for the GeForce RTX 3090.

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 21

What's NVIDIA doing with all those mentions of 21? I'm a kinda numerology nerd in some ways, so it helps decode at least some of it -- which I did so here. (GeForce) 256 x 21 (days/years celebration) = 5376 (GPU cores in purported GA102 GPU), and 5 + 3 + 7 + 6 = 21.
What about "ULTIMATE"? Well that's the other thing -- does NVIDIA mean this will be the "ultimate graphics card", or does it mean we could expect an ULTIMATE branded graphics card? GeForce RTX 3090 ULTIMATE has a nice ring to it... just saying.
Where can I tune into the GeForce special event? NVIDIA will be premiering the special event broadcast on September 9 at 9AM PT, hosted by none other than NVIDIA CEO and founder Jensen Huang. You can tune in right here.

More reading:

Micron confirms: GeForce RTX 3090 has 12GB of GDDR6X memory at 21Gbps 444

Traversal coprocessor: We have had more leaks on NVIDIA's next-gen GeForce RTX 3000 series than any family of graphics cards before it, with an interesting "traversal coprocessor" on the new GeForce RTX 3080 and GeForce RTX 3090 graphics cards. You can read more on that here.
NVCache: Ampere is meant to have something called NVCache, which would be NVIDIA's own form of AMD's HBCC (High Bandwidth Cache Controller, more on that here). NVCache would use your system RAM and SSD to super-speed game load times, as well as optimizing VRAM usage. You can read more on NVCache here.
Tensor Memory Compression: NVCache is interesting, but Tensor Memory Compression will be on Ampere, and will reportedly use Tensor Cores to both compress and decompress items that are stored in VRAM. This could see a 20-40% reduction in VRAM usage, or more VRAM usage with higher textures in next-gen games and Tensor Memory Compression decreasing that VRAM footprint by 20-40%.
How fast is the GeForce RTX 3090? Freaking fast according to rumors, with 60-90% more performance than the current Turing-based flagship GeForce RTX 2080 Ti. We could see this huge performance leap in ray tracing titles, but we'll have to wait a little while longer to see how much graphical power NVIDIA crams into these new cards. You can read more on those rumors here.
Power hungry: As for power consumption, GA102 reportedly uses 230W -- while 24GB of GDDR6X (which we should see on the new Ampere-based TITAN RTX) consumes 60W of power. You can read more on that here.
Production begins soon: NVIDIA is reportedly in the DVT (or Design Validation Test) range of its new GeForce RTX 3000 series graphics cards. Mass production reportedly kicks off in August 2020, with a media event, benchmarks, and more in September 2020 as I predicted many months ago. More on that here.