After the publication of the LPDDR5X memory standard earlier this summer, Samsung has now been the first vendor to announce new modules based on the new technology.

The LPDDR5X standard will start out at speeds of 8533Mbps, a 33% increase over current generation LPDDR5 based products which are running at 6400Mbps.

LPDDR Generations
  LPDDR3 LPDDR4 LPDDR4X LPDDR5 LPDDR5X
Max Density 32 Gbit 64 Gbit 32 Gbit
Max Data Rate 2133Mbps 4266Mbps 6400Mbps 8533Mbps
Channels 1 2 1
Width x32 x32 (2x x16) x16
Banks
(Per Channel)
8 8 8-16 16
Bank Grouping No No Yes
Prefetch 8n 16n 16n
Voltage 1.2v 1.1v Variable
(Max 1.1v)
Vddq 1.2v 1.1v 0.6v 0.6v

Samsung’s implementation notes 16-gigabit dies (2GB) on a 14nm process node, with the company explaining that the new modules will use 20% less power than LPDDR5. It’s also possible to allow for 64GB memory modules of a single package, which would correspond to 32 dies.

“Later this year, Samsung will begin collaborating with global chipset manufacturers to establish a more viable framework for the expanding world of digital reality, with its LPDDR5X serving as a key part of that foundation.”

We generally expect LPDDR5X SoCs and products to start being released for the 2023 generation of devices.

Related Reading:

Comments Locked

19 Comments

View All Comments

  • lemurbutton - Tuesday, November 9, 2021 - link

    Should be around the time Apple Silicon M3 Pro/Max launches.

    Would be curious to see what memory the M2 Pro/Max will ship with. The GPU cores are expected to increase because the A15 went from 4 to 5 cores. M2 should have 10 cores. M2 Pro should have 20. M2 Max should have 40. Would LPDDR5 still be enough?
  • NextGen_Gamer - Tuesday, November 9, 2021 - link

    I'm thinking that, at some point, Apple will have to move the Pro/Max versions to HBM2e memory. The standard M2 and M3 SoC's I'm sure would be fine with LPDDR5/X still, but those big-boy SoC's can only go so far. Since they already command a price-premium anyways, only being found in the MacBook Pro's and whatnot, it would make sense to move them to HBM2e.
  • nandnandnand - Tuesday, November 9, 2021 - link

    Sounds reasonable. If you are going to pay $5000, might as well get the best. HBM3 is also on the table.
  • schuckles - Tuesday, November 9, 2021 - link

    I don’t know. Apple gets most of the benefit of HBM power reduction just by doing in chip / on package / on substrate ddr. It’s not like they are using 25% of the package power on the ddr bus of the pro/max.
  • caribbeanblue - Tuesday, January 25, 2022 - link

    They will just use multiple M1 Max dies which already have the necessary memory controllers on them, and will increase memory bandwidth correspondingly.
  • blanarahul - Tuesday, November 9, 2021 - link

    They are already using 512 bit wide 6400 MHz memory bus with M1 Max. They are going to have to go with HBM/GDDR6 if they want more memory bandwidth.
  • schuckles - Tuesday, November 9, 2021 - link

    I agree, seems like a straightforward path to boost max memory bandwidth by another 33% while using 20% less power during normal operation
  • mode_13h - Tuesday, November 9, 2021 - link

    > M2 should have 10 cores. M2 Pro should have 20. M2 Max should have 40.

    That's an awful lot of cores, for a laptop.
  • Hrunga_Zmuda - Tuesday, November 9, 2021 - link

    With their energy budget, that doesn't sound like that big a deal. The M1 Max is an awful lot of performance for a laptop.
  • mode_13h - Wednesday, November 10, 2021 - link

    I was really just wondering what anyone would do with so many cores... in a laptop!

    I can understand 8 cores. Certain tasks parallelize well and it's more efficient to go multi-core than juice up the clock speeds. Plus, you might want a VM or two.

    But 20 or 40 cores... ??? Do most people even do stuff on such laptops that could harness so many? And we're not talking about those "mobile workstation" laptops, either. Just like your typical Powerbook user.

Log in

Don't have an account? Sign up now