SoylentNews is people
SoylentNews
DDR5 Memory Specification Released: Setting the Stage for DDR5-6400 And Beyond
We'll start with a brief look at capacity and density, as this is the most-straightforward change to the standard compared to DDR4. Designed to span several years (if not longer), DDR5 will allow for individual memory chips up to 64Gbit in density, which is 4x higher than DDR4's 16Gbit density maximum. Combined with die stacking, which allows for up to 8 dies to be stacked as a single chip, then a 40 element LRDIMM can reach an effective memory capacity of 2TB. Or for the more humble unbuffered DIMM, this would mean we'll eventually see DIMM capacities reach 128GB for your typical dual rank configuration.
[...] For DDR5, JEDEC is looking to start things off much more aggressively than usual for a DDR memory specification. Typically a new standard picks up from where the last one started off, such as with the DDR3 to DDR4 transition, where DDR3 officially stopped at 1.6Gbps and DDR4 started from there. However for DDR5 JEDEC is aiming much higher, with the group expecting to launch at 4.8Gbps, some 50% faster than the official 3.2Gbps max speed of DDR4. And in the years afterwards, the current version of the specification allows for data rates up to 6.4Gbps, doubling the official peak of DDR4.
Of course, sly enthusiasts will note that DDR4 already goes above the official maximum of 3.2Gbps (sometimes well above), and it's likely that DDR5 will eventually go a similar route. The underlying goal, regardless of specific figures, is to double the amount of bandwidth available today from a single DIMM. So don't be too surprised if SK Hynix indeed hits their goal of DDR5-8400 later this decade.
[...] JEDEC is also using the introduction of the DDR5 memory standard to make a fairly important change to how voltage regulation works for DIMMs. In short, voltage regulation is being moved from the motherboard to the individual DIMM, leaving DIMMs responsible for their own voltage regulation needs. This means that DIMMs will now include an integrated voltage regulator, and this goes for everything from UDIMMs to LRDIMMs.
JEDEC is dubbing this "pay as you go" voltage regulation, and is aiming to improve/simplify a few different aspects of DDR5 with it. The most significant change is that by moving voltage regulation on to the DIMMs themselves, voltage regulation is no longer the responsibility of the motherboard. Motherboards in turn will no longer need to be built for the worst-case scenario – such as driving 16 massive LRDIMMs – simplifying motherboard design and reining in costs to a degree. Of course, the flip side of this argument is that it moves those costs over to the DIMM itself, but then system builders are at least only having to buy as much voltage regulation hardware as they have DIMMs, and hence the PAYGO philosophy.
"On-die ECC" is mentioned in the press release and slides. If you can figure out what that means, let us know.
See also: Micron Drives DDR5 Memory Adoption with Technology Enablement Program
Previously: DDR5 Standard to be Finalized by JEDEC in 2018
DDR5-4400 Test Chip Demonstrated
Cadence and Micron Plan Production of 16 Gb DDR5 Chips in 2019
SK Hynix Announces Plans for DDR5-8400 Memory, and More
(Score: 3, Informative) by takyon on Wednesday July 15 2020, @06:17PM (2 children)
Comments by TFA author about "on-die ECC":
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 2) by Rich on Thursday July 16 2020, @07:10AM (1 child)
As I understand, the "rowhammer" class of attacks still is a threat on present-day RAM, so some sort of error correction is due anyway. Any insights on this with the DDR5 proposals?
(Score: 2) by takyon on Thursday July 16 2020, @09:21AM
It isn't clear yet.
From 2018:
https://www.tomshardware.com/news/ecc-memory-not-safe-rowhammer,38144.html [tomshardware.com]
If it's discussed in the newly released spec, well, they are charging $369 for access to that information.
The same exact user is whining about a fictional Rowhammer3 on AnandTech and Phoronix comments. But it's not a bad bet to assume that more vulnerabilities will be found.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 4, Funny) by DannyB on Wednesday July 15 2020, @06:59PM (2 children)
Just in the nick of time. This September, Java 15 raises the maximum heap size from 4 TB to 16 TB of memory.
How do you even build a board with that much memory?
This will create new and interesting challenges for us developers to use up all that memory.
(but still 1 ms max GC pause time, if you have plenty of cpu cores)
The Centauri traded Earth jump gate technology in exchange for our superior hair mousse formulas.
(Score: 2) by takyon on Wednesday July 15 2020, @07:13PM
16x 1 TB DDR5 RDIMMs should do the trick, once they exist.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Wednesday July 15 2020, @08:20PM
Still less than Emacs though, so they're good to go.
(Score: 1, Funny) by Anonymous Coward on Wednesday July 15 2020, @09:39PM (1 child)
btw, whatever happened to rambus cocksuckers?
(Score: 3, Informative) by takyon on Wednesday July 15 2020, @10:50PM
They are still around and making their presence known:
RAMBUS Claims They are First to Have Working DDR5 [eteknix.com]
Rambus and Gigadrive Form Joint Venture to Commercialize Resistive RAM [soylentnews.org]
Rambus Demonstrates GDDR6 Running At 18 Gbps [anandtech.com]
Rambus Unveils PCIe 5.0 Controller & PHY [anandtech.com]
Rambus Develops HBM2E Controller & PHY: 3.2 Gbps, 1024-Bit Bus [anandtech.com]
I have linked this blog post of theirs a couple times:
DDR5 vs DDR4 – All the Design Challenges & Advantages [rambus.com]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 3, Insightful) by shortscreen on Thursday July 16 2020, @12:37AM (2 children)
Overclockers will be sad that they can't crank up the memory voltage via software. That is, unless they get the inevitable special overclocker DIMMs with blue LEDs that also have voltage tweaking as an additional feature.
Shady vendors can put slow chips on a DIMM and raise the voltage to pass it off as a higher speed grade.
(Score: 2) by takyon on Thursday July 16 2020, @02:21AM
There's already plans for DDR5-8400, which is way outside of the spec. Clearly some sort of overclocking will be possible.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Friday July 17 2020, @03:48PM
This is nothing new. So, you got cheap DDR4 at 2933 or 3200 on the desktop while laptops get cheap 2400 or maybe 2666.
The desktop is "cheating" by running at 1.35V instead of 1.2V but it's faster, everyone knows it, it's warrantied and it costs about the same.
You go buy cheap slower DDR 2400 for the desktop if you want to.