Additional Details on Micron’s DDR3L-RS, DDR4-RS, and Other Memory
by Jarred Walton on September 22, 2012 7:02 PM ESTEarlier this week we posted a short write-up about Micron’s new DDR3L-RS memory. We didn’t have a lot of technical detail to go on at the time, but Micron offered us a chance to chat with them on the phone and we were able to get more information about DDR3L-RS as well as their other memory products. Memory is something many of us take for granted in our PCs and other computing devices, but there’s a lot more going on in the market than you might expect.
If you need the least expensive memory possible, DDR3 is currently the way to go. On the other hand, if you’re making a mobile device, finding memory that uses less power even if it costs more might be the best option. Naturally, there are plenty of other options that fall somewhere in between those extremes; Micron provided us with the following chart showing where the various memory types fall in terms of price vs. power requirements.
Starting at the top with LPDDR3 and other LPDDR products, their specialized nature is what gives them both their low power usage as well as their higher cost—consider how most tablets and smartphones only ship with 1GB LPDDR or less right now. Chiefly this comes because of the complexity of the devices; for example, the memory might be integrated into an SoC, or placed in a PoP package. The result is that while you can get the best power characteristics out of LPDDR, the volume is much lower as it’s generally not used in high volume markets like laptops and PCs with 8GB or more RAM. We haven’t seen any laptops that use LPDDR so far (at least, not that I’m aware of), but Intel reportedly has LPDDR3 support on their Ultrabook roadmaps, which would allow for improved battery life as well as smaller/thinner designs.
At the other end of the spectrum we have DDR3, a commodity memory where low price is generally the primary consideration. These devices are mass produced so economies of scale along with less difficult targets (e.g. 1.5V and DDR3-1600 speed) allow them to reach lower price points. Right now, for example, you can find a kit of 8GB DDR3-1600 CL9 SO-DIMMs for around $40 (and under $35 for DDR3-1333 and/or CL11).
One step up from DDR3 in terms of power efficiency is DDR3L, which targets 1.35V instead of 1.5V. Power scales linearly with voltage and current (P = V * I), and reducing the voltage typically reduces the current required for the chip as well, resulting in a substantial reduction in power draw. Getting chips that will run at a lower voltage is mainly a matter of binning, along with improvements in process technology, so the costs are very similar to regular DDR3. Sticking with the previous example, the same DDR3-1600 CL9 kits cost about 10% more if you get 1.35V DDR3L. Note that most DDR3L laptop kits will also run fine at 1.5V, but if you want to run at 1.35V you’ll generally need a laptop specifically designed to utilize the lower voltage—Apple’s MacBook Pros for instance use 1.35V CL11 memory.
Straddling the line between LPDDR3 and DDR3L, we have Micron’s new DDR3L-RS memory. The RS suffix stands for “Reduced Standby”, and through a process of binning along with a few extra features, Micron is able to cut standby power use for a system by around 25%. DDR3L-RS is also backwards compatible with the DDR3 standard, so there’s no change necessary at the controller level—all the extra work happens in the memory devices. Micron couldn’t discuss specific prices of their various memory types, but they did suggest that at the component level DDR3L-RS should cost around 20% more than DDR3L. In terms of power efficiency, Micron provided the following information showing their expected power savings:
One of Micron’s key features in reducing the amount of power used in standby mode is TCSR: Temperature Controlled Self Refresh. Most systems are specified to run the RAM at up to 85C when active, but in sleep mode the temperatures drop substantially and open the door for some additional power savings. In the case of Micron’s DDR3L-RS, once the temperature hits 45C or less, they can reduce how frequently RAM needs to be refreshed and thereby reduce the power draw. It's also important to remember that DDR3L-RS won't perform any better than DDR3L in active use; its benefits as the name implies are only when the memory/system is in standby.
We should note that while we’re talking about Micron’s specific memory, DDR3L-RS, it is expected that the other major memory manufacturers (e.g. Hynix, Elpida, Samsung, etc.) will have similar RAM technologies, though the specifics of how they save power may vary among the suppliers.
What about future memory technologies like DDR4? Micron also discussed some of their upcoming designs that leverage DDR4 with us, and like the switch from DDR2 to DDR3, the change from DDR3 to DDR4 will necessitate new memory controllers and will not be backwards compatible. One of the biggest changes with DDR4 is that the standard voltage drops from 1.5V (DDR3) down to 1.2V, enabling power savings over even DDR3L. Micron will also have DDR4-RS memory available, and we’ll likely see products start to use that (e.g. some tablets) as soon as late 2012/early 2013. While Intel hasn’t officially made any statements in regards to Haswell’s memory technology, the emphasis on reducing power use would make that an ideal time for Intel to switch from a DDR3 controller to DDR4—we should know more sometime in the coming months.
Wrapping up, obviously there’s no single “silver bullet” memory technology that works best for all markets. Paying a price premium for DDR3L or DDR3L-RS on a desktop just to save a couple watts of power doesn’t really make sense, while on laptops and in particular Ultrabooks the power savings could definitely be worthwhile. Like other DRAM manufacturers, Micron looks to offer a broad selection of DRAM devices for the whole array of options. The end result is that customers can choose based on cost, form factor, power, etc. and find the best balance of features and pricing for their product. Margins on memory products have become razor thin over the years, so anything that can help companies like Micron find a way to improve their bottom line is obviously something they will pursue; currently, the ultrathin computing initiative—tablets, Ultrabooks, sleekbooks, etc.—is really pushing for improvements in memory technology.
Facebook
Twitter
RSS
24 Comments
View All Comments
CyberAngel - Sunday, September 23, 2012 - link
quadraticallyJarredWalton - Sunday, September 23, 2012 - link
I was going off P = V * I, but didn't consider the reduction in current as part of the package. Generally speaking, reducing voltage also reduces current (I believe), so yes it's quadradically reduced power in most cases. However, if current remains constant, power scales linearly with voltage. I have updated the text to mention current just to keep everyone happy. :-)Penti - Sunday, September 23, 2012 - link
I = V / R R = V / I V = I*R P = V^2 / RIf a chip uses 5 W at 1.5V it draws 3.33A. If it follows Ohm's law it would be 4.05W at 1.35V. I.e. 19% reduction. That's the same number memory manufacturers use. If we don't have any laws of physics it would be just at 4.5W. I.e it's not linearly. Double to voltage quadruple the power. Current is a function of voltage and resistance. Resistance is virtually constant, but dependable on temperature. Of course if you like to design a chip drawing 200W at 1.15V it will draw very high currents at up to 174A from the power source so it's not like we lower voltage to lower the current draw. A 1.75V 200W chip would be at 114A. That would be a lot easier to power. It stems from manufacturing, materials, leakage etc. If we can lower it below the normal threshold you do save power of course. You do save even more power by constructing it in another way so you utilize less space (silicon, also why you process shrink stuff), uses other properties like reducing leakage etc. Like transistor types. That could be from materials or physical differences etc. If you shrink stuff then of course you won't be able to feed it with the same voltage as you lower the threshold voltage of the transistors due to oxide thickness. Thus it depend largely on design. There is no good reason to feed 30nm DRAM at 1.5V. It's not because it's low power however. One of the first DDR3 DRAM's was produced at 80 and 90 nm. About 90 nm was the target back then. If you like to save power you also have to have other power saving features like being able to shut off part of the logic. LPDDR2 can use up to about 70% less power then a DDR3-module alone. LPDDR2 still is powered by a 1.2V operating voltage. Most of the reduction comes from other stuff then the lower voltage there.
MrSpadge - Sunday, September 23, 2012 - link
That's quadratically, not exponential. But thanks for catching the "linear" error in the article and your further posts, saves me some time :)jjj - Saturday, September 22, 2012 - link
"The result is that while you can get the best power characteristics out of LPDDR, the volume is much lower as it’s generally not used in high volume markets like laptops and PCs with 8GB or more RAM. "Right so most decent smartphones right now are using LPDDR2.
In Q2 some 150-155 mil smartphone units shipped ,25 mil tablets and 87mil traditional PCs.
About a week ago IHS said that in Q2 PC RAM was 49% of the market while smartphones and tablets were 14.1%. and they expect Q4 2013 to be 42.8 % for PC and 26.7% for mobile devices.
So yeah not really used in " high volume markets".
SunLord - Sunday, September 23, 2012 - link
Well if you consider that just about all those smartphones and tablets usually only have 1 chip maybe 2 for some of the new high end SoC compared to 8 chips per so-dimm in laptops and 8-16 chips per dimm in desktop those pcs have a lot more ram chips compared to the smart phones.So with global smartphone and tablet sales were just under 300Million last year so that probably 310M LPDDR2 chips compared to global desktop/laptop sales of around 370M which works out to about 4.4+Billion DDR3 chips which is a far more massive amount and its not even taking into consideration servers
Beenthere - Saturday, September 22, 2012 - link
There won't be a rush to DDR4 because it's primarily designed for servers though it may be useful in some portable devices also. The anticipated change in topology to point-to-point for DDR4 means you install all the memory when you build and you replace all the memory if you want to upgrade to higher speed/density DRAM.Since current typical desktops are not bottlenecked with DDR3 running at ~1333 MHz, there really is no need for the added bandwidth of DDR4 for years to come. With DDR3 being converted to low voltage similar to DDR4 and higher frequency - again similar to DDR4, the updated DDR3 DRAM is an advantage and cheaper than DDR4 will be.
http://en.wikipedia.org/wiki/DDR4_SDRAM
UpSpin - Sunday, September 23, 2012 - link
If I read the Wiki article correctly it makes no sense to use DDR3 at all, because DDR4 is much better suited for smartphones and tablets:- lower power consumption from the start on
- higher bandwidth, which is important, if you consider that both CPU and GPU use the same memory and at the moment, memory bandwith is the biggest limiting factor in smartphones (see iPhone 5 boost thanks to improved memory bandwidth)
- The to point-to-point topology is the current way SoCs connect to DIMM, or am I wrong? So no change here.
- Support for 3D stacking, the only way to increase memory in smartphones/tablets is stacking. So only DDR4 allows to increase the amount of memory, without increasing required space and power draw (long routes if you don't stack it)
So DDR4 will consume less power (see above chart), be cheaper and allows higher densities (3D stacking), and will remove the current bottleneck (low bandwidth). And because manufactures had to redesign their SoC completely for ARM A15 and new GPU and DDR4 is available already, I hope they just skipped DDR3 completely.
Ultrabooks will also benefit, because they also integrate both CPU and GPU in a single die and thus share the same memory, which is a limiting factor for IGPs right now.
Penti - Sunday, September 23, 2012 - link
Mobile DDR already uses 3D-stacking. 3D stacking has also been used on Deskop and Server DDR3. None 3D stacking is used in any DRAM before the 3D-technology for that matter. DRAM usually comes in DDP and QDP-packages i.e. several dies.ssj3gohan - Sunday, September 23, 2012 - link
Regular DDR3, 1.5V stuff, according to Micron runs at 7mA per package in self-refresh mode. For my 5.9W ivy bridge desktop computer (http://ssj3gohan.tweakblogs.net/blog/8217/fluffy2-... with 16GB RAM (32 packages in two SODIMMs) that works out to be, theoretically, 0.336W. i actually went and measured this, and it turned out to be roughly 0.4W so that is pretty much spot on by Micron. The 'extra' power use is probably down to interface termination current and frankly measurement error.With DDR3L-RS, it's running at 1.2V and the current requirement is roughly 2 times lower. That means that for the same configuration, we're looking at 0.135W. Theoretically about 0.2W lower, in practice maybe even more because of the termination and interface drivers also requiring less current.