Micron Reports on GDDR5X Dev Progress – Volume Production This Summer

Engineers from Micron Development Center in Munich (also known as Graphics DRAM Design Center) are well known around the industry for their contribution to development of multiple graphics memory standards, including GDDR4 and GDDR5. The engineers from MDC also played a key role in development of GDDR5X memory, which is expected to be used on some of the upcoming video cards. Micron disclosed the first details about GDDR5X in September last year, publicizing the existance of the standard ahead of later JEDEC ratification and offering a brief summary of what to expect. Since then the company has been quiet on their progress with GDDR5X, but in a new blog post they have published this week, the company is touting their results with their first samples and offering an outline of when they expect to go into volume production.

The GDDR5X standard, as you might recall, is largely based on the GDDR5 technology, but it features three important improvements: considerably higher data-rates (up to 14 Gbps per pin or potentially even higher), substantially higher-capacities (up to 16 Gb), and improved energy-efficiency (bandwidth per watt) thanks to 1.35V supply and I/O voltages. To increase performance, the GDDR5X technology uses its new quad data rate (QDR) data signaling technology to increase the amount of data transferred, in turn allowing it to use a wider 16n prefetch architecture, which enables up to 512 bit (64 Bytes) per array read or write access. Consequently, GDDR5X promises to double the performance of GDDR5 while consuming similar amounts of power, which is a very ambitious goal.

In their blog post, Micron is reporting that they already have their first samples back from their fab – this being earlier than expected – with these samples operating at data-rates higher than 13 Gbps in the lab. At present, the company is in the middle of testing its GDDR5X production line and will be sending samples to its partners this spring.

Thanks to reduction of Vdd/Vddq by 10% as well as new features, such as per-bank self refresh, hibernate self refresh, partial array self refresh and other, Micron’s 13 Gbps GDDR5X chips do not consume more energy than GDDR5 ICs (integrated circuits) — 2–2.5W per component (i.e., 10–30W per graphics card), just like the company promised several weeks ago. Since not all applications need maximum bandwidth, in certain cases usage of GDDR5X instead of its predecessor will help to reduce power consumption.

GDDR5X memory chips will come in new packages, which will be slightly smaller (14×10mm vs. 14×12mm) compared to GDDR5 ICs despite the increase of their ball count (190-ball BGA vs. 170-ball BGA). According to Micron, denser ball placement, reduced ball diameter (0.4mm vs. 0.47mm) and smaller ball pitch (0.65mm vs. 0.8mm) make PCB traces slightly shorter, which should ultimately improve electrical performance and system signal integrity. Keeping in mind higher data-rates of GDDR5X’s interface, improved signal integrity is just what the doctor ordered. The GDDR5X package maintains the same 1.1mm height as the predecessor.

Micron is using its 20 nm memory manufacturing process to make the first-generation 8 Gb GDDR5X chips. The company has been using the technology to make commercial DRAM products for several quarters now. As the company refines its fabrication process and design of the ICs, their yields and data-rate potential will increase. Micron remains optimistic about hitting 16 Gbps data-rates with its GDDR5X chips eventually, but does not disclose when it expects that to happen.

All of that said, at this time the company has not yet figured out its GDDR5X product lineup, and nobody knows for sure whether commercial chips will hit 14 Gbps this year with the first-generation GDDR5X controllers. Typically, early adopters of new memory technologies tend to be rather conservative. For example, AMD’s Radeon HD 4870 (the world’s first video card to use GDDR5) was equipped with 512 MB of memory featuring 3.6 Gbps data-rate, whereas Qimonda (the company which established Micron’s Graphics DRAM Design Center) offered chips with 4.5 Gbps data-rate at the time.

The first-gen GDDR5X memory chips from Micron have 8 Gb capacity, hence, they will cost more than 4 Gb chips used on graphics cards today. Moreover, due to increased pin-count, implementation cost of GDDR5X could be a little higher compared to that of GDDR5 (i.e., PCBs will get more complex and more expensive). That said, we don’t expect to see GDDR5X showing up in value cards right away, as this is a high-performance technology and will have a roll-out similar to GDDR5. At the higher-end however, a video card featuring a 256-bit memory bus would be able to boast with 8 GB of memory and 352 GB/s of bandwidth.

Finally, Micron has also announced in their blog post that they intend to commence high-volume production of GDDR5X chips in mid-2016, or sometime in the summer. It is unknown precisely when the first graphics cards featuring the new type of memory are set to hit the market, but given the timing it looks like this will happen in 2016.

Corsair and G.Skill Introduce 128 GB (8×16 GB) DDR4-3000 Memory Kits

An average personal computer nowadays is equipped with 8 GB or less of DRAM, according to analysts from DRAMeXchange. Due to the requirements of Microsoft Windows 10 operating system, 8 GB may be enough for general-purpose computing. But there are PCs, particularly at the high-end desktop and workstation level, which need a lot of memory either for software, computation, RAM disks or even RAM caches to the point where motherboard manufacturers are now including such software in their bundles. To fulfill demand from owners of high-end desktops, Corsair and G.Skill this month unveiled their 128 GB quad-channel DDR4 memory kits consisting of eight DRAM modules.

Corsair and G.Skill’s 128 GB DDR4 memory kits are rated to run at 3000 MT/s per pin data-rate (DDR4-3000) and are subsequently designed for Intel’s X99 platform where the quad memory bus allows for up to 96 GB/s of bandwidth with 4 or 8 DIMMs.  These quad-channel kits consist of eight 16 GB unbuffered memory modules, which are based on 8 Gb DRAM chips made by Samsung using its 20 nm fabrication process. The memory sticks fully support Intel XMP 2.0 SPD profiles and can automatically set their clock-rates when installed into appropriate PCs.

Corsair’s Black Vengeance LPX 128 GB DDR4-3000 memory kit comes in with CL16 18-18-36 latency settings as well as the higher specification 1.35 V voltage for DDR4. The modules are equipped with black aluminum heat-spreaders to aid with cooling. Corsair also supplies their Vengeance Airflow cooling system, a removable 40mm fan cooling bracket, with the kit. Corsair’s Black Vengeance LPX 128 GB DDR4-3000 kit costs $1174.99 without tax and is currently available from the company’s online store with the official name of CMK128GX4M8B3000C16.

Meanwhile G.Skill’s Ripjaws V 128 GB DDR4-3000 set of DRAM modules for high-end desktop features surprisingly low latencies of CL14 14-14-34, as well as the higher 1.35V voltage. G.Skill’s Ripjaws V memory come with black or red aluminum heat-spreaders, and we assume these kits also come with extra fan cooling similar to G.Skill’s other high end kits. G.Skill’s Ripjaws V 128 GB DDR4 memory kit will be priced at $999.99 when it becomes available later this month under the SKU name F4-3000C16-16GVK.

It is noteworthy that despite of more aggressive timings and potentially higher real-world performance, G.Skill’s 128 GB DDR4 memory kit costs less than Corsair’s 128 GB DDR4 set of modules. The two companies are addressing a relatively small segment of the market with their 128 GB DRAM kits, hence, the competition between Corsair and G.Skill is inevitable. The reason for the high price for both kits comes down to binning – the ICs used for these are typically sold by the IC manufacturer as a certain bin (e.g. DDR4-2400 low voltage) and then they are individually tested by the memory stick manufacturer to fit within certain frequency ranges. At DDR4-3000 C14 for example, the process of testing might only produce one memory kit per 10000 ICs tested (educated guess) – and then the modules have to be tweaked to ensure they run together. We always recommend buying a single kit for a PC, especially of high speed memory, because the modules are designed to work together, whereas two separate kits hold no guarantee, especially if the secondary and tertiary sub-timings are close to the grain (typically these are slightly loosened for larger kits).

At present both Corsair and G.Skill market their 16 GB DDR4-3000 memory modules as solutions for overclockers because highest JEDEC data rate validated by Intel’s Haswell-E processors is 2133 MT/s. As JEDEC’s DDR4 memory standard supports data-rates up to 3200 MT/s, eventually we might see high-speed 16 GB+ memory sticks becoming normal for workstations with memory speed-limited workloads.

Source: Corsair, G.Skill

Analyzing Intel-Micron 3D XPoint: The Next Generation Non-Volatile Memory

The current mainstream memory technologies, namely DRAM and NAND, have been around for decades. While clever engineering work has allowed the two to scale below 20nm geometries, both technologies have their inherent architectural limitations. But what if you were to take the best characteristics of DRAM and NAND and create a whole new type of memory? Meet Intel’s and Micron’s new 3D XPoint, the next generation non-volatile memory technology. It is said to provide up to a thousand times better endurance and performance than modern NAND technology does, while being non-volatile and very scalable at the same time. Read on for our full analysis.