Bookmark and Share


Samsung Electronics, the world’s largest maker of various dynamic random access memory, said on Thursday that it had started mass production of GDDR5 memory chips using 50 nm process technology. The new fabrication process allows the devices to reach unprecedented clock-speeds and trim power consumption.

“Our early 2009 introduction of GDDR5 chips will help us to meet the growing demand for higher performance graphics memory in PCs, graphic cards and game consoles. Because GDDR5 is the fastest and highest performing memory in the world, we’re able to improve the gaming experience with it across all platforms,” said Mueez Deen, director of mobile and graphics memory, Samsung Semiconductor.

The new GDDR5 memory chips are not only capable of working at up to 7GHz clock-speeds, but also use 1.35V voltage, a reduction from 1.425V – 1.575V voltage of currently available GDDR5 memory chips from Samsung. Moreover, the new GDDR5 devices are cheaper to produce: by adopting 50nm class technology, Samsung expects production efficiency to rise 100% over 60nm class technology.

7GHz GDDR5 chips could provide 224GB/s or 448GB/s peak bandwidth when connected to a processor using 256-bit or 512-bit memory bus, respectively. It is unclear which of the forthcoming graphics cards will utilize 7GHz GDDR5 chips, but since both ATI, graphics products group of Advanced Micro Devices, and Nvidia Corp. are projected to release their new high-end graphics solutions later this year, it is definite that high-speed memory will be in demand.

Samsung’s new GDDR5 chips are now available in 32Mb x32 configuration (128MB) and are also configurable as a 64Mb x16 device.

Samsung expects GDDR5 to account for over 20% of the total graphic memory market in 2009. The company also said it plans to expand the 50-nm process technology throughout its graphics memory line-up this year.

Tags: Samsung, GDDR5, 50nm


Comments currently: 3
Discussion started: 02/12/09 08:37:49 PM
Latest comment: 02/14/09 03:06:22 AM
Expand all threads | Collapse all threads


Damn thats fast! 0_o
0 0 [Posted by: digitalrurouni  | Date: 02/12/09 08:37:49 PM]

I'm not a memory expert, but as far as I can recollect the same trick is used all over again - the base clock stays low (~200 Mhz?) and agressive prefetching is used to achieve those insane equivalent speeds - which results in higher latency. I'm not really sure that any GPU can make use of such bandwidth right now, but tech must keep on moving, anyway.
0 0 [Posted by: npp  | Date: 02/13/09 06:50:45 AM]
- collapse thread

You got the idea.

For a serial processor, yeah, high latency is a bad thing. Fortunately GPUs are parallel and thrive off of sheer bandwidth.

BTW this RAM is for next-gen cards. Think: GTX380 or HD 5870. They'll love these chips.
0 0 [Posted by: cheeseman  | Date: 02/14/09 03:06:22 AM]


Add your Comment

Related news

Latest News

Tuesday, July 22, 2014

10:40 pm | ARM Preps Second-Generation “Artemis” and “Maya” 64-Bit ARMv8-A Offerings. ARM Readies 64-Bit Cores for Non-Traditional Applications

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones

Monday, July 21, 2014

10:32 pm | PQI Debuts Flash Drive with Lightning and USB Connectors. PQI Offers Easy Way to Boost iPhone or iPad Storage

10:08 pm | Japan Display Begins to Mass Produce IPS-NEO Displays. JDI Begins to Mass Produce Rival for AMOLED Panels

12:56 pm | Microsoft to Fire 18,000 Employees to Boost Efficiency. Microsoft to Perform Massive Job Cut Ever Following Acquisition of Nokia