by Anton Shilov
02/23/2010 | 11:16 AM
Advanced Micro Devices said this week that it would not cut-down feature set of its forthcoming AMD Opteron processors code-named Magny-Cours with eight or twelve processing engines. This means that all next-generation multi-core chips from AMD will have the same functionality and features and will only differ in terms of performance, which is supposed to improve value of the microprocessors to customers.
“When you look at our existing AMD Opteron processors as well as our upcoming new products (Magny-Cours), you see two big themes: we don’t compromise on features and we are going to deliver the feature customers care most about: value. As you look at our products, you’ll see that as you move down the stack, from those fastest processors that everyone lusts after (but few buy) down to the value-priced processors, you’ll see the same set of features. You won’t see AMD artificially limiting the capabilities or punishing a focus on low power,” said John Fruehe is the director of product marketing for server/workstation products at AMD.
Although making promises like those that AMD does at present hardly makes a lot of sense when performance, price and power consumption of the chips are not clear, it looks like the company wants to make it clear that all of its chips will work equally efficiently. By contrast, AMD’s arch-rival Intel Corp. from time to time cuts down certain features from less expensive desktop or server central processing units (CPUs) for marketing or power consumption reasons.
“Scaling back the features for marketing purposes is really not a very appetizing prospect for us. We believe that you should have the full set of features, no matter where you buy in the stack. You won’t see us scale back the memory speeds. You won’t see us scale back the I/O speeds. You won’t see us pull features. That is not our way to bring products to market,” said Mr. Fruehe.
According to preliminary information from unofficial sources, the mainstream line of twelve-core AMD Opteron 6000-series processors will consist of three chips operating at 1.90GHz, 2.10GHz and 2.20GHz. Besides, AMD will also release highly efficient (HE) and special edition (SE) AMD Opteron 6000 microprocessors with twelve processing engines functioning at 1.70GHz and 2.30GHz, respectively. In addition, AMD plans to launch three standard-voltage eight-core chips at 2.0GHz, 2.30GHz and 2.40GHz frequencies along with two HE eight-core processors at 1.80GHz and 2.0GHz clock-speeds. In order to ensure maximum stability, all AMD’s twelve-core processors will come with reduced clock-speed of integrated memory controller and L3 cache (1.80GHz) compared to six-core and quad-core products.
Even though clock-speeds of AMD Opteron 6000-series processors are not high, AMD will still increase thermal design power of Maranello server platform. Based on preliminary information, AMD G34 (1944-pin) CPUs will have 85W, 115W or 140W TDP, which is somewhat higher compared to TDPs supported by existing multi-processor platforms.
“You will see our processors vary in price based on core count and clock speed. You won’t see us pull a feature that you count on, and you really won’t see us pull features that could impact software images across different servers. Can every processor company in the x86 server business claim this?” asked the director of product marketing for server/workstation products at AMD.
The company hopes that its next-generation servers will still offer great performance-per-watt despite of higher power consumption of twelve-core and eight-core chips thanks to increased amount of cores and other innovations of the next-gen platforms. In particular, the new processors will have HyperTransport bus speed increased to 6.4GT/s, which will increase performance scalability, as well as Cool Speed and C1E technologies to reduce power consumption. In addition, it can be expected that Globalfoundries 45nm silicon-on-insulator process technology will improve a bit by the end of Q1 2010.
AMD Opteron “Magny-Cours” processor will be the first chip for the AMD G34 “Maranello” platform designed for Opteron processors 6000-series with up to 16 cores, quad-channel memory interface, 2 or 4 sockets, up to 12 memory modules per socket and some server and enterprise-specific functionality. Magny-Cours microprocessors feature two six-core or quad-core dies on one piece of substrate.
Even though AMD promises not to cut-down features of its high-end AMD Opteron 6000-series microprocessors for 4-way servers, the company will still disable support of quad-rank DIMM, low-voltage DDR3 and C1E technology by standard voltage by quad-core AMD Opteron 4000-series microprocessors for 2-way servers.
AMD Opteron “Magny-Cours” processors are being shipped for revenue now.