Nanotech - The Circuits Blog

Updated at 2:15 p.m. PDT: adding information about Dell system.

The main message of the new Core i5 chip is simple: it's cheap--even cheaper than Intel chips based on older technology.

(Credit: Amazon)

The i5, which brings Intel's new "Nehalem" microarchitecture into the mainstream PC market, immediately makes many, if not most, of the older desktop processors obsolete. Consumers need look no further than pricing on sites like Amazon. The i5-750 lists for $250, while the older--based on Intel's last-generation "Core 2" microarchitecture--Q9650 lists for $319.

The official pricing from Intel in quantities of 1,000 units makes the price gap even more stark: $196 for the i5 and $316 for the Q9650.

"The new Core i7's and Core i5's bring pricing to more mainstream levels, with the Core i5-750 at a 1KU (1,000 units) price of $196, which is well below the Core 2 Quad Q9650 at $316," said Intel spokesman George Alfs.

"We are very serious about bringing all new Core processors to new price points and you'll see this trend continue with Westmere," he said, referring to Intel's upcoming processors based on a next-generation 32-nanometer manufacturing process.

Comparing the old with the new, some consumers might point out that the older Q9650 has, for example, more on-chip memory and a higher clock speed than the Core i5. But the writing is on the wall: consumers will almost always opt for new over old when new is less expensive.

On Tuesday, Dell began offering the Studio XPS 8000 tower with the Core i5 starting at $799 and packing 4GB of "Dual Channel DDR3 memory" and a 500GB hard disk drive, among other features. Adding a 20-inch monitor hikes this to $979.

The message is more muddled, however, for the updated Core i7 processors because they maintain the same "i7" identifier as their predecessors--first launched in November--but offer different features that are not readily apparent to less-sophisticated buyers and potentially vexing for some savvy consumers.

"It gets confusing for the more technically knowledgeable buyer, and for us as system builders," said Kelt Reeves, president of enthusiast PC maker Falcon Northwest. "Buying a Core-i7 950 model? Well then you can have a maximum of 12 gigs (gigabytes) of triple channel memory and you buy your memory in sets of 3 sticks. Buying a Core i7-870? Well then your memory is installed in pairs and the max you can have is 8 gigs," he explained.

Reeves continued. "For instance, if you're a heavy Photoshop user having 12 gigs of the fastest memory might be very important to you," he added, saying in that case a consumer would want to opt for a Core i7 900 series over the newer 800 series.

There are other gotchas too. On the i5 processors a feature called hyperthreading is not included, as CNET's Rich Brown pointed out Tuesday. Hyperthreading effectively doubles the number of tasks--or processing threads--a chip can do. "Heavy multitaskers and those who use multithreaded software will feel the loss here," Brown said.

The Hot Chips conference in Palo Alto, Calif this week is focusing on high-end chips for servers and scientific computers, with IBM's upcoming Power7 as a standout.

On Tuesday, IBM will give a presentation on its next-generation server chip, the Power7. IBM documentation describes the chip as having up to eight cores. A dual-chip module holds two processors for a total of 16 cores, according to IBM.

Each core has a rated performance of 32 gigaflops, providing 256 gigaflops per processor--one of the fastest chips to date based on this scientific-centric performance benchmark.

Power7 will be used in the National Center for Supercomputing Applications "Blue Waters" supercomputer, the first system of its kind to sustain one petaflop performance on a range of science and engineering applications, according to the NCSA. A petaflop is one quadrillion floating point operations per second.

Power7 "will be the first of a powerful new system design from IBM. The design includes extensive research and development in new chip technology, interconnect technology, operating systems, compiler, and programming environments," according to the NCSA.

Other chips to be described at the conference include the Sparc64 VIIIfx: Fujitsu's new 8-core processor for Peta scale computing. Sun will discuss its "next-generation multi-threaded processor Rainbow Falls" and AMD will spell out its Magny Cours processor, 12-core chip.

Intel will present a paper on its upcoming Nehalem server processor.

Intel will also discuss Moorestown, an upcoming version of the Atom processor targeted at mobile Internet devices and smartphones. Intel will also give a presentation entitled "Understanding the Intel Next Generation Microarchitectures (Nehalem and Westmere) transitioning into the Mainstream."

Core i3, i5, i7. A straightforward, if not insipid, branding scheme, right? Wrong. Those alphanumeric identifiers are fighting words.

Last week, Intel announced a new branding scheme for its upcoming processors. In a blog, spokesman Bill Calder wrote that the branding will be "simplified into entry-level (Intel Core i3), mid-level (Intel Core i5), and high-level (Intel Core i7)." Intel calls the "i" suffix an identifier.

The upcoming Lynnfield chip for desktop PCs, for example, will be available as either Intel Core i5 or Intel Core i7 depending upon the feature set and capability. The upshot of the new branding is to make it easier for less tech-savvy consumers to readily identify classes of Intel chips based three simple identifiers, according to Calder.

But judging by the tenor of many of the comments attached to Calder's brand structure blog, you would think the chipmaker had committed high treason.

In the minds of some, it did. The shortcomings of the current naming scheme notwithstanding, many tech-savvy consumers have gotten used to it. For example, Core 2 Quad means a chip built on the Core 2 architecture with 4 processing cores. Core 2 Duo indicates two cores.

One of the most common criticisms cited in the comments section is that i3, i5, and i7 are too vague. "Above all, I'd like to see...at a glance how many cores and what features they have (or have not)," one comment said. Another comment suggested that Intel add more identifiers. For example, Intel Core i5 4100, where 4 is the number of cores and 100 is a speed rating.

Yet another idea was this: Intel/name/number/year, where "name" is the product name, "number" is a bigger-is-better ranking, and "year" the year the architecture was released.

And another: "Either ditch the Celeron, Pentium and Xeon names completely or embrace them completely. These are fairly well known as the 'good, better, best'."

AMD launched its first six-core processor, which will compete against Intel's "Dunnington" chip.

The Cray XT5m supercomputer will use the Istanbul chip

(Credit: Cray)

The "Istanbul" Opteron processor is for high-end server computers that use two, four, and eight processors or "sockets." Intel has been shipping a six-core processor for this market since September of last year and will bring out a processor based on its new Nehalem architecture for this segment later this year.

Among other new features, AMD is touting an Istanbul technology called HT Assist. The previous way of retrieving data from the processor's memory was "like checking every room in your house for your car keys," said Pat Patla, AMD's general manager of the server workstation group, in a phone interview last week. With HT Assist, "You know where your car keys are," he said. "It's much more efficient and takes out a lot of traffic," Patla added.

Patla also said that AMD is ahead of schedule with this chip, in stark contrast with the company's ill-fated quad-core Barcelona processor, which saw repeated delays. "Within almost a 15-month period of time we were able to design the product, tape out the product (final stage before production), verify the product and launch," Patla said about Istanbul.

Systems based on six-core AMD Opteron processors are expected to be available beginning this month from server suppliers including Cray, Dell, Hewlett-Packard, IBM and Sun Microsystems, AMD said Monday.

Istanbul boasts up to 34 percent more performance-per-watt over the previous-generation quad-core processors.

Intel on Tuesday said it will ship a server chip that contains up to eight processing cores later this year, while IBM showed off a high-end server in the works that uses eight such chips, yielding 64 cores.

Intel's Nehalem-EX architecture supports up to eight processors and each processor can integrate up to eight cores.

(Credit: Intel)

Intel's Nehalem-EX processor, in production later this year and expected to be shipping in high-end server systems by early 2010, will feature up to eight cores inside a single chip that supports 16 threads, according to Boyd Davis, Intel's general manager of the Server Platforms Marketing Group, speaking at a teleconference on Tuesday.

Using threads, Intel essentially doubles the amount of work that can be done on each processing core.

IBM, which participated in the conference, discussed a server currently under development that uses 64 Nehalem-EX cores (eight processors) and can handle 128 threads, according to Alex Yost, vice president IBM BladeCenter. "We're very excited today to be the first to demonstrate Nehalem-EX," Yost said.

Nehalem-EX will also double the memory capacity with up to 16 memory slots per processor socket, and offer four high-bandwidth "QuickPath" Interconnect links.

Intel also said the currently-shipping Nehalem server chip is making market gains. Intel's currently-available Xeon 5500, the first server processor based on Intel's Nehalem architecture, will be "greater than half of shipments" for Intel's high-volume two-processor (aka, "two-socket") server shipments by August, according to Davis.

"Customer acceptance has been quite strong," Boyd said. "From an introduction at the very end of March to representing the majority of our shipments in the market for two-processor servers by the August time frame," he said.

Intel showed off a prototype server that can accommodate four eight-core Nehalem-EX processors.

(Credit: Stephen Shankland/CNET)

Intel's prototype Nehalem-EX server accommodates eight of these memory cards. They'll use relatively conventional DDR3 memory rather than the FB-DIMM technology Intel's current Xeon 7300 systems.

(Credit: Stephen Shankland/CNET)

Updated at 10:00 a.m. PDT with correction about launch of Nehalem-EX processor.

Intel is expected to announce details of an 8-core processor for the high-end server market next week.

The chip itself will not actually ship in systems until late 2009 or early 2010.

The 8-core "Nehalem-EX" Xeon processor is designed for servers that can use more than two processors (referred to as "sockets" in server argot). Currently, Intel is shipping Nehalem Xeon processors for servers with two sockets.

Nehalem is the same architecture used in Intel's Core i7 desktop processor line.

The Nehalem-EX is expected to become Intel's top-line Xeon processor. Currently, the six-core "Dunnington" processor, based on Intel's older Penryn architecture, is Intel's highest-performance chip for multi-processor servers.

Nehalem-EX packs 2.3 billion transistors and its eight cores are capable of executing 16 threads (or tasks) at the same time. The chip also has "integrated power gates" for lowering power-consumption.

The announcement of details is slated for May 26. Boyd Davis, Intel's general manager of Server Platforms Marketing Group, will host the roll-out event and "discuss how this new server product raises the standard in cost-effective RISC replacement solutions," according to an Intel note about the event.

Advanced Micro Devices' server roadmap is solid but its mainstream mobile lineup is languishing.

AMD's six-core "Istanbul" processor release was moved up to June

(Credit: AMD)

First, the good news. These days AMD is walking the talk. This is a radical change from the AMD of 2007-2008, which always seemed to have a hopper full of Intel-vanquishing paper processors that, if they did materialize, disappointed.

Fast forward to AMD's Tuesday earnings announcement, when the company said it was actually moving up the introduction its most sophisticated processor, the six-core Istanbul, to June.

And AMD has proved its silicon mettle at large server customers such as IBM and Sun Microsystems--the latter's executive vice president John Fowler had nothing but praise for AMD processors in high-end Sun server systems.

In a "Global Webcast" on server technology Wednesday, Patrick Patla, a vice president in AMD's server and workstation business, revealed a strong roadmap, saying that 8- and 12-core "Magny-Cours" processors will appear in 2010. "We're currently working on new processors which we expect will deliver more than 35 times the performance of the original single-core AMD Opteron processor released in 2003," Patla said in a statement.

Intel, of course, will also bring out many-core processors, but AMD is keeping pace, and, according to people who should know, like Sun's Fowler, maybe more than keeping pace.

Now, the bad news. This post today on CNET's Crave blog says it all: "One of our biggest issues with HP's Pavilion dv3z was its AMD processor, keeping it from beating out the performance of comparable 13- and 14-inch laptops with Intel Core 2 Duo CPUs." This is, by no means, the first review that expresses this sentiment. In short, AMD mobile platforms consistently come up short in the high-profile, burgeoning laptop market. Will AMD close the gap in 2009?

Maybe one answer to that question is AMD's Neo chip that powers the low-cost, ultra-thin HP Pavilion dv2 laptop. More than a Netbook but less than a mainstream laptop, this kind of sleek mobile device could eventually eclipse the high-end Netbook segment.

AMD CEO Dirk Meyer said Tuesday that the single-core Neo processor will get a dual-core sibling dubbed "Congo" by summer. A dual-core processor in this low-cost, MacBook-Air-for-the-masses category is a compelling proposition. AMD needs to stay ahead of the game, especially when Intel brings out chips for this category in the May-June timeframe.

A start-up founded by former Sun Microsystems computer scientists is tapping IBM and Intel hardware to accelerate the enormous server workloads of burgeoning Web 2.0 businesses.

Menlo Park, Calif.-based Schooner Information Technology announced Monday that it is readying a server appliance based on Intel's newest Nehalem processors and its solid-state drives. The first products are due by the end of May with volume shipments in the third quarter of 2009.

Hewlett-Packard and Fusion-io said recently that they are working on analogous technology and had achieved extremely high performance using Fusion-io's solid-state drives running on HP servers.

Schooner Information Technology's President and CEO John R. Busch was formerly research director of computer system architecture and analysis at Sun laboratories. Chairman and CTO Tom McWilliams was a lead engineer at Sun, working on server architecture and advanced CAD tools. Prior to that, McWilliams was a director in the MIPS division of Silicon Graphics. Both men were involved in moving Sun to multicore server architectures, according to Busch.

The company is funded by CMEA Capital and Redpoint Ventures. The current total investment is $15 million.

In a phone interview Monday, CEO Busch explained that the company has set out to fuse standalone high-performance server technologies into a faster organic whole. "Computer companies are pretty much selling boxes while others are selling networking. They're basically just selling component technologies," he said. "If you just speed up the processor or speed up the interconnect or add in flash drives, it will have a small effect."

"The observation I had when we started the company was that we really need to make a shift and we really need to put the middleware application and (our) new operating environment together with these technologies--tightly coupled with parallel flash memory and with Intel multicore processors. As opposed to loosely coupled, in order to bring their real inherent benefits through," Busch said.

In the wake of reported merger talks with IBM, Sun Microsystems executive vice president John Fowler talked about Big Blue as a rival.

Sun Microsystems executive vice president John Fowler

(Credit: Sun Microsystems)

Fowler, in a phone interview Friday, discussed IBM as a competitor in the server computer market and the competitive differences between Intel and Advanced Micro Devices.

"IBM is obviously a classic competition space where we bring to bear all of our technology innovation," Fowler said. "In this particular case, how we've incorporated and done networking technology, how we've incorporated and done flash (memory). The fact that we have an open operating system that runs across our RISC platforms and our x86 platforms and that we have an open storage offering on the 7000," he said.

Fowler continued, "Those are all great discussions to go have with IBM customers. Those are all things that IBM just doesn't have and represent a significant amount of value."

In a question about comparing IBM--which also offers a RISC architecture server platform (PowerPC) as well Intel and AMD--to other competitors like Hewlett-Packard and Dell, Fowler said: "They (IBM) are more similar (to Sun) than the others and interested in more of the same things."

And how does Sun compete with larger rivals? "We have for decades now innovated in a marketplace where companies like HP and IBM have been consistently bigger than us. What we do is incorporate new technologies or invent new technologies more quickly," he said.

Fowler also spoke of some of the differences between Sun offerings on AMD and Intel platforms. He began by saying that a recent Intel Nehalem server chip announcement covers only two-socket (a socket accommodates one processor) servers and that Intel's product is an "entry level" product. "It's a solid product but it really applies to the volume entry-level systems in the world today," he said.

Fowler continued: "Obviously, AMD and (Sun) SPARC cover a much broader range of application scale. In AMD's case, extending up to eight-socket servers and in SPARC's case up to 64-socket servers," he said.