Sign up for our Technology Newsletter 
Dell Intros Ivy Bridge Xeon Servers
Dell has become the first to announce servers using Intel’s latest Ivy Bridge Xeon E3 processors.
Intel launched its single socket Ivy Bridge Xeon E3 processors a month after it wowed everyone with its dual-core Sandy Bridge Xeon E5 processors, and it has taken Dell only another month to announce the first servers to make use of Intel’s latest nearline server chip. Dell’s Poweredge C5220 microserver uses Xeon E3 1200 series processors that have thermal design power (TDP) down to 17W.
Dell is pitching its Poweredge C5220 servers towards high performance computing, cloud deployments and content delivery networks. While Dell calls the Poweredge C5220 a microserver, that really isn’t a reference to its size or density, but rather the fact that it is a single socket server.
Dell offers the Poweredge C5220 with either 17W or 45W TDP Intel processors supporting DDR3-1600 memory. The firm claims close to double the performance over previous generation single socket servers, mainly due to a 50 per cent increase in density.
Trinity Launching On Desktops This Summer
Comments Off on Trinity Launching On Desktops This Summer
AMD is expected to introduce its new mobile Trinity APU in a week or so and now we are hearing some timeframes for desktop parts as well.
According to Digitimes, desktop Trinity parts are coming in August, while Brazos 2.0 chips are expected in June. There is no word on Trinity ULV parts yet and we believe they will be the most interesting of the lot.
Is Moore’s Law Dead?
May 11, 2012 by admin
Filed under Uncategorized
Comments Off on Is Moore’s Law Dead?
Physicist Michio Kaku says that Moore’s law will be dead within about 10 years.
Kaku, professor of theoretical physics at City University of New York told BigThink.com that we are already seeing a slowing down of Moore’s Law. He said that computing power simply cannot maintain this rapid, exponential rise using standard silicon technology.
Kaku said that the latest CPUs from Intel, which use a unique three-dimensional design, do continue roughly doubling processors. But he points out that the new design is nonetheless proof that the Law is winding down.
The two basic problems are heat and leakage and that is why the age of silicon will eventually come to a close.
By continuing to shrink the parts that go into processors, heat becomes concentrated. At a point in the near future, the heat generated will be so intense that the chip will melt. You can literally fry an egg on top of the chip, and the chip itself begins to disintegrate, he said.
Asus Stakes Claim In Tablets Market
May 4, 2012 by admin
Filed under Around The Net
Comments Off on Asus Stakes Claim In Tablets Market
Asus has won critical acclaim for its early tablets, including the quad-core Transformer Prime TF201 that was debuted last year and the new quad-core Transformer Pad TF300T that will be available in some U.S. retail stores next week.
And it continues to improve its place in the crowded market — jumping into the Top 5 list of suppliers for the first time in 2011, according to the latest research from IDC. Asus ranked fifth in tablet shipments in both the U.S. and world in 2011 with a 2.3% share of the U.S. tablet market and a 2.5% share of the worldwide market.
Apple’s iPad remained at the top of both lists.
Though Asus still lags somewhat behind the market leaders, its influence on the market is clearly on the rise and could move it further up the tablet list, analysts say.
For instance, along with launching an array of powerful higher-priced new tablets, Asus is reported to be Google’s choice to supply a low-priced $200 device said to be called Google Play. The Google tablet would run Ice Cream Sandwich and a Tegra 3 quad-core processor, reports say.
The reports suggest Google is looking to launch the new tablet at its Google I/O developer conference in late June.
Intel’s Itanium Processor Available Until 2022
Comments Off on Intel’s Itanium Processor Available Until 2022
HP will have access to Intel’s Itanium processor until 2022, according to Intel’s Kirk Skaugen.
Skaugen, who used to be VP of Intel’s Datacentre and Connected Systems Group, testified under oath during the HP versus Oracle lawsuit that HP and Intel had an arrangement that “enabled HP to have access to the Itanium microprocessor through 2022, and that HP could extend even longer”. Skaugen’s testimony was submitted as part of court documents filed on 23 April and gives some indication of how long both HP and Intel were expecting to push Itanium.
Oracle, which has been involved in an increasingly bitter spat with HP over Itanium and the hiring of its current co-president and former HP CEO Mark Hurd, claims Intel’s Itanium was on its way out, alleging that HP mislead it and customers, leading to its decision to drop support for the IA64 architecture.
However, according to HP, “Oracle resorts to mischaracterising HP’s statements, taking them out of appropriate context, or misstating the relevant timeline.”
HP claims Skaugen’s comments show that when HP said Itanium had a long future it wasn’t lying. “By any measure, all of HP’s statements regarding the length of its roadmap and the future of the Itanium microprocessor were true,” HP said in documents filed with the court.
Skaugen’s comments are something of a surprise, not because they show an agreement between HP and Intel – HP already admitted that one existed – but the length of Intel’s commitment to HP on Itanium and the fact that HP could extend it.
HP’s language suggests that it, as the customer, had the upper hand in the contract with Intel, with the firm saying, “Intel was committed by contract to continue producing Itanium microprocessors”. Skaugen’s testimony in court supports HP’s claim.
AMD Shows Off New Radeon Chips
AMD has now officially updated its HD 7000M lineup of mobile GPUs with HD 7700M, HD 7800M and HD 7900M GPUs that are all based on AMD’s 28nm GCN architecture. The first in line that should show up in notebooks is the flagship HD 7970M GPU.
As noted, the entire lineup is based on AMD’s 28nm GCN architecture, but as always, the naming scheme of the AMD mobile parts has nothing to do with the desktop ones. The HD 7900M, codename Wimbledon, is actually based on 28nm Pitcairn destkop GPU that features 1280 stream processors, 80 texture units, 32 ROPs and up to 2GB of GDDR5 memory paired up with a 256-bit memory interface.
The HD 7800M series, codename Heathrow and the Radeon HD 7700M series, codename Chelsea, are both based on Cape Verde desktop GPU but with a slight twist. While Heathrow, HD 7800M series features fully enabled Cape Verde GPU with 640 stream processors, 40 texture units and 16 ROPs, the Chelsea HD 7700M series is based on a “crippled” Cape Verde core with 512 stream processors and 32 texture units. Both the HD 7800M series and the HD 7700M series will feature up to 2GB of GDDR5 memory paired up with a 128-bit memory interface. Of course, we expect at least two SKUs for both HD 7700M and HD 7800M series.
AMD also decided to ditch the PCI-Express 3.0 support on the HD 7700M series mainly as this one is aimed at lower-performance platforms that are all about power saving, and performance gain was simply too low.
For now, AMD has only shed precise details regarding the HD 7970M GPU that is based on the fully enabled Pitcairn GPU with 1280 stream processors that will end up clocked at 850MHz for the GPU and 4.8GHz for 2GB of GDDR5 memory paired up with a 256-bit memory interface.
As you can notice, the HD 7970M has lower GPU clocks than its HD 7870 desktop counterpart, but AMD also decided to keep the memory at the same level resulting in 153.6GB/s of memory bandwidth. According to AMD slides, HD 7970M should end up to be anywhere between 30 and 60% faster than HD 6990M and anywhere between 16 and 76 percent when compared to Nvidia current high-end GTX 675M GPU.
Is Samsung Pursuing The Server Market?
Comments Off on Is Samsung Pursuing The Server Market?
It is certain that Korean electronics giant Samsung will soon be entering the server chip market.
Reports are coming in that the company has been picking up key server personnel from Intel and AMD. Samsung has been focused on developing ARM chips and stayed clear of the x86 architecture used by Intel and AMD.
But the companies latest hires seem to indicate that might change.
Samsung’s latest recruits include veterans of the chip business like Jim Mergard, Frank Helms, who is a Fusion APU architect, Brad Burgess who designed the Bobcat APU and Patrick Patla (VP of AMD’s server business). Patla was behind the success of the Opteron chip set and has done well using the x86-server system.
AMD’s Trinity To Have Fewer Cores
AMD’s soon to launch A10 5800K is a 100W quad-core Trinity 32nm CPU with 3.8 GHz base clock and 4.2GHz maximal clock possible with AMD turbo core dynamic overclocking technology.
The A10 5800K has 4MB of L2 cache, supports DDR3 1866, dual graphics configurations as well as AMD’s new FM2 socket. The fun part is new HD 7660D GPU that works at 800MHz and comes with 384 shader units. The current APU market leader A8 3870K that works at 3GHz has HD 6550 graphics with 400 cores running at 600MHz.
AMD claims that new Radeon cores from Trinity CPU including A10 5800K are more efficient and this is the main reason why you have fewer cores that can deliver superior performance. The other reason is that with 800MHz core clock they can probably process more data, meaning that HD 7660D of A10 5800K should end up quite a bit faster than the Llano A8 3870K.
All these Radeon cores are a key feature of the Vision Engine that accelerates GPU enabled applications. AMD also tells the world that Trinity is DirectX 11 compatible, supports Direct compute and the new A series of processors, including the A10 5800K all the way to dual-core A4 5300, should not have any issues playing Blu-ray 3D. The GPU part of Trinity supports AMD V, UVD3 as well as Open CL acceleration.
AMD 7990 Specs Unveiled
As we draw close to the official Nvidia GTX 680 Kepler GK104 launch day it is no wonder that we’ll hear more and more about AMD’s upcoming dual-GPU HD 7990 graphics card and Chinese site INPAI has shed some new light on the dual-headed beast. Apparently, the HD 7990 will pack two Tahiti XT chips squeezed together on one PCB.
According to the post at INPAI, the upcoming HD 7990 features two 28nm GCN Tahiti XT chips, same one found on the HD 7970 graphics card. This means that we are looking at a card that will have 2048 stream processors and 3GB of memory per GPU. Appearently, these two Tahiti chips will end up clocked at 850MHz while memory will work at 1250MHz (5GHz effective).
Did HP Really Need Intel?
Back in the day HP said it chose Intel to co-develop Itanium due to its process technology, as it didn’t have the cash to fund next generation fabs.
Intel has often been the butt of Itanium jokes, but HP invested as much, if not more into the enterprise. Kirk Bresniker, CTO of HP’s Business Critical Systems told The INQUIRER that HP needed Intel’s manufacturing capability. Bresniker said HP’s decision to partner with Intel was due to the firm realising it couldn’t afford the VLSI manufacturing process iteration needed for developing competitive chips.
Bresniker said, “It is really an extension of the CISC processor that led us to partner with Intel on the Itanium. We knew we weren’t going to be able to maintain the investment levels neccessary to continue to fund deep sub-micron fabs.”
Until HP’s foray with Itanium, the firm was known for its PA-RISC systems, some of which Bresniker designed himself back in the early 1990s. Bresniker said, “We got to the point of microprocessor development and more importantly the economics of fabrication environments and realised we were facing transition to the deep sub-micron [fabrication processes] and potentially writing billions and billions of dollars worth of cheques for fabrication, and part of the impetuous for us to partner with Intel on the Itanium design was that we wanted to have access to the world’s number one microprocessor silicon fabrication.”
While HP continued with PA-RISC chips well into the new millennium, HP’s decision to offload the work of actually producing chips onto Intel could be seen as shrewd move, and one that firms such as AMD did a decade later. The cost of process node iteration is getting ever higher, which is something that Intel itself admits.
Not surprisingly, Bresniker wouldn’t be drawn on the demise of Itanium, though HP did announce Project Odyssey late last year, which effectively mixes and matches the firm’s Itanium kit with Intel Xeon servers. He did admit that the firm had to go towards x86 in the mission-critical market, Bresniker was quick to point out that while Intel is porting more features from the Itanium chip, not everything will be moved over.