|
 |
 |
 Processors |
 Motherboards |
 Chipsets |
 Memory |
 Graphics Cards |
 Storage |
 Cases and Cooling |
 Mobile |
 Systems |
 Displays |
 Shows and Expos |
 .:Latest Topics |
:: PC Perspective . Processor
AMD Llano APU Displayed at GlobalFoundries Technology ConferenceProcessor - Sep 01, 2010 | 03:00 PM AMD was on stage today at the first annual Global Technology Conference hosted by former AMD production arm GLOBALFOUNDRIES talking about their move to 32nm process technology and their partnership with the foundry. Most importantly though we saw the first public showing of the AMD Llano Fusion desktop platform APU!! 
AMD's Bulldozer Architecture Preview: New from the Ground UpProcessor - Aug 24, 2010 | 12:02 AM The original Athlon architecture has been refined and reused for the past decade at AMD. Now we finally received the first glimpse of a brand new architecture from AMD, and one that they hope will again place them at the forefront of innovation. Bulldozer is AMD's best hope to relive the success of the original Athlon and Athlon 64, and it has a good chance of doing so for the company. 
Intel Demonstrates 50 Gbps Silicon Photonics, Could Replace Electronic SignalsProcessor - Jul 27, 2010 | 03:00 PM Today Intel held a press briefing to discuss a milestone in the development of silicon photonics reaching a reproducible 50 Gbps link between two modules. Silicon photonics is the process of creating, modulating and reading photons via lasers for communications entirely on manufactured silicon. The technology and work being done is very complex but the ideas are pretty simple: create a piece of silicon that can create a laser to transmit data and another chip that can receive and decode the data at a low cost.
Jeremy Hellstrom | Source: InsideHW | Subject: Processor
If you are in the mood for some really big performance charts drop by InsideHW for comparisons of just about every x86/64 processor on the market currently. Covering Athlon II and Phenom II, Celerons, i3, 5 and 7 as well as some Pentium chips you can see how they all match up. To sum up the article InsideHW breaks up the results they saw by a number of common uses for PCs, from gaming to surfing and Facebook and they offer a top contender in each category.
"Do you remember the times when CPUs were chosen based on whether you were assembling a gaming PC, in which case you'd opt for cheaper Celerons and Durons, or a heavy workload PC, in which case you'd buy the more expensive Pentiums and Athlons? It came to our mind on multiple occasions during this test. Those days of single-core CPUs, when Megahertz were the only relevant thing, and the difference between home and professional models consisted in the difference in the quantity of cache memory, only served to underline the problems that modern-day users are facing when assembling a new PC…" Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: PC Perspective | Subject: Processor
ARM has finally taped out their 28nm process dual core Cortex A9 HKMG processor at Global Foundries; the processor will run between 2-2.5GHz and offer improved performance over their past generation which was 40nm and ran about 500MHz slower.
At first you might assume this means very little to the enthusiast but that is a mistaken assumption as the number of ARM processors currently in use is estimated to be about 20 billion or about 90% of all embedded 32-bit RISC processors in use. ASUS has demonstrated a Win7 based tablet that is powered by a ARM v7 processor so these new dual core, low voltage, low heat processors will be powering a device that you own very soon. A second example would be nVIDIA's Tegra which is also designed with the help of ARMs architecture and represents something very different from the usual embedded applications that ARM is associated with. As it has just taped out we do not have a lot of information on the architecture or the capabilities of this chip but you can read about the current generation single core 40nm TSMC made Cortex A9 at ARMs website here. It's the processor you didn't even know you loved and it is about to get better.
Click on the picture for a larger version of the slide.
Ryan Shrout | Source: PC Perspective | Subject: Processor
We don't know much about it, but at the first annual Global Technology Conference hosted by GLOBALFOUNDRIES, AMD's Chekib Akrout showed the first images of the upcoming Orochi core processor:
Here is what we know for sure about the upcoming Orochi processor: it is going to be the second 32nm product from AMD after the upcoming Llano Fusion core is built, it uses a set of 4 Bulldozer modules that bring 8 processing cores and 8 threads with integration of AMD's unique SMT alternative.
If you haven't read details about the new Bulldozer core and what it has to offer, definitely check out our recent preview of the processor based on information revealed at the Hot Chips conference last month. Nothing else was shared about the Orochi CPU in particular but we thought the hardware porn was worth the mention!
A good way to express what Bulldozer is can be summed up as “slimmed
down, but double wide”. For each traditional core, AMD has instituted a
dual ALU design with robust floating point and SSE units. Each core
can handle two threads, like SMT, but actually has separate execution
units which each process individual threads without sharing execution
resources.
Each unit features a single fetch and decode stage. The decode stage is comprised of four units, but we do not yet know their inner workings. In the previous K7/K10.5 generations of parts, there are three complex decode units. On the Intel side with Core 2 and Nehalem, there are three simple decode units and a single complex. AMD also did not cover subjects such as macro-ops and macro-op fusion. AMD has beefed up their decode stage significantly though. It simply had to, because it is now feeding dual integer schedulers and a floating point scheduler feeding 2 x 128 bit FMACs and MMX units. 
Fetch, decode, floating point/SSE, and the L2 cache are the shared components. Since most workloads are integer based, AMD doubled the integer units. These 128 bit packed integer pipes are a step above what was offered in the Phenom II. In theory, there should be a sizeable per clock increase in integer and floating point apps on Bulldozer over the Phenom II. When something is more heavily threaded, then we will see dramatic improvements in performance. Each integer core features its own L1 D-cache. AMD has again not clarified how much L1 or L2 cache there is for each discrete unit, or L3 cache sizes for the entire processor.
Ryan Shrout | Source: AnandTech | Subject: Processor
It was bound to happen, really. Anandtech has posted a very in-depth preview of the upcoming Intel Sandy Bridge architecture that includes benchmarks of an engineering sample they "located" somewhere without Intel's blessing. But who cares, right? Let's see what the numbers are telling us!
The Core i5-2400 used in this test is still an early sample, and as such the results might not be indicative of final performance. This 3.1 GHz processor will not ship with HyperThreading enabled later this year but the early samples had it enabled for motherboard vendor testing. CPU performance is compelling in that it offers a noticeable performance increase over currently existing products at similar price points if Intel sticks to the leaked roadmaps that we have seen over the past few months.
Probably more interesting is the updated integrated graphics that gets a very dramatic redesign from previous Intel CPUs. In Anandtech's testing the single core version that resided in this i5-2400 sample was able to outperform the low-end of discrete graphics like the AMD Radeon HD 5400:
This is definitely an article you should be reading if you have any interest in what Intel will be offering you early next year!!
Jeremy Hellstrom | Source: AnandTech | Subject: Processor
The ION platform has been languishing, partly because the low powered Atom processor is only good for certain tasks as its abilities are limited. Of course its power draw and heat output are also very low and it could handle streaming video at a high enough quality to attract some users. Now Zotac is pairing the ION chipset with an Intel CULV processor, which provides more power at the cost of heat and energy costs. AnandTech examines the changes that this brings to the system and if it can bridge the difference between a full system and an Atom plus ION system.
"Last year one of the most exciting product combinations to grace consumer shelves was the NVIDIA’s ION platform teamed up with an Intel Atom processor. The ultra-low power consumption, low heat output and ability to play HD video better than competing solutions of the time made it a difficult combo to ignore. ZOTAC took full advantage of this and successfully filled a niche demand with a slew of ION based products, offering various levels of plug and play functionality." Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: AnandTech | Subject: Processor
32 cores and 64 threads equals an incredible amount of processing power but it comes at a high cost. A basic Dell server using X5670's will run you about $9000 and a higher end model using X7550's can be as much as $32,000. These servers are not intended for home use and really are not optimized to give great gaming by dropping in a high end GPU or four; in the data centre they do represent the current pinnacle of power. Part of AnandTech's review deals with the challenge of designing a way to actually test these beasts and max out there computational potential, using scenarios that would be found in business applications as these servers are intended to be used. Follow the link to see 32 cores running at 100% load but be prepared to never look at your own Task Manager in the same way again.
"Lots of people base their server form factor choice on what they are used to buying. Critical database applications
equal a high-end server. Less critical applications: midrange server. High-end machines used to find a home at larger
companies and cheaper servers would typically be attractive to SMEs. I am oversimplifying but those are the clichés
that pop up when you speak of server choices."
Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: X-Bit Labs | Subject: Processor
The biggest difference between the various lines of processors that AMD has offered us over the past year lies in the cache and the number of processors. Only the Phenoms have L3 cache and the L2 cache for almost all CPUs is 512KB regardless of the number of cores, only the Athlon II X2 has a model featuring 1024KB of L2 cache. X-bit Labs wanted to examine the differences in performance, specifically the difference between a dual core with 1024KB of L2 cache and a triple core with 512KB.
The survey says "16 benchmarks out of 57 preferred fewer cores and larger L2 cache". "Today we shall try to answer the following question. What's better: a triple-core CPU with 512KB L2 per core or a dual-core CPU with 1024KB L2 per core? On the one hand, the former has an additional core. On the other hand, each core of the latter has twice as much cache available. This isn't as obvious as it may seem." Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: The Tech Report | Subject: Processor
The Westmere-EP is a 32nm process and it's current representative is the Xeon 5680, a 6 core 3.33GHz CPU though it will hit 3.6GHz via Intel's TurboBoost technology. If you want a dual CPU PC then the Xeon family is currently the way to go, though you might not choose to use the board that The Tech Report did. The focus of TR's review was not the ultimate in high end machines, instead they chose to test it using low power components as it would have in a server role, though they did go all out and use an SDD for storage. Drop by to see just how well this new CPU will fit into its role in server rooms, especially considering it can be a drop-in upgrade from a 5500 series Xeon.
"Intel's new Xeons promise two more cores per socket at the same clock speed and power envelope. How good are they? We collected a host of systems in Damage Labs to find out." Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: Legit Reviews | Subject: Processor
With one of the best overclocking tricks since the one with the pencil, many unassuming Athlon II X3's hide an unused but perfectly functional 4th core under that shiny heatspreader. The flick of a BIOS setting will tell you if you can try to pull this trick off, though stability testing is very important as some of the cores on these CPUs are disabled for a good reason. Legit Reviews did exactly that with an AMD Athlon II X3 445 and it turned out that they were unlucky enough to receive one with a faulty fourth core. They did get plenty of benchmarks from the three cores that worked though.
"The AMD Athlon II X3 445 was a processor I was looking forward to working with when the opportunity had presented itself. Not necessarily because it is going to be the fastest processor out there, but it only has three active cores. Whats that mean? Well, that most likely there is a fourth core under the heat spreader. With any luck I would be able to unlock the fourth core, and be able to test the AMD Athlon II X3 445 as a Quad Core as well. Fortunately there was a fourth core on the AMD Athlon II X3 445..." Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: Techware Labs | Subject: Processor
As Josh rightly pointed out in this weeks Tech Dirt, AMD been digging around in people's pockets for every available penny they can get any time they so much as looked at purchasing a GPU from them. nVIDIA seems poised to change that, so we will see if the Radeons can get back down to at least their MRSP. The same cannot be said on the CPU side, as they continue to offer a range of fairly inexpensive processors at several levels of performance. Two examples are the Regor based Athlon II x2 260 at $79 and the $70 X2 255, either of which are a great way to start an inexpensive system ... if you don't want to gamble on unlocking the cores of a Phenom II for an extra $10. Drop by Techware Labs to see them in action.
"Dual core chips still have a place in the PC market place. Whether you need something for your HTPC or a new chip for mom, dual cores provide an affordable way to get performance. Join us as we take a look at AMD's latest offering the Athlon II 255 and 260." Here are some more Processor articles from around the web:
Josh Walrath | Source: PC Perspective | Subject: Processor
Over at Coolaler some enterprising young soul was able to lay hands on a Sandy Bridge sample from Intel. For those not in the know, Sandy Bridge is the codename for the new Tick/Tock architecture from Intel. This is the next, big step forward from the previous Nehalem architecture, and promises to again increase per clock performance as well as add in on-die graphics capabilities. Much is not known about this processor, but a few things were unveiled with this post.
Sandy Bridge is going to be a 32 nm product with 32K of L1 data and 32K of L1 instruction cache, 256 KB of L2 cache per core, and a nice 6 MB of L3 cache to be shared around the native four core design. The sample in question is running at a unimpressive 2.5 GHz, but this is not surprising considering it is an early engineering sample. Clockspeeds will obviously be increased substantially when the product is released to desktops in late Q4 2010 or Q1 2011. The integrated graphics portion is a complete mystery though, but considering the big leap that Intel made with Clarkdale's integrated graphics, this area could have received much needed attention and a jump up to DX 11 standards. Rumors have it that many of the former Larrabee staff were transferred to the integrated graphics design teams, and hopefully we should see a product which will at least come close to AMD's Fusion products in terms of graphical performance and functionality. The tests run on this sample were fairly limited, and what little we could gather is that Sandy Bridge does improve upon the performance per clock of Nehalem, but it does not blow it out of the water. I believe that right now Intel has extracted about as much performance from a X86 CPU as they possibly can, considering the manufacturing technology of today. Certainly in the future more performance can be extracted due to relaxed transistor budgets afforded by smaller processes, as well as clockspeed increases and improvements in cache transistor designs that will be released in coming years. This at least creates and opening for AMD to potentially match Intel in terms of per clock performance with their upcoming Bulldozer core, but I certainly do not expect AMD to take any kind of performance lead in most applications. About the only two areas where AMD could potentially score big is in integer limited performance, where their cluster based SMP (two ALU units per core) may outperform what Intel has to offer, and if AMD can get much more software support for their "APU" concept which would leverage the integrated graphics core as a massively parallel floating point and streaming unit. 
Still, Intel is in a far superior position, as this will be their second architecture on a 32 nm process. AMD has yet to release any 32 nm parts, and we have only seen a publicly shown sample of the Ontario APU, which is the low power variant (Bobcat) of their Fusion platform. And that particular sample was reportedly fabricated on TSMC's 40 nm process. 32 nm Llanos are sampling with partners, but this quad core Phenom II variant certainly will not turn heads in terms of per clock performance compared to Sandy Bridge. The saving grace for this particular part could very well be the integrated graphics portion, which is going to be very well supported by AMD and their Catalyst driver program due to its design being essentially lifted from the HD 5000 series of graphics chips. Remember as well that this is an engineering sample, and we do not know what kind of turbo boost we will see in clockspeeds, as well as how it handles individual core speeds depending on loads. It could be another step up from the current Nehalem designs, and thereby increase overall performance across multiple applications (far greater than what we see from this 2.5 GHz clocked core). Intel could also have disabled some advanced functionality in these engineering samples, which would again decrease overall performance as compared to a final, shipping product. But from what we see so far, Sandy Bridge will not disappoint in the traditional CPU market.
Jeremy Hellstrom | Source: X-Bit Labs | Subject: Processor
X-bit Labs had a chance to speak with Sebastian Steibl, the director of Intel Labs Braunschweig, who is working with the team designing Intel's 48 core CPU, aka the Single Chip Cloud Computer. They touch on what architecture these cores use and just how those cores have been designed to be used. It may be a little technical for anyone not interested in L2 cache coherency and the benefits and drawbacks of having heterogeneous core frequencies but for those who are interested this proves to be a good read. As a bonus they asked a few questions of Jon Peddie about the future of discreet CPUs and GPUs and inserted them in the interview.
"Intel’s 48-core single chip cloud computer (SCC) unveiled just about half a year ago has quickly gained a lot of attention to itself as not only the world’s first x86 processor with 48 processing engines, but also as a potential successor of the infamous Larrabee. Today we are speaking with one of the co-designers of the SCC in order to find out more about the ambitious project that is not supposed to come alive. In addition, we have an independent expert Jon Peddie to tell us about the future of CPUs and GPUs." Here are some more Processor articles from around the web:
Ryan Shrout | Source: PC Perspective | Subject: Processor
At Computex 2010 in Taipei this morning AMD held the first public demonstration of a Fusion-based APU (accelerated processing unit): the combination of a traditional CPU and GPU architecture on a single monolithic die. The pairing of high performance serial processing units and parallel graphics processing cores provides the potential for a dramatic shift in the computing and system marketplace.
Rick Bergman, SVP of the AMD Products Group was on hand to give the first public demo of the Fusion APU that brings "power-efficient processors that combine CPU, GPU, video processing and other accelerator capabilities in a single-die design to efficiently power the most popular and demanding consumer experiences, from HD video applications to media-rich Internet experiences to DirectX 11 games." Bergman suggested with the continued migration of users into multimedia consumption and media creation roles the need for the Fusion-based APU has increased dramatically. AMD hopes that the "PC experience will evolve dramatically" when the first Fusion APUs are formally announced in the first half of 2011. 
A generic APU design Specifics were light, but I do know that the APU being shown was not the desktop variant built on 32nm technology but rather was the 40nm Ontario core built at TSMC and aimed at the Atom markets. AMD obviously feels they have a strong advantage in this market with the APU as they are putting focus on it rather than the Llano-based notebook and desktop parts that were originally billed to be the first Fusion parts available. If you were reading this hoping to get a hint of the clock speeds, die sizes or shader counts for Ontario you and I are both disappointed - AMD continues to hold that very close to the chest. 
Bobcat cores will be used in the Ontario APU AMD's APU demonstration comes one day after Intel's Dadi Perlmutter went on stage at Computex and briefly showed the upcoming Sandy Bridge CPU+GPU chip at work. Though details were light, the Sandy Bridge GPU performance was compared to a "current generation mainstream discrete graphics card" and the side by side comparisons of Mass Effect 2 were honestly impressive. Though no more specs were given (frame rate, the discrete GPU being used as comparison) the fact is that after seeing the gaming demonstration I think AMD will have a run for its money in 2011 to have the fastest integrated processor for graphics. 
Details were light, but in real time the graphics on this demo looked great on both systems The demonstration of the APU hardware wasn't the only point brought home by AMD at the press conference. The company also announced the formation of the AMD Fusion Fund that would be making investments in companies that display an innovative solution to take advantage of AMD Fusion APUs and their unique hardware capabilities. By simply submitting an application to AMD a software development house that can showcase the ability to drive demand for the combined CPU+GPU product could get funding to continue their work. Obviously AMD's intent is to increase the demand for both GPU-accelerated applications as well as heterogeneous programs that could take advantage of both CPU and GPU components of the APU. Even though we are essentially combining a current generation processors with the integrated graphics of an IGP motherboard chipset (in very basic terms) there are new complications and new benefits this merger provides and AMD is demonstrating the desire to find them. Fusion has been a long time coming and the product as we know it now is very different from our first glances of it back in 2006. A shift unlike any other in the consumer market is about to take place and the world of the discrete graphics card will never be the same. But we are also hopeful that the world of computing will never be the same as well as the APU (and Intel's versions of a similar idea) will drive innovation in a market that desperately needs it. It might be uncertain for consumers right now but for AMD the future is most definitely Fusion.
TAIPEI, Taiwan — June 2, 2010 — At Computex 2010, AMD (NYSE: AMD) today delivered the first public demonstration of an AMD Fusion processor, initiating the accelerated processing era. The AMD FusionTM Family of Accelerated Processing Units (APUs) represents a significant shift in processor architecture and capabilities, combining high-performance serial computing and parallel graphics processing cores onto a single die to improve visual and data-intensive tasks that are pervasive in today’s computing environments. A video of today’s demonstration can be found here.
Rick Bergman, senior vice president and general manager, AMD Products Group, provided an APU technology demonstration during a press conference today. This demonstration provided a sneak peek into the upcoming seismic shift in the computing industry: power-efficient processors that combine CPU, GPU, video processing and other accelerator capabilities in a single-die design to efficiently power the most popular and demanding consumer experiences, from HD video applications to media-rich Internet experiences to DirectX 11 games. The AMD Fusion Family of APUs represent a distinctly powerful processing approach to the evolving digital consumer landscape, where more than 28 billion videos are watched each month online and a thousand pictures are uploaded to social networking sites every second. “Hundreds of millions of us now create, interact with, and share intensely visual digital content,” said Rick Bergman, senior vice president and general manager, AMD Product Group. “This explosion in multimedia requires new applications and new ways to manage and manipulate data. Low resolution video needs to be up-scaled for larger screens, HD video must be shrunk for smart phones, and home movies need to be stabilized and cleaned up for more enjoyable viewing. When AMD formally launches the AMD Fusion family of APUs, scheduled for the first half of in 2011, we expect the PC experience to evolve dramatically.” Software Development Consumers are hungry for applications that run faster and make digital media easier to enjoy, and a new wave of software innovation is taking place as AMD software partners take advantage of AMD APUs and GPUs to enable better experiences across an ever-widening set of content. Microsoft joined AMD on stage at Computex and discussed how AMD Fusion APUs can enable improvements to applications such as Microsoft Windows 7 and DirectX 11, and how CPU and GPU collaborative computing can enable superior PC experiences. “While visual computing has made incredible strides in recent years, we believe that the AMD Fusion family of APUs combined with Windows 7 and DirectX 11 will fundamentally change how applications are developed and used,” said Steven Guggenheimer, corporate vice president, original equipment manufacturer division, Microsoft. “Applications such as Internet browsing, watching HD video, PowerPoint and more can enable more immersive, visually rich, and intuitive experiences for consumers worldwide.” In addition to Microsoft DirectX with DirectCompute, software developers can also build enhanced applications using OpenCL via the ATI Stream SDK, which further underscores AMD’s commitment to industry standards. AMD Fusion Fund At Computex 2010, AMD also unveiled the “AMD Fusion Fund,” a vehicle to make strategic investments in companies developing innovative solutions that will take advantage of the forthcoming AMD Fusion family of APUs. Additional details were disclosed on the AMD Fusion Fund in a separate announcement. Extending Consumer and Graphics Leadership at Computex With the launch of the AMD Fusion Family of APUs planned for the first half of 2011, AMD is positioned to extend the leadership it already has in the consumer PC market with VISION Technology from AMD and in the graphics market with the award winning ATI RadeonTM family of GPUs. At Computex, both product lines were on display for attendees, including many of the brand new VISION-powered SKUs recently launched by OEMs. In May, AMD announced a significant platform refresh for its VISION Technology, with as many as 135 new ultrathin and mainstream notebooks arriving on the market.
Ryan Shrout | Source: PC Perspective | Subject: Processor
Who would have thought that just a couple short years ago one of the most interesting news stories we would hear about at Computex is about the expansion of manufacturing capabilities at a foundry company? But that is definitely the case and not just because of the details of the press conference itself but because of the implications the information presented indicates for the future of our industry.
Today at an event occurring in the early hours of the first day of Computex 2010, GLOBALFOUNDRIES, once the manufacturing arm of AMD and now a completely independent and global foundry company, detailed expansion at both facilities in Dresden, Germany and in upstate New York. Dresden, currently the drive behind the 45nm processors from AMD today and the upcoming 32nm Llano processors of 2011, will be receiving up to an additional $1.5 billion investment for a new addition to the corporate campus that will bring the total output from the facility to 80,000 wafer starts per month. The focus of the new portion will start with 40nm technologies in Q3 of 2011 but will eventually transition to 28nm and below. GLOBALFOUNDRIES is taking an aggressive stance to the construction of the facility in order to have it up and running in about 12 months in order to take advantage of the huge market opportunity for 40nm process customers that has been created by less than stellar yields and output from TSMC. 
Fab 1 in Dresden expansion plans GLOBALFOUNDRIES CEO Doug Grouse also discussed an expansion to the upstate New York campus that will add an additional clean room to a facility that is still yet to have its initial construction completed. The company plans to complete the shell of the addition during the current construction phase and then have it outfitted with production tools in early 2013. The focus of this new addition in NY will be 22nm and 20nm technologies targeted for that time frame. This new investment will be worth in excess of $2 billion when fully tooled and outfitted and will increase the wafer start capability to upwards of 60,000 per month. 
Fab 8 in NY expansion plans Finally, GLOBALFOUNDRIES is committing today to establish a semiconductor complex in Abu Dhabi, UAE on a 3km area of land near the international airport. While the company is still leaving room for placing either a manufacturing facility or a research facility there (or both) GLOBALFOUNDRIES wouldn't begin construction until the Dresden and NY projects are completely putting ground breaking at a time frame of 2012 or so. The decision will be made on what type of facility to build will be based on customer demand and technology transitions of the time but this announcement will not apparent preclude any additional expansions to the NY fab campus over and above the ones announced today. GLOBALFOUNDRIES indicated that its 32nm SOI production was on schedule as were the rest of the technologies on the roadmap and even had some 28nm wafers on hand to display as the first foundry to show High-K metal gate at this process. The addition of $3.5 billion plus investment into the GLOBALFOUNDRIES production plans indicates two things to us. First, the company continues to answer questions from critics about its ability to keep up with demand for process technologies into the future and its ability to handle the workloads associated with design wins for high profile customers. Even with these additions TSMC will have the edge in total available output with the high performance technologies but the gap is closing and if nothing else GLOBALFOUNDRIES appears to be winning the PR/marketing war. The addition in Dresden specifically for 40nm is particularly interesting because GLOBALFOUNDRIES originally had no plans for such a process but the demand from customers and the potential for sustainable profits apparently convinced them to develop it. Interestingly, it was apparently too late to get AMD's own Ontario 40nm Fusion part that will be built at TSMC later this year. 
Based on the technologies that GLOBALFOUNDRIES is focusing and building capacity for, it is obvious to me that they are betting a large portion of their future capital on the world of SoCs and designs from ARM, etc. Larger, high performance components from both AMD's CPU and GPU design teams are going to be built with GLOBALFOUNDRIES (Llano on 32nm and the next gen on 28nm are already in the works) but the capacity being outfitted (and purchased through mergers) is targeted at lower power, smaller, mobile designs. AMD's own Fusion parts are heading that direction, as is Intel with Sandy Bridge and Atom processors, and we already know how dominant ARM has been in this field. It is obvious that the market for consumer component-based designs is slowly fading into a smaller, more niche market as we get into the 2013 and 2014 time frames and designs for devices like cell phones and tablets are going to dominate. Have we reached a consensus that we have enough performance for everyday computing and now we demand more than just "faster" out of our devices? That is an editorial for another day... 
GLOBALFOUNDRIES has come from nothing to becoming a large player in the technology world in a very short time. When AMD discussed its "asset light" goals some years ago it was hard to see the manufacturing division of the company becoming so large and so relevant this quickly. They appear to have done it though and with fast paced and continued investments like the $4 billion discussed here today, the foundry market will continue to be interesting into the future.
Jeremy Hellstrom | Source: Intel Press Room | Subject: Processor
SANTA CLARA, Calif. and HAMBURG, Germany, May 31, 2010 - During the International Supercomputing Conference (ISC), Intel Corporation announced plans to deliver new products based on the Intel Many Integrated Core (MIC) architecture that will create platforms running at trillions of calculations per second, while also retaining the benefits of standard Intel processors. Targeting high-performance computing segments such as exploration, scientific research and financial or climate simulation, the first product, codenamed "Knights Corner," will be made on Intel's 22-nanometer manufacturing (nm) process – using transistor structures as small as 22 billionths of a meter – and will use Moore's Law to scale to more than 50 Intel processing cores on a single chip. While the vast majority of workloads will still run best on award-winning Intel Xeon processors, Intel MIC architecture will help accelerate select highly parallel applications. Industry design and development kits codenamed "Knights Ferry" are currently shipping to select developers, and beginning in the second half of 2010, Intel will expand the program to deliver an extensive range of developer tools for Intel MIC architecture. Common Intel software tools and optimization techniques between Intel MIC architecture and Intel Xeon processors will support diverse programming models that will place unprecedented performance in the hands of scientists, researchers and engineers, allowing them to increase their pace of discovery and preserve their existing software investments. The Intel MIC architecture is derived from several Intel projects, including "Larrabee" and such Intel Labs research projects as the Single-chip Cloud Computer. "The CERN openlab team was able to migrate a complex C++ parallel benchmark to the Intel MIC software development platform in just a few days," said Sverre Jarp, CTO of CERN openlab. "The familiar hardware programming model allowed us to get the software running much faster than expected." "Intel's Xeon processors, and now our new Intel Many Integrated Core architecture products, will further push the boundaries of science and discovery as Intel accelerates solutions to some of humanity's most challenging problems," said Kirk Skaugen, vice president and general manager of Intel's Data Center Group. "The Intel MIC architecture will extend Intel's leading HPC products and solutions that are already in nearly 82 percent of the world's top supercomputers. Today's investments are indicative of Intel's growing commitment to the global HPC community
Jeremy Hellstrom | Source: The Tech Report | Subject: Processor
AMD's Black Edition processors are very popular among overclockers as the unlocked multiplier allows for a lot of leeway when you are adjusting clock speeds. Intel has not really been supportive of their fans in that way, though the lock down did happen for a very good reason that protected the majority of consumers. For that small subset of users that do overclock the newly announced 2.93GHz Core i7 875K and 3.2GHz i5 655K are a bright light on the horizon as both processors have unlocked multipliers. See how that overclocking ability effects the price to performance ratio at The Tech Report.
"Intel has unlocked a pair of affordable CPUs, making overclocking dead simple. Join us as we welcome the K series to the 4GHz+ club." Here are some more Processor articles from around the web:
Ryan Shrout | Source: PC Perspective | Subject: Processor
Though Computex doesn’t start for another few days, some initial bits of information continue to flow in. One notable piece was that VIA was planning to demonstrate the first monolithic dual-core version of its Nano processor for the mobile technology markets. While we had indeed previously known about a dual-core variant of Nano, it was always expected to be a dual-chip package – similar to what we saw with the first Intel quad-core processors under the Core 2 Quad brand.
The benefits of building the processor on a single package include performance and power efficiency improvements – both critical for the fickle mobile market. Using the 40nm process node at TSMC, VIA is expecting the new dual-core Nano to compete directly with the updated Atom processor lines, though I doubt we’ll see it go after the cell phone markets that Intel is pushing Moorestown towards. Back in our initial comparisons of the VIA Nano and Intel Atom processors the performance of the Nano impressed but has basically remained stagnant during several iterations of Atom. 
A current generation VIA Nano processor VIA is a company in desperate need of a product that will help maintain its relevancy in the western markets. It continues to do very well in the emerging markets of India and China with lower power desktop products but those efforts just don’t apply very well to the US consumer. An improved Nano dual-core processor with better power efficiency could help VIA take ground in the battle for netbooks, tablets and other small format computing devices. And don't forget, VIA has an ARM-based design they are pushing as well. 
VIA Nano single-core die shot PC Perspective will bring you more information about the Nano updates as we find it! Keep an eye on the PC Perspective Computex 2010 page for all the latest news and notes from the show floor.
Ryan Shrout | Source: PC Perspective | Subject: Processor
It should come as no surprise to the industry that AMD will be demonstrating Fusion-ready systems at Computex that are the first to combine a traditional microprocessor architecture and a modern graphics processing core on a single die for mainstream processing. The details of the Fusion processors have been discussed by AMD at various microprocessor conferences.
The Llano part will be the notebook-market APU (accelerated processing unit) that combines four x86 processor cores with a Redwood-class (~400 stream processors) DirectX 11 capable performance GPU, all on the same monolithic die. Both functional units will access the same DDR3 memory controller (nothing specific on number of channels) and each of the four cores will have a 1MB L2 cache to itself but these APUs will not have any L3 cache. Clock speeds are expected to reach the 3.0+ GHz mark. AMD claims Llano will have a maximum TDP of 25-50W (remember this is a mobility processor) and will operate at voltages ranging from 0.8v to 1.3v. AMD says that the entire chip will have drastically improved power management features and clock gating to improve power efficiency. 
This core will be built on GlobalFoundries 32nm SOI process using High-K metal gate transistors as expected; AMD will actually be the only 32nm SOI customer GlobalFoundries is likely to have. Interestingly, this marks the first time an ATI GPU will be built outside of TSMC’s process and will be the first step in what we all thought would happen when AMD bought ATI – GPUs manufactured by AMD (now GlobalFoundries). It would seem obvious after this product is released that the market for sub-$100 graphics cards, both mobile and discrete, will somewhat fall away. OEMs are already weighing the benefits of integrated technologies like this that will not only offer pricing advantages but also offer more than enough performance for the majority of the high-performance computing the everyday consumer demands. Ontario is a much lower power and lower performance part that is being target at the Atom markets, some of which may overlap with Intel’s goals for Moorestown. Ontario will include a pair of x86 processing cores based on the Bobcat architecture as well a much lower performance DX11-ready GPU in a SoC (system on a chip) design with a TDP under 1 watt per core. AMD confirmed that they had internal samples of these new Ontario processors already and planned to be in “volume production” in late 2010. What might be most interesting is not WHAT it is, but WHERE it is made. We have learned that the Ontario Fusion solution will be built on a 40nm process and thus we can conclude this AMD-designed part will not be built by GlobalFoundries, the previously-AMD-owned foundry company, but rather by TSMC on the 40nm G technology. GlobalFoundries does not have a 40nm process node today (UPDATE: GlobalFoundries does have a 40nm process currently but it is targeted at low power devices like cell phones, not high performance required for processors like Ontario) or planned until late in 2011 so they were out of the discussion and that leaves TSMC as the only other fabrication facility with experience with AMD designs.
Bobcat cores will be used in the Ontario APU - and built at TSMC This does introduce some interesting ideas around intellectual property and producing AMD processors cores outside of its own or GF fabs for the first time. We have to say we are just as surprised by this move by AMD as we were when we learned Intel was going to be allowing Atom-based SoCs to be produced by TSMC – now both companies appear to be taking the same route for this specific mobility market. We are looking forward to getting more information from AMD on these topics and more at Computex this week. Stay tuned to our Computex coverage page for the latest.
Jeremy Hellstrom | Source: PC Perspective | Subject: Processor
All of AMD's Athlon series just got an update, the X4 series gets the 640 and 610e, the X3 adds the 445 and the 415e and the X2 now has a 260 and 245e. The 'e' you see in those names denotes a 45W processor, great for low voltage and low heat. Josh got his hands on the 3.0 GHz X4 620 and the 3.2 GHz X2 260, which will cost $122 and $76 respectively. Neither processor is going to overtake the Core i7 series for the performance crown but AMD is gaining a lead in the performance per dollar category. You also get a chance to see the new MSI 890GXM-G65in action. "AMD is milking the Athlon II processors for all they are worth, and it is helping AMD to keep afloat in the face of a very aggressive and competent Intel. While the Phenom II processors are faster, they are more expensive to produce. The Athlon II takes all the experience learned from manufacturing the larger, more complex Phenom IIs and puts it into a compact package that sells for a low price, yet performs nearly as good as anything else out there." Here are some more Processor articles from around the web:
Jeremy Hellstrom | Source: Neoseeker | Subject: Processor
Arriving today are 6 new processors from AMD, mostly a refresh of some of their older and lower end CPUs. Neoseeker has two of them, the Athlon II X4 610e and X4 640. The 640 is clocked at 3.0GHz and the 610e at 2.4GHz, both are quad core 45nm parts and neither are unlockable Black Editions. That doesn't mean you can't overclock, as Neoseeker hit 3.85GHZ on the 640 which is a nice jump from the 635. The X4 610e was tested a little differently, since it is low voltage they tested how low the voltage could go while keeping the stock speed and the answer would be 0.992V. Read the whole review here.
"The low-power Athlon II X4 610e, like its brothers, really shines in rendering and general computing at the 45W mark." Here are some more Processor articles from around the web:
Want more news? Check out our Archive |
.:Latest News
.:Latest Reviews
|
 |
Legal - Contact - Advertising |  |
.:Processor Reviews
