Subject: Processors | January 18, 2012 - 04:10 PM | Jeremy Hellstrom
Tagged: xbox next, IBM, amd, Power PC, southern islands, xbox 720, oban
SemiAccurate has been doing some digging into the hardware that will power the next XBox, perhaps a bit more successfully than Microsoft would like. This builds on the rumours that they had collected in December of 2011 and confirms that the next generation console is only a partial win for AMD. Oban is the code name for the CPU, which is being fabbed by GLOBALFOUNDRIES for the most part and will be a variant model of IBM's Power PC architecture and not an x86 based chip. AMD will provide a Graphics Core Next Southern Islands GPU to provide the graphical power, terrible news for NVIDIA's bottom line over the next several years as they lose out on at least one platform of the coming generation. This will continue to sting as unlike PCs, consoles are not refreshed several times over a year and the current hardware will likely be powering the XBox Next for years to come.
From what SemiAccurate has gathered, Microsoft have ordered a huge run of the chips which will power the console and should guarantee availability in the Spring of 2013 which is the current predicted release date for the console. Considering the low yields from GLOBALFOUNDRIES lately this seems likely a move to ensure that even a large amount of bad silicon will not have a major impact on their ability to provide deep supplies of XBox Next for retailers.
"If you crave more info about the upcoming XBox 720/Next, there is finally some concrete info. The one nice thing about this job is that proud parents like to talk, and that is exactly where this story begins."
Here are some more Processor articles from around the web:
- AMD A8 3870K @ Guru3D
- AMD Llano A8 3870K @ LostCircuits
- A8-3870K vs. Core i3-2105 CPU Review @ Hardware Secrets
- AMD A8-3870K: Black Edition Llano @ Bjorn3D
- The Workstation & Server CPU Comparison Guide @ The Tech Report
- Intel Sandy Bridge Shines With Mesa 8.0 @ Phoronix
- Intel SNA Architecture Is Constantly Evolving @ Phoronix
- Title: Intel's Medfield Still A Botched Binary Mess Under Linux? @ Phoronix
Subject: General Tech, Processors | January 11, 2012 - 06:21 PM | Scott Michaud
Tagged: windows, processor, microsoft, cpu, bulldozer, amd
Let us take a little break from the CES news with a load of bull -- a download of Bulldozer. If you have an eerie sense of being in this situation before then you may in fact have a bad memory as it did in fact happen and it was only about a month ago. Microsoft released an update in mid-December to optimize their operating systems for AMD Bulldozer technology; that patch disappeared without any listed reason. As of today, we have access to both the patch as well as most of the reason for the delay in the first place.
You know: part of me wants to see a Bulldozer go 100MPH, and another part of me fears greatly.
The first order of business is to explain to those who have an AMD FX series, Opteron 4200 series, and/or an Opteron 6200 series processor how to increase their potential performance: KB 2646060 and KB 2645594 each contain a patch which will optimize Windows to the Bulldozer architecture for most users when both are applied.
It turns out that Microsoft pulled the Bulldozer update last month when discussions with AMD revealed that the patch would not provide the promised performance increases for most users. The problem specifically centers on the Core Parking feature within Windows 7 and Windows Server 2008 R2: after the hotfix in December was applied, Core Parking would still interfere with Bulldozer’s design by attempting to save power and sleep cores that were unused without understanding that Bulldozer cores are not typical cores. With Core Parking disabled for Bulldozer-based CPUs either through this hotfix or by changing your performance profiles to “High Performance” from the often default “Balanced” you would allow Bulldozer to run as it actually desires to run. According to how these bulletins are worded, should you have been on “High Performance” profile back in December before the hotfix was pulled you would have experienced what should only have been available starting today.
These performance increases are not for every application, however. AMD has stated that applications which are more sparsely multithreaded would benefit most from the update.
Workloads that are moderately threaded have the potential for uplift. This could include virtualization, database, or transactional environments that are “bursty” – a mixture of light and heavy transactions, or legacy applications that are by nature not very threaded. The more heavily threaded the application, the less the likely the uplift.
My intuition knowing this as well as the Core Parking issue is that once Windows finally wakes the Bulldozer core, your experience is maximal with the December patch; applications which only temporarily become multithreaded either do not wake the proper portions of the processor or wake the processor in time to be of maximum benefit.
It appears as if the removal of the hotfix last month was simply because AMD believed that while the patch was effective, it would not be correctly applied to the vast majority of customers without a second hotfix and thus give the appearance of little to no real benefits.
Subject: Cases and Cooling, Processors | January 10, 2012 - 08:13 PM | Tim Verry
Tagged: VLIW-4, trinity, piledriver, CES, APU, amd
Today at CES, the Consumer Electronics Show, we got to see a demonstration by AMD of an AMD powered computer running dual monitors. Only, it was not just a dual monitor pushing desktop computer. In a surprise twist, AMD took the side panel off of the desktop computer to reveal that it was actually a laptop computer using their next generation AMD Trinity APU that was driving the game on one display, and the windows desktop on the other display. Even more, on the laptop screen itself, it was playing a 720p video.
Here you can see the two displays that the Trinity powered laptop was driving with Dirt 3 on the left monitor and the Windows desktop on the right one where a video conversion was happening in the background. AMD did not get into any details regarding the transcode, however.
This is the "desktop" computer case that they opened up to reveal that it was, in fact, a Trinity laptop that was driving all the displays.
A die shot of the upcoming Trinity APU (Accelerated Processing Unit) sitting next to a USB flash drive. Specifications of the Trinity APUs have not yet been released by AMD; however, if this leak holds true the Trinity APUs will have either two or four Piledriver CPU cores and TDP (thermal design power) of 65 W, 100 W, and 125 Watts (depending on particular chip). Clock speeds will further vary between 2.2 and 3.8 GHz at stock speeds (will run a bit faster with Turbo Core 3.0). The GPU aspect will be clocked between 563 MHz and 711 MHz and is based on the VLIW4 technology of the Cayman graphics Cards (69xx). They estimate that it will deliver up to 30% more performance versus current Llano chips and will support all the fancy new X86 instruction sets like AVX and AES-NI. A nice boost and hopefully the real specifications will come close to this (or be even better, of course).
Update: Another interesting bit of information is that AMD will have a low power Trinity APU with a TDP of 17 watts and will supposedly deliver the same level of performance as the current Llano chips (that draw twice the power).
Update: AMD has stated Trinity will deliver a 25% increase in CPU performance and a 50% increase in GPU performance versus current Llano APUs. Stay tuned to PC Perspective for more Trinity info as it develops.
PC Perspective's CES 2012 coverage is sponsored by MSI Computer.
Follow all of our coverage of the show at http://pcper.com/ces!
Subject: Processors | January 5, 2012 - 05:12 PM | Jeremy Hellstrom
Tagged: westmere, linux, ironlake, Arrandale
It has been a busy year for Phoronix as not only have they had a lot of hardware to review, also Linux developers have been quite busy this year updating drivers and the base kernel. They decided that the beginning of 2012 was a perfect time to reflect on the effects of these changes, specifically the graphics driver that powers the Intel Core i3-330M. The results are mixed, with one driver version excelling in a single task but lagging in others. On the plus side, the performance never stays consistent which gives hope that there is still room for improvement and the performance has not plateaued.
"Back in December I posted historical Intel Sandy Bridge benchmarks looking at the graphics performance over the course of 2011 that this latest-generation of Intel hardware has been supported under Linux. In this article are some similar Intel OpenGL benchmarks of each quarter going back to the end of 2010, but this time it is for the previous-generation Intel Ironlake hardware."
Here are some more Processor articles from around the web:
- Intel Core i7 3820 Review: $285 Quad-Core Sandy Bridge E @ AnandTech
- Desktop CPU Comparison Guide @ TechARP
- Mobile CPU Comparison Guide @ TechARP
- AMD A8-3870K - Radeon HD 6550D Graphics @ Phoronix
- AMD A8-3870K Llano APU Linux Overclocking @ Phoronix
- AMD Fusion A8-3870K Overclocking and Dual Graphics Performance Review @ HardwareHeaven
- AMD A8-3870K Graphics With Gallium3D @ Phoronix
Subject: Processors, Mobile | January 5, 2012 - 01:47 PM | Jeremy Hellstrom
Tagged: WonderMedia, PRIZM WM8950, SoC, arm, cortex-a9
Taipei, Taiwan, January 5, 2012 -WonderMedia Technologies, Inc, a leading innovator of energy-efficient, feature-rich system-on-chip platforms, today announced the WonderMedia PRIZM WM8950 SoC.Combining a highly energy-efficient ARM Cortex-A9 core running at 800MHz with advanced graphics and stunning 1080p video playback capabilities, the WonderMedia PRIZM WM8950 further extends WonderMedia's leadership in the fast-growing Android media tablet market.
Prizm WM8950The next generation WonderMedia PRIZM platform and software suite provides customers an easy migration path to a Cortex-A9 SoC and offers support for the latest Android 4.0 and Windows CE 7.0 operating systems. With its high-performance, and feature-rich design, the WM8950 is also optimized for a wide range of innovative system design applications, including smartbooks, Smart TV, SmartStream for wireless display and multimedia streaming, networked projectors, digital signage, and thin clients.
"Our line of PRIZM platforms has spurred the explosive growth in the Android media tablet market," commented Tzumu Lin, President and CEO, WonderMedia Technologies, Inc. "The new WM8950 delivers greater computing power and richer multimedia experiences to meet the ever growing global demand for affordable smart mobile devices."
PRIZM WM8950 Highlights:
- 800MHz ARM Cortex-A9 processor
- ARM Mali-400 3D graphics Processor
- Multi-standard 1080p video decoding engine
- H.264 video encoding
- DDR3/LPDDR2 DRAM interface
- Multiple video interface including HDMI, LVDS, and DVO
- Flexible networking and peripheral interface
- Advanced hardware security engine
- Android 4.0 and Windows CE7.0 support
A New Chip for a New Year
When Intel launched the Sandy Bridge-E platform in November, there were three processors listed on the specification sheet. The Core i7-3960X is the flagship, 6-core processor with the ~$1000 price tag, the Core i7-3930K still had 6-cores but a much lower cost and similar clock speeds and the Core i7-3820 was the only quad-core option and was listed for a Q1 release. We reviewed the Core i7-3930K in December and found that it offered nearly the same performance as the more expensive unit at about half the price.
Today we are getting a preview of the Core i7-3820 that will be released likely in early February and will come with a much more reasonable price tag of $285 to fill out the LGA2011 socket. The question that we must ask then is can the quad-core Core i7-3820 compete against the currently available quad-core Sandy Bridge parts that fit in the widely available LGA1155 socket? We not only have to consider performance but also the features of each platform as well as the total cost.
Same Feature Set, New Die
While most of the features of the Core i7-3820 are going to be identical to those of the previous SNB-E processors we have seen, there are some important differences with this chip. Let's see what is familiar first. The Core i7-3820 is based on the Sandy Bridge-E design that works on the LGA2011 socket and the X79 chipset and motherboards currently on the market. It includes a quad-channel memory controller and 40 lanes of PCI Express that are actually capable of PCIe 3.0 speeds. HyperThreading is still enabled so you are getting the benefit of being able to run twice as many threads as you have cores.
There are some very important changes on this CPU as well though starting with a quad-core design. This directly pits this Sandy Bridge-E part against the currently existing Sandy Bridge processors running on the Z68/P67 chipset and LGA1155 socket. Also, the L3 cache on the Core i7-3820 is at 10MB, 5MB less than the Core i7-3960X and 2MB less than the Core i7-3930K. We are basically talking about a processor that bridges the gap between the original SNB and newer SNB-E parts and it creates some interesting battles and comparisons.
Subject: Processors | December 28, 2011 - 07:15 PM | Tim Verry
Tagged: pine trail, netbook, Intel, cedar trail, atom, 32nm
Intel has been pumping out quite a few new processors lately, with new Sandy Bridge-E CPUs, a new Sandy Bridge i7 2700K processor coming out, and now a new line of Atom CPUs sneaking in the news right before the new year! Not to mention, they are also working on Ivy Bridge.
The new Atom CPUs are of the Cedar Trail variety and succeed the older Pine Trail-M Atom processors. Currently, there are three Cedar Trail chips that will be available as soon as January in OEM systems including the N2600, N2800, and D2700 CPUs. Further, the new chips are 32nm and have a 22x22mm package size. These little chips are destined to power netbooks, tablets, embedded devices (think medical devices, ruggedized tablets, machinery). Yes, Intel still believes in netbooks, and feels as though emerging markets will keep the market alive and growing as people want for cheap computers that are able to get them on the web. While the netbook is losing popularity in the US, Intel expects the South American, Eastern European, and African markets to see great interest in the netbook platform. Their netbook plans involve three price tiers with accompanying use cases including netbooks at $200 with minimal features and a price to match that enables people to access the web all the way to $400+ netbooks with lots of features that would fill out the market up to where the Ultrabook territory begins at around $900.
The new Cedar Trail processors improve upon the previous gen Atom chips by quite a bit, according to Intel. The graphics aspect in particular has been improved such that 1080p HD Youtube and HD Netflix streams are playable at at least 24 FPS. Something that early netbooks using Intel's integrated graphics will never be able to do. Intel further estimates a 50% lower TDP and a 28% processor performance increase over the Pine Trail chips. Further, the new Cedar Trail chips have more cache at 2 x 512 L2 cache(s), higher clockspeeds, lower TDP, higher C-State (C6 vs C4E)/lower power usage in sleep mode, a 200MHz higher clocked graphics card (400MHz vs 200 MHz), and increased memory speeds (DDR3 800 and 1066 vs DDR3-667). The fastest Nxx chip, the N2800 manages a .2GHz clock speed increase while also knocking off 2 watts from the TDP versus the previous top N570.
Needless to say, Cedar Trail is looking very good, on paper at least. The individual chip specifications are listed below.
|CPU Clock Speed||Graphics Clock Speed||TDP|
|N2600||1.60 GHz||400 MHz||3.5 W|
|N2800||1.86 GHz||640 MHz||6.5 W|
|D2700||2.10 GHz||640 MHz||10 W|
What are your thoughts on the new Cedar Trail chips, do you think they will provide enough "oomph" to make new netbooks desirable again? Some more information can be found here and straight from Intel here.
Subject: General Tech, Processors | December 28, 2011 - 02:33 AM | Scott Michaud
Tagged: intel atom, Intel, atom
Intel’s Atom processors were created as a tier below their Celeron product line. Netbooks, then running VIA Nano processors, have started to gain popularity since their introduction in late 2007. Intel’s Atom processors took the place of the VIA parts since that time. In 2009, Intel has stated that they have seen approximately twenty percent of their sales of notebook processors replaced with sales of their cheaper Atom processors. Intel still maintains the Atom processor line, but apparently with new goals in mind.
Up and ATOM!!!
According to Digitimes, the demand for Intel’s Atom processors has declined recently. Intel, in response, decided to market that tier of parts to embedded and server customers for use in network-attached storage devices and very low-end servers. Intel is also rumored to have plans to shrink the process size of Atom in 2013 to 22nm and again shrink process size to 14nm in 2014. The upcoming 32nm Atom processor is expected by the second quarter of 2012.
Subject: Processors, Mobile | December 26, 2011 - 01:44 PM | Ryan Shrout
Tagged: Intel, Medfield, Android, x86, SoC
Intel hopes that 2012 will finally be the year they see mainstream phones with Intel inside. Despite Intel's attempts to tell us otherwise for the past several generations, the upcoming Medfield design is the first truly serious attempt to enter the phone and tablet market currently dominated by the many ARM-based partners of phone manufacturers all over the world. A recent post over at Technology Review discussed the advantages that Medfield offers over previous Intel Atom-based designs with Steve Smith, Intel's VP of Architecture.
First shown at the Intel Developer Forum in San Francisco this past September, a Medfield-based reference design has many believing in what some thought was impossible but what others thought was inevitable: Intel x86 cores in a phone that matters. Why the change from many in the analyst space? Medfield is the first option from Intel that is truly a single-chip solution, removing design space concerns and power consumption issues that previous Atom-architecture solutions were saddled with.
Intel showed Technology Review the Android-based reference phone running Gingerbread.
The phone prototype seen by Technology Review was similar in dimensions to the iPhone 4 but noticeably lighter, probably because the case was made with more plastic and less glass and metal. It was running the version of Google's operating system shipping with most Android phones today, known as Gingerbread; a newer version, Ice Cream Sandwich, was released by Google only about a month ago.
Intel has a lot of experience in the consumer markets though it took a shift inside the company to really put the focus on phones and tablets over netbooks and convertible-notebooks. At the recent showing not only did they have the reference design phone but also an iPad-like tablet device running Ice Cream Sandwich, another key to the consumer's dollar. And as you can clearly see in the diagram below, there is a lot of money being made that Intel wants in on. A LOT.
Source: Technology Review, IDC
Intel will also enjoy a process technology advantage over the competition with current Medfield SoCs built on the company's internal 32nm process and the upcoming 22nm technology promises even more power consumption advantanges. ARM designs are built at different foundries including Samsung and TSMC and while they are competitive, no one can keep up with Intel on this front.
Anandtech also had some interesting information to share from an investor conference earlier this month about the power consumption and performance levels of Medfield.
The diagram shows that power consumption on Medfield should be competitive with the current ARM-based SoC leaders in the market today. Areas like 3G standby, basic audio playback and video playback should be accomplished with minimal power draw in order to have battery life extended to at least current expectations. The performance graphs here on Browser Mark and "Graphics" are impressive as well though obviously we have a TON of missing information to really make the graph meaningful. Anand puts it well:
Barring any outright deception however, there seems to be potential in Medfield.
I tend to believe that Intel is too smart to misjudge a product to investors, but remember how impressive the initial performance results of Larrabee were for years?
I am hopeful and excited for Intel's mobility plans in 2012 as other information we have seen looks impressive. Let's see what CES has to offer.
Speed Bumps and Unlocked Processors
AMD has announced the latest members of their fairly successful APU series for both the desktop and the mobile markets. The original release in June of this year saw the first fully integrated 32 nm APUs from AMD. These proved to be quite popular with their decent CPU performance and outstanding integrated graphics speed and quality. The launch was not entirely smooth for AMD though, even though the company had been shipping products to partners and OEMs for some months.
The desktop saw limited SKUs, and the availability of the top end parts was disappointing to say the least. AMD and their partners at GLOBALFOUNDRIES were not able to produce enough usable chips to supply demand. Quantities were tight throughout the summer, and the mobile market did not see as big of a boost for AMD as was hoped. AMD did get a lot of new business though, as the thermal and power envelopes of these A-series chips were able to match that of Intel.