Sunday, February 22, 2009

Embedded systems

Embedded systems
Alchemy (processor) and Geode (processor)In February 2002, AMD acquired Alchemy Semiconductor and continued its line of processor in MIPS architecture processors, targets the hand-held and Portable media player markets. On June 13, 2006, AMD officially announced that the Alchemy processor line was transferred to Raza Microelectronics Inc.[20]In August 2003, AMD also purchased the Geode business which was originally the Cyrix MediaGX from National Semiconductor to augment its existing line of embedded x86 processor products. During the second quarter of 2004, it launched new low-power Geode NX processors based on the K7 Thoroughbred architecture with speeds of fanless processors 667 MHz and 1 GHz, and 1.4 GHz processor with fan, of TDP 25 W. This technology is used in a variety of embedded systems (Casino slot machines and customer kiosks for instance), several UMPC designs in Asia markets, as well as the OLPC XO-1 computer, an inexpensive laptop computer intended to be distributed to children in developing countries around the world.For the past couple of years AMD has been introducing 64-bit processors into its embedded product line starting with the AMD Opteron processor. Leveraging the high throughput enabled through HyperTransport and the Direct Connect Architecture these server class processors have been targeted at high end telecom and storage applications. In 2006 AMD added the AMD Athlon, AMD Turion and Mobile AMD Sempron processors to its embedded product line. Leveraging the same 64-bit instruction set and Direct Connect Architecture as the AMD Opteron but at lower power levels and in smaller footprint packages[citation needed], these processors were well suited to a variety of traditional embedded applications. Throughout 2007 and into 2008 AMD has continued to add both single-core Mobile AMD Sempron and AMD Athlon processors and dual-core AMD Athlon X2 and AMD Turion processors to its embedded product line and now offers embedded 64-bit solutions starting with 8W TDP Mobile AMD Sempron and AMD Athlon processors for fan-less designs up to multi-processor systems leveraging multi-core AMD Opteron processors all supporting longer than standard availability.In April 2007, AMD announced the release of the M690T integrated graphics chipset for embedded designs. This enabled AMD to offer complete processor and chipset solutions targeted at embedded applications requiring high performance 3D and video such as emerging digital signage, kiosk and Point of Sale applications. The M690T was followed by the M690E specifically for embedded applications which removed the TV output, which required Macrovision licensing for OEMs, and enabled native support for dual TMDS outputs, enabling dual independent DVI interfaces.
......................................................................

Sunday, February 1, 2009

Intel Core 2 Duo E7200

Intel Core 2 Duo E7200

It has been surprising and frankly somewhat satisfying to see the success to date of the "Wolfdale" core architecture in Intel’s Core 2 lineup of processors. "Wolfdale" is Intel’s first 45nm based dual-core design, and is an interesting product beyond first glance. When it was first announced, there was a bit of doubt throughout the industry of the chip’s viability in the market. The doubt was that it would be foolish of Intel put major weight behind a dual-core processor architecture, as the future is clearly quad-core processors and beyond. More cores are always better, right? Why would Intel continue pushing dual-core models when quad-core models are getting more and more cost efficient? That's a fair question.
Once we got our first “Wolfdale” chips in for testing, we realized that a modern dual-core processor can still bring exceptional performance for today’s high-end machines. While quad-cores definitely have more appeal for the heavy multi-tasking power user, dual-cores can typically accomplish most tasks with performance to spare, but can do so with much lower power consumption and heat production. In addition, if your application of choice is only coded to handle two processor cores, it can actually run faster on a highly-clocked dual-core compared to a mid-range clocked quad-core. While the wave of multi-core software is rising every day, we still haven’t seen truly compelling numbers in main stream applications showing that quad-core processors are worth their premiums for many average, everyday computing scenarios.


Intel's Core 2 Duo E7200 Processor Engineering Sample

In any case, despite initial availability issues, Intel’s new E-series processors based on this new “Wolfdale” architecture ended up being a big hit, and chips are still in high-demand, months after their initial release. Now that the product line is beginning to become available in volume, we’re seeing Intel flesh out their Core 2 Duo product line with the entry of the first truly low-cost “Wolfdale” processor release to date, Intel’s Core 2 Duo E7200. The Core 2 Duo E7200 is a bit of an anomaly on Intel’s dual-core product lineup, much like the Core 2 Quad Q9300 is the anomaly for the quad-core product lineup. Both the E7200 and Q9300 are the only models in their lineups which have half the L2 cache of other models in their families. In addition, both run at 2.5 GHz clock speeds and both are priced very competitively for the performance they offer. The E7200 is a based on a simple, modern dual-core design, whereas the Q9300 uses two of these processor dies to create a quad-core version . Beyond the raw core count, the chips are nearly identical. With the release of the Core 2 Duo E7200, it’s now possible to get a fast “Wolfdale” dual-core chip for about $130, about half the price of today’s high-end “Wolfdale” models. Let’s see if this newbie can live up to the high-bar set by preceding 45nm dual-core parts from Intel.


The Core 2 Duo E7200 is the slowest speed 45nm dual-core processor released to date at 2.53 GHz stock speed, the next closest offering being the Intel Core 2 E8200 model, which runs at a slightly higher clocked 2.66GHz with a full 6MB of cache. The E7200 processor runs at 5% slower clock speed and has half the cache at 3 MB, but also costs 30% less overall, and is currently selling for around $130, which would be considered high-end Celeron territory previously.

The Wolfdale architecture is based on Intel’s latest 45nm manufacturing process, which means the chip runs cool and doesn’t consume a lot of power, even under heavy loads. Intel rates the Core 2 Duo E7200 with a TDP of 65W, although we feel that even this is somewhat conservative, as the chip ran close to room temperature with very low noise cooling.

Each of the two processor cores have an individual 64k of L1 cache and they share a pool of 3 MB L2 cache. 3 MB is the smallest amount of cache offered in an Intel 45nm product to date, and this is essentially why many enthusiasts will stay away from this processor. At the same clock speed, we can expect a Wolfdale with a full 6MB of cache to perform 5-10% better under intensive applications (such as gaming). However, 3 MB of cache is still plenty big for many work loads, and as you’ll see in our following benchmarks, the chip still performs within pretty similar levels of the other Wolfdale chips (with larger caches) we’ve seen to date in most scenarios. The positive side of having less cache means that the chip is physically less complex, and in addition to the power and heat benefits that come along with that, it also means that the chips are extremely good overclockers, as we’ll look into later.

In addition to having a smaller amount of cache, the E7200 also runs down a notch on its frontside bus speed from the standard 1333 MHz on other current Core 2 Duo chips to a modest 1066 MHz front side bus speed. Don't be too concerned about this, as Core 2 Duo dual-core chips rarely have the opportunity to saturate the front side bus, and you can easily shoot it back up to 1333 MHz if you so choose (and frankly, we would recommend so). It’s still a very modern processor architecture, with support for 64-bit processing, SSE4, top-notch power management and Execute Disable support. However, we should note that the E7200 does not have support for hardware virtualization acceleration and does not support Intel’s TXT (Trusted Execution Technology), which ties in with their enterprise level security efforts. As this chip is targeted for consumer applications, this isn’t all that surprising. Although it’s disappointing to see, as these chips will undoubtedly make their way into low-cost / low-power / low-heat dual-core servers as well.
In order to run the new Core 2 Duo E7200, you will need to ensure that your motherboard and BIOS revision support Intel’s 45nm processors, as a BIOS update is needed for most systems that are older than six months. Most platforms in the last year will have BIOS support for this chip but we'd advise double-checking with the manufacturer. The latest round of motherboards based on a new chipset architecture like Intel's P35, X38,and X48 or NVIDIA's nForce 7x0i series should support this chip out of the box.

The Core 2 Duo E7200 processor is ripe for overclocking, as it runs at a lower-than-standard front side bus speed for the processor lineup, has very low power consumption by default, and has a low core voltage level. It’s also manufactured on the same 45nm process technology which Intel is using to produce their 3.0 GHz+ Core 2 Duo/Quad core models and sells for upwards of 10x the price of the humble E7200.
For testing, we used the massive Coolermaster Gemini II cooling system with low-speed cooling fans. With this cooler, our Core 2 Duo E7200 chip idled at around 91ºF, just a about 15ºF over room temperature. Under heavy loads (at stock speeds) the chip would hit thermals of about 110ºF, which is very good considering the performance possible from this CPU. For overclocking, we cranked the vCore up to 1.4V, which increased the chip’s heat output levels and pushed peaked temperatures up to about 135ºF under heavy load.
The end result of our overclocking efforts was a Core 2 Duo E7200 chip running at stable levels at 3.5 GHz clock speed. That’s a 38% overclock with a simple boost to the vCore levels and upping the front side bus speed – what we would consider to be very easy work. Our E7200 chip was actually able to hit 3.8 GHz at its peak, but was not as stable we demand from our overclocks. At 3.5 GHz, we were able to run through all of our tests without issue and pass all of our stress tests. This is a higher overclock than we were able to obtain with the Core 2 Quad Q9300 processor, which is basically the E7200 x 2, showcases that a simple, less complex design can allow for higher peak clock speeds.

Power consumption is rather impressive with the Core 2 Duo E7200, as shown by our charts below. As always, tests are done with a hardware A/C wattage meter testing the full system power consumption (not just the processor). The platforms are identical across the board, only changing out the processors (and motherboards for AMD platforms). Testbed details can be seen on the following page.

As expected, the Core 2 Duo E7200 consumes less power compared to other members of the Wolfdale family, as it has a lower clock speed and a lower vCore level. At stock speeds, our testbed platform idled at 146W and peaked at 159W, which are very tolerable levels. When overclocking, our system peaked out at 205W, which is pretty impressive for a 3.5 GHz dual-core system with an 8800 GT graphics card, 4 GB of memory, and a 10,000 RPM hard disk. The low power consumption and support for SSE4 extensions means that this chip will likely be popular with the home theater PC (HTPC) crowd.

At its stock clock speed, the Core 2 Duo E7200 isn’t all that impressive in terms of raw computing power. Its arithmetic processing and memory bandwidth levels are low with respect to the rest of our test pack, but that’s to be expected as it’s the lowest priced option in here. The multimedia numbers are quite good, and of course the inclusion of SSE4 support helps here. As you’ll notice, the 2.53 GHz dual-core E7200 chip is able to perform at the same speed as a 3.0GHz E8650, which is based on Intel’s prior generation dual-core technology, a solid sign of improvement in multimedia performance with this new architecture.
When overclocked, the Core 2 Duo E7200 at 3.5 GHz is able to outperform Intel’s Core 2 Duo E8500, the fastest member of the “Wolfdale” family to date.


The Core 2 Duo E7200 puts up solid, albeit not too exciting numbers in our synthetic tests. It’s interesting to see its 3DMark06 numbers drop in higher than AMD’s quad-core Phenom processors, which is impressive for Intel’s dual-core lineup. The Phenom’s overtake the E7200 in Cinebench when all four cores are used, but not by a huge margin. When overclocked, this $130 dual-core chip can give performance levels similar to a low-end quad-core machine in many scenarios, such as these. Intel Core 2 Duo E7200.

......................................................