Graphics Battle

As a CPU reader, you’re probably aware of the perpetual battle between NVIDIA and ATI for dominance in the desktop graphics arena. There’s a significant third party, however, pushing myriad pixels in the mobile space: Intel. However, Intel’s presence in the mobile graphics field is felt more in the mid- and low-range markets. Both NVIDIA and ATI have various mobile graphics processors available, targeted at different groups of customers. But Intel has only a few options that are integrated into its mobile chipsets. Because we’re now transitioning between AGP and PCI-E, there are many options available in mobile graphics. There are AGP and PCI-E IGPs, numerous discrete AGP GPUs, and multiple discrete PCIE GPUs, as well. At the low end, Intel’s Mobile 8xx and mobile 9xx chipsets with a “G” designation in their model numbers have either Intel’s Extreme Graphics 2 (AGP) or Intel’s Graphics Media Accelerator 900 (PCI-E) integrated into the chipset. These GPUs have limited 3D functionality when compared to NVIDIA’s and ATI’s solutions, but they’re well-suited to 2D operations or video playback. Competing with Intel at the low end are ATI’s entire line of Mobility Radeon IGPs (AGP) and their Radeon Xpress 200M chipsets (PCI-E) for Intel and AMD processors. ATI’s Mobility Radeon 7500/9000/9200 discrete AGP GPUs also compete at the low end, alongside its Mobility Radeon X300 with HyperMemory (PCI-E). Additionally, NVIDIA offers the GeForce4 440 Go and GeForce FX Go 5200 for low-priced and thin-and-light AGP notebooks and also has the new GeForce 6200 with TurboCache available for similar PCI-E notebooks. These discrete AGP graphics processors all have varying amounts of frame buffer memory, while the low-end PCI-E parts from NVIDIA and ATI use limited amounts of discrete graphics memory in conjunction with proprietary technologies, TurboCache and HyperMemory, to dynamically share system memory.

Mobile CPU

Virtually all of today’s notebooks are powered by processors built by Intel, AMD, or Transmeta. In the same way AMD and Intel duke it out in the desktop space, they constantly try to one up each other in the mobile space, as well. Transmeta’s processors, however, have been relegated to low-power niche products, such as OQO’s Model 1, and aren’t as ubiquitous as Intel’s or AMD’s products. Both Intel and AMD have multiple specialized processor lines available for mobile platforms, and each company’s desktop processors are used in some DTR notebooks, as well. But just as the mobile space has been divided into desktop replacement, performance thin, mainstream thin-and-light, and value mainstream segments, so has the type of processor each segment uses. In the DTR space, the Athlon 64 and Pentium 4 are the most widely used processors, with a portion of the market belonging to high-end Pentium M processors. These DTR notebooks are large and heavy and don’t offer much in terms of battery life, but they are high-performance products. Many DTR notebooks are just as powerful as similarly configured full-sized machines, and you could even use them as a desktop replacement. At the other end of the spectrum, AMD’s Mobile Sempron and Intel’s Celeron M processors hold the majority of the value mainstream and a portion of the mainstream thin-and-light market. Both the Mobile Sempron and Celeron M processors share much of the same architecture as their higher-powered counterparts, but they tend to have lower clock speeds, slower buses, or less L2 cache. Yet, the recent influx of inexpensive notebooks is due in part to the widespread availability of the Mobile Sempron and Celeron M processors. As of press time, Compaq/HP, Dell, Gateway, and others have sub-$750 notebooks available, and they’re all powered by Mobile Sempron or Celeron M

AOpen XC Cube Mini

Anxious to get in on the mini market, AOpen recently unveiled a Mac Mini look-alike with some fairly impressive tech specs. The MZ855-II centers on Intel’s 855GME mobile chipset and supports Pentium M processors at speeds of up to 2.13GHz. It boasts a pair of memory slots, each capable of accommodating 1GB of DDR400 RAM. Half-sized AGP and PCI slots give the platform some semblance of expandability. USB 2.0, IEEE 1394, Gigabit Ethernet, and a number of digital audio ports combine to enable much better connectivity than the Apple offering, too. And although AOpen makes significant progress in the style and size department, the MZ855-II is about twice as deep and nearly two times heavier than the Mac Mini. It’d be hard to argue that AOpen has bested Apple with the XC Cube mini design. Though attractive, the XC Cube Mini still looks very much like a PC.Then there’s the issue of price; the MZ855-II XC Cube Mini retails for roughly $449. After adding a processor, hard drive, optical drive, and memory to that price, you may very well be topping a grand. On the other hand, Apple sells a fully loaded Mini for $599. Perhaps more exciting than this 855GME-based XC Cube Mini is the Intel mobile 915-based version that’s in the works and expected out within a month or so. Featuring PCI Express connectivity and a more advanced I/O controller, you can expect it to at least support SATA hard drives.

AMD’s new Sempron 3400

AMD’s new Sempron 3400+ is an example of mainstream hardware. Endowed with many of the same building blocks that give Athlon 64 and Opteron processors their mojo, Sempron enables somewhat reduced performance at drastically lower prices. The 3400+, for instance, sells for less than $150 in Socket 754 trim. The other models naturally fall in place behind, dipping as low as $74 for a 64-bit Sempron 2600+ designed for the same interface. The 3400+ runs at 2GHz and features a larger 256KB L2 cache, up from 128KB and half the size of many Athlon 64 processors. An integrated 64-bit memory controller supports up to DDR400 modules in single-channel configurations, while the 800MHz HyperTransport bus shuttles up to 3.2GBps of data bi-directionally. Because the 3400+ centers around AMD’s 90nm SOI Palermo core, it incorporates a number of features inherent to the latest Athlon 64 processors. SSE3 instruction support is one, as is an improved memory controller with mismatched DIMM recognition, better memory mapping, and improved load handling. A recent upgrade to 64-bit software support represents one more feature shared between Sempron and pricier AMD chips. Manufacturing advances have shrunk the Sempron die down to 84mm² and reduced heat dissipation to 62W. Although it can’t compete with AMD’s Athlon 64 lineup, Sempron 3400+ offers a reasonable compromise between price and performance. It isn’t for the hardcore gamer or media encoding junkie, but everyday business apps run just fine. While many may question the longevity of Socket 754, think of the Sempron 3400+ as a means to save money here and now.

Intel Pentium M 770

Intel’s Pentium M is a silent beast. On one hand, the very fastest 2.13GHz model boasts a thermal design power of 27W, which is a fraction of the dual-core Pentium D chips. It will even run passively without a cooling fan under the right conditions. The Pentium M architecture is efficient at accelerating game performance and posts numbers that rival Pentium 4 and Athlon 64 systems. The Pentium M combines efficiency and muscle to power through software without relying on inflated clock frequencies. The architecture is tweaked for speeds up to 2.13GHz. It wields a 2MB L2 cache as a means of shifting reliance from memory accesses and further augmenting overall performance. The processor interfaces with a 533MHz FSB through Intel’s 915-series of mobile chipsets. Performance clearly improves by virtue of the faster bus speed and extra bit of operating frequency. Of course, it’s still important to establish a balance. Although the Pentium M is small and runs at a reduced clock rate, Intel still charges upwards of $650 for a boxed retail processor, and that’s on top of the other platform components you’ll need for a complete Socket 479 system. Yet, if you’rebuilding an ultra-quiet home theater machine or a DIY notebook, the Pentium M 770 is your fastest option.

AMD Athlon 64 FX-57

Another month, another speed bump from AMD. But when I’m dealing with super-high-end expensive gamingtargeted processors, do I care? Of course I do. Enter the new AMD Athlon 64 FX- 57. I apologize right now for those of you with the FX-55 that was top of the charts for a few weeks. It’s still silly fast but not quite as fast as this replacement. Based upon the very successful Athlon 64 939-pin platform, and with 1MB of L2 cache at its disposal (as is the norm for these FX’s), the FX-57 clocks at 2.8GHz. AMD is rapidly approaching the 3GHz mark that Intel set sometime ago, but the only difference is that, clock for clock, the boys in green are so much more efficient in almost every department. AMD didn’t just bump up the clock speed by 200MHz for the FX-57 over the FX-55, it threw in a few extra tweaks for even more performance. You certainly get what you pay for when spending around $1,000 (actually $1,031 in quantities per thousand) on a CPU. But who in his right mind would do such a thing anyway? The introduction of the FX-57 will not mean the death of the FX-55, and even the price remains the same at $827, so you need not feel too bad if you shelled out that amount recently. However it’s rather odd for the price to remain stagnant, especially forAMD, and it certainly won’t please AMD enthusiasts that have supported the company through thick and thin. The core itself is a revised K8 core (codenamed San Diego) with a die shrink thrown in for good measure, down from 130nm to 90nm. It’s similar to previous Athlon 64 X2 cores based on the San Diego, with all of the same E-K8 revisions with both SOI process and Dual Stress Liner technologies. The die shrink obviously helps FX processors run faster, suck up less power, and even cost less to manufacture than the 130nm cores. The FX-57 can achieve its increased clock speed while maintaining the same 104-watt TDP as the FX-55 (in large part due to this die shrink), and the nominal core voltage sits at 1.4 volts. Having said that, I didn’t have any real luck overclocking the FX-57, which was disappointing considering the die shrink. It kind of ran at 3GHz with improved cooling, but perhaps later revisions will run above and beyond that mark more stably. As with other Rev E chips, the FX-57 AMD has SSE3 instructions, along with a more flexible memory controller (letting you use different size DIMMs on the same channel) and improved memory mapping (more efficient use of memory space) and loading (you can now populate memory with double-bank DIMMs with no sacrifice in performance).
My benchmarks helped to demonstrate how potent the FX-line remains at gaming. No Intel CPU can come close in that department. Games such as UT2004 or Painkiller are especially CPU-bound and benefit from the bumps that FX-57 provides; of course any FX processor performs extremely well and it just depends on how badly you want those bragging rights. Furthermore, just when I thought playing Battlefield 2 with a dual-core X2 and SLI 7800 GTX cards was the “right” thing to do for the price, the performance figures bring me back down to my gaming roots. Yes folks, the FX-57 is fast, and it absolutely makes me forget about all things multithreaded; therefore, I would say that 99.9% of you should go for the more forward-looking dual-core technology in the form of the Athlon 64 X2 4800+, especially since it’s so much more affordable. However, for that 1% that needs 1,500HP, a two-stage nitrous setup, and single-digit quarter-mile runs for gaming, go for the FX-57. Just remember that it’s part of a dying breed in terms of single-core processors.

GeForce 7800 GTX

After a solid few weeks of gaming and benchmarking NVIDIA G70-powered GeForce 7800 GTX cards, I’m here to tell you what I think. First, let me get the obvious commentary out of the way: These puppies are wickedly fast. So much faster and more efficient than the previous 6800 series of GPUs from NVIDIA, that under the right conditions, a single 7800 GTX can outperform a pair of 6800 Ultras in SLI mode. By that I mean don’t spend $1,000 on graphics cards for previous-generation DX7 and/or DX8 class games. The 7800 GTX is geared for DX9 and then some at high resolutions with all of your favorite AA/AF eye candy turned on. It also eats more complex pixel shaders and is an AA monster. Clearly it’s built for upcoming DX9 games based on PS3.0 and FP32. The early demos I saw included Huxley, UT- 2007, and Age Of Empires III and looked absolutely stunning running with SLI. NVIDIA says that the 7800 GTX is not a “refresh” of the 6800, and instead the pixel and vertex shader pipelines are redesigned and gentrified, whilst at the same time decreasing power consumption. The core is still based upon a 0.11-micron process, and even though the clock speed is bumped up 5MHz to 430MHz, it has 302 million transistors (up from 222 million), indicating some serious intentions under the hood. The pixel pipelines are also up from 16 to 24, and even the vertex pipelines are up to eight. That’s some serious horsepower and torque, assuming that they were the same as the 6800s pipes, which of course they aren’t. Peak pixel fill rate remains at 6.8Gpps (gigapixels per second), but the pixel texture fill rate is now capable of 10.32Gtps (gigatexels per second), which is up from 6.8Gtps on the 6800. Using a 256-bit memory interface, the clock speed for the GDDR3 memory is up to 1.2GHz, and peak memory bandwidth is up from 35.2GBps to 38.4GBps, which takes over from the previous champ (Radeon X850 XT PE) that boasted 37.8GBps. But all those numbers that don’t sound too impressive, save the fact that efficiency in the graphics pipeline is improved, NVIDIA actually spec’d out over 1,300 different shader operations from games past, present, and future to find out which were the most popular, required optimizations and could be improved by redesigning the shader units. Obviously NVIDIA isn’t overly free with its secret sauce, but some things have been revealed, such as the scalar unit in the vertex pipeline that’s now 20% more efficient and the vertex texture fetch unit that’s received a boost. Also, the floatingpoint- texture processors are now jazzed up, which improves HDR capabilities.

Also, the FP32 shader units are now able to cope with two Vect4 MAD (multiply-add) operations per clock. Visually, there’s a new AA mode, called Transparency Adaptive AA, which takes AA further on to alpha blended areas that previously didn’t get the AA goodness in more traditional AA modes, such as multisampling. So, for example, all those chainlink fences set one behind the other that used to look blurry in Half-Life 2, now look more realistic. One area in which I was particularly excited to see NVIDIA move forward is video acceleration features, dubbed Pure- Video. ATI has long been king in that department, but now NVIDIA has listened to end-user and OEM feedback and stepped up to the plate. The programmable video processor and its HDTV and MPEG-2 features are accelerated and sport 2:2 pull-down correction. The only issue is that I’m still awaiting the Force- Ware Release 75 driver in order to dabble, test, and play with PureVideo. And I’m especially happy to hear about its apparent CPU-utilization tweaks. I look forward to seeing more games with DX9c-style Shaders, normal maps, HDR rendering, and floating-point textures. Clearly the best solution available is the 7800 GTX. ATI delayed its next-gen R520 architecture, but even so, it’s going to take something mighty special from the company to even catch up to NVIDIA.

XDR2 From Rambus.

The PlayStation 3 will have rockin’ XDR memory chips that transfer data at a rate of 3.2GHz; however, the new memory interface from Rambus, dubbed XDR2, will transfer data at 8GHz. Richard Warmke, director of Rambus product marketing, says games with intense 3D graphics should benefit from the faster speed. But will this enable memory to close the gap with processor speeds? It will help, Warmke says, but not enough to satisfy power users. Consequently, Rambus will license the technology to memory chipmakers that are expected to debut XDR2 DRAM chips in 2007. The XDR2 interface includes several innovations that enable data to overcome noise problems in computer pathways. XDR2 also includes an innovation known as microthreading, which can quadruple memory speed through architectural improvements alone. With microthreading a processor will fetch data in chunks of eight bytes at a time, instead of the typical 32 bytes, which is more efficient.

A4Tech USB Battery Charger

Of course you never forget to charge and pack your notebook, but what about all of those other electronic devices you just can’t function without, like your digital camera, MP3 player, and electric razor? Once the batteries drain, you’re out of luck until you can find a wall outlet or purchase more batteries. With the A4Tech portable USB battery charger ($16; www.ebay.com), however, you don’t have to subject yourself to the mercy of your powerhungry electronics while on the road. This inconspicuous device plugs into a USB port of your notebook and uses the notebook’s power to charge up to two AA or AAA nickel metal hydride or nickel-cadmium rechargeable batteries. Speaking of batteries, A4Tech’s handy charger includes two AA and two AAA nickel metal hydride rechargeable batteries. Green LEDs indicate the charging status of the batteries, and a built-in USB port lets you connect other USB devices or daisy-chain multiple chargers. But don’t worry about needlessly draining your notebook battery, the device shuts off automatically once the batteries are fully charged.

Blog Searching Made Easy by Google

Are you having trouble finding the best of the best political blogs? Searching for a good physical fitness blog, perhaps? Or just looking for some juicy celebrity gossip? (We all know it happens.) Check out Google Blog Search (blogsearch.google.com). Just type in the topic of your choice and click Search Blogs, and Google will display lists of links to blogs relating to your topic, as well as individual postings from all over the Internet. Click the Advanced Search link to specify further specifications for your blog search, including the option to search blogs in English, Spanish, German, Korean, Russian, Hebrew, and many other languages. As with the basic Google Search, users can employ SafeSearch to filter out adult content from their blog search. Google Blog Search is also available with a Blogger-style interface (search.blogger.com).

New Yahoo!!

Yahoo! Mail is getting faster, and it has many more features, too. Yahoo! released the beta version of its new Yahoo! Mail in September. The limited number of current Yahoo! Mail users who are testing the latest version of Yahoo! Mail are guaranteed a faster email experience with enhanced functionality, all of which is made possible through technologies such as DHTML (Dynamic HTML), XML (Extensible Markup Language), and SOAP (Simple Object Access Protocol).“The beta gives people a faster and more dynamic way to experience Yahoo! Mail, yet continues to offer the same features they rely upon today: great antispam and virus protection, tons of storage—it’s all there,” says Ethan Diamond, director, product development, Yahoo! Mail. The new version of Yahoo! Mail is available on Firefox and Internet Explorer. In addition to its increased speed, the beta gives users a fast and easy-to-use interface that functions like a desktop client, lets users view multiple email messages at the same time, provides keyboard shortcuts and right-click menus, and allows users to scroll all message headers in a folder rather than page by page.

Logitech Introduces New Bluetooth Headset.

Logitech’s new Bluetooth-based Mobile Traveller Headset ($79.99; www.logitech.com) features seven hours of continuous talk time and as many as 300 hours of standby time. The headset includes a call/answer button, and it comes with voice dialing, three-way calling, and a mute feature. To help eliminate interference from wind noise and car AC/heating fans, Logitech developed its patent-pending WindStop technology, which is also a feature of the new headset. An over-ear design with a flip clip fits on either ear, and the headset weighs less than 1 ounce.