The personal computer industry is about to undergo a major metamorphosis, due in large part to the recent efforts of Intel and AMD. For the last few decades, mainstream desktop processor manufacturers have focused on increasing the performance of their CPUs by raising their clock speeds, augmenting the base architectures with specialized instruction sets, and by making changes to the pipeline to achieve a higher IPC (among other things). Advances have also been made in I/O technologies like AGP, PCI Express and HyperTransport, along with a plethora of other platform enhancements that have also been employed to increase overall system performance. But even with all of these changes, most modern consumer-class desktop processors have remained fundamentally similar to their ancestral counterparts, in that they've all been comprised of a single processor core. With the recent introduction of Intel's Pentium Extreme Edition 840 and Pentium-D processor, AMD's dual-core Opterons, and now the Athlon 64 X2 line of processors, personal computers can be outfitted with a single CPU that has not one, but two available processor cores.
Having dual-cores incorporated into a single central processor, basically doubles the amount resources available to the operating system and applications being run on the PC. It's essentially the same as having a matched pair of processors installed into a single system. But to fully harness all of the extra horsepower afforded by the second core, the operating system and applications must be programmed specifically to be multi-threaded. Owners of dual-processor or HyperThreading enabled machines also know that there are benefits to having dual-processors (or simulating the effects of dual-processors) when multi-tasking with numerous single-threaded applications as well. However to date most developers haven't focused on delivering multi-threaded applications because there simply hasn't been a huge demand for them on the desktop. Of course, there's been a demand in the enterprise, and with users of specialized animation and imaging applications, but let's be honest "Joe Sixpack" couldn't care less until recently. Now though, with the two most dominant players in X86 processors having announced and revealed functional dual-core processors, the development community has much more incentive to produce applications that will benefit from the new abilities of these more powerful CPUs.
A few weeks ago, we previewed Intel's Pentium Extreme Edition 840 Processor and quantified some of the benefits of Intel's dual-core architecture. The 840 is basically two HyperThreading enabled Prescott cores, each with 1MB of L2 cache, running at 3.2GHz. This design is recognized as 4 virtual processors by the operating system (in our case Windows XP Pro), and even though it's clocked significantly lower than some other high-end Pentium 4 processors, in certain scenarios, like 3D rendering for example, the Pentium Extreme Edition 840 is measurably faster than its single-cored brethren. Today we've got AMD's first dual-core desktop processor on the test bench, the Athlon 64 X2 4800+. This CPU is essentially a pair of Athlon 64 4000+ processor die, incorporated into a single CPU. The A64 X2 4800+ is clocked at 2.4GHz, and sports all of the same functionality as the recently released Rev. E (Venice Core) Athlon 64 processors. Read on to get more familiar with this new powerhouse, and see how she performed with some of today's popular benchmarks and applications.
The Athlon 64 X2 - Dual-Core Processor
|Specifications: Socket 939 Athlon 64 X2 4800+
|Is Two Better Than One?
|AMD64 - When utilizing the AMD64 Instruction Set Architecture, 64-bit mode is designed to offer:
_•_Support for 64-bit operating systems to provide full, transparent, and simultaneous 32-bit and 64-bit platform application multitasking.
_•_A physical address space that can support systems with up to one terabyte of installed RAM, shattering the 4 gigabyte RAM barrier present on all current x86 implementations.
_•_Sixteen 64-bit general-purpose integer registers (per core) that quadruple the general purpose register space available to applications and device drivers.
_•_Sixteen 128-bit XMM registers (per core) for enhanced multimedia performance to double the register space of any current SSE/SSE2 implementation.
Integrated DDR memory controller:
_•_Allows for a reduction in memory latency, thereby increasing overall system performance.
An advanced HyperTransport link:
_•_This feature dramatically improves the I/O bandwidth, enabling much faster access to peripherals such as hard drives, USB 2.0, and Gigabit Ethernet cards.
_•_HyperTransport technology enables higher performance due to a reduced I/O interface throttle.
Large level one (L1) and level 2 (L2) on-die cache:
_•_With 128 Kbytes of L1 cache and 512K or 1MB of L2 cache per core, the AMD Athlon 64 and Athlon 64 X2 processors are able to excel at performing matrix calculations on arrays.
_•_Programs that use intensive large matrix calculations will benefit from fitting the entire matrix in the L2 cache.
_•_A 64-bit address and data set enables the processor to process in the terabyte space.
_•_Many applications improve performance due to the removal of the 32-bit limitations.
|Processor core clock-for-clock improvements:
_•_Including larger TLB (Translation Look-Aside Buffers) with reduced latencies and improved branch prediction through four times the number of bimodal counters in the global history counter, per core, as compared to seventh-generation processors.
_•_These features drive improvements to the IPC, by delivering a more efficient pipeline for CPU-intensive applications.
_•_CPU-intensive games benefit from these core improvements.
_•_Introduction of the SSE2 instruction set, and now SSE3 (Rev. E and Athlon 64 X2) which along with support of 3DNow! Professional, (SSE and 3DNow! Enhanced) completes support for all industry standards.
_•_32-bit instruction set extensions.
_•_AMD's Fab 30 wafer fabrication facility in Dresden, Germany
_•_.13 micron SOI (silicon-on-insulator) - Newcastle
_•_.09 micron SOI (silicon-on-insulator) - Winchester
_•_.09 micron SOI (silicon-on-insulator) - Venice & A64 X2
_•_Newcastle Core - 144mm2
_•_Winchester & Venice - 84mm2
_•_Toldeo - 199mm2
_•_Newcastle Core - Approximately - 68.5 million
_•_Winchester Core - Approximately - 68.5 million
_•_Venice Core - Approximately - 68.5 million
_•_Toldeo Core - Approximately - 233.3 million
_•_1.40v (Winchester, Venice & Toledo)
To test the AMD Athlon 64 X2 4800+ CPU, we used an NVIDIA nForce 4 SLI based motherboard from Asus, the A8N-SLI Deluxe and it was equipped with a pre-release BIOS that supported AMD's new dual-core processors. Other than the BIOS though, this setup is no different hardware wise than what's already available in the retail channel.
Some Pictures of the test-bed with our oversized Thermaltake cooler
Unlike Intel's dual-core Pentium-D and Pentium Extreme Edition processors, which will require a new motherboard that's based on a compatible chipset like the i955, i945 or the nForce 4 Intel Edition, AMD's new dual-core "X2" processors will work with virtually any socket 939 platform as it exists today. The only caveat is that your motherboard will need a BIOS upgrade to support an Athlon 64 X2. We suspect most major motherboard manufactures will have updated BIOS files available in time for the official release of the Athlon 64 X2.
The rest of the components used in our test bed consisted of a Corsair TWINX1024-3200XLPRO 1GB memory kit, capable of 2-2-2-5 timings, a GeForce 6800 GT, a 10K RPM Western Digital "Raptor" hard drive, and a the large Thermaltake cooler you see here. With what is basically a pair of 4000+ processors under the hood, the Athlon 64 X2 4800+ requires a robust cooling solution. This copper / aluminum hybrid cooler with heat-pipes, which is reminiscent of the cooler
supplied with the Athlon 64 FX-55
, was up to the task though, at least it was if the temperatures being reported in the hardware section of the motherboard BIOS were accurate, which we're fairly certain they were. More on this to follow.