The current dual-core systems from Intel and AMD both use SMP (Symmetrical Multi-Processing). SMP is the most common approach to creating a multi-processor system, in which two or more separate processors work together on the same motherboard. The processors coordinate and share information through the system bus, and the processors arbitrate the workload amongst themselves with the help of the motherboard chipset and the operating system. The operating system treats both processors more or less equally, assigning work as needed.

Each processor within the die functions as an independent unit, and it is the responsibility of the motherboard and chipset to coordinate them and exchange data between them when necessary.

Intel's approach forces both processors to communicate through the Northbridge and FSB outside the processor die which is different from the internal communication method used by AMD for its dual-core Athlon 64 processors.

There's been some contention that this method is less efficient and will bring down the performance slightly, but we'll see some actual numbers later in the article to settle that question. What we can say for sure is that Pentium D processors like higher memory speeds, regardless of actual FSB speed; dual channel PC2-8000 DDR memory is the bare minimum realistically.

What Does Multiprocessing Offer?

Software can be defined as either single-threaded or multi-threaded, with a 'thread' being a set of operations the processor is performing on a portion of a given program at a given time. Single-threaded programs are designed to allow only a single thread to be operated on at any one time, whereas multi-threaded programs can separate different portions of the work that needs to be done by the processor(s) into different threads.

Single-threaded programs derive no major benefit from multiple-processor systems, but the system as a whole benefits because the second processor can continue to operate while the first processor is using 100% of its time on a single program or error. Multi-threaded applications can share their load between multiple processors, allowing different parts of the program to be executed at the same time, and thus derive the largest benefit from dual-processor systems.

Dual processor systems also gain from a general decline in latency. Simply put, while there is no current way to share the current operating system load evenly between two processors, the second processor can step in and keep the system running smoothly while the first is maxed out to 100% burning a CD or encoding a file (or from a software error).

To sum up, any modern operating system will see at least some benefit from running with more than one processor, even when that OS is running single-threaded applications only. This benefit will be specific to certain applications and conditions though, not a general speed boost.