The ultimate guide to motherboards

As with every other component, motherboards have come a long way from the original IBM PC of 1981. If you're old enough to remember the first De Lorean DMC-12, perhaps the original PC XT mobo still casts some dark shadow over your memory? At the time there were certainly wonders to behold; these days, they simply look a mess with integration the last thing on the designers mind and all the IO having to be decidedly off-board.

The XT had all the same parts as today's mobos, they just worked a little slower. Instead of having a dedicated, integrated chipset, the XT used discrete off-the-shelf components: clock generators, DMA controller, interrupt handler, keyboard and bus controllers, a system timer and a real-time clock, along with the CPU, FPU, ROM and system memory.

That's eleven individual integrated-chips along with all the additional components, adding up to one expensive mother of a board. What we might recognise today as a motherboard didn't appear until 1986, when a company called Chips and Technology offered a single-chipset solution, by rolling most of the previous parts into one. Requiring only a few support chips, it simplified mobo design, reduced costs and started the trend of ever-greater integration.

Twin-chipsets still the way?

Almost all motherboards still use a twin-chipset design, commonly called the North and Southbridges. Intel tries to call it a hub these days, but we're not sure why. This dual-chipset design balances functionality and manufacturing considerations. The Northbridge is the high-speed part, sitting between the processor bus and connecting it to the graphics, memory – okay, so not AMD – and interface buses. At the other end of the motherboard the Southbridge handles all the I/O; ATA interfaces, USB, networking and PCI; any legacy interfaces such as floppy drives and ISA slots are often done via a 'Super IO' chip.

Chipset makers often offer alternative Southbridges for premium and budget boards. It's amazing just how robust this design has been. Nvidia, with SLi, has offered single-chip solutions in the past, but the vast majority of solutions share features across the pair. In future, it does seem the Southbridge will be relegated to less and less importance, but it'll still be around at least into 2010, even if it's just to provide extra PCI Express lanes.

How do the two bridges communicate? And how do they manage to transfer such large amounts of data so quickly? Originally, the PCI bus was used; perfectly adequate when hard drives ran at 8MB/s and the parallel port was most people's idea of a high-speed external interface. But with data demands rising exponentially, Intel introduced a dedicated quad-pumped Hub Bus that ran twice as fast as PCI.

High-speed busses

This turned out to be a revolutionary move: nowadays, every manufacturer uses this type of high-speed link. It's essential with so much data flying around the Southbridge. Intel now uses DMI running at 2GB/s, VIA has V-Link, SiS has MuTIOL, while HyperTransport has been used by AMD and NVIDIA. Despite the different technologies, they all do the same thing: transport data at high-speed. This bus system also helps simplify motherboard design as the serial links require far fewer physical connections, resulting in faster design turnaround, fewer physical layers and a reduced price.

It is a recurring theme in modern motherboards that old parallel buses are being replaced by new, shiny, high-speed serial ones. The only major exception to this rule being the memory bus: see Memory Tech opposite for more details on that. Other than speed – reason enough for users to love serials – from a manufacturing point of view, it simplifies design and cuts costs, as fewer numbers of lines are easier to route around the motherboard and result in less layers being required, significantly reducing costs.

The most obvious addition to the serial party was the introduction of PCI Express. This usurped PCI and AGP as the primary way to add graphics and expansion cards to a PC. Attached to the South or Northbridge, PCIe slots have multiple lanes associated with them – 16 for a graphics slot offering 4GB/s worth of bandwidth. The new PCIe v2 standard doubles this rate and is supported by all the latest chipsets.

Multiple-GPUs

The biggest side-effect of PCIe was the ability for motherboards to suddenly support multiple graphics cards. As you probably know, Nvidia offers SLI and AMD CrossFire – the latter also supported on Intel's high-end X chipsets – for multiple GPU support. Whilst twin 16-lane PCIe slots have been the norm until now, expect new boards to offer three slots, as we move into a period where GPUs will be used for physics acceleration. So it's a case of more the merrier… Oh, our poor steaming power supplies!