PCI was the first industry-wide expansion slot solution; it appeared in 1992 when Intel led the industry to produce a universal expansion solution. Intel’s initiative grew into an alliance known today as PCI-SIG (PCI Special Interest Group). PCI-SIG proceeded to standardise other PCI derivatives, including PCI Express or PCIe. Despite the similarity of the names, PCIe topology is radically different from that of PCI.

When PCI became too slow for high-end video cards, the AGP slot was developed exclusively for their use. This in turn was replaced by PCIe. Whereas PCI is a parallel bus, PCIe is a point-to-point connection. All devices on a PCI bus share the same data path, so a bottleneck can occur if too many devices attempt to transmit data simultaneously. By contrast, each PCIe slot is connected to the CPU board chipset using a dedicated lane; each lane is a high-speed serial communications link. Lanes can be grouped to form higher bandwidth connections. The ‘x’ that follows the description of a PCIe connection specifies the number of lanes it uses.

A PCIe card fits into a slot of its physical size or larger (maximum x16), but may not fit into a smaller PCIe slot (e.g. a x16 card in a x8 slot). Some slots use open-ended sockets to permit physically longer cards and negotiate the best available electrical connection. The number of lanes actually connected to a slot may also be less than the number supported by the physical slot size.

AMP’s AVC8000S 8x D1 Video Frame Grabber is built on to a single PCIe x1 board.