PCI-X

It uses a modified protocol to support higher clock speeds (up to 133 MHz), but is otherwise similar in electrical implementation.

PCI-X I/Os are registered to the PCI clock, usually through means of a PLL to actively control I/O delay the bus pins.

Some devices, most notably Gigabit Ethernet cards, SCSI controllers (Fibre Channel and Ultra320), and cluster interconnects could by themselves saturate the PCI bus's 133 MB/s bandwidth.

These extensions were loosely supported as optional parts of the PCI 2.x standards, but device compatibility beyond the basic 133 MB/s continued to be difficult.

Developers eventually used the combined 64-bit and 66-MHz extension as a foundation, and, anticipating future needs, established 66-MHz and 133-MHz variants with a maximum bandwidth of 532 MB/s and 1064 MB/s respectively.

It was an effort to codify proprietary server extensions to the PCI local bus to address several shortcomings in PCI, and increase performance of high bandwidth devices, such as Gigabit Ethernet, Fibre Channel, and Ultra3 SCSI cards, and allow processors to be interconnected in clusters.

PCI-X 2.0 makes additional protocol revisions that are designed to help system reliability and add Error-correcting codes to the bus to avoid re-sends.

[12][13] ServerWorks was a vocal supporter of PCI-X 2.0[14] (to the detriment of the first generation PCI Express) particularly through its chief Raju Vegesna,[15] who was however fired soon thereafter for roadmap disagreements with the Broadcom leadership.

[16] In 2003, Dell announced it would skip PCI-X 2.0 in favor of more rapid adoption of PCI Express solutions.

[17] As reported by PC Magazine, Intel began to sideline PCI-X in their 2004 roadmap, in favor of PCI Express, arguing that the latter had substantial advantages in terms of system latency and power consumption, more dramatically stated as avoiding "the 1,000-pin apocalypse" for their Tumwater chipset.

The theoretical maximum amount of data exchanged between the processor and peripherals with PCI-X is 1.06 GB/s, compared to 133 MB/s with standard PCI.

PCI-X also improves the fault tolerance of PCI, allowing, for example, faulty cards to be reinitialized or taken offline.

Many 64-bit PCI-X cards are designed to work in 32-bit mode if inserted in shorter 32-bit connectors, with some loss of speed.

PCI-X should not be confused with the similar-sounding but incompatible PCI Express, commonly abbreviated as PCI-E or PCIe.

PCIe is a serial point-to-point connection with a different physical interface that was designed to supersede both PCI and PCI-X.

The serial interface of PCIe suffers fewer such problems and therefore does not require such complex and expensive designs.

PCI-X slots take quite a bit of space on motherboards, which can be a problem for ATX and smaller form factors.

A Dual Port Gigabit Ethernet Network Card for single PCI-X slot to save on PCI-X slots and use the full potential of the PCI-X 64-bit bus.
A 8 port SATA host bus adapter for PCI-X from Lsi logic .
HP VISUALIZE fx10 Pro video card for PCI-X
3.3 V and 5 V keying of 64-bit PCI cards (both PCI and PCI-X). While most 64-bit PCI-X cards are universal and are backward compatible with common 32-bit 5 V PCI slots, PCI-X slots are 3.3 V and will not accept 5 V-only PCI cards.
64-bit PCI-X card partially inserted in 32-bit PCI slot, showing compatibility
A MOTU PCIX-424 audio interface card, which was also released in standard PCI and PCIe variations.