In 1996 Intel implemented the PCI standard which ran a maximum speed of 33MHX the same speed as the VLB (Vesa Local Bus).
Macintosh had already implemented this standard. By design the PCI is more configurable than this ISA; it allowed users to determine tradeoffs based on individual needs, such as a higher throughput or a faster response time.
In the Macintosh system the PCI bus architecture was important because it allowed ships to communicate directly with the process of building into the connection stream or any hardware buffers.
All the PCI devices have a set of 256 registers, and each of theses registers contain information. Some of that information is fixed, for example device ID and the information can re reconfigured, like address maps or interrupt types. One of the main advantages to this approach was that with so much information stored on the PCI device itself, the PCI should be able to detect, identify, and configure the device when the PCI component is installed into the motherboard, and the computer is booted up.
Because of the registers, even early PCI device had support for PnP (Plug and Play) operations. In the ISA technology users had to manually move shunts around specific jumper pins on the device itself. In addition unlike the ISA bus that requires one more unique IRQ per ISA slot device, PCI devices can, at least in theory share an IRQ, however because of the having of resources can lead to hardware conflicts, this can be easier said than done. Certain types devices share better than other devices and device resources manipulation is not for newbie technicians.
