Why use Serial communication?

If you pay attention, you can see that all current technologies tend to translate from parallel to serial. The new IDE standard for hard drives is also Serial (Serial ATA - SATA). Then the PCI bus also switched to a string with PCI Express standard. The SCSI interface is also turning into a chain.

The only difference between serial communication and parallel communication is that this standard only transmits one bit over a period of time, while parallel communication transmits multiple bits over a period of time. This difference makes parallel communication often faster than chain communication.

However, this is not true in all cases. Chain communication can also be faster than parallel communication if the bits are moved at a faster rate. For example, SATA ports can achieve data transfer rates up to 150Mbps, while traditional IDE ports only stop at 133Mbps.

There are several reasons for moving devices from parallel communication to chain communication. In parallel communication, since many bits are moved over a period of time, each bit needs a wire. For example, in a 32-bit interface, 32 wires are needed for data transmission, not to mention the wires for the signal. The higher the number of bits transferred each time, the more wires needed, making it difficult to produce cables and installation costs. In string communication, only two wires are needed, so the communication process between the two devices will be much simpler.

The higher the data transfer rate, the easier it will be for electromagnetic interference. Each conductor can become an antenna, collecting a lot of noise from the outside environment, affecting the data transmission process. In parallel communication, since many wires are used, the electromagnetic interference process is very serious. In string communication, on the other hand, since only two wires are involved in the data transfer process, the problem is much easier to solve, and thus can ensure the safety of the wire.

There is still a small problem to mention here. That is even when we say in parallel communication, all bits are transmitted at the same time, but that does not mean the bits are at the receiver (receiver) at the same time. If in a small device, this is not very important, but if it is a high-speed device, the difference in time to receive the bits will cause the device to waste time (because Wait all the incoming bits). And this can seriously affect the performance of the device because data transmission only takes place in very short periods of time.

Another difference is that parallel communication is a half-duplex process, while chain communication is a two-way (full-duplex) process. More specifically: in parallel communication, there is only one path to transfer data between the transmitter and receiver. Since there is only one line, it is impossible to simultaneously transmit and receive data at the same time. In string communication, since only two wires are used, manufacturers often add two more wires (a total of 4 wires). Two conductors for data transmission and the other two wires for data reception. In other words, there will be two parallel lines in the communication, one for transmission, and the other for receiving data. This capability helps the transmission and reception of data take place simultaneously.

The above architectural differences help chain communication can transmit data twice as much as parallel communication.