Main Memory

Main Memory

                                                             
Early form of computer main memory were mercury delay lines, which were mercury tubes, which stored the data in the form of ultrasonic waves and cathode-ray tubes, which used to store data on the tubes screen as data. Magnetic drum invented in 1948 used iron oxide coating on the rotating drum to store data and programs as magnetic patterns.

In a binary computer, any bistable device (some which can be placed in one of two states) can represent two potential bit values ​​of 0 and 1 and thus can serve as a computer memory.

Magnetic-core memory, the first relatively cheap RAM device, appeared in 1952. This two-dimensional wire was made of tiny, donut-shaped ferrite magnets at intersection points of the grid. These stars carried the currents to change the direction of the magnetism of each core, while the third string threaded through the donut detected its magnetic orientation.

The first integrated circuit (IC) memory chip was shown in 1971. The IC store stores a little bit in combination with the transistor-capacitor. The capacitor is charged for representing 1 and there is no charge for 0; The transistor switches between these two states. Since a capacitor charge is decayed gradually, the IC memory is the dynamic RAM (DRAM) in which its stored values ​​should be periodically refreshed (Either 20 milliseconds each). There is also a static RAM (SRAM), which does not need to be refreshed. Although faster than DRAM, SRAM uses more transistors and thus is more expensive; Its primarily used for CPU internal registers and cache memory.

In addition to the main memory, computers for computer displays typically have special video memory (VRAM) to keep graphic images, which are called bitmaps. This memory is often a dual port - a new image can be stored at the same time as its current data is being read and displayed.

It takes time to specify an address in the memory chip, and since memory is slower than a CPU, therefore there is a benefit to memory that can quickly move a series of words after the address is specified. One such design is known as synchronous DRAM (SDRAM), which was widely used until 2001.

Even so, data transfer through "bus" - a set of stars that connects the CPU to memory and peripheral devices-is an obstacle. For this reason, CPU chips now have cache memory - small amounts of fast SRAM. Cache contains copies of data from blocks of main memory. A well designed cache can be done from 85-90 percent of memory references in general programs, which leads to manifold speed in data access.

The time between the two memory reading or writing (cycle time) for the initial core memory was approximately 17 microdes (one millionth of a second) and in the beginning of the decade of 1970, there was approximately 1 microsound for the core. The first DRAM had about half the cycle cycle of Microsoft, or 500 nanoseconds (one billionth of a second), and today it is 20 nanoseconds or less. An equally important measure is the cost per bit of memory. The first DRAM stores 128 bytes.

 (1 byte = 8 bits) and spends around $ 10, or $ 80,000 per megabyte (millions of bytes). DRAM can be purchased for less than $ 0.25 per megabyte in 2001. This huge drop in cost has resulted in large scale manipulation and visualization of possible graphical user interfaces (GUIs), display fonts of word processors, and scientific computers.