what is ram memory

what is ram memory

The PC must have main memory. Main memory is high-speed memory that the CPU can read from or write to. "High speed" means, less than a microsecond to read/write. The other name for such memory is RAM, Random Access Memory-a particular kind of chip on circuit boards.

Memory is easy to pick out on a circuit board. It's packaged either as a "bank" of eight or nine small chips, or it's a mini-circuit board with several square chip mounted on it, called a SIMM (Single Inline Memory Module). Memory is always organized into banks either eight or nine discrete chips, or a SIMM. Most motherboards have room for four banks of memory. As each SIMM is the equivalent of nine chips built in single independent card, SIMMs make replacing bad memory easier.

RAM chips are distinguished by their following characteristics:

•  Package type, which can be:
•  DIPs, which are normal looking chips. DIP means Dual Inline Package. It is just a simple IC with two sets of legs inline. These ICs are directly inserted in sockets on the motherboard. Laptop and old 8088 to 386 computers use this package.
•  Single Inline Pin Packages, or SIPPs (a small circuit board with pins across its bottom).
•  Single and Double Inline Memory Modules (SIMMs and DIMMs respectively), which are small circuit boards with an edge connector across the bottom.
•  PCMCIA cards. This package is used in laptop computers. In this package the memory is built in a card which is inserted in the PCIMCIA slots of laptop computers.
•  Various proprietary vendor-specific chip packages.
•  Access speeds, measured in nanoseconds. Common access times are 60, 70, 80, 90, and 100ns for normal memories, and as low as 8 ns for expensive high-speed memories. Lower numbers are faster.
•  Size, which ranges from 16K to 128MB For memory modules.

Static Versus Dynamic RAM
As computers are built faster and faster, all of the components must get faster and faster. Design a system around a 100 MHz chip, and you need a lot of 100 MHz components including RAM. Unfortunately, about the fastest type of common (that is, cheap) RAM doesn't come any faster than 20 MHz.

Static RAM
The simplest kind of memory to understand is called static RAM (SRAM). It's called "static" RAM because when you put data into it, the data stays there.

To build that kind of memory, you've got to build about six transistors into each bit storage location. That kind of memory-SRAM-can bc quite fast, but also quite expensive. If we used SRAM for our PC memory, then there'd be no trouble with getting memories that kept up with our CPUs. But, on the other hand, we would be able to afford those computers, as SRAM is about 10 times more expensive than the DRAM that we're used to buying.

Dynamic RAM
The economical answer to SRAM was Dynamic RAM (DRAM). Each DRAM built of a single transistor and a capacitor, in contrast to SRAM's six transistor. DRAM has two problems from the point of view of a PC designer. First, the "dynamic" in its name means that it forgets the data that you give it almost as you can give it the data. That means that DRAM-based systems require refresh circuitry to get around this problem. Second, the way that DRAM is built to be cheaper is that DRAM is organized not simply into a set of addresses; rather, each bit in a DRAM has a row address and a column address.

SDRAM
Synchronous Dynamic Random Access Memory, (SDRAM) is a new variant of DRAM that includes an on-chip burst counter. This burst counter can be used to increment column addresses and helps increase SDRAM the speed of burst accesses.

Aside from the facts that faster is always better and speed is pursued for its own sake, the reason behind the SDRAM is that CPUs are getting faster. With the increasing of speed, they demand faster memory, in order to function at Its maximum potential. With SDRAM, the CPU and RAM are locked together same clock. Thus, the speed of the RAM and the CPU are linked, or synchronized.

Tweet