The memory is DDR (Double Data Rate) is the default, which replaced the traditional memories SDR SDRAM (further referred to as “memories SDRAM” or, still, as “memories DIMM”), being very well received by the market especially in the segment of personal computers. But, what made it so well-accepted? What are the features that the differ from the previous standard? What are the benefits? The answers to these and other related questions you gives the following.
The emergence of the memories DDR
At the time that the Pentium III processor, of Intel, was one of the the main products of the kind in the market, the default rate of the FSB (Front Side Bus- essentially, the speed at which the processor communicates with the RAM was 133 MHz, equivalent the 1.064 MB per second. However, it is known that, generally, the the chipset of the motherboard does not use the frequency of the FSB to communicate with the memory, but the speed of this last. On this occasion, the standard for speed of memories also was of 133 MHz (the well-known memories SDRAM PC-133), which also provided a transfer rate of 1.064 MB per second. It is possible to notice that there was a certain “balance” in the communication speeds between the components of the computer.
However, with the release of chips such as the Pentium 4, the Intel and the Athlon from AMD, this “balance” left exist, because the FSB of the processors started to have more speed, as the memories continued to be in default PC-133, maintaining the frequency of 133 MHz. These conditions, this means that the computer as a whole can’t take advantage of all the processing resources.
For users of the Pentium 4 until there was an alternative: use the memories of the type Rambus (or RDRAM). This type it was faster that the memories PC-133, but had some drawbacks: it worked only with processors Intel had a very high price and the motherboards that supported the memories Rambus also they were very expensive.
In this same period, the memories DDR were already reality, but Intel tried to popularize the memories Rambus, which made it “ignore” the existence of the first. The AMD, in turn, needed an efficient alternative that could to work fully with their new processors. The company just betting in the memories DDR, and from hence, this type became popular, especially because Intel, soon after, had to adhere to the idea.
But the simple appearance of the memories, the DDR was not a immediate solution to the problems of speed between memory and FSB. Only with the release of memories Dual-Channel DDR is that the solution has effectively become effective. The issue that will be discussed later.
The operation of the memories DDR
The memories of the DDR are very similar memories the SDR SDRAM. The latter work synchronously with the processor, avoiding the delay problems existing in previous technologies. The great the spread of the technology in the DDR, however, is in its ability to perform twice as many operations per clock cycle (in a few words, the speed with which the processor prompts for operations – understand more in this article about processors). Thus, while a memory SDR SDRAM PC-100 works 100 MHz, for example, a module DDR with the same frequency causes this corresponds to the double, that is, the 200 MHz.
But, how is this possible? In the memories, the data are stored in spaces called cells. These are arranged in a kind of matrix, that is, they are oriented in a scheme that recalls rows and columns. The intersection of a row with a column forms what we know as memory address.
Usually, in the operations of reading and recording, is only cannot access a row at a time. But the memories of DDR have a “trick”: they access two different positions, but both on the same line. That is why this technology is unable to perform the fold of operations per cycle, one at the beginning this and another at the end.
Because of this characteristic, the memories DDR now have a different default naming. In the modules SDR SDRAM, are expressions as PC-100 and PC-133, where the number indicates the frequency. Thus, a comb PC-133 states that the device works at 133 MHz. In the memories DDR, this also occurs, but considering the characteristic of duplicity by cycle. Thus, a module DDR-200, for example, works in the true, at the rate of 100 MHz. But, in the nomenclature an alternative, such as PC-1600, for example, the amount of megabytes transferred per second is that it is considered to be.Note the table:
Worth mentioning that these values of transfer are theoretical, that is, they indicate the maximum range. In practice, a number of factors can influence the speed of transfer. But, even being theoretical, how is this calculation done?
Is simple: in its operations, the memories DDR can transfer up to 64 bits per time, that is, 8 bytes. Just then multiply this value by the frequency of the more memory the amount of operations cycle-by-cycle. Thus, the calculation of a module DDR-400 is the following:
8 (64-bit) x 200 (frequency) x 2 (operations per cycle) = 3.200
The final result is given in megabytes per second.
Although very similar to the kinds of memories the SDR SDRAM, the memories DDR have another differential considerable: work with 2.5 V, against the 3.3 V of the first. Therefore, reduce energy consumption, an aspect especially important in portable devices, such as notebooks.
The physical aspects of the memories DDR
Visually, it is easy to distinguish the memories DDR of memories the SDR SDRAM. The first have only a division in the fitting of the module, between the terminals contact, while the second have two. In addition also, the memories DDR use 184 terminals, against 168 pins of the standard SDR SDRAM.
In relation to the tunnel (learn more about this in the article Memories ROM and RAM), the chips DDR generally use the standard TSOP (Thin Small Outline Package), but it is also possible to find versions in CSP (Chip Scale Package), although more rare.
One can consider the Dual-Channel as a solution that alleviates the fact that memories do not keep the speed of the processors. For this, the scheme makes the memories DDR transfer twice as much data at a time. So, to 3,200 MB per second can be make of 6,400 MB per second.
This is possible because in the chipset motherboard- or even within the processors, in the case of some models that are more current – there is a special circuit called memory controller, which is responsible for all aspects access and use of this. In Dual-Channel, this driver causes the memories DDR can transfer twice as much data at a time, that is, instead of 64-bit, transfer 128 bits (16 bytes). With this, the calculation of the previous topic shall be:
16 (128 bytes) x frequency x 2 (operations per cycle)
To enable the scheme to Dual-Channel in a computer, it is must have a chipset that is compatible (or, if it is the case, a processor). In addition, it is recommended to have one or two pairs of modules memory is identical (or, at least, with the same specifications). The equality decreases the risk of problems. At this point, an interesting tip is to acquire a kit for Dual-Channel, which provides two memory sticks DDR themselves to work in this mode.
Refer to the manual of the motherboard to know in which slots the modules must be installed to activate the mode Dual-Channel, so as to know if it is need to change some parameter in the BIOS setup.
The memories DDR had great acceptance in the market, however, as the evolution of technology not to, especially with regard to the processors, new standards had to be released to keep up with the speeds of the chips more recent: these are memories DDR2 and DDR3.