Saturday, March 20, 2010

What is RAMDAC ?

 RAMDAC (random access memory digital-to-analog converter) is a microchip that convertsdigital image information in to the analog data needed by a computer display that makes use of analog inputs such as CRTdisplays. A RAMDAC microchip is built in to the video adapter in a computer. It combines a tiny static RAM (SRAM) containing a color table with digital-to-analog
converters that alter digital picture information in to analog signals that are sent to the display's color generators, for each primary color - red, green, & blue. Depending on the number of bits used & the RAMDAC-data-transfer rate, the converter will be able to support different computer-display refresh rates. With CRT displays, it is best to work over 75 Hz & seldom under 60 Hz, in order to minimize flicker and with LCD displays, flicker is not a controversy. In a cathode ray tube (CRT) display, an analog signal is sent to each of electron guns. With displays using other technologies, the signals are sent to a corresponding mechanism.

How RAMDAC Works ?
The SRAM part of the RAMDAC contains a color palette table. A logical color number in the digital information input to SRAM is used to generate separate values obtained from the table - for each of red, green, & blue - that are output to of digital-to-analog converters. The analog signal output from the converter is input directly to the display electron guns or other picture projecting mechanisms.
For displays with true color, the digital color information is fed directly to the DACs, bypassing the SRAM table, which is not needed.

Wednesday, March 17, 2010

FeRAM ( Ferroelectric RAM )

Ferroelectric capacitors is a molecular dynamics simulator for bulk and thin-film ferroelectrics and relaxors possess the characteristics necessary for a nonvolatile memory cell, that is they have stable states corresponding to the binary levels in a digital memory, and they retain their states without electrical power. Although similar characteristics exist in ferromagnetic cores of a core memory, ferroelectric capacitors switch faster and they may even be fabricated on a single VLSI chip.


Each cell in the core array can be selected by simultaneous X and Y current pulses. The cells that are receiving (or none) of these current pulses stay unselected. The cell that is receiving both X and Y current pulses flips to its opposite magnetization state or stay unchanged, depending on its preliminary magnetization state and the current pulse directions. A similar situation exists in the array of FE capacitors. In fact, the above statements can be applied to the capacitor array if the terms current and magnetization are replaced by their counterparts voltage and polarization, respectively.


The capacitor array forms the simplest and the densest cell structure among ferroelectric nonvolatile memory designs. However, more complex structures are necessary due to nonideal characteristics of ferroelectric capacitors. of these characteristics is the dearth of a one-of-a-kind switching voltage. Partial switching can occur within a range of applied voltage. If this range is wide , the cells that are in the same row (or column) as the selected cell can be disturbed by X (or Y) voltage pulses. The effects of these partial switchings in an unselected cell can add up and damage the stored information.


In Physics : 

  • Molecular dynamics (MD) simulation with a first-principles-based model Hamiltonian
  • Ferroelectrics
  • Relaxor
  • Phase transition
  • Narrow film
  • Electrodes
  • Capacitor
  • Hysteresis loop
  • Nose-Poincare thermostat
  • Optical phonon
  • Evaluate forces in reciprocal-space; k-locality of the force matrix
  • Coarse-graining; reduction of the number of degree of freedom

Tuesday, March 16, 2010

Window RAM (WRAM)

Window RAM (WRAM), unrelated to Microsoft Windows, is high-performance video RAM that is dual-ported & has about 25% more bandwidth than VRAM but costs less. This type of RAM developed by Samsung Electronics that supports two ports. This allows a video adapter to fetch the contents of memory for display simultaneously that new bytes are being pumped in to memory. This ends in much faster display than is feasible with conventional single-port RAM.

  • It's features that make it more efficient to read knowledge for use in block fills & text drawing.
  • It can be used for high resolution (such as 1600 by 1200 pixels) using true color. It is used in the Matrox Millenium video card.

Multibank Dynamic RAM (MDRAM)

Multibank Dynamic RAM (MDRAM) is a high-performance RAM, developed by MoSys, that divides memory into multiple 32 kilobytes parts or "banks" that can be accessed individually. Traditional video RAM is monolithic; the entire frame buffer is accessed at one time. Having individual memory banks allows accesses to be interleaved concurrently, increasing overall performance. It's also cheaper since, unlike other forms of video RAM, cards can be manufactured with just the right amount of RAM for a given resolution capability instead of requiring it to be in multiples of megabytes.

MRAM (Magnetoresistive Random Access Memory)

        MRAM (Magnetoresistive Random Access Memory) is a system of storing knowledge bits using magnetic charges in lieu of the electrical charges used by DRAM (dynamic random access memory). Scientists define a metal as magnetoresistive if it shows a slight alter in electrical resistance when placed in a magnetic field. By combining the high speed of static RAM & the high density of DRAM, proponents say MRAM could be used to significantly improve electronic products by storing greater amounts of knowledge, enabling it to be accessed faster while consuming less battery power than existing electronic memory.
        MRAM, however, retains knowledge after a power supply is cut off. Replacing DRAM with MRAM could prevent knowledge loss & enable computers that start instantly, without waiting for program to boot up.
        Unlike conventional RAM chip technologies, in MRAM information is not stored as electric charge or current flows, but by magnetic storage elements. The elements are formed from two ferromagnetic plates, each of which can hold a magnetic field, separated by a narrow insulating layer. of the plates is a permanent magnet set to a specific polarity, the other's field can be changed to match that of an outside field to store memory. This configuration is called a spin valve and is the simplest structure for a MRAM bit. A memory device is built from a grid of such "cells".
        The simplest system of reading is accomplished by measuring the electrical resistance of the cell. A specific cell is (usually) chosen by powering an associated transistor which switches current from a supply line through the cell to ground. Due to the magnetic tunnel effect, the electrical resistance of the cell changes due to the orientation of the fields in the plates. By measuring the resulting current, the resistance inside any particular cell can be determined, & from this the polarity of the writable plate. Usually if the plates have the same polarity this is thought about to mean "1", while if the plates are of opposite polarity the resistance will be higher & this means "0".

Sunday, March 14, 2010

Things to remember before decide to buy RAM

Capacity
To ram does the more capacity the better performance, with a minimum was 512MB for various needs, for gaming you need 1GB of RAM for medium performance, but 2/3 GB for high one.
While the graphic design needs 2-4GB for a optimal performance. 
If you have money left, it is recommended to buy another component rather than upgrading RAM, trust me, it won't be any difference for using 1GB with 2GB RAM for a daily use.


Brand
Kingston enough for normal use is compatible with various motherboard, for performance it is not recommended to take local brand because the chipset would be easily damaged.


Gaming RAM from cheap to expensive one:
- Deam 2Gb PC 10600 / 1333
- Memory DDR3 V-GEN 2Gb PC 10600 > many people say that the performance and durability is good
- Musclepower 2GB DDR3 Stiletto
- PATRIOT PSD2 4G 800KH (2 X 2G) Module Kit Signature Line KH Series (Premium Heatsink)
- Memory DDR3 TEAM Elite 2Gb*2 TED340961333HC9DC with the Elite Heat Spreader, so it wont be a matter if -      you use it for more than 10 hours
- Gskill DDR3 PC10600 4GB NT Dual Channel F3-10666CL9D-4GBNT


Chipset
The best chipset for 128-512GB ram chip is tccd, windbond, infeon, micron, and Hynix, or maybe elpida, samsung, qimondawhile for 1GB UCCC, Micron db-5, infeon CE6 
To select a ram, highly recommended, prefer to choose the chipset rather than brand.



Top-Ranked RAM Manufacturers :


  • Infineon
  • Hynix
  • Samsung
  • Micron
  • Rambus
  • Corsair




Note: if you want to check the price, visit rakitan.com for details



What is CAS Latency in RAM?

CL stands for CAS Latency. It is a programmable register in a SDRAM which sets a series of time cycles in between a distribution of a READ authority as well as when a interpretation comes out. Smaller series for CL indicates faster SDRAM inside of a same frequency.
There is some case that your computer and on the RAM said that it had latency of 8-8-8-24. When I looked in the bios and on cpuz it said that it was walking at 9-9-9-24. Regrettably, this is because you make the error of buying RAM at 1600mhz when my motherboard will only support upto 1333mhz. It still works but runs at 1333mhz.
When you open the memory tab from cpu-z, there will be some number. You can read it from top to down :

667 --> your memory is walking on 1333MHz

7
7
7
20 --> The latency is 7-7-7-20

Friday, March 12, 2010

RAM's Heat Spreader

As memory gets faster, heat becomes more of an issue because DRAM transistors generate heat when they switch. The faster they switch, the more heat is produced, because as transistors get warmer, they get slower, meaning that performance can be sometimes less-than-optimal if you have pushed a memory module far with improper cooling.

Here are the basic materials involved:
Heat spreaders
Rubbing alcohol
Cotton swabs (Q-Tips)
For this editorial I will be using solid copper generic heat spreaders. I'm using the thermal tape that came with the heat spreaders.



Steps :

1. There isn't much in preparing for this. It's a simple process to put in the heat spreaders and you don't need much in the way of tools.
2. In case, you need to remove all the stickers from the module
3. Neat it with rubbing alcochol until all of the residue is gone. This will make positive optimal heat transfer, but will take much longer. DO NOT use a citrus remover such as goo-gone, as the citrus can destroy yourDIMMs.
* In case you use any kind of silver compound, make absolutely positive that you aren't connecting any of the legs on the DIMM chips. Usually, stick with the tape that is already installed in the heat spreader.
1
2
3

Thursday, March 11, 2010

Video RAM (VRAM)

Video Ram is a dual-port memory with one random access port and one serial port access .


The standard DRAM used for video cards usually does not have bandwidth to handle the demands of running a card at high resolution & color depths, with acceptable refresh rates. The main reason why is the competing access factors for the video memory: the processor writing new knowledge to the memory, & the RAMDAC reading it lots of times per second in order to send video signals to the monitor.


To address this essential limitation, a new type of memory was created called video RAM or VRAM. As the name implies, this memory is specifically tailored for use in video systems. The essential difference between VRAM & standard DRAM is that VRAM is dual-ported. This means that it's access paths, & can be written to & read from simultaneously. The advantages of this are of coursework sizable given what the video card does: lots of times per second a new screen picture is calculated & written to the memory, & lots of times per second this memory is read & sent to the monitor. Dual-porting allows these operations to occur without bumping in to each other.
VRAM provides substantially more bandwidth than either standard DRAM or EDO DRAM; double in lots of cases. It is more suited for use in systems requiring high resolution & color depth displays. The only reason that it has not replaced standard DRAM entirely is of coursework: cost. VRAM is more complex & requires more silicon per bit than standard DRAM, which makes it cost more.


Note: Don't confuse VRAM with the generic term "video RAM" or "video memory", which refer to the memory in the video subsystem in general.

Dual Ported - RAM

The difference between single port RAM & dual port RAM is that single port RAM can be accessed at address at time, thus you can read/write memory cell in the coursework of each clock cycle. Dual port RAM has ability to simultaneously read & write different memory cells at different addresses.
SPRAM makes use of a 6 transistor basic ram cell, while the dual port ram cell makes use of 8 transistor cell for memory.

So the area of SPRAM is much smaller than the area of DP RAM cell. To gain the advantages of both spram (less area) & DPRAM ( high speed) Pseudo dual port ram is introduced. which can read & write the information in the same clock, using rising & falling edges for the operations respectively, & using spram memory cell for the storage of information.

DP RAM as I felt it was a great way for one
different controllers to access the same information
independently, which would also reduce my coding lines to make a
simpler method.

Both clocks in to the RAM are from the same source, so I get a
warning that there is no phase difference between the rising edges of
the clocks (apparently it needs 0.1ns).
This is the picture of dual ported-RAM that I pick from google :

Wednesday, March 10, 2010

Common problems that often occur in RAM

Why does a mistake occurs after I install new RAM?
You can look at the manual booknya. For a quantity of the BIOS: In case you are installing (eg) 2-4 Mb 2-8 Mb Simms and Simms, you ought to put bigger RAM in bank.


After adding RAM, why RAM is detected still the same?
After you install the RAM, you ought to re-do the BIOS Setup. The trick, the first time you turn on the computer until the BIOS logo appears, press Del (for some computers may Ctrl + Alt + Del or other key combination). When completed, select Save Settings and Exit. 


Why do posts appear Parity Check Error?
Writing a parity check error may occur if there is destroy to of the RAM, of the DMA not working, or because there is parity boot virus on your hard disk.

After I install new RAM, why frequent Blue Screen?
On computers that use the bus 66 MHz (Pentium 133 or higher), it takes the RAM with a speed of 60 ns or higher. You can only use RAM with a speed that is less than 60 ns but you need to change settings in the BIOS Wait State. Change the setting from 0 to 1.


Why does Windows detect the RAM is less than the actual thing? 

1. Phsically damaged. To overcome this you must check the physical condition of the RAM. Open your computer case, then take off the RAM. Neat if dirty RAM & socket. Also check if there is rust. You ought to also make positive that the RAM of the same material (golden or silver). If different, you ought to replace the RAM.
2. In some BIOS, ROM shadowing & 16 bit drivers, sometimes wearing some (read: tiny) RAM before Windows starts. There's also some BIOS ROM shadowing feature that will copy the contents of the technique & video BIOS ROM in to RAM in order to improve computer performance (On DOS & Windows three.1). This is because the speed is lower ROM than RAM. But in the event you are using Windows 95/98/NT with minimal memory ROM shadowing feature ought to be turned off (from the CMOS setup), because Windows 95/98/NT never access the ROM so that RAM can be used by Windows.
3. The use of the VGA card on board. If for example you have 32 Mb RAM & three MB VGA card, the RAM can be used as much as 30 Mb (32 Mb - three Mb). So you ought to not use the VGA card on board.
4. There's drivers or programs that run from config.sys or autoexec.bat that works using the RAM. The solution: Edit the autoexec.bat & config.sys files using Notepad. Then edit the line that "suspicious" or are roughly used to load the program. When you see you add the word REM (but in the event you add the word REM then the program won't di-load/dijalankan). When you wear a 4x CD-ROM driver from TEAC, change the parameters xmssize = -1 in the autoexec.bat to xmssize = 0.
5. Virtual device driver walking from technique.ini. To fix this open the technique by clicking Start - Run - type SysEdit. Select the Technique.ini. Then edit the line that roughly used to load virtual device driver. Another alternative is to generate a new Technique.ini file.
6. There's lines in the Technique.ini file Maxphyspage refer to the Technique.ini file, whether or not there is a line Maxphyspage. If so, delete that line.
7. Errors in the CMOS settings. On some computers there's features of the Memory Hole at 16 Mb or Hold to 15 Mb of RAM. If the feature is enabled so in the event you have a bigger RAM of 15 Mb or 16 Mb, Windows will still detect as much as 15 Mb or 16 Mb. To overcome this, disable these features.
8. Load Ramdrive.sys from config.sys Check your Config.sys file for a line containing "Ramdrive.sys." In the event you have this line, it means you are using a RAM drive. To disable the RAM drive, remove or disable the line in the Config.sys file contains That "ramdrive.sys." To disable the line, use a text editor (Such as Notepad) to edit the Config.sys file & place a semicolon (;) at the beginning of the line. Open the config.sys file with Notepad. Check on the file, whether there is a line ramdrive.sys. If so, delete the line

Tuesday, March 9, 2010

How to calculate transfer data rate of memory & software

In calculating time information transfer is to make use of a RAM unit of nanosecond (ns), or also called as the time necessary by the RAM to transmit tiny bit of information to the processor.
. As an example they will calculate the time of transfer of DDR3 RAM which in this example they use DDR RAM PC 12 800 means a bus of 1600 Mhz.
. Next we'll convert the first unit of Hertz. Where Mhz =1.000.000 Hertz, which means that DDR3 RAM with a bus at 1600 MHz =1.600.000.000 Hertz. So it can be summed up as 1600 Mhz = 1.600.000.000
. Next is to convert the units second to nanosecond (ns). second the same as1.000.000.000 ns (nanosecond). They must keep in mind again that billion seconds equal to nanosecond.
. Then they multiply the numbers: 1/1.600.000.000 x1.000.000.000 = 0625 ns. So RAM DDR3 PC 12800 has the time information transfer as much = 0625 nanosecond

Next is they will calculate the transfer rate of DDR3 RAM. DDR3 memory has a transfer speed times that of DDR2 RAM. Transfer rate is the capacity of information that may be sent a RAM to a processor in units of Megabytes / second (MB / s).
For example:

DDR3 RAM
. A 12 800 DDR3 RAM PC that has a memory clock rate of 200 Mhz.
. For DDR3 memory we'll use the following formula = transfer rate of (memory clock rate) x 4 (bus clock multiplier) x 2 (for information rate)  64 (the number of bits transferred) / 8 (number of bits / bytes).
. Then they enter the numbers stay perhitungnya be = (200 x 4 x 2 x 64) / 8.
. So the result will be equal to 12,800, meaning a DDR3 RAM with a memory clock of 200 MHz giving maximum transfer rate of 12 800 MB / s.
. With the Dual Channel know-how is the transfer rate of 12 800 MB / s would be multiplied by, & produces 25 600 MB / s

DDR2 RAM
. A PC 6400 DDR2 RAM which has a clock rate of 800 Mhz
. Width information (width) of a RAM is 64-bit, or converted in to bytes, equal to 8 bytes. That is byte = 8 bits.
. Transfer Rate = Bus (MHz) x Width Information (Byte)
. Transfer Rate = 800 MHz x 8 Bytes = 6400 MB / s. means a DDR2 RAM with 800 MHz memory clock give maximum transfer rate of 6400 MB / s.
. With the Dual Channel know-how is the transfer rate of 6400 MB / s would be multiplied by, & produces 12 800 MB / s
Through the example above they can conclude that it's a speed of DDR3 RAM 2x transfer information faster than DDR2 RAM.

PC report the memory modules from a RAM.dimana each RAM has a maximum transfer rate (multiplied by 8 from its actual clock)
nah to find the module / "PC" (multiplied by 16 from each actual clocknya). after it is multiplied by the result was the reply will be made in memory modules from a RAM.namun plenty of results that have rounding in order to facilitate our numbers to call.

Excess DDR3 RAM
Higher bandwidth (up to 1600 MHz)
Improved performance on less power.
On a laptop computer computer, battery will be more long lasting.
Operational memritambahan to improve performance, efficiency & timing margins
Allows some high density, low-voltage module choice for servers, desktops, laptops & applications.
Lack of DDR3 RAM
DDR3 memory modules are not backward compatible to DDR2 based motherboards
Cost yg expensive than DDR2 RAM

example of "PC" that they use the generation of DDR memory (DDR1, DDR2 & DDR3):
pc-1600, pc-2100, pc-2700, pc-3200, pc-4200, pc-5300, pc-6400, pc-8500, PC3-10600 (DDR3), PC3-12800 (DDR3)

Example:
DDR 200: that is, actual clock 100 MHz x 16 = 1600 MB / s called PC-1600
DDR 266: that is, actual clock 133 MHz x 16 = 2133 MB / s called PC-2100 (Rounding)
DDR 333: that is, actual clock 166 MHz x 16 = 2666 MB / s called PC-2700 (Rounding)
DDR2 400: that is, actual clock 200 MHz x 16 = 3200 MB / s called PC2-3200
DDR3 800: that is, actual clock 400 MHz x 16 = 6400 MB / s called PC3-6400.
This procedure for calculating MEMORY MODULE "PC". If you want to know the maximum speed, just multipy the clock speed with 8.

Where your RAM should be placed in mainboard?

This is where your RAM should be placed.

Note:
- At the contact point, which consists of hundreds of points, separated by a special indentation. Often known as a notch. Its main function, to prevent installation errors on the type of memory modules available DIMM slots on the motherboard.

Monday, March 8, 2010

Different RAM Types and its uses

EDO RAM

EDO RAM, which stands for Extended Information Out RAM came out in 1995 as a used type of memory available for Pentium based systems. EDO is a modified type of FPM RAM which is often often called Hyper Page Mode\. Extended Information Out refers to fact that the information output drivers on the memory module are not switched off when the memory controller removes the column address to start the next cycle, unlike FPM RAM. Most early Penitum based systems use EDO.

FPM RAM

FPM RAM (Quick Page Mode RAM) is a kind of Dynamic RAM (DRAM). The term of Quick Page Mode comes from the capability of memory being able to access information that is on the same page & can be done with less latency. Most 486 & Pentium based systems from 1995 & earlier use FPM Memory.

SDRAM (Synchronous DRAM)

All systems used to ship with two.3 volt, 168-pin SDRAM DIMMs. SDRAM is not an extension of older EDO DRAM but a brand spanking new type of DRAM altogether. SDRAM started out jogging at 66 MHz, while older quick page mode DRAM & EDO max out at 50 MHz. SDRAM can scale to 133 MHz (PC133) officially, & unofficially up to 180MHz or higher. As processors get faster, new generations of memory such as DDR & RDRAM are necessary to get proper performance.

DDR (Double Data Rate SDRAM)

DDR fundamentally doubles the rate of knowledge transfer of standard SDRAM by transferring data on the up & down tick of a clock cycle. DDR memory operating at 333MHz actually operates at 166MHz * five (aka PC333 / PC2700) or 133MHz*2 (PC266 / PC2100). DDR is a five.5 volt know-how that makes use of 184 pins in its DIMMs. It is incompatible with SDRAM physically, but makes use of a similar parallel bus, making it simpler to implement than RDRAM, which is a different technology.

Rambus DRAM (RDRAM)

Despite it is higher cost, Intel has given RDRAM it's blessing for the consumer market, & it will be the sole choice of memory for Intel's Pentium two. RDRAM is a serial memory know-how that arrived in flavors, PC600, PC700, & PC800. PC800 RDRAM has double the maximum throughput of elderly PC100 SDRAM, but a higher latency. RDRAM designs with multiple channels, such as those in Pentium two motherboards, are currently at the top of the heap in memory throughput, when paired with PC1066 RDRAM memory.

DIMMs vs. RIMMs

DRAM comes in major form factors: DIMMs & RIMMS.

DIMMs are 64-bit parts, but if used in a motherboard with a dual-channel configuration (like with an Nvidia nForce chipset) you must pair them to get maximum performance. So far there are not lots of DDR chipset that use dual-channels. Usually, in case you need to add 512 MB of DIMM memory to your machine, you pop in a 512 MB DIMM if you have got an obtainable slot. DIMMs for SDRAM & DDR are different, & not physically compatible. SDRAM DIMMs have 168-pins & run at 3.3 volts, while DDR DIMMs have 184-pins & run at 2.5 volts.

RIMMs use only a 16-bit interface but run at higher speeds than DDR. To get maximum performance, Intel RDRAM chipsets need the use of RIMMs in pairs over a dual-channel 32-bit interface. You need to plan more when upgrading & purchasing RDRAM.

Memory Modules

Types of Memory Modules :
1. S I M M 
Short for Single In-Line Memory Module, that module or memory chip embedded on side of the circuit PCB. The memory of this type only have a total of leg (pin) as lots of as 30 and 72.
SIMM 30 pins in the type of FPM DRAM, widely used in processor-based systems 386 and 486 generations of the finish of the first generation. 30 pins SIM card capacity 1MB, 4MB and 16MB.
Meanwhile, 70-pin SIMMs can be either FPM DRAM and EDO DRAM shared with the late-generation processor 486 and Pentium. 70 pin SIMM is produced on the capacity of 4MB, 8MB, 16MB, 32MB, 64MB and 128MB.

2. D I M M 
Short for Dual In-Line Memory Module, that module or memory chip embedded on both sides of the PCB. DIMM memory is produced in three different forms, namely the amount of feet 168 and 184.
DIMM 168-pin can be either Rapid Page, EDO and ECC SDRAM. While the type of DIM 184 pin DDR SDRAM.

3. SODIMM 
Short for Small Outline Dual In-Line Memory Module. Memory is fundamentally similar to DIMMs, but the difference is in their use. If the DIMM is used on a PC, then SO DIMM used on laptop computer / laptop.
SODIMM produced in types, first type has a total of 72 pin, and only 144 numbered pieces

4. RIMM / SORIMM 
RIMM and SORIMM is a kind of memory that is made by Rambus. RIMM is fundamentally the same with similar SORIMM DIMM and SODIMM.
Because the use of Rambus know-how, which prioritizes famous speeds, this memory is so hot so rapid Rambus parties will need to add aluminum to help remove the heat generated by this memory.

Sunday, March 7, 2010

Chip Packaging

Chip Packaging
Or in the Indonesian language is the chip packaging. Is the outer layer of the physical building blocks of each memory chip. Most often used, in the TSOP DDR memory modules (Narrow Tiny Outline Package). On RDRAM & DDR2 using CSP (Chip Scale Package). A quantity of the chip to memory modules prior to make use of DIP (Dual In-Line Package) & SOJ (Tiny Outline J-lead).

DIP (Dual In-Line Package)
This type of memory chip used when memory is installed directly on the motherboard PCB. DIP included in the section of through-hole parts, which can be mounted on the PCB through holes are obtainable for the legs / pins. Type of DRAM chips can be soldered or attached with the socket. SOJ (Tiny Outline J-Lead) Chip DRAM SOJ type, called demikan because it's a pin formed like huruh AJA. SOJ included in surfacemount part, meaning that this part is installed on the side surface on the PCB.

TSOP (Narrow Tiny Outline Package)
Included in surfacemount part. His name is in accordance with the shape & physical size are much thinner & smaller than a SOJ.

CSP (Chip Scale Package)
If the DIP, SOJ & TSOP using the legs / pins to connect it to the board, CSP is no longer using a PIN. Connections using a BGA (Ball Grid Array) contained at the bottom of the part. DRAM chip parts are put in to use in the RDRAM (Rambus DRAM) & DDR.

Pictures of RAM

This is the example picture of RAM from the old one to the new one.



From Top To Down:
1.DIP
2.SIPP,
3.SIMM 30 pin
4.SIMM 72 pin
5.DIMM, 
6.DDR DIMM.

Terms of RAM

What Does The terms of RAM?

So many names and specific terms used in RAM. Sometimes it can be confusing. But it does not matter, after you read the following brief explanation. It can be used as a guide, at least to read the specifications and take into account the ability of the product concerned.

Speed
Speed ​​or velocity, is increasingly becoming an important factor in the selection of a memory module. CPU speed increases, coupled with the development of the use of dual-core processors, making the RAM should have a faster ability to serve the CPU.
There are several important parameters which will affect the speed of a memory.

Megahertz
The use of this term, beginning in the era of the triumph of SDRAM. Memory speeds, ranging expressed in megahertz (MHz). And still used, even to the DDR2.
Calculation based on the time interval (period) is needed between each clock cycle. Usually in the order of nanosecond time. As an example in the memory with the actual clock speed of 133 MHz, would require 8ns access time for one clock cycle.
Then shifted by the presence of DDR SDRAM (Double Information Rate). With a major development in the ability to send information over twice as much. DDR send information twice in one clock cycle.
Most products begin using effective clock speed, the result of multiplying two times the information that is sent. These are more accurately referred to as DDR Rating.
The same thing happened to DDR2. Is the result of the development of the DDR. With the main advantages of low voltage power supply that reduces heat during operation. DDR2 memory chip capacity also increased dramatically, enabling a DDR2 chip has a capacity of up to three GB. DDR2 also increased the speed compared to DDR.

PC Rating
In DDR modules, often found in terms such as PC3200. For the DDR2 module, PC2-3200. Where did this figure come from?
Commonly known as PC Rating for DDR and DDR2 modules. As this example is a DDR module with a clock speed of 200 MHz. Or for DDR rating called DDR400. With 64-bit bus width, then the information that can be transferred is 25 600 megabits per second (= 400 MHz x 64-bit). With one byte = 8 bits, then rounded to two.200MBps (Mebabyte per second). Throughput figure is used as the value of the PC rating. Additional figures ³ A2A, both on the PC Rating Rating maupu DDR, only to tell the difference between DDR and DDR2.

CAS Latency
Acronym derived from the abbreviation CAS column address strobe or column address select. Meaning they are the same, namely the specific location of an array of information on DRAM modules.
CAS Latency, or also commonly abbreviated CL, is the amount of time required (in units of clock cycles) during the delay time between the information request sent to the memory controller for the read, until the memory module successfully remove the information output. The lower the CL specification owned a RAM module, with the same clock speed, will result in faster memory access.


Peak Bandwith

Here you see the maximal peak bandwidth of the widely known RAM types. The figures illustrates the absolutely maximal transfer from RAM to the L2-cache - in peaks, not as continuously transferred.
RAM typeMax. peak bandwidth
FPM176 MB/sec
EDO264 MB/sec
SD528 MB/sec



RAM Parts

KNOW YOUR PARTS RAM
Physically, this one PC components including components with small size and simple. Compared to other PC components.
At first glance, it is only in the form of a small piece of PCB, with several additional components of black. With additional points of contact points, to interact with the motherboard memory. This is in between.

1. PCB (Printed Circuit Board)
In general, green colored PCB board. On the PCB is some component of memory chips installed.
PCB itself is composed of several layers (layer). At each layer is attached path or circuit, to supply power and information. In theory, the more the number of layers used on the memory PCB, will increasingly cross-sectional area available in designing the route. This allows the distance between point and line width can be set more freely, and avoid noise interference antarjalur on the PCB. And overall will make the memory modules are more stable and faster performance. That is why in some ads for memory products, emphasized the number of layers on the PCB used memory module products.

2. Contact Point
Often also called the finger contacts, edge connectors, or leads. When the memory module is inserted into the memory slot on the motherboard, this is the part that connects between the motherboard information to and from memory modules. This connector is usually made of copper or gold. Gold has a better conductivity value. But the consequences, with a more expensive price. Better choice of memory module connectors adapted to the materials used on the motherboard memory slot. Two different metals, coupled with electrical current flow allows the PC to work more corrosive reaction.
At the contact point, which consists of hundreds of points, separated by a special indentation. Commonly referred to as a notch. Its main function, to prevent installation errors on the type of memory modules available DIMM slots on the motherboard. For example, DDR module has a notch is 73 mm from one end of the PCB (the front). While DDR2 has a notch at a distance of 71 mm from the tip of the PCB. For SDRAM, more easily distinguished, with the three notch on the contact point.
DRAM (Dynamic Random Access Memory)
The components shaped black boxes mounted on a PCB module is called DRAM memory. Called dynamic, because it only holds information in short periods of time and must be refreshed periodically. While the types and forms of DRAM or memory chip itself is quite diverse.

Dual Channel Technology

Dual Channel RAM is a motherboard feature that allows increased RAM bandwidth becomes wider.With Dual Channel, then the available bandwidth will be doubled, compared with the installation of a single-chip RAM (doubling the value of architecture).The use of Dual Channel RAM, requires two pieces are identical,and it is being applied to the platformDDRPC3200,PC4200&PC5300. Formerly known as the Dual Channel RDRAM.With dual channel,then the bandwidth that can be provided by DDRPC3200 at 6.4Gb/s, whereas if using a single channel,which is only 3.2Gb/sonly.
The Difference
Dual Channel is a technology on the motherboard chipset, not the RAM.Each of DDR in use can be configured in Dual Channel mode or Single.In the past, DualChannel configuration, requires two pieces of RAM that are identical (both in terms of type,capacity,FSB, to thetypeIC). But at present, there isa know-how called Intel Flex Memory Technology at Intel motherboard platform that allows two types of RAM that are not identical (type should remain identical) to be installed in Dual Channel configuration.

Saturday, March 6, 2010

History of RAM / Memory

INTRODUCTION..
       The development of micro computers, or more often referred to as a PC (Personal Computer) that such a rapid course not separated from the human need for information to be processed by the PC and of course the development of technology, especially technology hardware, software, as well as functions or algorithms used in processing information that is processed.
        Twenty years of PC celebrations which falls in August 2001 , that if we look today we are at a time when the PC has become an inseparable part of our lives. If the initial discovery, the PC is still regarded as expensive goods, are now almost everyone already have them. That said, people who are not familiar with computers will be branded as a person who stutters technology.
         If at that time the Intel 8088 processor could only walk with three speeds, 77 MHz which is used to drive word-processing program in creating and editing documents, simple spreadsheets to do the job and business accounting, and simple database program and some educational programs and games are also still very simple. Now, PC can run with Intel Pentium4 2GHz speeds, even this new new â Intel Corp. through the arena its Intel Developer Forum, Intel has been showing a demo of processor speed two, 5GHz! A leap of technological invention is quite fantastic.
         However, the development of the ability of the PC is not always determined by the development of the processor alone. The other factors, such as technology, chipset, memory, VGA card, the media store, and so forth. All devices with each other grow, evolve into a better direction for build a PC system.
         For that, through this paper, the author tries to give a little information about the evolution of the memory device on a PC. But before stepping on the subject matter, it must be stressed before discussing the scope of this paper. Evolution of memory that the author discussed in this paper covers only the main memory (main memory) type of RAM (Random Access Memory) used in micro-computer (PC).
         The rapid development of processor capability must be balanced with the increased memory capabilities. As a container for information / information needed by the processor as well as a container for the results of calculations performed by the processor, memory capabilities in managing information is vital. Useless to a PC system with high-speed processor, if not matched by a commensurate memory capabilities.
Inaccurate fusion processor with the memory capacity can lead to inefficiencies for both. Say we have a processor that can process information flow of 100 instructions per second, while we have the memory with the ability to distribute information to the processor by 50 instructions per second. Then what happened? The system will experience a bottleneck. Processor must wait for information from memory. The instructions should be done within one second to three seconds because of limited memory capacity.


COMPUTER SCIENCE (History of RAM / Memory)
1. R A M RAM stands for Random Access Memory, was discovered by Robert Dennard and produced on a large scale â by Intel in 1968, long before the PC is found by IBM in 1981. From here is the development of RAM starts. At the beginning of creation, the RAM requires two.0 volt voltage to be able to walk on two frequencies, 77MHz, with a memory access time (access time) of about 200ns (1ns = 10-9 seconds).
2. D R A M In 1970, IBM created a memory called DRAM. Own DRAM stands for Dynamic Random Access Memory. Dynamic Named because this type of memory at any given time interval, always update the information or content validity. DRAM has a working frequency range, between the two, 77MHz to 40MHz. 
3. FP RAM Rapid Page Mode DRAM, or abbreviated with FPM DRAM is found around the year 1987. Since it was first launched, this type of memory directly dominate the memory market, and people often call this type "memory DRAM" , without naming the FPM. This type of memory works just like an index or table of contents. Meaning Page itself is part of the memory contained on a row address. When the system requires the contents of a memory address, FPM stay retrieve information about them based on indices that have been owned. FPM allows faster transfer of information on line (row) of the same from the previous memory types. FPM works in the frequency range 16MHz to 66MHz with an access time of 50ns. In addition, FPM is able to process information transfer (bandwidth) of 188.71 Mega Bytes (MB) per second. FPM memory is widely used in systems based on Intel 286, 386 and bit 486. 

4. EDO RAM In 1995, were created memory type Extended Information Output Dynamic Random Access Memory (EDO DRAM), which is a refinement of the FPM. EDO memory read cycle can shorten it so as to improve its performance around twenty percent. EDO has an access time that is varied enough, which is about 70ns to 50ns and working at a frequency of 33MHz to 75MHz. Although EDO is a refinement of the FPM, but both could not be installed simultaneously, because of differences in ability. EDO DRAM memory used in systems based on Intel 486 and Pentium and earlier generations.




5. PC66 SDRAM At the turn of 1996 to 1997, Kingston creates a memory module which can work on the speed (frequency) the same bus / sync with a working frequency of the processor.That's why Kingston's call this type of memory as a Synchronous Dynamic Random Access Memory (SDRAM). SDRAM is then better known as the PC66 because working at 66MHz bus frequency. Unlike previous memory types that require a fairly high working voltage, voltage SDRAM only requires registration for two, 3 volts and has an access time of 10ns. With the best ability at the time and has been mass produced, not only by Kingston alone, PC66 memory quickly became the standard time memory. Processor-based systems like Intel Pentium Socket 7 Classic (P75 and P266MMX) and kompatibelnya from AMD, WinChip, IDT, and so can work very fast by using these PC66 memory.Even Intel Celeron II early generation still uses PC66 SDRAM memory system.

6. PC100 SDRAM Lapse of a period of one year after PC66 mass produced and used, Intel made a new standard type of memory that is an extension of PC66 memory. The new standard was created by Intel to balance the system chipset i440BX with one system slot is also created by Intel. This chipset is designed to work on the bus frequency at 100MHz. This chipset was developed by Intel at the same time to be paired with the latest Intel Pentium II processor that works on a 100MHz bus. Because the bus system operates at a frequency of 100MHz while Intel still wants to use SDRAM memory systems, SDRAM memory method is developed that can work at 100MHz bus frequency. Like its predecessor PC66, SDRAM memory is then known as PC100. By using the working voltage of two, 3 volt, PC100 memory has an access time of 8ns, shorter than PC66. In addition, PC100 memory capable of flowing information of 800MB per second. Almost identical to its predecessor, PC100 memory has brought a change in computer systems. Not only is it one slot-based processor that uses PC100 memory, socket-based system 7 was updated to be able to use PC100 memory. Then came the so-called system Tremendous Socket 7. Examples of processors that use socket Super7 is AMD K6-2, Intel Pentium II generation of late, and early-generation Intel Pentium II Intel Celeron II and the early generations.
7. DR DRAM In 1999, Rambus memory system by creating a new architecture and a revolutionary, totally different from SDRAM.Oleh Rambus memory architecture, memory is called Direct Rambus Dynamic Random Access Memory. With only the voltage of three, 5 volts, which works on RDRAM 800MHz system bus via a bus system called the Direct Rambus Channel, capable of flowing information at one, 6GB per second! (1GB = 1000MHz). Unfortunately the sophistication DRDRAM not be utilized by the system chipset and processor at that time so the memory is less received support from various parties. One more thing that makes this memory is less desirable because the price is very expensive. 


8. PC800 RDRAM Still in the same year, Rambus has also developed an other memory types with the same ability with DRDRAM. The difference only lies in the necessary working voltage.If DRDRAM require voltages of three, 5 volts, the voltage RDRAM PC800 work on two, 3 volt. The fate of RDRAM memory is almost the same as DRDRAM, less desirable, if not utilized by Intel. Intel has managed to create a very high-speed processor requires a memory system that is able to keep up and work together well. SDRAM memory type was not worth it anymore. Intel needs more than that. With dipasangkannya Intel Pentium4, RDRAM name skyrocketed, and the longer the price is getting down.


9. PC133 SDRAM In addition to the development of PC800 RDRAM memory in 1999, SDRAM memory has not been abandoned, even by the Vikings, even more enhanced capabilities. As the name implies, PC133 SDRAM memory works at 133MHz bus frequency with an access time of 7.5 ns and is able to stream information for one, 06GB per second.Although developed to work on PC133 bus frequency of 133MHz, but memory is also capable of running at 100MHz bus frequency, although not as good as the ability possessed by PC100 at that frequency.





10. PC150 SDRAM SDRAM memory development is increasing greatly after Mushkin, in 2000 succeeded in developing memory chips capable of working at 150MHz bus frequency, although in fact there has been no official standards regarding the frequency of the bus system or chipset for this. Still with the working voltage of two, 3 volt, PC150 memory has an access time of 7ns and be able to stream information for one, 28GB per second. Memory is intentionally created for purposes of overclockers, but the user gaming applications and two-dimensional graphics, desktop publishing, and computer servers to take advantage of the presence of PC150 memory.



11. DDR SDRAM Still in 2000, successfully developed the SDRAM ability memory to be doubled. If the regular SDRAM can only run once every single instruction clock cycle frequency of the bus, the DDR SDRAM is able to run two instruction in the same time.The technique used is to use the full single frequency wave. If the regular SDRAM only do the instructions on the positive wave, the DDR SDRAM running good instructions on the positive wave and negative wave. Because of this memory is called DDR SDRAM which stands for Double Information Rate Synchronous Dynamic Random Access Memory. With DDR SDRAM memory, system bus with a frequency of 100 and 133 MHz will work effectively at a frequency of 200 and 266 MHz. DDR SDRAM was first used in ultra-speed AGP graphics card. While the use of the processor, AMD Thunderbird was the first to use it.

12. DDR RAM In 1999 two large companies Intel and AMD microprocessors competing in the increase in CPU clock speed. But the obstacles, because when upgrading to 133 MHz memory bus needs Memory (RAM) will be greater. And to solve this problem then be made to DDR RAM (double the transfer rate information), which was originally used on a graphics card, because now you can use only 32 MB to 64 MB capacity. AMD was the first company to use DDR RAM on the motherboard. Differences DDR2 with DDR






13. DDR2 RAM When the memory type DDR (Double Information Rate) felt starts to slow down the rapid performance processors and graphics processors, DDR2 memory is the presence of logical advances in memory technology refers to the addition of speed and anticipation of increasingly wide access point triangle processor, memory, and interface graphics (graphic card) that comes with double the computing speed. The main differences between DDR and DDR2 is the speed of information and increased latency to reach two-fold. This change is intended to produce a maximum speed in a computing environment that the faster, both on the processor or graphics. In addition, DDR2 voltage requirement also decreased. If the DDR voltage requirement was recorded three, 5 Volt, the DDR2 this need only reach one, 8 Volt.That is, technological advances in this DDR2 requires less power to write and read in memory. Technology DDR2 himself first used on several graphics interface device, and was finally introduced its use in RAM technology. And DDR2 technology is not compatible with DDR memory, so its use could only be done on a computer that does support DDR2.
14. DDR3 RAM DDR3 RAM has a reduced power demand of about 16% compared to DDR2. This is because DDR3 is already using 90 nm technology so that only the power required konsusmsi one.5v, less when compared to DDR2 and DDR one.8v three.5v. In theory, owned by the RAM speed is indeed quite stunning. He was able to transfer information with an effective clock of 800-1600 MHz. In the 400-800 MHz clock, much higher than DDR2 at 400-1066 MHz (200-533 MHz) and DDR at 200-600 MHz (100-300 MHz). DDR3 prototype of which has 240 pins. It's actually been introduced long ago in early 2005. 
However, the products themselves actually appear in mid 2007 in conjunction with a motherboard that uses Intel's P35 Bearlake chipset and the motherboard supports DIMM slots 

EVOLUTION MODULE In addition to experiencing the development of the capability, memory module processing techniques were also developed. From a simple system, until RIMM SIMM. The following brief explanation. 1. S I M M Short for Single In-Line Memory Module, that module or memory chip embedded on one side of the circuit PCB. The memory of this type only have a total of leg (pin) as many as 30 and 72. SIMM 30 pins in the form of FPM DRAM, widely used in processor-based systems 386 and 486 generations of the end of the first generation. 30 pins SIM card capacity 1MB, 4MB and 16MB. Meanwhile, 70-pin SIMMs can be either FPM DRAM and EDO DRAM shared with the late-generation processor 486 and Pentium. 70 pin SIMM is produced on the capacity of 4MB, 8MB, 16MB, 32MB, 64MB and 128MB. 2. D I M M Short for Dual In-Line Memory Module, that module or memory chip embedded on both sides of the PCB, berbalikan each other. DIMM memory is produced in three different forms, namely the amount of feet 168 and 184. DIMM 168-pin can be either Fast Page, EDO and ECC SDRAM, with capacities ranging from 8MB, 16MB, 32MB, 64MB and 128MB. While the form of DIM 184 pin DDR SDRAM. 3. SODIMM Short for Small Outline Dual In-Line Memory Module. Memory is basically similar to DIMMs, but differ in their use. If the DIMM is used on a PC, then SO DIMM used on a laptop computer / laptop. SODIMM produced in two types, first type has a total of 72 kakai, and only 144 numbered pieces 4. RIMM / SORIMM RIMM and SORIMM is a type of memory that is made by Rambus. RIMM is basically the same with similar SORIMM DIMM and SODIMM. Because the use of Rambus technology, which prioritizes famous speeds, this memory is so hot so fast Rambus parties will need to add aluminum to help remove the heat generated by this memory.