SDRAM Speed and Naming Scheme

SDRAM has had many different versions throughout the years the supported SDRAM Types of eMachines are as follows:

• PC66 = Memory has a maximum speed of 66 MHz
• PC100 = Memory has a maximum speed of 100 MHz
• PC133 = Memory has a maximum speed of 133 MHz

Note: The speed is the MAXIMUM speed of the memory, due to its nature; SDRAM can slow down to operate slower speeds if needed.

Double Data Rate SDRAM (DDR)

DDR SDRAM is an incremental upgrade from Standard SDRAM. DDR uses the same ideas of SDRAM and adds a “Double Edge Clock Cycle”. The Double Edge Clock Cycle adds an extra communication per cycle so information can be going to the processor as data is coming from the processor.

DDR Speed and Naming Scheme

DDR is the most common type of memory in eMachines currently. There are a number of different speeds. The naming scheme states the bandwidth of the chip unlike SDRAM where the naming scheme detailed the speed.

• PC2100 = 2.1 Gigabytes per second @133x2 MHz (AKA DDR266)
• PC2700 = 2.7 Gigabytes per second @166x2 MHz (AKA DDR333)
• PC3200 = 3.2 Gigabytes per second @200x2 MHz (AKA DDR400)
• PC3700 = 3.7 Gigabytes per second @233x2 MHZ (AKA DDR466)

Dual Channel DDR SDRAM

A recent advancement in memory technology is the introduction of Dual Channel DDR. Dual Channel enabled motherboards use two separate memory controllers to access the memory modules. The motherboard takes the information being sent to memory and cuts it in half then sends half the data to one memory module and the other half to the other module. This effectively cuts the time required to send or retrieve a piece of data in half. Dual channel requires two memory chips of equal size and speed to work.

Rambus Dynamic Random Access Memory (RDRAM)

Rambus dynamic random access memory is a radical departure from the previous DRAM architecture. Designed by Rambus, RDRAM uses a Rambus in-line memory module (RIMM), which is similar in size and pin configuration to a standard DIMM. What makes RDRAM so different is its use of a special high-speed data bus called the Rambus channel. RDRAM memory chips work in parallel to achieve a data rate of 800 MHz, or 1,600 MBps. Emachines only shipped one system with Rambus the T4130 released Q3 2001.

Processors

The Central Processing Unit (CPU) is the heart of the computer. The power and speed of the process will determine how the programs on the computer will perform. CPUs are actually very simple in their design. The processor receives data from the system memory. Some of that data is information some of that data is instructions. The processor will perform the basic mathematic instructions it receives then moves the data back into memory to store it. The focus of this module will not be on the technical details of how a processor works. This module will discuss the types of processors available in the market and what the differences are.

AMD vs. Intel

There has been much debate about which high-end processor is the best purchase in today's highly competitive market, both performance-wise and value-wise. There are two current main competitors in this market, AMD and Intel. Some prefer AMD processors because they offer excellent value for money, while others prefer Intel processors, due to their high performance and comparatively low heat output. However from an objective view point, both companies are very similar, it is a matter of personal opinion which company is “BETTER”.

Do Megahertz Matter?

This has been something that has plagued AMD for some time now. Clock for clock, AMD's flagship processor, the Athlon XP outperforms Intel's flagship processor, the Pentium 4, by quite a significant margin. While hardware enthusiasts may realize this, the average consumers likely will not, and end up purchasing a processor just because it boasts a higher megahertz rating. Intel has used this as a marketing tool for quite some time now, however, with the relatively recent release of the Athlon XP processor, AMD has attempted to wipe out this confusion by introducing a marketing based rating system called "Quantispeed".

The Quantispeed rating system is an estimation of real-world performance. It is NOT a comparison to a similarly clocked Intel Pentium 4 processor. For example, the 1800+ Athlon XP does the same amount of work as a Pentium 1.8GHz but actually runs at 1.53GHz.

High End Vs. Low End

Intel and AMD both offer high end processors like the Intel Pentium and the AMD Athlon. They also offer lower end or “Value” like the Intel Celeron and the AMD Durron. The most significant difference between both types (aside from price) is their Level 2 (L2) Cache. L2 Cache is the loading zone for the processor, As information is sent from RAM, it is received in the L2 Cache. More L2 cache means that the CPU can store more Information on the chip and not have to wait for the RAM. Value CPU’s typically have half the L2 Cache as their High end counterparts.

AMD Athlon64

A lot of controversy has surrounded the release of the Athlon 64. The Athlon64 is a 64 bit processor. Currently Microsoft has not released an operating system that is capable of fully utilizing the Athlon64. Currently the biggest advantages of the Athlon64 are that the chip can address 4 million times the amount of memory that the Athlon XP can. Secondly the Athlon64 only uses 1.4Volts, is cooler, and costs less to run.

Setting the Clock

Many computers have achieved speeds around 2.5 to 3.0 Gigahertz. These speeds can be misleading. The front side bus is the pathway that connects a Memory module to the processor. The FSB is typically a fraction of the speed of the processor for example the FSB for a Pentium 4 2GHz (2000MHz) is actually 400Mhz. At first glance this can be very confusing, but if you think of the FSB as a road that has a speed limit of 266 miles per hour, nothing has been said about the number of people allowed in the car. After all, the important factor in your computers speed is the number of people or Data that you can transport. Much like a car, the computer can adjust the number of people in the car by turning up the “Clock Multiplier”.

Intel Pentium 4 2GHz (2000MHz) = 400FSB x 5 Clock Multiplier

Over Clocking

Now you understand what the clock multiplier is, we can talk about over clocking. Over clocking is the practice of forcing a higher value for the clock multiplier. This will make the computer run faster. This is not recommend however because when you turn up the speed on the processor you must also turn up the voltage, which in turn turns up the heat and shortens the life of the processor.

Caution: Over Clocking the processor voids any warranty through eMachines.

Under Clocking

Under clocking is the opposite of over clocking. Under clocking is the practice of lowering the clock multiplier to cause the computer to slow down. This is done for one of two reasons. When you under clock the processor, you can turn down the voltage and lower the temperature of the processor. AMD and Intel both implement this idea for two very different reasons.

Intel Thermal Protection:

Intel chips can sense if the processor is beginning to overheat. If the computer detects a high temperature, the motherboard will under clock the processor in an attempt to cool off the processor.

AMD Power Now!

AMD Power Now is typically used on AMD Notebooks. Power Now watches how hard you are pushing the processor, if you are not pushing the processor, it will under clock to save battery power. Power Now re-evaluates the processor 7 times a second, so the speed loss is not typically noticed by the customer.

CPU Sockets

A CPU Socket is a plastic housing with an array of pin holes. The naming scheme has changed throughout the years. Until the release of the Athlon64, there where two naming schemes. Intel’s naming scheme Pin Grid Array (PGA) the PGA system details the number of pin in the socket. For example the PGA 478 has 478 pins. AMD had a style of socket the “Socket A” for use with all the Athlon XP processors. With the release of the Athlon64, AMD has adopted Intel’s PGA scheme.

Note: AMD and Intel Processors are not compatible with each other, if your system uses an Intel Chip, the only chips that will fit that board will be Intel chips and vise versa.

Zero Insertion Force (ZIF) Sockets

Socket A and PGA Sockets are both considered Zero Insertion Force (ZIF) sockets. A ZIF socket is one in which it has a lever and when the lever is up the CPU will fall out very easily, when the lever is down the chip is locked down.

CPU Heatsinks and CPU Fans

Modern processors can reach temperatures close to 700 degrees Fahrenheit, hot enough to destroy the processor and motherboard. To address this issue, computers will have a heat sink and fan to help cool the processor.

A heat sink is an aluminum block that has fins cut into it. The aluminum heat sink is attached directly to the processor. Aluminum can dissipate heat very quickly. Commonly aided by a small fan (CPU FAN) that forces air across the fins, the heat sink assembly can keep the processor below 170 Fahrenheit.