Microprocessor Types and Specifications

Processor Manufacturing

Processors are manufactured primarily from silicon, the second-most common element on the planet (only the element oxygen is more common). Silicon is the primary ingredient in beach sand; however, in that form it isn't pure enough to be used in chips.

The manner in which silicon is formed into chips is a lengthy process that starts by growing pure silicon crystals via what is called the Czochralski method (named after the inventor of the process). In this method, electric arc furnaces transform the raw materials (primarily quartz rock which is mined) into metallurgical-grade silicon. Then to further weed out impurities the silicon is converted to a liquid, distilled, and then redeposited in the form of semiconductor-grade rods, which are 99.999999 percent pure. These rods are then mechanically broken up into chunks and packed into quartz crucibles, which are loaded into the electric crystal pulling ovens. There the silicon chunks are melted at over 2,500° Fahrenheit. To prevent impurities, the ovens are normally mounted on very thick concrete cubes often on a suspension to prevent any vibration which would damage the crystal as it forms.

Once the silicon is melted, a small seed crystal is inserted into the molten silicon, and slowly rotated (see Figure 3.3). As the seed is pulled out of the molten silicon, some of the silicon sticks to the seed and hardens in the same crystal structure as the seed. By carefully controlling the pulling speed (10 to 40 millimeters per hour) and temperature (approximately 2,500° F) the crystal grows with a narrow neck that then widens into the full desired diameter. Depending on the chips being made, each ingot is approximately eight or 12 inches in diameter and over five feet long, weighing hundreds of pounds.

Figure 3.3 Growing a pure silicon ingot in a high-pressure, high-temperature oven.

The ingot is then ground into a perfect 200mm- (eight-inch) or 300mm-diameter cylinder, with normally a flat cut on one side for positioning accuracy and handling. Each ingot is then cut with a high-precision diamond saw into over a thousand circular wafers, each less than a millimeter thick (see Figure 3.4). Each wafer is then polished to a mirror-smooth surface.

Figure 3.4 Slicing a silicon ingot into wafers with a diamond saw.

Chips are manufactured from the wafers using a process called photolithography. Through this photographic process, transistors and circuit and signal pathways are created in semiconductors by depositing different layers of various materials on the chip, one after the other. Where two specific circuits intersect, a transistor or switch can be formed.

The photolithographic process starts when an insulating layer of silicon dioxide is grown on the wafer through a vapor deposition process. Then a coating of photoresist material is applied and an image of that layer of the chip is projected through a mask onto the now light-sensitive surface.

Doping is the term used to describe chemical impurities added to silicon (which is naturally a non-conductor), creating a material with semiconductor properties. The projector uses a specially created mask, which is essentially a negative of that layer of the chip etched in chrome on a quartz plate. The Pentium III currently uses twenty or more masks to create six layers of metal and semiconductor interconnects.

As the light passes through a mask, the light is focused on the wafer surface, imprinting it with the image of that layer of the chip. Each individual chip image is called a die. A device called a stepper then moves the wafer over a little bit and the same mask is used to imprint another chip die immediately next to the previous one. After the entire wafer is imprinted with chips, a caustic solution washes away the areas where the light struck the photoresist, leaving the mask imprints of the individual chip vias (interconnections between layers) and circuit pathways. Then, another layer of semiconductor material is deposited on the wafer with more photoresist on top, and the next mask is used to produce the next layer of circuitry. Using this method, the layers and components of each chip are built one on top of the other, until the chips are completed.

The final masks add the metallization layers, which are the metal interconnects used to tie all the individual transistors and other components together. Most chips use aluminum interconnects today, although many will be moving to copper in the future. The first commercial PC chip using copper is the Athlon made in AMD's Dresden fab. Copper is a better conductor than aluminum and will allow smaller interconnects with less resistance, meaning smaller and faster chips can be made. The reason copper hasn't been used up until recently is that there were difficult corrosion problems to overcome during the manufacturing process that were not as much a problem with aluminum. As these problems have been solved, there will be more and more chips fabricated with copper interconnects.

A completed circular wafer will have as many chips imprinted on it as can possibly fit. Because each chip is normally square or rectangular, there are some unused portions at the edges of the wafer, but every attempt is made to use every square millimeter of surface.

The standard wafer size used in the industry today is 200mm in diameter, or just under eight inches. This results in a wafer of about 31,416 square millimeters. The Pentium II 300MHz processor, for example, was made up of 7.5 million transistors using a 0.35 micron (millionth of a meter) process. This process results in a die of exactly 14.2mm on each side, which is 202 square millimeters of area. This means that about 150 total Pentium II 300MHz chips on the .35 micron process could be made from a single 200mm-diameter wafer.

The trend in the industry is to go to both larger wafers and a smaller chip die process. Process refers to the size of the individual circuits and transistors on the chip. For example, the Pentium II 333MHz through 450MHz processors were made on a newer and smaller .25 micron process, which reduced the total chip die size to only 10.2mm on each side, or a total chip area of 104 square millimeters. On the same 200mm (8-inch) wafer as before, Intel can make about 300 Pentium II chips using this process, or double the amount over the larger .35 micron process 300MHz version.

The Pentium III in the 600MHz and faster speeds is built on a .18 micron process and has a die size of only 104 square millimeters, which is about 10.2mm on each side. This is the same size as the older Pentium II, even though the newer PIII has 28.1 million transistors (including the on-die L2 cache) compared to only 7.5 million for the Pentium II.

In the future, processes will move from .18 micron to .13 micron, and from 200mm (eight-inch) wafers to 300mm (12-inch) wafers. The larger 300mm wafers alone will allow for more than double the number of chips to be made, compared to the 200mm mostly used today. The smaller 0.13-micron process will allow more transistors to be incorporated into the die while maintaining a reasonable die size allowing for sufficient yield. This means the trend for incorporating L2 cache within the die will continue, and transistor counts will rise up to 200 million per chip or more in the future. The current king of transistors is the Intel Pentium III Xeon introduced in May 2000 with 2MB of on-die cache and a whopping 140 million transistors in a single die.

The trend in wafers is to move from the current 200mm (eight-inch) diameter to a bigger, 300mm (12-inch) diameter wafer. This will increase surface area dramatically over the smaller 200mm design and boost chip production to about 675 chips per wafer. Intel and other manufacturers expect to have 300mm wafer production in place during 2001. After that happens, chip prices should continue to drop dramatically as supply increases.

Note that not all the chips on each wafer will be good, especially as a new production line starts. As the manufacturing process for a given chip or production line is perfected, more and more of the chips will be good. The ratio of good to bad chips on a wafer is called the yield. Yields well under 50 percent are common when a new chip starts production; however, by the end of a given chip's life, the yields are normally in the 90 percent range. Most chip manufacturers guard their yield figures and are very secretive about them because knowledge of yield problems can give their competitors an edge. A low yield causes problems both in the cost per chip and in delivery delays to their customers. If a company has specific knowledge of competitors' improving yields, it can set prices or schedule production to get higher market share at a critical point. For example, AMD was plagued by low-yield problems during 1997 and 1998, which cost it significant market share. It has since solved the problems, and lately it seems Intel has had the harder time meeting production demands.

After a wafer is complete, a special fixture tests each of the chips on the wafer and marks the bad ones to be separated out later. The chips are then cut from the wafer using either a high-powered laser or diamond saw.

After being cut from the wafers, the individual die are then retested, packaged, and retested again. The packaging process is also referred to as bonding, because the die is placed into a chip housing where a special machine bonds fine gold wires between the die and the pins on the chip. The package is the container for the chip die, and it essentially seals it from the environment.

After the chips are bonded and packaged, final testing is done to determine both proper function and rated speed. Different chips in the same batch will often run at different speeds. Special test fixtures run each chip at different pressures, temperatures, and speeds, looking for the point at which the chip stops working. At this point, the maximum successful speed is noted and the final chips are sorted into bins with those that tested at a similar speed. For example, the Pentium III 750, 866, and 1000 are all exactly the same chip made using the same die. They were sorted at the end of the manufacturing cycle by speed.

One interesting thing about this is that as a manufacturer gains more experience and perfects a particular chip assembly line, the yield of the higher speed versions goes way up. This means that out of a wafer of 150 total chips, perhaps more than 100 of them check out at 1000MHz, while only a few won't run at that speed. The paradox is that Intel often sells a lot more of the lower-priced 933 and 866MHz chips, so it will just dip into the bin of 1000MHz processors and label them as 933 or 866 chips and sell them that way. People began discovering that many of the lower-rated chips would actually run at speeds much higher than they were rated, and the business of overclocking was born. Overclocking describes the operation of a chip at a speed higher than it was rated for. In many cases, people have successfully accomplished this because, in essence, they had a higher-speed processor already—it was marked with a lower rating only because it was sold as the slower version.

An interesting problem then arose: Unscrupulous vendors began taking slower chips and remarking them and reselling them as if they were faster. Often the price between the same chip at different speed grades can be substantial, in the hundreds of dollars, so by changing a few numbers on the chip the potential profits can be huge. Because most of the Intel and AMD processors are produced with a generous safety margin—that is, they will normally run well past their rated speed—the remarked chips would seem to work fine in most cases. Of course, in many cases they wouldn't work fine, and the system would end up crashing or locking up periodically.

At first the remarked chips were just a case of rubbing off the original numbers and restamping with new official-looking numbers. These were normally easy to detect. Remarkers then resorted to manufacturing completely new processor housings, especially for the plastic-encased Slot 1 and Slot A processors from Intel and AMD. Although it may seem to be a huge bother to make a custom plastic case and swap it with the existing case, since the profits can be huge, criminals find it very lucrative. This type of remarking is a form of organized crime and isn't just some kid in his basement with sandpaper and a rubber stamp.

Intel and AMD have seen fit to put a stop to some of the remarking by building overclock protection in the form of a multiplier lock into most of its newer chips. This is usually done in the bonding or cartridge manufacturing process, where the chips are intentionally altered so they won't run at any speeds higher than they are rated. Normally this involves changing the bus frequency (BF) pins on the chip, which control the internal multipliers the chip uses. Even so, enterprising individuals have found ways to run their motherboards at bus speeds higher than normal, so even though the chip won't allow a higher multiplier, you can still run it at a speed higher than it was designed.

I recently installed a 200MHz Pentium processor in a system that is supposed to run at a 3x multiplier based off a 66MHz motherboard speed. I tried changing the multiplier to 3.5x but the chip refused to go any faster; in fact, it ran at the same or lower speed than before. This is a sure sign of overclock protection inside, which is to say that the chip won't support any higher level of multiplier than it was designed for. Today, all Intel Pentium II and III processors are multiplier locked, which means the multiplier can no longer be controlled by the motherboard. This means that overclocking can be accomplished only by running the motherboard at a higher bus speed than the processor was designed for. My motherboard at the time included a jumper setting for an unauthorized speed of 75MHz, which when multiplied by 3x resulted in an actual processor speed of 225MHz. This worked like a charm, and the system is now running fast and clean. Many new motherboards have BIOS or jumper settings which can be used to tweak the motherboard bus speeds a few MHz higher than normal, which is then internally multiplied by the processor to even higher speeds. Note that I am not necessarily recommending overclocking for everybody; in fact, I normally don't recommend it at all for any important systems. If you have a system you want to fool around with, it is interesting to try. Like my cars, I always seem to want to hotrod my computers.

The real problem with the overclock protection as implemented by Intel and AMD is that the professional counterfeiter can still override it by inserting some custom circuitry underneath the plastic case enclosing the processor. This again is particularly a problem with the slot-based processors, since they use a case cover that can hide this circuitry. Socketed processors are much more immune to these remarking attempts. To protect yourself from purchasing a fraudulent chip, verify the specification numbers and serial numbers with Intel and AMD before you purchase. Also beware where you buy your hardware. Purchasing over online auction sites can be extremely dangerous since it is so easy to defraud the purchaser. Also the traveling computer show/flea market arenas can be a hotbed of this type of activity.

Fraudulent computer components are not limited to processors; I have seen fake memory (SIMMs/DIMMs), fake mice, fake video cards, fake cache memory, counterfeit operating systems and applications, and even fake motherboards. The hardware that is faked normally works, but is of inferior quality to the type it is purporting to be. For example, one of the most highly counterfeited pieces of hardware is the Microsoft mouse. They sell for $35 wholesale and yet I can purchase cheap mice from overseas manufacturers for as little as $2.32 each. It didn't take somebody long to realize that if they made the $2 mouse look like a $35 Microsoft mouse, they could sell it for $20 and people would think they were getting a genuine article for a bargain, while the thieves run off with a substantial profit.