In many fields, the term granularity refers generally to the size of the components that make up a system. The term can refer to any system and has been applied to almost any branch of science, engineering, and manufacturing. Home computers have reached the pinnacle of granularity with owners able to upgrade and reconfigure their computers unlike any other time in home computing history.
Granularity is a useful tool in describing any system. For example, take the government system of the United States. As a system, granularity at the Federal level is coarse because there is only one component. Add the state governments and the granularity becomes finer. It becomes even finer when the county level is considered and furthermore at the city/local level. Even a system of measurement can be granular. A mile broken down into inches is finer or more granular than that same mile broken down into feet.
In computer programming, it is useful to describe the code in terms of granularity. Long strings of code are coarsely granulated but the same code broken down into smaller units or modules is finer. However, granularity comes with a price. As a system becomes more granular, the complexity of the system increases. As the complexity increases, efficiency is often lost to friction between and among the smaller parts making up the system.
Today’s computers are finely granulated. This is evidenced in the multitude of manufacturers making motherboards, video cards, sound cards, CPUs, RAM, hard drives, optical drives, etc. Granularity of computers is highly dependent on the standards set forth by the most influential companies in the industry. For example, the PATA (or IDE) standard for connecting devices such as hard drives and DVD-ROMs to a computer has been in place since the earliest home computers.
Early home computers such as the Commodore 64 and the 8-bit series of ATARI computers were coarsely granulated. Few upgrades and reconfigurations were possible with these systems resulting in longer innovation times for new technologies to enter the market. Later, the IBM PC standard greatly enhanced the granularity of computers and paved the way for such companies as Intel and Microsoft to dominate the hardware and software industries.
One problem with finer and finer granularity is that standards take over and stagnate advancements. The PATA connectivity standard has been argued to be an anachronism in modern computing. Only recently has the SATA standard replaced the old standard. Manufacturers are less likely to undertake the high costs and risks of designing and making new products that may not be profitable. They would rather stick to tried and true profit centers during a boom in computer popularity as was seen in the 1990s, than risk losing market share from a mistake in forecasting the popularity of a new standard. Consequently, standards are hard to break.
Another problem with fine granularity in home computers is lower competition which usually means higher prices for consumers. As standards are set in place by market conditions, competition among hardware manufacturers becomes fierce as technological advancement stagnates. Manufacturers must compete on cost rather than product features, eventually leading to takeovers, swallowing up of smaller manufacturers, and as is often the case, a two-horse race among competitors. Examples of this phenomenon include nVidia versus ATI, Intel versus AMD, and Microsoft versus Apple.
Although finer granularity increases the customizability of home computers, this option comes at a price of slower technological advancement and fewer competitors in the home computer hardware and software markets. The result is higher prices for consumers and an uncertain future for computer innovations. The question is: what balance of fine granularity and proprietary hardware and software makes the industry profitable at fair prices to the consumer?