Major PC makers, such as dell, now offer 64-bit computers. If you are getting ready for a new computer, you may wonder whether a 64-bit machine is for you.

The difference between a 32-bit and a 64-bit computer is the number of zeros or ones (”bits”) it uses to represent a number internally. A 32-bit integer can take any value between 0 and 232-1, or 4,294,967,295 (inclusive). A 64-bit integer can take any value between 0 and 18,446,744,073,709,551,615 (that’s eighteen quintillion, four hundred forty-six quadrillion and change). So a computer with a 32-bit CPU can handle numbers up to about 4 billion, and a 64-bit CPU can handle numbers that are a billion times larger than that. That’s the fundamental difference.

So does that mean you need a 64-bit computer to calculate the size of the Federal debt? Not exactly. A 32-bit CPU can represent numbers over four billion just by smooshing (”concatenating,” as computer people would say it) two 32-bit integers together. But a 32-bit CPU, the main brain of the computer, only works with numbers up to 32 bits. This is most important when the CPU has to retrieve a piece of data from memory. In order to figure out where to get the information it needs, the CPU uses what’s called a memory address, which works like a mailbox number. The CPU fetches data by grabbing whatever is stored in the “mailbox” at a given address. A 64-bit CPU has a lot more memory addresses available.

What this means to you, the intelligent but non-technical user, is that a 64-bit CPU can support more than 4 GB of RAM.

That’s because each byte of memory has its own address, and a 32-bit machine can only interpret addresses between 0 and 4,294,967,295 (4 billion, two hundred ninety-four million and change). Any memory beyond 4GB would not be addressable and hence not usable. Incidentally, this is also the reason older versions of Windows (Windows 95 and 98) only supported hard disk partitions up to 4 GB.

So, you only need a 64-bit machine if you are planning to have more than 4 GB or RAM. Obviously a 64-bit computer is more upgradeable. This flexibility does come at a price: important kinds of software, namely operating systems and device drivers, are not compatible between 32-bit and 64-bit machines. You would need a special version of Windows (or Linux) and all new device drivers in order to use a 64-bit computer. The operating system is not a big deal because most users buy computers with an operating system pre-installed. The device drivers really matter, for two reasons.

First, there’s a good chance any devices you already have, such as printers, scanners, or that really high-end liquid-nitrogen-cooled video card, don’t have 64-bit drivers and can’t be installed on a 64-bit computer. Second, 64-bit computers are quite new on the consumer market, and device manufacturers may not have 64-bit drivers for their products (or they may have drivers that are full of bugs).

There are also compatibility issues with other kinds of software. I have a 64-bit laptop at work, and boy do I wish I didn’t. Several programs I want to use, such as certain Firefox plug-ins, aren’t available for 64-bit machines. Moreover, when I write software on the 64-bit machine, the finished program won’t run on the 32-bit computer across the hall. You do not want these headaches. I would recommend home users stay away from 64-bit computers for at least one more obsolescence cycle (about 3 years).