Diagnosing problems with stubborn PCI cards

Brien Posey reveals basic steps that you can take to correct most PCI problems.

Have you ever bought a nifty new PCI-based device, only to have difficulty installing it? Although Windows’ Plug and Play code has improved over the years, it’s still not perfect. Combining this fact with the possibility that the problem could be hardware-related means that some PCI problems can be a real mess to sort out. Fortunately, you can take some basic steps to correct most PCI problems.

What is PCI?
PCI is the bus architecture that was released at the same time that the original Pentium CPU came out, even though it appeared on some 486 motherboards, as well. The purpose of PCI was to overcome troublesome hardware conflicts that occurred with the older ISA architecture. A PC that has PCI slots reserves certain system resources, such as IRQs and memory addresses, for the device that’s plugged into a given PCI slot.

However, on many systems, the PCI slots also support Plug and Play. The system is free to assign any unused resources to the card, rather than being limited to using the resources that are reserved for the card’s slot.

Why are there PCI problems?
If the computer is smart enough to assign unused or reserved resources to a PCI card, you may wonder why configuration problems exist at all. There are several reasons for this fact. For example, the card may be preset at the factory to use resources that are already occupied. Another possible cause is that an ISA card (an older type of expansion card) is holding a resource and not reporting it to the system. Finally, Windows could be misidentifying the card.

Will Windows detect the card?
After you install a new PCI card, Windows should recognize that a new device has been added to the system. It should prompt you to supply a driver for the card. If this prompting doesn’t happen, it’s the first sign of trouble.

If Windows doesn’t detect the new card, open Control Panel and double-click the System icon. When you see the System Properties sheet, select the Device Manager tab. You can look for the device under Device Manager. It may appear in the correct section for the device’s type. It also may appear under Other Devices. If you find the device listed under Device Manager, delete it and reboot Windows. Now, Windows should detect the device.

If Windows still doesn’t detect the device, it’s a sign of deeper problems. To avoid confusing Windows, shut down the PC and remove the card. Once the card has been removed physically from the computer, allow Windows to boot up. When the boot sequence has completed, wait for a few minutes; then, turn the computer off.

At this point, I recommend that you check the CMOS setup on the PC. Many PCs have a setting under the CMOS editor that allows you to select Yes or No for the Plug And Play Operating System option. If you’re still having problems, set this option to No. Occasionally, having the Plug And Play Operating System option set to Yes can cause some bad values to be assigned to the card, especially on cheaper PCs.

Once you’ve set the Plug And Play Operating System option to No, allow the PC to boot. Then, shut down the PC and insert the card. Now, when you boot, Windows should identify the card.

Driver problems
Perhaps Windows has been identifying the card all along, but the card just doesn’t work correctly. For example, suppose you’re trying to install a network card. Windows may identify the card, but even after supplying the drivers, you may not be able to attach to the network.

If this happens, the first thing that you should do is download the latest driver from the Internet and install it. If that still doesn’t correct the problem, the next thing you should do is start reading more carefully.

When you provide the driver disk, Windows will identify the driver and give you a choice of installing that driver or another driver. You should look carefully at this window because Windows often lists a driver file other than the one that you’ve specified. For example, suppose that you point Windows to a driver file that’s located on drive A. Windows will read the driver from drive A, but it may try to install a driver that’s located on drive C because Windows slightly misidentifies the hardware that you’re installing. It sees the filename and date of the driver file that you’ve provided and realizes that it has a file with an identical name and a newer date. Therefore, it tries to use the file that comes with Windows.

To get around such a problem, select the option to use another driver. Now, Windows will list the driver from the disk that you’ve provided.

Brien M. Posey is an MCSE who works as a freelance technical writer and as a network engineer for the Department of Defense. If you’d like to contact Brien, send him an e-mail . (Because of the large volume of e-mail he receives, it's impossible for him to respond to every message. However, he does read them all.)

The authors and editors have taken care in preparation of the content contained herein, but make no expressed or implied warranty of any kind and assume no responsibility for errors or omissions. No liability is assumed for any damages. Always have a verified backup before making any changes.