วันอาทิตย์ที่ 2 พฤศจิกายน พ.ศ. 2551
Expansion Cards Part 1: (of a 3 part series)
PCI The expansion slots available on motherboards allow for a variety of upgrades in a computer system, but matching the appropriate card to an available slot needs to be addressed before making any purchasing decisions. The most common types of expansion cards for modern computer systems can be broken down into three formats: PCI, AGP, and PCI Express. Each of these formats will be addressed separately in this three part series of Tech Tips, starting with PCI. The letters "PCI" stand for Peripheral Component Interconnect, and is the term used to describe a bus that connects components directly to the system's memory and to the system's processor through the "frontside bus." When discussing communications on a motherboard, the term "bus" has nothing to do with the big yellow thing that takes the kids to school. There may be several buses in a computer, and like the PCI bus, they are all responsible for managing the communication "traffic" from different devices to the processor. The frontside bus is a high speed connection that manages the processor's communication with items such as hard drives, memory, and PCI devices, while not burdening the processor with all of the management responsibilities. First developed by Intel in the early 1990s, PCI was spawned from even earlier (and slower) bus architectures such as ISA (Industry Standard Architecture) and VL-Bus (VESA Local), which were common back in the 1980s and 1990s. The original specifications for the PCI Bus had a speed of 33 MHz, with a 32-bit bus width, and a maximum bandwidth of 132 MB per second. There have been a few revisions to the PCI standard which have significantly increased these specifications, taking it to 66 MHz, 64-bit, and 512 MB per second, respectively. The 32-bit and 64-bit versions have different physical features, and most motherboards only offer 32-bit connections. The original power specification had PCI devices operating on 5V DC, and with the revisions came the capability for devices to continue using 5V, as well as now being able to operate on 3.3V DC. A simple explanation of 32-bit and 64-bit can be had by continuing the analogy of buses and traffic. Think of each bit as a lane of traffic on the communication path. Think of a 32-bit bus as having 32 lanes of traffic, and a 64-bit bus having 64 lanes of traffic. Just as a greater number of cars can travel simultaneously on a road with more lanes, more data can be transferred on a bus with a larger bit count. Motherboards can support multiple slots sharing one PCI Bus, and although not particularly common, can include more than one PCI bus. Depending on the form factor size of the motherboard, and other features that may be taking up space on the board, one can expect to have one to six PCI slots on a typical motherboard. For example, the mATX format features just two 32-bit PCI slots, while the ATX format features six 32-bit PCI slots. A 32-bit PCI card features 124 pins for mating with a slot on a system's motherboard, and will fit into either a 32-bit or 64-bit slot (although data transfer will be 32-bit in either type of slot). A 64-bit PCI card features 184 pins for mating with the appropriate slot on a system's motherboard, but can generally fit into a 32-bit slot as well, as long as features on the motherboard do not interfere. When installed in a 32-bit slot, data transfer on a 64-bit card will be limited to 32-bit. The Intel STL2 Dual Socket 370 Server Board w/VRM The Intel STL2 Dual Socket 370 Server Board is a good reference for comparing 32-bit and 64-bit PCI slots. Looking at the lower left corner of the motherboard shows four 32-bit PCI slots and two 64-bit PCI slots. Subsequent installments in this series of Tech Tips will look at AGP and PCI Express, each of which has its own unique physical features. Although the different format PCI cards may be interchangeable, PCI, AGP, and PCI Express cards do not work (or fit) in any other type of slot. Most PCI cards will be of the 32-bit variety, and the selection of items available is fairly extensive. Graphics cards, sound cards, network cards, RAID controllers, TV tuners, modems, and USB/Firewire controllers are all common items that may be added to a system through the use of a PCI card. Many of the items listed in the previous paragraph can be found integrated on modern motherboards, but these onboard devices offer no upgrade ability. PCI devices provide plug and play installation, allowing a user to install (or remove) a device with ease. For example, an inexpensive 2-channel sound card may be good enough for someone initially, but down the road they may decide that something like the 7.1 channel Sound Blaster Audigy 2 offers the sound quality they really want. Upgrading is a matter of powering down the system, swapping the cards, rebooting, and installing the new software/drivers (OK, perhaps a bit over simplified). The good thing about PCI cards is that, even if you do have a board with built-in feature (such as built-in sound mentioned above), your motherboard's BIOS will usually lets you disable that feature if you did want to add an upgraded card (such as the Audigy sound card mentioned in the example above), or the card can complement the feature already built-in (such as an IDE RAID card). The one area that drove the development of AGP is the performance of PCI based graphics cards. The demands of fast-paced video games, and other graphically intensive applications, require a great deal of bandwidth, which just wasn't available on the PCI Bus. Considering that all of the devices on the PCI Bus share the bandwidth available, an even faster, dedicated bus was required to handle just the graphics data. PCI graphics cards are still available though, and make for an easy way to add a second display to a system currently operating on an AGP or PCI Express graphics card. Final Words The PCI slot has been around for a while, and seems to have a place in at least the near future of computer architecture. AGP and PCI Express offer performance benefits that the PCI standard cannot match, but for many applications, the performance offered by PCI is more than adequate. Be sure to check out the next Tech Tips in this series for the basics of AGP. <a target="_new" href="http://www.geeks.com">Computer</a> <a target="_new" href="http://www.geeks.com/pix/techtips.htm">tech tips</a> and <a target="_new" href="http://www.geeks.com/pix/techtips.htm">computer help</a>
Expansion Cards Part 1: (of a 3 part series)
PCI The expansion slots available on motherboards allow for a variety of upgrades in a computer system, but matching the appropriate card to an available slot needs to be addressed before making any purchasing decisions. The most common types of expansion cards for modern computer systems can be broken down into three formats: PCI, AGP, and PCI Express. Each of these formats will be addressed separately in this three part series of Tech Tips, starting with PCI. The letters "PCI" stand for Peripheral Component Interconnect, and is the term used to describe a bus that connects components directly to the system's memory and to the system's processor through the "frontside bus." When discussing communications on a motherboard, the term "bus" has nothing to do with the big yellow thing that takes the kids to school. There may be several buses in a computer, and like the PCI bus, they are all responsible for managing the communication "traffic" from different devices to the processor. The frontside bus is a high speed connection that manages the processor's communication with items such as hard drives, memory, and PCI devices, while not burdening the processor with all of the management responsibilities. First developed by Intel in the early 1990s, PCI was spawned from even earlier (and slower) bus architectures such as ISA (Industry Standard Architecture) and VL-Bus (VESA Local), which were common back in the 1980s and 1990s. The original specifications for the PCI Bus had a speed of 33 MHz, with a 32-bit bus width, and a maximum bandwidth of 132 MB per second. There have been a few revisions to the PCI standard which have significantly increased these specifications, taking it to 66 MHz, 64-bit, and 512 MB per second, respectively. The 32-bit and 64-bit versions have different physical features, and most motherboards only offer 32-bit connections. The original power specification had PCI devices operating on 5V DC, and with the revisions came the capability for devices to continue using 5V, as well as now being able to operate on 3.3V DC. A simple explanation of 32-bit and 64-bit can be had by continuing the analogy of buses and traffic. Think of each bit as a lane of traffic on the communication path. Think of a 32-bit bus as having 32 lanes of traffic, and a 64-bit bus having 64 lanes of traffic. Just as a greater number of cars can travel simultaneously on a road with more lanes, more data can be transferred on a bus with a larger bit count. Motherboards can support multiple slots sharing one PCI Bus, and although not particularly common, can include more than one PCI bus. Depending on the form factor size of the motherboard, and other features that may be taking up space on the board, one can expect to have one to six PCI slots on a typical motherboard. For example, the mATX format features just two 32-bit PCI slots, while the ATX format features six 32-bit PCI slots. A 32-bit PCI card features 124 pins for mating with a slot on a system's motherboard, and will fit into either a 32-bit or 64-bit slot (although data transfer will be 32-bit in either type of slot). A 64-bit PCI card features 184 pins for mating with the appropriate slot on a system's motherboard, but can generally fit into a 32-bit slot as well, as long as features on the motherboard do not interfere. When installed in a 32-bit slot, data transfer on a 64-bit card will be limited to 32-bit. The Intel STL2 Dual Socket 370 Server Board w/VRM The Intel STL2 Dual Socket 370 Server Board is a good reference for comparing 32-bit and 64-bit PCI slots. Looking at the lower left corner of the motherboard shows four 32-bit PCI slots and two 64-bit PCI slots. Subsequent installments in this series of Tech Tips will look at AGP and PCI Express, each of which has its own unique physical features. Although the different format PCI cards may be interchangeable, PCI, AGP, and PCI Express cards do not work (or fit) in any other type of slot. Most PCI cards will be of the 32-bit variety, and the selection of items available is fairly extensive. Graphics cards, sound cards, network cards, RAID controllers, TV tuners, modems, and USB/Firewire controllers are all common items that may be added to a system through the use of a PCI card. Many of the items listed in the previous paragraph can be found integrated on modern motherboards, but these onboard devices offer no upgrade ability. PCI devices provide plug and play installation, allowing a user to install (or remove) a device with ease. For example, an inexpensive 2-channel sound card may be good enough for someone initially, but down the road they may decide that something like the 7.1 channel Sound Blaster Audigy 2 offers the sound quality they really want. Upgrading is a matter of powering down the system, swapping the cards, rebooting, and installing the new software/drivers (OK, perhaps a bit over simplified). The good thing about PCI cards is that, even if you do have a board with built-in feature (such as built-in sound mentioned above), your motherboard's BIOS will usually lets you disable that feature if you did want to add an upgraded card (such as the Audigy sound card mentioned in the example above), or the card can complement the feature already built-in (such as an IDE RAID card). The one area that drove the development of AGP is the performance of PCI based graphics cards. The demands of fast-paced video games, and other graphically intensive applications, require a great deal of bandwidth, which just wasn't available on the PCI Bus. Considering that all of the devices on the PCI Bus share the bandwidth available, an even faster, dedicated bus was required to handle just the graphics data. PCI graphics cards are still available though, and make for an easy way to add a second display to a system currently operating on an AGP or PCI Express graphics card. Final Words The PCI slot has been around for a while, and seems to have a place in at least the near future of computer architecture. AGP and PCI Express offer performance benefits that the PCI standard cannot match, but for many applications, the performance offered by PCI is more than adequate. Be sure to check out the next Tech Tips in this series for the basics of AGP. <a target="_new" href="http://www.geeks.com">Computer</a> <a target="_new" href="http://www.geeks.com/pix/techtips.htm">tech tips</a> and <a target="_new" href="http://www.geeks.com/pix/techtips.htm">computer help</a>
Recordable DVD Formats Explained
Picking your way through the ton of information available on recordable DVD formats can be a nightmare. To help you out, we've done our best to distill it into this summary. There are five recordable versions of DVD; DVD-R for General, DVD-R for Authoring, DVD-RAM, DVD-RW, and DVD+RW. None of the formats is fully compatible with the other although there are drives which will read, and in some cases write to more than one format. DVD-R for General and DVD-R for Authoring are essentially DVD versions of CD-R. And DVD-RW is a DVD version of CD-RW. All three formats can be read in standard DVD-ROM drives and in most DVD video players. The difference between DVD-R for General and DVD-R for Authoring is that DVD-R for General is a format intended for widespread consumer use and doeasn't support 'professional' features such as piracy protection or duplication in mass duplicators. The Pioneer DVD-RW drive which is the most popular PC device for writing to DVD uses the DVD for General format. And as as the case with CD, DVD-RW is essentially the same as DVD-R except that it can be erased and written to again and again. DVD-RAM is slightly different as it is a sector based disc which mounts on the desktop of a PC when inserted into a drive. Files can then be copied to it in the same way as any other mounted media. Some single-sided DVD-RAM discs can be removed from their caddy and inserted in a DVD-ROM drive which will then be able to read the content of the disc. There are DVD video recorders which use the DVD-RAM format. This enables themn to pull off clever tricks like timeshifting ? where you can watch the beginning of a programme you have recorded while you are still recording the end on the same disc. DVD+RW is the newest format and not supported by the DVD Forum, the body which sets the standards for DVD. However, it is supported by some of the biggest electronics and computer manufacturers, and is therefore likely to stick around. It is also the format used by Philips in its DVD video recorders. Despite not being authorised by the DVD Forum, DVD+RW is claimed by its supporterd to be compatible with more DVD video players than DVD-R and DVD+RW writers are found in PCs from quite a few manufacturers. In early 2006, things are set to get even more complicated with the introduction of two new formats, <a target="_new" href="http://www.the-hdtv-tuner.com/hd-dvd-vs-blu- ray.html">HD-DVD and Blu-ray</a>, but that's a whole other story... Kenny Hemphill is the editor and publisher of <a target="_new" href="http://www.the-hdtv- tuner.com">The HDTV Tuner</a>, a site which aims to cut through the confusion surrounding HDTV and provide surfers with up to date, accurate and easy to read information on HDTV.
Intel Celeron vs Pentium
The Celeron and Pentium Processors are two of Intel's best selling CPUs. They are found in a majority of home computer systems. When comparing the two processors it should be first understood that there are different types of Pentium processors - the original Pentium all the way to the Pentium 4 (the latest Pentium processor). The Celeron processors are more or less the same, although you will find them in a wide variety of speeds. The Intel Celeron processor was always designed to be a low-cost alternative to the Pentium processor line. It is much like a car company that offers various priced cars from the luxury sedan to the economy compact. The Celeron is simply a downgraded Pentium, that almost anyone can afford (it is essentially the compact). To begin, Celeron chips have a smaller L2 cache 9128kb compared with 512kb in the Pentium 4 Northwood, which translates into slower processing speeds. In fact, current Celerons have a clock speed limit of about 2.0GHz, where as the Pentium for is capable of speed in excess of 3.0GHz. In addition, the Pentium runs at a lower core voltage because it is more energy effecient (1.75V vs. 1.5V). In summary, the Pentium 4 is more powerful than the most advanced Celeron processor on the market. However, Intel has planned it to be this way. Many applications will work just great with a Celeron processor, despite a little less power than the Pentium 4. It is a way to save a little cash when buying a new pc - but don't forget the saying "you get what you pay for." Celeron processors are of good Intel quality, but they will never be as good as the Pentium. This Celeron vs. Pentium review was brought to you by SciNet Science and Technology Search Engine. SciNet is not affiliated with or specifically endorses the Celeron or Pentium processors or the manufacturer, Intel Corp. Please consult the Celeron and Pentium product information and configuration before you purchase either processor. It is also a good idea to seek other up-to-date product reviews and information as necessary. Bradley James is a senior editor at <a target="_new" href="http://www.scinet.cc">SciNet.cc</a>, a website containing many helpful consumer electronics review articles. For more information on Celeron and Pentium processor technology, please visit our <a target="_new" href="http://www.scinet.cc/articles/celeron-vs-pentium/celeron-pentium.html">Celeron vs Pentium</a> webpage.
วันเสาร์ที่ 1 พฤศจิกายน พ.ศ. 2551
Tips To Select Proper Hard drive
Day by day computers are getting advanced. Capacity and performance are increasing with every passing year. This is true for the primary storage also. Latest Hard disks are getting bigger in capacity and with higher rotational speed can access the data faster. This is good news for computer users. The digital revolution has brought with it the need to store vast amounts of data. The new Hard disks are ready to take the challenge. The good news is, these Hard disks with higher capacity are affordable. Thanks to a very competitive market and demanding consumers. Hard disk technology is suitable for the bandwidth demands of today's applications. The 7200 rpm hard disk drives with a minimum storage space of 40 GB and above are now more common than ever before. The demand for storage in PCs has boomed. With a lot of multimedia files being used today, including MP3, Flash files and movies-even today's entry level 40 GB hard disks aren't enough for users. On the other hand, 80 ? 400 GB, 7200 rpm drives are also available and are preferred by professionals. Depending on your need, you should select the proper hard drive. Ask yourself few questions, before going for the selection of new hard drive. This strategy will be useful in long run. Think of the following points. Hard drive storage capacity, speed, brand and price. First select the Hard drive storage capacity. 1) Also what software, you will loading in your computer.<br> 2) What type of files, you will be loading. If you are a basic computer user, go for the 40 GB hard drive which is sufficient for OS and file storage. The text files require very less storage space however multimedia files require large storage space. If you are a gamer or a graphic designer, go for minimum 80 / 120 GB hard drive. If you are movie or song lover who would be storing lot of movies then go for higher capacity of 240 GB depending on your budget. Let us select the proper speed. The most common speeds which are available in the market are 5400 and 7200 RPM. The 5400 RPM means hard disk will be rotating at a speed of 5400 revolutions per minute. Now there is not much price difference between 5400 and 7200 rpm. It is better to go for 7200 RPM hard drive. It is better to go for standard brand. Seagate, Maxtor, Western Digital, Samsung & Hitachi are the standard brands available in the market. The above suggestions are for your internal hard drives. If you have to carry data frequently, you can go for external hard drive. These drives can be connected to computer through USB port. There is absolutely no difference between external hard drive and internal hard drive if you consider the operation or drive mechanics. USB interface has made external hard drive installation very user friendly. External hard drives give lot of portability. However these external hard drives have higher access time compared to desktop drive because of USB interface. But because of USB interface, CPU utilization is less in the range of 8.4 to 10. This means less load on CPU. Iomega, Freecom, Transcend are some of the manufacturers of external hard drives. These drives require external power supply and are most suitable for desktop use. When you are selecting external hard drive, check whether the manufacturer is offering a carry bag, data cable and power supply if it is externally powered. Also check the warranty offered by the manufacturer. Now a days, some of the manufacturers offer warranty for 3 years. Krishna Pai is a webmaster and a writer. Get Computer Tutorials, reviews, Hardware & Software Tips, How to guides at <a target="_new" href="http://www.cheapest-computer-hardware-software.com/">http://www.cheapest-computer-hardware-software.com/</a> For more Computer tips join his "Build Computer ezine" at <a target="_new" href="http://www.cheapest-computer-hardware-software.com/build_computer_ezine.html">http://www.cheapest-computer-hardware-software.com/build_computer_ezine.html</a>
Power and Your PC
The power coming into a computer is the most critical component, and it may be one of the most overlooked. It is just taken for granted that it will always be there and working properly. A top of the line processor and ultra powerful video card do nothing if a system does not receive the ample, stable power it demands. Having quality components providing and regulating the power supplied to a computer is critical, and this brief overview looks at a few areas worth consideration. Power Supplies Computer power supplies take the high (110V or 220V) DC voltages from an electrical outlet and convert it to the various lower AC voltages required for a system to operate. The typical voltages required inside a computer are 3.3V, 5V, and 12V, where the 3.3V and 5V lines are generally used to power circuitry, and the 12V line provides power to run items such as hard drives, optical drives, and cooling fans. Power supplies are sold in terms of their total power output, in terms of wattage. Choosing the correct power supply means not only finding one that will provide enough power for all of the components connected to the system, but also one that is the correct size physically, has enough connections for typical drives and fans, and if necessary, that also has special connections for things such as Serial ATA drives and modern video cards. Choosing a power supply with enough power shouldn't be much of a problem, as having more power than you need is never a bad thing. Just because a power supply is rated for 400W, or perhaps 600W, does not imply that it is drawing that at any given time. It just indicates the total power available to the various lines inside the computer. For those interested in getting a good idea of their minimum power requirements, this Power Wattage Calculator is a convenient reference. In addition to checking out the total wattage of a power supply, looking for strong amperage ratings on the 3.3V, 5V, and 12V lines is also recommended, as power supplies with identical total power ratings may distribute the power to the various lines in different quantities. Power supplies come in a few different physical sizes, but the most common are designed to fit the standard ATX and micro ATX (mATX) form factor cases. A typical ATX power supply, such as this Echo-Star 680W unit, measures 3.25" x 6" x 5.5" and features two cooling fans to not only cool the power supply, but to also help draw hot air out of the computer. A typical mATX power supply, such as this A-Power 320W unit, measures 2.5" x 5" x 4" and due to the smaller size features just one cooling fan. mATX cases are generally much smaller than ATX cases, and therefore have smaller power supplies, with generally lower power ratings, and fewer connectors. The connectors on a power supply also deserve consideration. Most power supplies come with what looks to be an electric octopus of wires hanging off the back surface, and you need to make sure that somewhere in that tangled bundle are all of the connectors you need. The power supply should at least have as many connections as the number of drives, cooling fans, and other items found in the case. Up until recently power supplies had a fixed number of connections, and if you needed more, you needed to use splitters to distribute the power to all the components. Modular power supplies, such as the Ultra X-Connect 500W unit, are now available that eliminate that "electric octopus" all together, and allow the end user to connect just the cables they need. The flexibility of a modular power supply design not only lets you customize the connections to your needs, it also makes for a simple and tidy installation, since there are no extra wires dangling inside the case. The selection of a high quality power supply may cost more money up front, but down the road it could wind up saving money. Many manufacturers now offer power supplies that consume less energy thanks to high quality internal components, advanced designs, and active power factor correction. These units are now able to provide the same power to the components in a computer, but due to increased efficiency, draw less power from the electrical outlet. Surge Protectors Surge protectors are intended to protect your electronics from a brief increase in voltage caused by such things as lightning, rolling blackouts, and heavy drawing electrical equipment. A surge protector reacts to divert the extra electricity to ground, and thus protects your expensive computer equipment from damage. A surge is any increase lasting three nanoseconds or longer, so a surge protector needs to react quickly. Most surge protectors also include a fuse (or breaker), and if the surge is too great to be handled without interruption, the fuse will blow. Although the fuse may be destroyed, it's a small loss compared to what it may have saved. Surge protectors come in all shapes and styles. Some basic models can even be found at your local dollar store, but offer no more than a few outlets connected to a breaker. No serious protection is obtained, but many people just want more outlets, not protection. More serious surge protectors will probably cost a bit more than a dollar, but will offer some peace of mind that your equipment is actually being protected. In addition to protecting from electrical surges, some devices include extra features such as conditioning to filter out line noise and ports to protect other lines such as cable television, telephone, and networking. The Fellowes Smart Surge Power Strip protects up to 10 devices from surges, as well as offering line conditioning and ports to protect your phone line. A highly appealing feature of such a surge protector is that 4 of the ports are designed to accept bulky AC adaptors. For those with surge protectors that weren't designed to be this user friendly, there is still hope in the form of Power Strip Saver Cables. Basically just 7 inch long extension cords, these items can come in very handy when trying to connect multiple AC adaptors to a more traditional surge protector. Uninterruptible Power Supplies Many people familiar with Uninterruptible power supplies (UPS) know that they can keep a computer up and running during a total power failure, but don't know what else they do. Most UPSes will also provide protection from voltage surges and sags (when the voltage drops below normal), as well as protection from the possibility of a shift in the electricity's frequency. UPSes come in two varieties, standby and continuous, although standby versions are far more common and less expensive. A standby UPS allows the devices connected to it to run from the normal electrical connection until a loss of power is detected, at which point it quickly (in a matter of milliseconds) switches to the battery backup of the UPS. A continuous UPS always has the devices connected to it running off of battery power, while the batteries are recharged by the normal electrical connection. UPSes are sold in terms of their capacity, in terms of VA (voltage multiplied by amperage). This implies that devices connected to the UPS can draw a maximum of so much amperage at a specific voltage. A run time after a power failure is also generally specified for a UPS based on a full load being placed on the device. Selecting a UPS needs to be based on the intended use? A smaller unit, such as the Fenton 600VA, would be adequate for powering a typical desktop computer, monitor, and smaller peripherals such as a printer and lighting for what they rate as 15-23 minutes at full load. If multiple systems need to be powered, or perhaps there are plans for future expansion that will add to the power demands, a larger unit such as the Tripp Lite 1500VA may be more appropriate. The first two units are intended to be set in close proximity to the devices to be powered, perhaps on the floor behind a desk, but if you are seeking to add a UPS to a server, there are also rackmount solutions such as the Opti-UPS 1100VA. No matter the application, sizing a UPS may seem overwhelming. One manufacturer, APC, has created a handy UPS Selector Application which will take some of the guess work out of choosing the right UPS for any particular application. Final Words A computer system is only as strong as its weakest feature, and many times that distinction falls on the power supply and related components. By choosing a quality power supply, surge protector, and perhaps a UPS, one can make sure that they have adequately strong and stable power to keep their system running now, as well as down the road when upgrades may increase the demand on their system. Jason Kohrs<br> <a target="_new" href="http://www.geeks.com">Computer Geeks</a> <a target="_new" href="http://www.geeks.com/pix/techtips.htm">tech tips</a> and <a target="_new" href="http://www.geeks.com/pix/techtips.htm">computer advice</a>
Expansion Cards Part 1: (of a 3 part series)
PCI The expansion slots available on motherboards allow for a variety of upgrades in a computer system, but matching the appropriate card to an available slot needs to be addressed before making any purchasing decisions. The most common types of expansion cards for modern computer systems can be broken down into three formats: PCI, AGP, and PCI Express. Each of these formats will be addressed separately in this three part series of Tech Tips, starting with PCI. The letters "PCI" stand for Peripheral Component Interconnect, and is the term used to describe a bus that connects components directly to the system's memory and to the system's processor through the "frontside bus." When discussing communications on a motherboard, the term "bus" has nothing to do with the big yellow thing that takes the kids to school. There may be several buses in a computer, and like the PCI bus, they are all responsible for managing the communication "traffic" from different devices to the processor. The frontside bus is a high speed connection that manages the processor's communication with items such as hard drives, memory, and PCI devices, while not burdening the processor with all of the management responsibilities. First developed by Intel in the early 1990s, PCI was spawned from even earlier (and slower) bus architectures such as ISA (Industry Standard Architecture) and VL-Bus (VESA Local), which were common back in the 1980s and 1990s. The original specifications for the PCI Bus had a speed of 33 MHz, with a 32-bit bus width, and a maximum bandwidth of 132 MB per second. There have been a few revisions to the PCI standard which have significantly increased these specifications, taking it to 66 MHz, 64-bit, and 512 MB per second, respectively. The 32-bit and 64-bit versions have different physical features, and most motherboards only offer 32-bit connections. The original power specification had PCI devices operating on 5V DC, and with the revisions came the capability for devices to continue using 5V, as well as now being able to operate on 3.3V DC. A simple explanation of 32-bit and 64-bit can be had by continuing the analogy of buses and traffic. Think of each bit as a lane of traffic on the communication path. Think of a 32-bit bus as having 32 lanes of traffic, and a 64-bit bus having 64 lanes of traffic. Just as a greater number of cars can travel simultaneously on a road with more lanes, more data can be transferred on a bus with a larger bit count. Motherboards can support multiple slots sharing one PCI Bus, and although not particularly common, can include more than one PCI bus. Depending on the form factor size of the motherboard, and other features that may be taking up space on the board, one can expect to have one to six PCI slots on a typical motherboard. For example, the mATX format features just two 32-bit PCI slots, while the ATX format features six 32-bit PCI slots. A 32-bit PCI card features 124 pins for mating with a slot on a system's motherboard, and will fit into either a 32-bit or 64-bit slot (although data transfer will be 32-bit in either type of slot). A 64-bit PCI card features 184 pins for mating with the appropriate slot on a system's motherboard, but can generally fit into a 32-bit slot as well, as long as features on the motherboard do not interfere. When installed in a 32-bit slot, data transfer on a 64-bit card will be limited to 32-bit. The Intel STL2 Dual Socket 370 Server Board w/VRM The Intel STL2 Dual Socket 370 Server Board is a good reference for comparing 32-bit and 64-bit PCI slots. Looking at the lower left corner of the motherboard shows four 32-bit PCI slots and two 64-bit PCI slots. Subsequent installments in this series of Tech Tips will look at AGP and PCI Express, each of which has its own unique physical features. Although the different format PCI cards may be interchangeable, PCI, AGP, and PCI Express cards do not work (or fit) in any other type of slot. Most PCI cards will be of the 32-bit variety, and the selection of items available is fairly extensive. Graphics cards, sound cards, network cards, RAID controllers, TV tuners, modems, and USB/Firewire controllers are all common items that may be added to a system through the use of a PCI card. Many of the items listed in the previous paragraph can be found integrated on modern motherboards, but these onboard devices offer no upgrade ability. PCI devices provide plug and play installation, allowing a user to install (or remove) a device with ease. For example, an inexpensive 2-channel sound card may be good enough for someone initially, but down the road they may decide that something like the 7.1 channel Sound Blaster Audigy 2 offers the sound quality they really want. Upgrading is a matter of powering down the system, swapping the cards, rebooting, and installing the new software/drivers (OK, perhaps a bit over simplified). The good thing about PCI cards is that, even if you do have a board with built-in feature (such as built-in sound mentioned above), your motherboard's BIOS will usually lets you disable that feature if you did want to add an upgraded card (such as the Audigy sound card mentioned in the example above), or the card can complement the feature already built-in (such as an IDE RAID card). The one area that drove the development of AGP is the performance of PCI based graphics cards. The demands of fast-paced video games, and other graphically intensive applications, require a great deal of bandwidth, which just wasn't available on the PCI Bus. Considering that all of the devices on the PCI Bus share the bandwidth available, an even faster, dedicated bus was required to handle just the graphics data. PCI graphics cards are still available though, and make for an easy way to add a second display to a system currently operating on an AGP or PCI Express graphics card. Final Words The PCI slot has been around for a while, and seems to have a place in at least the near future of computer architecture. AGP and PCI Express offer performance benefits that the PCI standard cannot match, but for many applications, the performance offered by PCI is more than adequate. Be sure to check out the next Tech Tips in this series for the basics of AGP. <a target="_new" href="http://www.geeks.com">Computer</a> <a target="_new" href="http://www.geeks.com/pix/techtips.htm">tech tips</a> and <a target="_new" href="http://www.geeks.com/pix/techtips.htm">computer help</a>
สมัครสมาชิก:
บทความ (Atom)