PC expansion slots have come a long way since the old eight bit ISA slots, and now we’re going to take a look at their newest incarnation, the PCI Express slot. In this article I’ll be covering some of the technical details of PCI Express (or PCI-E for short) and how it stacks up with what is currently available on the market, and what this means to average users and power users alike.
Overview of Current PCI Technology and its Shortcomings
The PCI bus in most of our PCs is a relic of the i486 when Intel flexed its muscles for the second time, the first being the famous Intel Inside logo. IBM and everybody else had pigeonholed PCs to word processing and very light duty computing; leaving the more powerful tasks to the UNIX workstation with their fast RISC processors and speedy expansion busses needed for handling high-end I/O. To replace ISA, IBM invented the MCA bus and VLB and EISA were explored as another option as they are backwards compatible with 8 and 16 bit ISA, but this trio had one flaw – they only ran at half of the processor bus speed, around 8.33 MHz, the same as the old ISA busses. This is where Intel introduced its PCI bus, running at full processor speed of 25 to 33 MHz and scaling beautifully as the i486 introduced processor multipliers on its DX2 and DX4 line at 50, 66, 75 and 100 MHz. But it would only really shine later when used with high-speed graphics cards on Pentium systems and later Pentium Pro/Pentium II systems.
By this time, of course, graphics cards had gotten powerful enough to maximize a PCI bus and AGP was introduced, and with the main burden off of the PCI bus the old 33 MHz 32 bit bus remained adequate as the one expansion bus beyond AGP in a system until late in the Pentium III line when the southbridge was given control of the bus from the northbridge and rather than the southbridge’s peripherals and all of the expansion slots sharing that single PCI bus the southbridge gained a high speed proprietary point to point connection, allowing more bandwidth to the southbridges peripherals, especially newer, bandwidth hungry hard disk drives. Of course, there were also higher speed busses available in the form of 64 bit PCI and higher clock speeds, but the additional cost of the slots and chipsets and the cards themselves put the kybosh on that idea for desktop PCs fairly quickly, leaving them to servers and workstations, both PC and by this time RISC computers even.
By now there are several variations of PCI busses, ranging in bandwidth from 100 MB/s up to 1066 MB/s, as shown by the table below (Table 1), note the overlap between standard PCI and PCI-X; there are some differences between the two although they for most people exactly the same. In the second table (Table 2) I’ve listed some common peripherals to give a reference to the amount of bandwidth they can consume:
| Bus: | Clock Speed | Bandwidth at 32 bit | Bandwidth at 64 bit |
| Standard PCI | 25 MHz | 100 MB/s | 200 MB/s |
| Standard PCI | 33 MHz | 133 MB/s | 266 MB/s |
| PCI-66 | 66 MHz | 266 MB/s | 533 MB/s |
| PCI-X | 66 MHz | N/A | 533 MB/s |
| PCI-X | 100 MHz | N/A | 800 MB/s |
| PCI-X | 133 MHz | N/A | 1066 MB/s |
Table 1
As you can see, with the average PC sharing a single 133 MB/s PCI bus for all of its expansion slots how bandwidth can easily be gobbled up by modern peripherals. Enter PCI Express.
| Peripheral | Possible Bandwidth | Typical Bus Interface(s) | Bus Speeds |
| Gigabit Ethernet Controller (Full Duplex) | 200 MB/s | 33 or 66 MHz, 32 bit | 133 or 266 MB/s |
| Dual Channel ATA/133 Controller | 266 MB/s | 33 or 66 MHz, 32 bit | 133 or 266 MB/s |
| Ultra 320 SCSI Controller | 320 MB/s | 66 MHz, 64 bit PCI or 66 to 133 MHz PCI-X |
533, 800 or 1066 MB/s |
| Quad Channel SATA/150 Controller | 600 MB/s | 33 or 66 MHz 32 bit | 133 or 266 MB/s |
| Dual Controller 1394a | 160 MB/s | 33 MHz 32 bit | 133 MB/s |
Table 2
To answer this rapidly increasing bandwidth shortcoming choking even servers with 1066 MB/s of bandwidth on each bus and multiple busses, PCI Express was created. Notice it is not called a PCI Express bus, as that is a misnomer. In a bus, like PCI, expansion cards plugged into one shared resource. PCI-E uses a point-to-point technology where each slot receives its own path to the northbridge or southbridge, very similar to the difference between using a hub on a network and using a switch. Also, unlike PCI where a number of bits are sent in parallel (32 or 64), PCI-E is a serial interface, like Ethernet. Since there are fewer bits being sent at once, PCI-E uses a different method to create speed by simply increasing the signaling rate. Where a PCI-X bus can run at up to 133 MHz, a single PCI-E runs at 2500 MHz, or 2.5 GHz, and can send data both to and from the controller at the same time in a full duplex manner, giving a maximum theoretical bandwidth of 500 MB/s, 250 MB/s in each direction.
PCI Express and Graphics
As if all this speed wasn’t enough, PCI-E also allows channels, or “lanes” to be teamed to create larger channels, allowing the slots to be used for graphics applications as well as standard expansion. Currently the graphics slot of choice for PCI-E is a 16 lane, or X16, slot providing 75 Watts of power (up from 50W for AGP Pro). With its 16 teamed lanes it allows for a maximum of 8 GB/s combined bandwidth, 4 GB/s each direction compared with the AGP 8X maximum of 2 GB/s total. While this is still overkill (most graphics cards today would suffer little performance penalty from being placed in even an AGP 4X slot), it does allow enough bandwidth to reintroduce a now disfavored technique that AGP was originally designed for and hasn’t done very extensively at all since the nVidia Riva 128: texturing in main memory, allowing for more realistic games on lesser graphics cards.
Another possibility is installing two PCI-E X16 slots on a single motherboard allowing two high-powered graphics cards to be used in a single system. Alienware is currently preparing such a system, although due to current chipset limitations when a second card is inserted it will run each slot with only eight lanes rather than the full sixteen, that will still be more than sufficient bandwidth for graphics cards for the next several years. In the future chipsets will be designed to allow 32 lanes directly off of the northbridge to handle the extra load, as graphics channels are preferred to run directly from the northbridge for better performance to memory. Eventually it will not be unreasonable to remove the southbridge entirely and have all PCI-E lanes controlled directly from the northbridge, allowing more bandwidth than today is currently available. As of now, both ATi and nVidia have added PCI-E 16X products to their lines and they are currently battling it out for top performer.
Left: ATi’s Radeon X700 PCI Express showing connector
Right: a bit beefier solution, the 3Dlabs
Realizm 800 professional card
PCI Express and Expansion Cards
Just as with graphics cards where 16 lanes are combined at once, likewise in normal expansion cards they can be teamed together as well. Currently, of course, It would be reasonable to install a SATA controller or Gigabit Ethernet, even dual port, into one single lane slot, in the future the bandwidth can be expanded by using PCI-E 2X, 4X or even faster slots. As an added bonus, if you purchase a PCI-E 1X SATA controller and next year you purchase a system with PCI-X 2X slots, they will plug into them. The reverse is not true, however: you cannot plug a PCI-E 16X card into a PCI-E 1X slot, due to the slot and card configuration as shown below in Picture 1, below.
Above: PCI-E 16X; Below: PCI-E 1X
As you can see in the above picture, to the left of the key there is 22 pins, and on the right there is 14 on the PCI-E 1X and 142 on PCI-E 16X. This key arrangement allows forward compatibility with future slots, although not backwards compatibility for future cards and older slots. It should also work to plug a hardware RAID controller using a 4X or 8X interface into the 16X graphics slot allowing for a very capable uniprocessor home server on the cheap, although the PCI-E specification does not require a slot designed for graphics to be capable of accepting other expansion cards so your mileage will vary on this capability. I believe that Via and SiS chipsets would be most likely to accept such a configuration over Intel and AMD, but there’s no hard data on this.
PCI Express also allows for more power to individual cards: up to 25W. Intel states that PCI-E 1X is for, “Gigabit Ethernet, TV Tuners, 1394a/b controllers, and general purpose I/O,” and I heartily agree. PCI-E is for next-generation PCs where rather than just using a monitor, keyboard, mouse, printer and the other peripheral devices that we are used to seeing in every computer, TV tuners, video editing, HD gaming and et cetera demanding more bandwidth than current PCs can provide. However, in normal gaming, PCI-E 1X slots won’t make a lick of difference until a normal PCI bus is close to saturated. The most important key will still be graphics, memory and CPU, and PCI-E 1X doesn’t touch those.
PCI Express for Servers and Workstations
Without delving into too much depth, PCI-E can be configured in 6 different slot sizes: 1X, 2X, 4X, 8X, 16X and 32X. For servers and workstations serving up multimedia content, this is a dream come true. Needless to say, with PCI-E 4X well outperforming the best PCI-X slots and PCI-E 8X being seen on some server motherboards today, PCI-E suits their needs as well.
SIR-808X PCI-Express 8-port SATA RAID Host Adapter with a PCI-E 4X interface
As notebooks are becoming more popular, you may ask and legitimately so what will be done with them to increase bandwidth, as portable systems have traditionally had even less bandwidth than even entry-level desktops. The organization that creates the PCMCIA standard has introduced a new card with pluggable connections to either a USB 2.0 connection or a PCI-E connection, depending on card need. For instance, a media reader might do best on USB 2.0 whereas a TV Tuner or extra graphics card for presentations would do better on a PCI-E connection.
By now you may be saying, PCI-E sounds well and good, but how does it affect me? There aren’t any PCI-E expansion cards available, so why should I care? Well, it is true that there are very, very few PCI-E cards available now. However, Creative Labs is due to launch a PCI-E based sound card, several companies are investigating hardware RAID 5 SATA controllers and eventually everything that you now use a PCI card for will be PCI-E. With the holiday season coming up, you can expect a massive number of products to enter the market quickly before Christmas to make your PCI-E enabled motherboard worthwhile, including new PCI-E boards for the AMD processors in addition to the current Via chipset.
The current chipsets supporting PCI-Express are the following: Intel 925X, Intel 915G, Intel 915GV, Intel 915GL, Intel 915P and the Intel 910GL. All of these are Intel chipsets with various features accepting the Socket T chips. However, within the next two months more chipsets and boards based on them should become available very quickly. The Intel 925XE will come out supporting a 1066 MHz bus, otherwise it is identical to the 925X. ATi has an integrated video chipset launching, the RX480, which supports PCI-E. Via is launching the K8T890 AMD K8 chipset supporting PCI-E. The other PCI-E is chipset for AMD coming to market quickly is from nVidia in the form of the nForce4 series, with the nForce4, nForce4 Ultra and nForce4 SLI.
Conclusion
PCI Express is the first major upgrade to desktop PC expansion since the days of the 486, and purchasing a system without it now would be foolish. I would definitely suggest to anyone purchasing a new motherboard to not even bother unless it’s PCI Express capable. While I realize that only a handful of chipsets today are PCI Express capable, it’s the difference between a PC potentially lasting up to five or more years and a PC lasting one or, at a stretch, two years. So for all of those (like me) who do not own a money tree and have the capability to replace their motherboard every year should only consider a PCI-E board.