When it comes to any interfaces in the context of computer systems, one must be very careful not to "run into" incompatible interfaces for the same components within the system.

Fortunately, when it comes to the PCI-Express interface for connecting a video card, there will be practically no incompatibility problems. In this article, we will analyze this in more detail, and also talk about what this very PCI-Express is.

What is PCI-Express for and what is it?

Let's start, as usual, with the very basics. PCI-Express interface (PCI-E)- this is a means of interaction, in this context, consisting of a bus controller and the corresponding slot (Fig. 2) on motherboard(to summarize).

This high-performance protocol is used, as noted above, to connect a video card to the system. Accordingly, there is a corresponding PCI-Express slot on the motherboard, where the video adapter is installed. Previously, video cards were connected via the AGP interface, but when this interface, simply put: “it was no longer enough”, PCI-E came to the rescue, oh detailed specifications which we are going to talk about now.

Figure 2 (PCI-Express 3.0 slots on motherboard)

Main features of PCI-Express (1.0, 2.0 and 3.0)

Despite the fact that the names of PCI and PCI-Express are very similar, their connection (interaction) principles are fundamentally different. In the case of PCI-Express, a line is used - a bidirectional serial connection, of the point-to-point type, there can be several of these lines. In the case of video cards and motherboards(we do not take into account Cross Fire and SLI), which support PCI-Express x16 (that is, most), you can easily guess that there are 16 such lines (Fig. 3), quite often on motherboards with PCI-E 1.0, one could observe the second x8 slot, for operation in SLI or Cross Fire mode.

Well, in PCI, the device is connected to a common 32-bit parallel bus.

Rice. 3. An example of slots with a different number of lines

(as mentioned earlier, x16 is the most commonly used)

For the interface, the throughput is 2.5 Gbps. We need this data to track changes in this parameter in different versions of PCI-E.

Further, version 1.0 evolved into PCI-E 2.0. As a result of this transformation, we got twice the bandwidth, that is, 5 Gb / s, but I would like to note that graphics adapters did not particularly benefit in performance, since this is just a version of the interface. Most of the performance depends on the video card itself, the interface version can only slightly improve or slow down data transfer (in this case, there is no “braking”, and there is a good margin).

In the same way, in 2010, with a margin, an interface was developed PCI-E3.0, at the moment it is used in all new systems, but if you still have 1.0 or 2.0, then do not worry - below we will talk about relatively backward compatibility of various versions.

In PCI-E version 3.0, the bandwidth has been doubled compared to version 2.0. Also there were made a lot of technical changes.

By 2015 is expected to be born PCI-E 4.0, which is absolutely not surprising for a dynamic IT industry.

Well, okay, let's finish with these versions and bandwidth figures, and touch on the very important issue of backward compatibility of different versions of PCI-Express.

Backwards compatible with PCI-Express versions 1.0, 2.0 and 3.0

This question is of concern to many, especially when choosing a video card for the current system. Since being content with a system with motherboard, which supports PCI-Express 1.0, there are doubts whether the video card will work correctly with PCI-Express 2.0 or 3.0? Yes, it will be, at least as promised by the developers who ensured this very compatibility. The only thing is that the video card will not be able to fully reveal itself in all its glory, but the performance loss, in most cases, will be insignificant.

On the contrary, you can safely install video cards with PCI-E 1.0 interface, in motherboards that support PCI-E 3.0 or 2.0, nothing is limited here at all, so don't worry about compatibility. If, of course, everything is in order with other factors, these include an insufficiently powerful power supply, etc.

In general, we have talked in some detail about PCI-Express, which will allow you to get rid of a lot of ambiguities and doubts about compatibility and understanding the differences in PCI-E versions.

When changing only one video card, be sure to take into account that new models may simply not fit your motherboard, since there are not just several different types of expansion slots, but also several different versions of them (applicable to both AGP and PCI Express). If you are unsure of your knowledge on this topic, read the section carefully.

As we noted above, the video card is inserted into a special expansion slot on the computer's motherboard, through this slot the video chip exchanges information with the system's central processor. On the motherboards ah, most often there are expansion slots of one or two different types, differing in bandwidth, power settings and other characteristics, and not all of them are suitable for installing video cards. It is important to know the connectors available in the system and buy only the video card that matches them. Different expansion slots are physically and logically incompatible, and a video card designed for one type will not fit into another and will not work.

Fortunately, not only ISA and VESA Local Bus expansion slots (which are of interest only to future archaeologists) and their corresponding video cards have faded into oblivion, but video cards for PCI slots have practically disappeared, and all AGP models are hopelessly outdated. And all modern GPUs use only one type of interface - PCI Express. Previously, the AGP standard was widely used, these interfaces differ significantly from each other, including the bandwidth provided by the possibilities for powering the video card, as well as other less important characteristics.

Only a very small part of modern motherboards do not have PCI Express slots, and if your system is so old that it uses an AGP video card, then you will not be able to upgrade it - you need to change the entire system. Consider these interfaces in more detail, these are the slots you need to look for on your motherboards. See photos and compare.

AGP (Accelerated Graphics Port or Advanced Graphics Port) is a high-speed interface based on the PCI specification, but designed specifically for connecting graphics cards and motherboards. The AGP bus, although better suited for video adapters than PCI (not Express!), provides a direct connection between the CPU and the video chip, as well as some other features that increase performance in some cases, for example, GART - the ability to read textures directly from random access memory, without copying them to video memory; a higher clock frequency, simplified data transfer protocols, etc., but this type of slots is hopelessly outdated and new products with it have not been released for a long time.

But still, for the sake of order, we will also mention this type. The AGP specifications appeared in 1997, when Intel released the first version of the description, including two speeds: 1x and 2x. The second version (2.0) introduced AGP 4x, and 3.0 introduced 8x. Let's consider all the options in more detail:
AGP 1x is a 32-bit channel running at 66 MHz with a bandwidth of 266 MB/s, which is twice the PCI bandwidth (133 MB/s, 33 MHz and 32 bits).
AGP 2x is a 32-bit channel that operates with double the bandwidth of 533 MB/s at the same frequency of 66 MHz due to data transfer on two fronts, similar to DDR memory (only for the direction "to the video card").
AGP 4x is the same 32-bit channel operating at 66 MHz, but as a result of further tricks, a quadruple "effective" frequency of 266 MHz was achieved, with a maximum throughput of more than 1 GB / s.
AGP 8x - additional changes in this modification made it possible to obtain a throughput of up to 2.1 GB / s.

Video cards with AGP interface and corresponding slots on motherboards are compatible within certain limits. Graphics cards rated at 1.5V do not work in 3.3V slots, and vice versa. However, there are universal connectors that support both types of boards. Video cards designed for the morally and physically obsolete AGP slot have not been considered for a long time, so to learn about old AGP systems, it would be better to read the article:

PCI Express (PCIe or PCI-E, not to be confused with PCI-X), formerly known as Arapahoe or 3GIO, differs from PCI and AGP in that it is a serial rather than a parallel interface, which reduces the number of pins and increases bandwidth. PCIe is just one example of the transition from parallel to serial buses, other examples of this movement are HyperTransport, Serial ATA, USB and FireWire. An important advantage of PCI Express is that it allows you to stack several single lanes into one channel to increase throughput. The versatility of the serial design increases flexibility, slower devices can be assigned fewer lines with fewer pins, and fast devices more.

The PCIe 1.0 interface allows data to pass through at 250MB/s per lane, nearly double the capacity of conventional PCI slots. The maximum number of lanes supported by PCI Express 1.0 slots is 32, which gives a throughput of up to 8 GB/s. And the PCIe slot with eight working lines is roughly comparable in this parameter to the fastest version of AGP - 8x. What is even more impressive when you consider the possibility of simultaneous transmission in both directions at high speed. The most common PCI Express x1 slots provide a single lane (250 MB/s) throughput in each direction, while PCI Express x16, which is used for video cards and which combines 16 lanes, provides up to 4 GB/s throughput in each direction.

Although the connection between two PCIe devices is sometimes made up of several lines, all devices support a single line at a minimum, but can optionally work with a large number of them. Physically, PCIe expansion cards come in and work fine in any slots with equal or more lanes, so a PCI Express x1 card will work fine in x4 and x16 slots. Also, a physically larger slot can work with a logically smaller number of lines (for example, it looks like a regular x16 connector, but only 8 lines are routed). In any of the above options, PCIe will choose the highest possible mode by itself, and it will work fine.

Most often, x16 connectors are used for video adapters, but there are also boards with x1 connectors. And most motherboards with two PCI Express x16 slots operate in x8 mode to create SLI and CrossFire systems. Physically, other slot options, such as x4, are not used for video cards. Please note that this only applies to physical layer, there are also motherboards with physical PCI-E x16 slots, but in reality with divorced 8, 4 or even 1 channels. And any video cards designed for 16 channels will work in such slots, but with lower performance. By the way, the photo above shows x16, x4 and x1 slots, and PCI is also left for comparison (bottom).

Although the difference in games is not so big. Here, for example, is a review of two motherboards on our website, which examines the difference in the speed of 3D games on two motherboards, a pair of test video cards in which operate in 8-channel and 1-channel modes, respectively:

The comparison that interests us is at the end of the article, pay attention to the last two tables. As you can see, the difference at medium settings is very small, but in heavy modes it begins to increase, moreover, a large difference was noted in the case of a less powerful video card. Take note.

PCI Express differs not only in bandwidth, but also in new power consumption capabilities. This need arose because the AGP 8x slot (version 3.0) can transmit only no more than 40-odd watts in total, which was already lacking for video cards of the then generations designed for AGP, which were equipped with one or two standard four-pin power connectors. Up to 75 watts can be transferred through the PCI Express connector, and an additional 75 watts is received through the standard six-pin power connector (see the last section of this part). Recently, video cards with two such connectors have appeared, which in total gives up to 225 watts.

Subsequently, the PCI-SIG group, which develops the relevant standards, introduced the main specifications for PCI Express 2.0. The second version of PCIe doubled the standard bandwidth from 2.5 Gb/s to 5 Gb/s, so that the x16 slot can transfer data at speeds up to 8 Gb/s in each direction. While PCIe 2.0 is compatible with PCIe 1.1, older expansion cards usually work fine in new motherboards.

The PCIe 2.0 specification supports both 2.5 Gb/s and 5 Gb/s transfer rates to ensure backward compatibility with existing PCIe 1.0 and 1.1 solutions. PCI Express 2.0 backwards compatibility allows legacy 2.5 Gb/s solutions to be used in 5.0 Gb/s slots, which will simply run at a slower speed. And devices designed to version 2.0 specifications can support 2.5Gbps and/or 5Gbps speeds.

Although the main innovation in PCI Express 2.0 is the speed doubled to 5 Gb/s, but this is not the only change, there are other modifications to increase flexibility, new mechanisms for program control connection speed, etc. We are most interested in changes related to the power supply of devices, as the power requirements of video cards are steadily increasing. PCI-SIG has developed a new specification to accommodate the increasing power consumption of graphics cards, it expands the current power supply capabilities to 225/300W per graphics card. To support this specification, a new 2x4-pin power connector is used, designed to provide power to high-end graphics cards.

Video cards and motherboards with PCI Express 2.0 support appeared on the market already in 2007, and now there are no others on the market. Both major GPU manufacturers, AMD and NVIDIA, have released new GPU and video card lines based on them, supporting the increased bandwidth of the second version of PCI Express and taking advantage of new power supply options for expansion cards. All of them are backwards compatible with motherboards that have PCI Express 1.x slots onboard, although in some rare cases there is incompatibility, so you need to be careful.

Actually, the appearance of the third version of PCIe was an obvious event. In November 2010, the specifications of the third version of PCI Express were finally approved. Although this interface has a transfer rate of 8 gigatransactions / s instead of 5 Gt / s in version 2.0, its throughput has again increased exactly twice compared to the PCI Express 2.0 standard. For this, a different encoding scheme was used for data sent over the bus, but compatibility with previous versions of PCI Express was preserved. The first products of the PCI Express 3.0 version were introduced in the summer of 2011, and real devices have just begun to appear on the market.

A whole war broke out among motherboard manufacturers for the right to be the first to introduce a product with PCI Express 3.0 support (mainly based on Intel chipset Z68), and the corresponding press releases were presented by several companies at once. Although at the time of updating the guide there are simply no video cards with such support, so it's just not interesting. By the time PCIe 3.0 support is needed, completely different boards will appear. Most likely, this will happen no earlier than 2012.

By the way, we can assume that PCI Express 4.0 will be introduced over the next few years, and a new version will also have once again doubled the bandwidth demanded by that time. But this will not happen very soon, and we are not interested yet.

External PCI Express

In 2007, the PCI-SIG group, which officially standardizes PCI Express solutions, announced the adoption of the PCI Express External Cabling 1.0 specification, which describes the standard for data transfer over an external PCI Express 1.1 interface. This version allows you to transfer data at a speed of 2.5 Gb / s, and the next should increase the throughput to 5 Gb / s. The standard includes four external slots: PCI Express x1, x4, x8 and x16. Senior connectors are equipped with a special tongue that facilitates connection.

The external version of the PCI Express interface can be used not only for connecting external video cards, but also for external drives and other expansion cards. The maximum recommended cable length is 10 meters, but this can be increased by connecting the cables through a repeater.

Theoretically, this could make life easier for laptop enthusiasts when using a low-powered integrated video core when running on batteries, and using a powerful external graphics card when connected to a desktop monitor. The upgrade of such video cards is greatly facilitated, there is no need to open the PC case. Manufacturers can make completely new cooling systems that are not limited by the features of expansion cards, and there should be fewer problems with power supply - most likely, external power supplies designed specifically for a specific video card will be used, they can be built into one external case with a video card using one cooling system. It may be easier to assemble systems on several video cards (SLI/CrossFire), and taking into account the constant growth in the popularity of mobile solutions, such external PCI Express should have gained some popularity.

They should have, but they didn't. As of autumn 2011, there are practically no external options for video cards on the market. Their circle is limited by outdated models of video chips and a narrow selection of compatible laptops. Unfortunately, the case of external video cards did not go further, and slowly died out. You can no longer even hear the victorious advertising statements from laptop manufacturers... Perhaps the power of modern mobile video cards has simply become enough even for demanding 3D applications, including many games.

There remains hope for the development of external solutions in a promising interface for connection peripherals Thunderbolt, formerly known as Light Peak. It was developed by Intel Corporation based on DisplayPort technology, and the first solutions have already been released by Apple. Thunderbolt combines the capabilities of DisplayPort and PCI Express and allows you to connect external devices. However, so far they simply do not exist, although the cables already exist:

In the article, we do not touch on outdated interfaces, the vast majority of modern video cards are designed for the PCI Express 2.0 interface, so when choosing a video card, we suggest considering only it, all data on AGP is provided for reference only. The new boards use the PCI Express 2.0 interface, combining the speed of 16 PCI Express lanes, which gives a throughput of up to 8 GB / s in each direction, which is several times more than the same characteristic of the best AGP. In addition, PCI Express operates at such a speed in each of the directions, unlike AGP.

On the other hand, products with PCI-E 3.0 support have not really come out yet, so it doesn't make much sense to consider them either. If we are talking about upgrading an old or buying a new board or changing the system and video board at the same time, then you just need to purchase boards with a PCI Express 2.0 interface, which will be quite sufficient and most common for several more years, especially since products of different PCI Express versions are compatible with each other .

#PCI Express

The PCI Express serial bus, developed by Intel and its partners, is designed to replace the parallel PCI bus and its extended and specialized variant AGP. Despite similar names, PCI and PCI Express buses have little in common. The parallel data transfer protocol used by PCI imposes limits on the bandwidth and frequency of the bus; the serial data transfer used in PCI Express provides scalability (the specifications describe implementations of PCI Express 1x, 2x, 4x, 8x, 16x and 32x). At the moment, the tire version with index 3.0 is relevant.

PCI-E3.0

In November 2010, the PCI-SIG organization, which standardizes PCI Express technology, announced the adoption of the PCIe Base 3.0 specification.
The key difference from the previous two versions of PCIe can be considered a changed coding scheme - now instead of 8 bits of useful information out of 10 bits transmitted (8b / 10b), 128 bits of useful information can be transmitted over the bus out of 130 bits sent, i.e. The payload ratio is close to 100%. In addition, the data transfer rate has increased to 8 GT/s. Recall that this value for PCIe 1.x was 2.5 GT/s, and for PCIe 2.x it was 5 GT/s.
All of the above changes resulted in a doubling of the bus bandwidth compared to the PCI-E 2.x bus. This means that the total bandwidth of the PCIe 3.0 bus in a 16x configuration will reach 32 Gb / s. The first processors to be equipped with a PCIe 3.0 controller were Intel processors based on the Ivy Bridge microarchitecture.

Despite more than three times the throughput of PCI-E 3.0 compared to PCI-E 1.1, the performance of the same video cards when using different interfaces does not differ much. The table below shows the benchmark results for the GeForce GTX 980 in various benchmarks. The measurements were carried out with the same graphics settings, in the same configuration. The PCI-E bus version was changed in the BIOS settings.

PCI Express 3.0 continues to be backward compatible with previous PCIe versions.

PCI-E 2.0

In 2007, a new specification for the PCI Express bus - 2.0 was adopted, the main difference of which is the doubled bandwidth of each transmission line in each direction, i.e. in the case of the most popular version of PCI-E 16x used in video cards, the throughput is 8Gb / s in each direction. The first chipset to support PCI-E 2.0 was the Intel X38.

PCI-E 2.0 is fully backward compatible with PCI-E 1.0, i.e. all existing devices with PCI-E 1.0 interface can work in PCI-E 2.0 slots and vice versa.

PCI-E 1.1

The first version of the PCI Express interface, introduced in 2002. Provided throughput of 500 MB / s per line.

Comparison of the speed of work of different generations of PCI-E

The PCI bus runs at 33 or 66 MHz and provides 133 or 266 MB/s bandwidth, but this bandwidth is shared among all PCI devices. The frequency at which the PCI Express 1.1 bus operates is 2.5 GHz, which gives a bandwidth of 2500 MHz / 10 * 8 = 250 * 8 Mbps = 250 Mbps information) for each PCI Express 1.1 x1 device in one direction. If there are several lines, to calculate the throughput, the value of 250 Mb / s must be multiplied by the number of lines and by 2, because. PCI Express is a bidirectional bus.

Number of PCI Express 1.1 lanes	Throughput in one direction	Total throughput
1	250 MB/s	500 MB/s
2	500 Mb/s	1 GB/s
4	1 GB/s	2 GB/s
8	2 GB/s	4 GB/s
16	4 GB/s	8 GB/s
32	8 GB/s	16 GB/s

Note! You should not attempt to install a PCI Express card in a PCI slot, and vice versa, PCI cards are not installed in PCI Express slots. However, a PCI Express 1x card, for example, can be installed and most likely will function normally in a PCI Express 8x or 16x slot, but not vice versa: a PCI Express 16x card will not fit into a PCI Express 1x slot.

Support for the PCI Express 3.0 interface in motherboards - a real advantage or a marketing ploy?

In recent months, motherboards from various manufacturers have begun to appear in the model range, in which support for the PCI Express 3.0 interface is declared. ASRock, MSI and GIGABYTE were the first to announce such solutions. However, at the moment, there are absolutely no chipsets, graphics and central processing units, which would support the PCI Express 3.0 interface.

Recall that the PCI Express 3.0 standard was approved last year. It has numerous advantages over its predecessors, so it is not surprising that graphics card and motherboard manufacturers want to implement it in their solutions as soon as possible. However, the currently existing chipsets from Intel and AMD are limited to supporting the PCI Express 2.0 standard. The only hope to take advantage of the PCI Express 3.0 interface in the near future lies with new Intel processors Ivy Bridge, the announcement of which is scheduled only for March-April next year. These processors have an integrated PCI Express 3.0 bus controller, but only graphics chips will be able to use it, since other components use the chipset controller.

Note that the matter is not limited to just replacing the processor. You need to additionally update the BIOS settings and chipset firmware. In addition, on motherboards with several PCI Express x16 slots, there is a problem with "switches" - small microcircuits that are located near each slot and are responsible for the operational reconfiguration of the number of dedicated lines. These "switches" must also be compatible with the PCI Express 3.0 interface. It should be noted that nForce 200 or Lucid bridge chips support only the PCI Express 2.0 standard and they cannot work with the PCI Express 3.0 specification.

The last argument is that at the moment motherboard manufacturers do not have engineering samples of new Intel Ivy Bridge processors or new graphics chips that support the PCI Express 3.0 specification at the hardware level. Therefore, the announced compatibility with this high-speed interface is theoretical and cannot, at the moment, be practically confirmed.

Thus, support for the PCI Express 3.0 specification by modern motherboards is purely a marketing ploy, the benefits from which the user will be able to get only in a few months by replacing the processor and updating the software components.

#PCI

Attention! This article is about the PCI bus and its PCI64 and PCI-X derivatives! Do not confuse it with the newer tire ("PCI Express"), which is completely incompatible with the tires described in this FAQ.

PCI 2.0- the first version of the basic standard, which received wide distribution, both cards and slots with a signal voltage of only 5V were used.

PCI 2.1- differed from 2.0 by the possibility of simultaneous operation of several bus-master devices (the so-called competitive mode), as well as the appearance of universal expansion cards capable of operating in both 5V and 3.3V slots. The ability to work with 3.3V cards and the presence of appropriate power lines in version 2.1 was optional. PCI66 and PCI64 extensions appeared.

PCI 2.2- a version of the basic bus standard that allows connection of expansion cards with a signal voltage of both 5V and 3.3V. The 32-bit versions of these standards were the most common slot type at the time the FAQ was written. 32-bit, 5V type slots are used.
Expansion cards made in accordance with these standards have a universal connector and are able to work in almost all later varieties of PCI bus slots, and also, in some cases, in 2.1 slots.

PCI 2.3- the next version of the general standard for the PCI bus, expansion slots that comply with this standard are not compatible with PCI 5V cards, despite the continued use of 32-bit slots with a 5V key. Expansion cards have a universal connector, but are not able to work in 5V slots of earlier versions (up to 2.1 inclusive).
We remind you that the supply voltage (not signal!) 5V is stored absolutely on all versions of the PCI bus connectors.

PCI 64- an extension of the basic PCI standard, introduced in version 2.1, doubling the number of data lines, and, consequently, the throughput. The PCI64 slot is an extended version of the regular PCI slot. Formally, the compatibility of 32-bit cards with 64-bit slots (subject to the presence of a common supported signal voltage) is complete, and the compatibility of a 64-bit card with 32-bit slots is limited (in any case, there will be a loss of performance), exact data in each case can be found in the specifications of the device.
The first versions of PCI64 (derived from PCI 2.1) used a 64-bit 5V PCI slot and ran at 33MHz.

PCI 66- an extension of the PCI standard introduced in version 2.1 with support for clock frequency 66MHz, as well as PCI64, allows you to double the bandwidth. Starting with version 2.2, it uses 3.3V slots (the 32-bit version is almost never found on a PC), cards have a universal or 3.3V form factor. (There were also solutions based on version 2.1, casuistically rare on the PC 5V 66MHz market, such slots and boards were only compatible with each other)

PCI 64/66- A combination of the above two technologies, it can quadruple the data transfer rate compared to the basic PCI standard, and uses 64-bit 3.3V slots, compatible only with universal and 3.3V 32-bit expansion cards. PCI64/66 cards have a universal (with limited compatibility with 32-bit slots) or 3.3V form factor (the latter option is fundamentally not compatible with 32-bit 33MHz slots of popular standards)
Currently, the term PCI64 means exactly PCI64/66, since 33MHz 5V 64-bit slots have not been used for a long time.

PCI-X 1.0- Expansion of PCI64 with the addition of two new operating frequencies, 100 and 133 MHz, as well as a separate transaction mechanism to improve performance when running multiple devices at the same time. Generally backwards compatible with all 3.3V and universal PCI cards.
PCI-X cards are usually implemented in 64-bit 3.3 format and have limited backward compatibility with PCI64/66 slots, and some PCI-X cards are in universal format and can work (although this has almost no practical value) in regular PCI 2.2 /2.3.
In difficult cases, in order to be completely confident in the performance of the combination of motherboard and expansion card you have chosen, in the case you need to look at the compatibility lists of the manufacturers of both devices.

PCI-X 2.0- further expansion of the capabilities of PCI-X 1.0, added speeds of 266 and 533 MHz, as well as parity error correction during data transfer. (ECC). Allows splitting into 4 independent 16-bit buses, which is used exclusively in embedded and industrial systems, the signal voltage is reduced to 1.5V, but the connectors are backward compatible with all cards using a 3.3V signal voltage.

PCI-X 1066/PCI-X 2133- projected future versions of the PCI-X bus, with resulting operating frequencies of 1066 and 2133 MHz, respectively, originally intended for connecting 10 and 40 Gbit Ethernet adapters.

For all variants of the PCI-X bus, there are the following restrictions on the number of devices connected to each bus:
66MHz - 4
100MHz - 2
133MHz - 1 (2, if one or both devices are not on expansion boards, but are already integrated on one board along with the controller)
266.533MHz and above -1.

That is why in some situations, in order to ensure the stability of several installed devices, it is necessary to limit the maximum frequency of the used PCI-X bus (usually this is done by jumpers)

CompactPCI- a standard for connectors and expansion cards used in industrial and embedded computers. Mechanically not compatible with any of the "common" standards.

MiniPCI- a standard for boards and connectors for integration into laptops (usually used for wireless network adapters) and directly to the surface. It is also mechanically incompatible with anything other than itself.

Types of PCI expansion cards:

Summary table of constructs of cards and slots depending on the version of the standard:

Summary table of compatibility of cards and slots depending on the version and design:

	Cards
Slots	PCI 2.0/2.1 5B	PCI 2.1 generic	PCI 2.2/2.3 universal	PCI64/5B (33MHz)	PCI64/universal	PCI64/3.3B	PCI-X/3.3B	PCI-X universal
PCI 2.0	Compatible	Compatible	Incompatible		Limited compatibility with performance loss	Incompatible
PCI 2.1	Compatible	Compatible	Limited compatible	Limited compatibility with performance loss	Limited compatibility with performance loss	Incompatible
PCI 2.2	Compatible			Limited compatibility with performance loss	Limited compatibility with performance loss	Incompatible	Incompatible	Limited compatibility with performance loss
PCI 2.3	Incompatible	Limited compatible	Compatible	Incompatible	Limited compatibility with performance loss	Incompatible	Incompatible	Limited compatibility with performance loss
PCIB 64/5B(33MHz)	Compatible	Compatible	Limited compatible	Compatible	Limited compatibility with performance loss	Incompatible	Incompatible	Limited compatibility with performance loss
PCI64/3.3B	Incompatible	Limited compatible	Compatible	Incompatible	Compatible	Compatible	Limited compatibility with performance loss	Limited compatibility with performance loss
PCI-X	Incompatible	Limited compatible	Compatible	Incompatible	Compatible