Lably
Skip to content

Accelerated Graphics Port

Expansion bus standard

  • 19 min read
  • 7 chapters

Related on Lably

Through Obsolescence, Computer and Microsoft Windows

Advantages over PCI

AGP is a superset of the PCI standard, designed to overcome PCI's limitations in serving the requirements of the era's high-performance graphics cards.

The primary advantage of AGP is its independence from the PCI bus, providing a dedicated, point-to-point pathway between the expansion slot(s) and the motherboard chipset. The direct connection also allows backward-incompatible changes such as a lower signal voltage, which in turn allows a higher clock speed.

The second major change is the use of split transactions, wherein the address and data phases (request and response phases) are separated. More than one transaction can be in-flight. This allows READ transactions to perform equally well to WRITE transactions, which is hard to achieve with just strictly serial transaction processing where the bus must remain idle (in a wait state) while waiting for the read data access to finish (on the other hand, with or without queues, a "fire and forget" write transaction can be complete when it is received and before the actual memory write completes). With AGP's queued transactions, the bandwidth realized can be close to the theoretical maximum bandwidth. AGP has a high priority queue and a low priority queue to allow for fine-grained optimization. This great improvement in memory read performance makes it practical for an AGP card to read textures directly from system RAM, while a PCI graphics card must copy it from system RAM to the card's video memory. System memory is made available using the graphics address remapping table (GART), which apportions main memory as needed for texture storage. The maximum amount of system memory available to AGP is defined as the AGP aperture. Thus a trade-off could be realized in which the newly gained bandwidth could be exchanged for a lower parts count (graphics cards with little or no onboard memory).

Third, AGP allows (mandatory only in AGP 3.0) sideband addressing, meaning that the address and data buses are separated, so the address phase does not use the main address/data (AD) lines at all. This is done by adding an extra 8-bit "SideBand Address" bus, over which the graphics controller can issue new AGP requests while other AGP data is flowing over the main 32 address/data (AD) lines. This results in improved overall AGP data throughput.

Finally, PCI bus handshaking is simplified. Unlike PCI bus transactions, whose length is negotiated on a cycle-by-cycle basis using the FRAME# and STOP# signals, AGP transfers are always a multiple of 8 bytes long, with the total length included in the request. Further, rather than using the IRDY# and TRDY# signals for each word, data is transferred in blocks of 4 clock cycles (32 words at AGP 8× speed), and pauses are allowed only between blocks.

Sign in to keep reading

6 more chapters are free with a Lably account.

More in Computing and communication