front side bus

(redirected from Memory bus)
Also found in: Wikipedia.

front side bus

(FSB) The bus via which a processor communicates with its RAM and chipset; one half of the Dual Independent Bus, the other half being the backside bus. The L2 cache is usually on the FSB, unless it is on the same chip as the processor In PCI systems, the PCI bus runs at half the FSB speed.

Intel's Pentium 60 processor used a bus speed and processor speed of 60 MHz. All later processors have used multipliers to increase the internal clock speed while maintaining the same external clock speed, e.g. the Pentium 90 used a 1.5x multiplier. Modern Socket 370 motherboards support multipliers from 4.5x to 8.0x, and FSB speeds from 50 MHz to a proposed 83 MHz standard. These higher speeds may cause problems with some PCI hardware.

Altering the FSB speed and the multiplier ratio are the two main ways of overclocking processors.

Toms Hardware - The Bus Speed Guide.

Toms Hardware - The Overclocking Guide.
Mentioned in ?
References in periodicals archive ?
However, the card is supposed to have a smaller memory bus of just 160-bit, compared to the 192-bit bus available to both the 3GB and 6GB variant of the currently available version of the GPU.
Winning the Big Lottery's People's Project and taking delivery of the Memory Bus - our own mini bus - which we use to offer day trips for people with dementia.
Assuming a 64-bit (x64) wide memory bus, this can be viewed as transmitting over 34GBs of data per second.
The minimum amount of memory - 64 MB, the frequency of the memory bus - 2133 MHz, video connector on the rear panel - VGA x 1, HDMI x 1 DVI x 1, built-in sound chip, integrated network card, the minimum number of connectors external PCI-E 16x - 2 pcs.
As AnandTech points out, the A7 processor chip on the iPad Air is capable of achieving this remarkable feat despite featuring a smaller 64-bit memory bus in comparison to the 128-bit memory bus on iPad 4's A6X chip.
Additional memory can be added by using a multi-module memory bus to form a multi-drop topology.
The ZOTACEeA GeForceEeA GTX 460 1GB graphics card features 1GB of high-speed GDDR5 memory connected to a 256-bit wide memory bus while the ZOTACEeA GeForceEeA GTX 460 Synergy Edition has 768MB of GDDR5 memory connected to a 192-bit memory bus for.
2]C and SPI bus analysis, DPOJET for jitter and eye diagram analysis, DDRA for DDR memory bus verification, SDLA for EQ/channel emulation and analysis, and SignalVu for frequency-domain display and analysis.
Elpida's product reduces the number of mounted memory chips by 50% besides power consumption, simplifies the high-speed memory bus design and conserves chip mounting space.
11 a/b/g) embedded core modules have a multilayered memory bus architecture that allows simultaneous data transfer between processor, memory, and peripherals.
Its primary benefit is to allow the memory bus to run at a faster speed than the actual memory chips can sustain, thus enabling higher peak throughputs.
With remote direct memory access (RDMA), an addition to the IP standard, computers can directly place information onto another computer's memory with minimal demand on the memory bus and on the CPU, replacing TCP/IP copy requirements that add latency and consume CPU and memory resources.