- What is CPU cache?
- What is FSB or Front Side Bus?
- What is Hyper Threading or HT technology?
- What is the CPU memory controller?
- What is the difference in CPU sockets?
- What is Registered Memory?
Q: What is CPU cache?
A: Cache, on chip: Super-fast and efficient memory integrated into the processor, which it uses to store short-term, frequently accessed data. This type of memory is so zippy because it runs at the full clock speed of the processor, meaning at the gigahertz level. Cache is the first place the CPU looks for instructions. If not found, then the CPU must rely on the much slower system RAM, which operates at the megahertz level. Most CPUs today have on-chip L1 (level 1) and L2 cache. Some also have on-chip L3 cache. The L2 cache size will often be one of the main differences between CPUs. As a general rule of thumb, the larger the L2 cache, the better the performance.
Q: What is FSB or Front Side Bus?
A: Front Side Bus is the main highway for data in a PC. It connects the processor, chip set, DRAM, and AGP socket. FSB is described in terms of its width in bits and its speed in MHz.
The FSB interface has been replaced by similar, but newer, technologies on many platforms. For example, Intel has QuickPath Interconnect - and AMD has HyperTransport; both are far faster than the older FSB architecture, and are no longer rated for speed in the same way.
Q: What is Hyper Threading or HT technology?
A: Developed by Intel, which allows the CPU to access previously unused resources to run more than one program (or thread; hence the name) at the same time. Previously, only computers with dual processors could do that, but with Hyper-Threading enabled, your single-processor PC will function in many ways as if it has two processors. Realistically, though, only expect about 1.5 times the performance. Hyper-Threading is still in its youth. As more and more applications are written to take advantage of it, you'll notice even better performance.
Q: What is the CPU memory controller?
A: Historically, the memory controller is integrated into the chipset on the motherboard, but newer CPU's have the memory controller directly on the chip. This improves memory performance significantly and helps lower the thermal output of the chipset.
Q: What is the difference in CPU sockets?
A: This is an area on the motherboard where the CPU is installed. AMD CPUs use the PGA (Pin Grid Array) type, meaning there are pins underneath the processor which are inserted into the socket. Intel CPUs on the other hand are LGA (Land Grid Array) type, meaning the pins are on the socket on the motherboard rather than on the processor.
Q: What is Registered Memory?
A: Registered memory: Registered modules have additional components (registers) placed between the incoming address and control information and the SDRAM components. These modules are typically used in Servers due to their added reliability (they place much less of an electrical load on the memory controller and therefore make it possible to have as many as 16 or 32 modules in a large system). A stick of memory that contains registers will actually hold data for one full clock cycle before it's passed on. A small performance hit is generally incurred as a result. Registered memory is all about scalability and stability.
Would you like to see a question and answer added to this page?
Email your question to us at !