Showing posts with label Build a Computer. Show all posts
Showing posts with label Build a Computer. Show all posts

Friday, August 23, 2013

Memory Standard Definition

Memory standard is referring to the type of memory being used. The current memory standard most commonly used is DDR3. This stands for double data rate 3. DDR3 is usually delivers speeds between 1066 MHz and 3000 MHz. For every new version of DDR memory there is a increase in attainable speeds. It is expected that DDR4 memory speeds will vary between 2133 MHz and 4266 MHz.

Thursday, August 22, 2013

Form Factor Definition

A form factor is an industry standard that ensures that all components of a computer fit together properly. The most commonly used form factor for custom PC builders is ATX. There are other form factors such as Micro ATX, extended ATX, HPTX, mini ITX and so on. A computer case will have a list of compatible form factors. Motherboards and power supplies must be the same form factor as the case.

Monday, August 19, 2013

Tops Reasons to Build a Computer

Lets face it. There is nothing more awesome than to be able to say that you built your own computer. But why would somebody want to go through the trouble in the first place?

  1. Experience

    Building a computer is actually an easy process once you learn what parts to buy. Everything is usually explained in the motherboard manual how to piece everything together. Seriously, just knowing how to take your computer apart makes things so much easier when you want to upgrade expand upon your current build.
  2. Long Term Cost

    Now days, computers have really become like appliances in our daily lives. We absolutely need them but we don't have to buy a new one every two or three years to stay up with the times. However, there are some parts that do need to be replaced from time to time. Places like Best Buy only care about making the computer as cheap as possible. This often restricts you in several ways such as having weak power supplies which can limit how good of a video card you can upgrade to if you decide your computer is too slow. When you build your own PC, you can design it so you can upgrade it later and it won't be stuck in a outdated state.
  3. Customize

    The most fun part of building a computer is choosing exactly what goes into it. If you want a gaming PC you can shift your funds to buy a better video card. If you need to do a lot of encoding you can invest in a better CPU. You can personalize it so it is the perfect computer for your needs.
  4. Easier to Repair

    If a part goes bad, it is super easy to replace because you know exactly what you bought in the first place. It's a nightmare when your motherboard goes bad. You have to disassemble the whole thing! If you have a pre-built computer, sometimes it's a huge hassle trying to get the power connectors from the chassis to match the motherboard. Whereas if you build your own computer, it's very easy to buy a motherboard with a standard form factor that actually fits everything right.
  5. Manufacture Warranty and Quality of Components

    When you build a computer, you know the reputation of all the parts you put into it. You don't have to settle for cheap junk. You can sort by ratings and still find one that's priced decently. Not only that but you have manufacture warranties to back your parts up. These can easily be up to three to five years. By picking and choosing your parts, you know you have a reliable computer. If something does go wrong, you know you don't have to worry about spending more money because you have a manufacture warranty on your side.
  6. Easier to Overclock!

    CPU's might not be improving as quickly anymore but they are still very easy to overclock. If three years go by and your computer is starting to get left behind, bring it up with the times by overclocking it. You can easily buy a cooler for $20 to $30 that will make this happen. I usually like to wait until my warranty on my CPU is over before I overclock it. That way I don't void it early. Once the warranty is up, overclocking can make a significant difference in your computer's performance.

Wednesday, September 5, 2012

Cas Latency or Timing Definition

The simplest way to think of cas latency (or timings) is by associating it with how much delay your ram will have. Timings represent the gap between the processor sending information to the RAM and the RAM responding to it. High cas latencies will result in slower RAM.

You can expect that as memory becomes faster it will naturally have higher timings. For example, memory with a speed of 1066 Mhz will likely have a cas latency of 5-5-5-15. Whereas, memory with a speed of 2133 Mhz will possibly have a cas latency of 9-11-10-28. Aim to find RAM with a high speed first. Then focus on finding one with the lowest cas latency.

Thursday, August 23, 2012

SLI and CrossFire Support Definition

When a video card has SLI or CrossFire support, more than one video card can be used in the same computer. Using multiple Nvidia video cards is referred to as SLI or "Scalable Link Interface". Using multiple AMD video cards is referred to as "CrossFire". This can greatly increase performance if the program being used supports SLI or CrossFire. Otherwise performance may even become worse.

This is convenient because a computer can be built with one video card at first and still have room to upgrade without getting rid of your old card. Using two of Nvidia's card usually requires the cards to be the same model with a few exceptions. AMD, on the other hand, allows a little more flexibility as long as the first two numbers in the card's name are the same. This means that a Radeon HD 7870 can be paired up with a 7850 but not a 7750. This chart at Tom's hardware explains more about what cards can be paired together.

Check beforehand whether your motherboard supports SLI or CrossFire. Your power supply may also need to support SLI or CrossFire when pairing some of the higher end cards.

Monday, July 23, 2012

Ram Amount Definition

The amount of ram is the amount of space, usually in gigabytes, that the ram has. The more ram a computer has, the more programs a computer can open at one time. Having too little ram will make the computer run slow. Most modern computers will have 2 to 16 gigabytes of memory. Programs and video games will often have the minimum requirements of ram listed on them.

Note: The reason why a computer runs slow when it doesn't have enough ram is because it is trying to find another way of storing the information. There is a portion of space set aside on the hard drive for this purpose. The downfall is hard drives are incredibly slow. When they are forced to step in, programs will take a long time to open.

Thursday, June 7, 2012

Video Card - Memory Type Definiton

Video cards use different versions of dedicated memory. GDDR5 (Graphics Double Data RAM version 5) is commonly used on most modern video cards. Newer versions have greater bandwidth and faster speeds allowing for better performance.

Note: The video card's memory type is completely independent from the motherboard's memory type and doesn't have to be the same.

Saturday, April 21, 2012

Video Card Dedicated Memory Amount

Video card dedicated memory is RAM that is embedded into the video card for its use only. Having more memory helps games with larger textures, higher resolutions and so forth to achieve better frame-rates. This memory is completely separate from the operating system's memory. Both are necessarily for smooth performance. Dedicated memory can range anywhere from 512MB's to 4GB's on a modern desktop computer.

Thursday, April 19, 2012

Video Card Interface Definition

A video card's interface is the type of connection it uses to attach to the motherboard. The most modern type of interface is PCI Express 3.0 x16. Every PCI-express version is compatible with the others but they will only run as fast as the weakest part. This could be the motherboard or the video card.

PCI Express Speeds:
PCI Express 3.0 = 8Gt/s
PCI Express 2.0 = 5Gt/s
PCI Express 1.0 = 2.5Gt/s

PCI Express cards also come in different lengths as indicated by the number of lanes. The most common amount of lanes for video cards are x16.

PCI Express Lanes:
PCI Express x16
PCI Express x8
PCI Express x4
PCI Express x1

DirectX Definition

DirectX contains a number of programs that handle multimedia for things like video games. Newer versions allow for more efficient processing and can greatly increase graphics quality and frame-rates. DirectX is embedded into the video card and cannot be updated without the purchase of a new video card. The latest version is DirectX 11 and is used by the Radeon HD 5000, 6000 and 7000 series and by the Geforce 400, 500 and 600 series.

Tuesday, April 17, 2012

Stream Processor Definition

Stream processors are what video cards use to process the image on the screen. Doubling the stream processors can achieve very close to double the performance. Each one is similar to a core on a CPU but much less powerful. There are several other major factors in determining the final performance of a video card.

Note: Not all stream processors perform the same. Nvidia usually uses fewer but more powerful stream processors than AMD. In the end, both company's video cards perform close to the same.

Monday, April 16, 2012

Socket Type Definition

A CPU socket is the location where a CPU is secured on the motherboard. A socket is what allows the processor to communicate with the rest of the computer. Sockets make it very easy to install and remove CPU's. Each CPU generation will require a specific socket type. The motherboard and the processor must use the same socket.

Intel's Socket Types:
LGA 1150 - Used by Haswell CPUs (4th Generation i3, i5 and i7's)
LGA 1155 - Used by Sandy Bridge (2nd Generation i3, i5 and i7's) and Ivy Bridge CPUs (3rd Generation i3, i5 and i7's)
LGA 2011 - Used by high end i7's

AMD's Socket Types:
AM3+ Used by Vishera, Zambezi and other CPU's
FM1 Used by Llano CPU's
FM2 Used by Trinity and Richland CPU's

Sunday, April 15, 2012

Architecture or Core Definition

The architecture or core is how the ISA or instruction set architecture is designed in the CPU. Newer architectures greatly help with efficiency and allow for more performance per gigahert. It is important to always buy the latest architecture. Here is a wiki article containing a list of Intel and AMD's architectures.

Monday, April 9, 2012

Hyperthreading Definition

Hyperthreading is a technique used by Intel to help with multitasking. It makes every physical core act as two logical cores or "threads". This means that a dual core will have four threads and a quad core will have eight total. Every thread is good at doing its own task. How does the CPU accomplish this? It does this by making certain parts of the CPU work more often that would normally be underutilized. This can nearly double performance in highly optimized situations. More realistically though, it only improves performance marginally.

Friday, April 6, 2012

Integrated Memory Controller Definition

An integrated memory controller helps manage data coming from the RAM. This often comes as dual or triple channel. Using triple channel adds an extra bus that allows for faster data transfers. The integrated memory controller is rated to run at specific speeds in MHz. It is possible to benefit from RAM that runs faster than the memory controller's speed if the motherboard supports higher speeds. However, the performance increases may diminish rapidly or perform worse.

Multi-Core Definition

Multi-Core CPU's have multiple processors in them. Most CPU's come with either 2, 3, 4, 6 or 8 cores. All the cores will run at the same speed. This means that if you have a quad core that runs at 3.3GHz, you will have four processors all running at 3.3GHz. Having more cores helps a lot with handling multiple tasks at one time. Some programs or games are built to fully utilize multi-core processors. Otherwise the other cores will remain idle with little or nothing to do.

Thursday, April 5, 2012

Manufacturing Tech Definition

Manufacturing tech is the die size of the CPU. This is usually measured in nanometers or nm. A CPU with a smaller die size will perform better per gigahert and use less energy.

Tuesday, April 3, 2012

Cache Definition

Cache is very fast memory that is located in the processor chip. Cache is located in three levels: L1 being the fastest, L2 and L3 being the slowest.
  • L1 is usually located on the processor die itself but is limited in the amount of space that can be used.
  • L2 is more cost effective but a little slower because it is farther out from the processor die. 
  • L3 can be located either on the CPU chip or the motherboard and is larger than L2 cache.