What is a processor?
The processor is one of the most important components in a modern computer. The terms processor, CPU, and chip generally refer to the Central Processing Unit. Every action and command your PC processes depends on interaction with the CPU, so its performance is crucial to the overall computer experience.
What do I need to look for in a processor?
There are thousands of different processors available so it is understandable that your options can be confusing, especially when considering how the different models compare in performance and which one is right for you. For example, you may have seen the word core or thread mentioned when talking about CPUs. This is the biggest change in processor fabrication, and has increased performance many times over. A core is basically a self-contained processor, and multiple cores combine to form the Central Processing functionality. Each core is given an overall task as threads, essentially splitting up the task into smaller parts, and telling each core which task to do and in which specific order. Much the same as the efficiency of an assembly line, with each core being given smaller tasks to complete the overall task in a faster time.
Not that long ago comparing processors was much easier. A single core processor that had higher megahertz (MHz) would make for a faster PC. For example a 733MHz processor (such as the Intel® Pentium® III Processor) would make a faster PC than a 600MHz processor.
Now, modern processors have multiple cores, higher thread counts and a completely different fabrication process, layering cores and adding cache memory to store and divide the list of tasks. To add, processor speeds are now measured in gigahertz (1000MHz is equivalent to 1GHz), and you will usually see 3GHz - 4GHz as the common speeds.
Intel's Thread Director is a great example of how manufacturers have progressed the technology. Intel Thread Director helps "prioritize and manage the distribution of workloads, sending tasks to the most optimized thread. This new feature is on by default and works in tandem with the operating system for intelligent workload distribution."
Today comparing processors is no longer about finding the highest speed processor possible in GHz. To get the best performance you need to take into account the whole specification of the processor including the number of cores it has, thread count, overclocking capability, and how that matches up with your workload. If you are creating 4K videos for YouTube, your needs will be much different from a gamer who just plays a first-person shooter online.
Here is a brief list of what the specifications mean:
Clock Speed: Example: 3.7GHz = 3700MHz. This equates to clock cycles, and the execution of instructions. A 3.7GHz CPU can perform 3.7 billion executions per second. Both Intel and AMD have invested huge amounts of money in developing their processors to maximise the output of each and every clock cycle and in doing so they have made the actual clock speed less important. As a rule, look for both higher core counts and higher clock speeds when comparing two CPUs, because this means increased performance overall.
Nanometres (nm): 1nm = 0.00000001 cm = one billionth of a metre. The easiest way of measuring the technological complexity of a CPU is to measure how many individual transistors can fit on the CPU itself. You will see many processors described as 5nm, 7nm, 10nm and 14nm. As the technology size in nm drops, the size of the transistors decreases, this allows a 5nm processor to have many more transistors in the same physical package as a 10nm chip and increase performance massively. The lower the nanometre the processor is generally the more advanced the processor is. There are other factors, such as distances, densities etc., so it is worthwile not getting hung up on this technical information, and focus on actual performance data.
L1/L2/L3 Cache: For the CPU to function it has to be fed with data to process, and a bottleneck can easily arise if the data is far away (relatively speaking), for example in the system's memory. This will take time to access as the data has to travel through many controllers and a physical distance to the CPU causing a delay. If the CPU doesn’t have to travel outside of itself to get the data, it can process information more effectively. When the memory is built into the processor it is referred to as cache. As a general rule the more cache a CPU has per core the better its performance in programs i.e. the CPU can store more data locally to process tasks.
Whilst this information is educational and useful, the best way to compare CPU benchmarks, and check how they perform doing the same tasks you will use them for.
What are 64-bit processors?
Almost all processors available to buy today are 64-bit capable. The reason processors and operating systems have moved from 32-bit to 64-bit is largely due to the limits in the amount of memory that a 32-bit CPU can access. A 32-bit CPU can only address 4GB (Gigabytes – 4096MB). For many uses this limit is fine however higher end computers often need 4GB or over to complete complicated tasks or get the best performance out of the very latest games.
A 64-bit processor does not face this problem as it allows the processor to access potentially 16 billion gigabytes of memory (16 Exabytes). Having a 64 bit data path also doubles the amount of data that can be fetched at any one time increasing the performance of a 64-bit processor when running 64-bit applications.
All of the current 64-bit processors are backwards compatible to 32-bit operating systems and programs, although if running a 32 bit operating system they will still suffer from the limit of 4GB of memory. Both CPU and Operating System need to be 64-bit to provide the enhanced compatibility and performance.
What is an APU?
Also known as integrated graphics, or on-board graphics, an APU is the combination of a processor with a graphics chip. APU stands for Accelerated Processing Unit. Combined onto the same silicon they improve data transfer rates between them. Today's APUs can run 1080p games on low settings adequately, but however advanced - they are still unable match the performance of a dedicated graphics card.
What is overclocking?
Overclocking is taking a computer component such as a processor and running at a specification higher than the shipped, factory settings. Every part produced by companies such as Intel and AMD are rated for specific speeds. They have tested the capabilities of the part and certified it for that given speed. Of course, most parts are underrated for increased reliability. Overclocking a part simply takes advantage of the unused potential already present within the specified parts, and all unlocked processors have parameters to work to for overclocking.
What are the benefits?
The primary benefit of overclocking is additional computer performance. Most individuals who overclock their system either want to try and produce the fastest desktop system possible or to extend their computer power on a limited budget. You can expect anywhere from 5% - 20% increase in performance, but the higher the increase, the more work you may have to put in. Modern motherboards allow for simple AI overclocking, which is basically a press of a button! However, unlocking the 20% or higher performance is more of an expert task, and requires a lot of experience to pull off.
Interested in overclocking? Read a more in depth article on overclocking here.