Central Processing Unit (CPU) or processor is the component which carries out the tasks different programs give to it. It can be described as the brain of the computer because it does the data processing.
Processors ability to perform tasks is affected by its clock rate, which is measured in Gigahertz (GHz). Gigahertz is a unit of frequency, the higher the GHz, the faster the CPU. The CPUs ability to perform tasks is also affected by the number of cores and threads it has. The more cores and threads the more work the processor is able to do. The size of the cache affects the CPUs ability to perform as well. Nowadays, the clock rate isn’t the only thing to check when buying new processors since the number of cores and threads is affecting the CPUs performance.
Here are the three main things that affect CPUs performance:
- Core means an independent Central Processing Unit in a single processor. If a CPU is said to have 6 cores, it means it has 6 independent processing units inside. Having multiple cores is the most efficient way to improve the performance of your computer.
- Threading is a technique used by cores to perform multiple tasks at the same time (side by side). Threads are used to add the CPUs processing power. Threading can be described as multitasking.
- Cache is the CPUs own fast memory area. The cache is located on the CPU and it helps it to perform better. Modern CPUs have multiple levels of caches, for example, L1, L2, L3.
Selecting between Intel and AMD
When selecting the right CPU for your custom build, you need to know what kind of things you need the PC for. If you mostly use your computer lightly, you don’t need a high-end (expensive) CPU. Maybe, in that case, you don’t need to even build your own computer. However, if you play a lot of new games or use applications that require a lot of processing power, then you need to focus on what kind of processor you are going to get to fill all your needs.
Examples of CPU intensive tasks:
- Resident Evil
- 3D Rendering
- Video Encoding
- Compression or decompression tasks
Another thing is that different programs benefit the more cores and threads they can use. If you do multiple tasks at the same time, your system will also benefit from having more cores and threads to run these tasks in. For example, running virtual machines is easier with many cores, but all games can’t always benefit from all the cores and threads in your system. This depends on how the game or program was built, some are able to use more cores or threads than others, this is something to keep in mind since it really depends on the software.
When selecting between Intel and AMD processors, I courage you to find more information about the tasks you usually do or would like to do. For example, check what the game developer/software company is suggesting as the minimum or default requirements for your processor. It will help you find out what kind of CPU you need to have to run the game or program smoothly.
Keep in mind, that there are differences between the CPU manufacturers. However, the differences aren’t that big anymore, many people have told me recently that they picked Intel or AMD processor just because they liked it. One thing to consider is, of course, the price of the CPU.
Processors – picking the right one for your custom build
I will be covering few things that might be confusing for people who are new to computers and especially processors. Let’s start with Intel’s processors and after that continue to AMD processors. I will be covering some basic information about the CPUs and what things to look for when selecting one. If you are planning to learn more about processors, keep reading, because this post will give you the basic knowledge about CPUs.
First of all, generation of the processor does matter. How much, it really depends. I won’t go deeper into the differences between the old and new CPU generations. Feel free to search more about that if you are interested. In short, the newer the generation, the newer the CPU.
We are focusing here on the Intel Core processors. Intel does have many others processor families (Xeon, Atom and so on) but the Core is the one intended for consumers, we are also focusing on that. Let’s first see how Intel names their processors.
Core i7, i5 or i3 – what’s the difference?
I see a lot of people talking about Core i3, Core i5, and Core i7 being the generations, but this is wrong. Those are just the product lines Intel has published. For example, there are Core i3, Core i5 and Core i7 CPUs in all the new generations (Coffee Lake, Kaby Lake, Skylake…). The main difference is that Core i3 has the smallest number of cores, threads, and cache. Core i7 has the highest number of cores, threads, and cache. Naturally, the Core i5 lands somewhere between the Core i3 and Core i7. This affects the CPUs processing power as we learned earlier. One thing to keep in mind is that Core i7, i5, and i3 processors have also other differences between them that only the number of cores, threads, and cache.
Remember that we are talking about desktop processors now, mobile processors don’t follow the chart above. The terms Hyper-Threading and Turbo Boost Technologies are explained later.
Intel does have Core i9 processors also, but the performance of those beasts is way more than you need in your home/gaming PC. The price is pretty good as well.
Clarifying Intel’s processor naming scheme
The first part of the Intel Core processors name comes from the processor family and the product line, which in this case is Core i7, i5 or i3. Now, let’s take a look at the 4-digit number after the product line part. The first digit means what generation the processor is. The three last digits just mark the product number, it’s nothing very important for us to know.
There is sometimes a suffix after the product number, you mostly see K or no suffix at all.
If there is no suffix, it means that the CPU is just a normal CPU without any special properties.
- K = CPU is unlocked (This means that it can be overclocked.)
- T = CPU has power-optimized lifestyle.
- U = Ultra-low power CPU.
- H = CPU has high-performance graphics integrated.
Let’s break down the “Intel Core i5-8600K” processors name using the information we learned above. It’s a generation 8, mid-range performance CPU and it is unlocked, meaning that it can be overclocked. Easy!
Now you know the Intel’s way of naming their Core family processors. It isn’t very complicated after all and when you memorize this you are able to browse through the CPU ads very quickly. This saves time when selecting your CPU since you don’t need to open every ad separately and maybe even google the specs. The most commonly used terms and abbreviations in the ads are explained later in this post.
On the AMD processors, we are observing the Ryzen family of CPUs. As well as Intel, AMD has different processor families, but as said we are focusing on the Ryzen. The reason is that these are the newest processors made for consumer use. Unlike Intel’s Core family of processors, all AMDs Ryzen processors are unlocked already. This means that every Ryzen CPU is possible to overclock.
Ryzen 3, 5 or 7 – what’s the difference?
Just as in Intel’s Core i3, i5 and i7 the Ryzen 3, 5 and 7 are the product lines of these processors. They are made for different use cases and labeled according to it.
Ryzen means the processor family. The number 3, 5 or 7 means the class of the CPU. Lowest class is indicated by the number 3 and it is the CPU for normal use. High-performance CPU is indicated with the number 5, and 7 is considered as enthusiast CPU with the highest performance.
Differences between AMD Ryzen 3, 5 and 7 processors are that Ryzen 3 has the smallest number of cores, threads, and cache. Ryzen 7 has the most cores, threads, and cache and Ryzen 5 processors go somewhere between those. Ryzen 3 processors are competing with Intel’s Core i5 processors, which is the model to buy if you want to just play. If you are planning to stream at the same time, aim for the next class of processors.
Just like Intel, AMD has also some beast CPUs with very high performance. AMD’s version is called Threadripper. These are also an overkill for your home system, in performance and in price.
Clarifying AMD’s processor naming scheme
AMD’s naming of the processor is hard to explain by typing and even harder to understand by reading. Below is a chart explaining the name of a Ryzen 5 1600X processor.
AMD also uses suffixes giving us a bit more information about the processors’ properties. If there is no suffix given, it just means that it’s a standard CPU without any special properties. AMD’s suffixes are explained below, the meaning of some of them is explained later in this post.
- X = High-Performance CPU with XFR
- G = CPU has integrated graphics
- T = low power CPU
- S = low power CPU with integrated graphics
Now you have learned how AMD names their Ryzen processors. In the end, it isn’t that hard and knowing it will help you find the one you are looking for so much faster.
Write down or memorize what is the socket and generation of your CPU you need to know these when you are selecting your motherboard!
Complex CPU terms and abbreviations explained
I have created a list including the most commonly used terms and abbreviations in processor ads. This is the data people who are new to buying processors or computers will actually benefit from, this helps to understand what they are buying. Making a decision is much easier when you know exactly what all the terms used to advertise these processors actually mean.
There are common terms that apply to both Intel and AMD processors, but there are also different technologies trademarked or used by these manufacturers. I have listed first the terms that apply to both of the manufacturers. After that there are manufacturer specific terms, I have marked them with the name Intel or AMD.
- Socket - The type of the socket this CPU uses and fits. (Motherboard must use the same socket type).
- Cores - The number of cores the CPU has.
- Threads - The number of threads the CPU has.
- Processor Base Frequency - The CPUs operating frequency, measured in GHz. Base frequency defines TDP.
- Maximum/Turbo Frequency - The CPUs maximum frequency, measured in GHz. The CPU is able to speed up to this frequency under a heavy workload.
- Cache - The amount of memory the CPU has.
- TDP/Wattage - TDP means the Thermal Design Point. TDP is the CPUs power usage in watts. Can sometimes be referred as wattage
- Memory Type (Support) - What type of memory it supports and what is the default memory speed it supports. (Coffee Lake supports only DDR4 type memory. Reaching faster memory speed is possible with overclocking. The memory needs to support higher speeds and the motherboard must be compatible with RAM overclocking.)
- EEC Memory Supported - This means that the processor supports Error-Correcting Code (ECC) memory. EEC memory detects and fixes internal data corruptions. (Motherboard must also support EEC memory)
- Processor Graphics - This tells which graphics processing circuitry is integrated into the CPU.
- Technology/Lithography (nm) - This refers to the size of transistors inside the CPU. They are measured in nanometers, the smaller the better.
- Memory Channels - Most common is dual-channel architecture, which means that memory sticks should be added or removed as pairs.
- Intel Smart Cache - Technology that makes the cache memory sharing between different execution cores faster.
- Intel Hyper-Threading Technology - This means that the cores in the CPU are using threads. One physical core is divided into two threads. This enhances the CPUs total processing power since it can run tasks side by side.
- Intel Turbo Boost Technology - This means that the processor is able to increase its clock speed over the maximum frequency if needed and if there is thermal and power headroom to do this. Gives an extra kick to your system during high workloads and saves energy when it’s not needed.
- Intel Optane Memory Support - Non-volatile memory between storage and system memory. This enhances the system’s performance by making load times faster (for example, file search or opening programs).
- AMD L2/L3 Cache - L3 cache is often shared between cores to improve memory sharing. L2 cache is a private cache dedicated to the needs of each particular core.
- AMD XFR - This means Extended Frequency Range, it allows the CPU to boost over the Turbo Frequency. It gives more power to your CPU momentarily if the CPUs temperatures allow it at that moment.
- AMD Thermal Solution - This is noted so you know to get thermal paste separately if needed. Some CPUs don’t come with thermal paste, which is used between the CPU and cooler.
- AMD Cooling solution - Some of the AMD's processors don't come with an air cooler. Make sure when ordering that you will have a cooling solution for your CPU.
Cooling your CPU
Cooling your CPU is very important and it should be done properly. If you don’t cool your processor enough it will affect the performance and lifespan of your CPU. If the processor is getting too hot while in use, it will throttle its speed. That said I think everybody understands that it’s a good idea to cool your CPU properly to avoid problems and get the results you paid for.
Benchmarking your CPU
Benchmarking your CPU is very popular, it’s a way of knowing how well your CPU is performing in the tasks. It is also an easy way to compare different CPUs. Benchmarking doesn’t affect your computers ability to perform faster and better, it is simply a method to find out how it performs in different situations. Running benchmarks in your system can also reveal different problems in your system. For example, that you don’t have enough cooling for your CPU. Benchmarking can be a useful tool for troubleshooting the performance issues of your system.
Popular CPU benchmarking software are CineBench, GeekBench, and RealBench.
CineBench tests your CPUs ability to do rendering, it is a very good benchmark to gauge the performance of the CPU and compare it to other CPUs. This is called render benchmarking.
GeekBench and RealBench are testing the CPUs performance in a so-called real-world situation. Real-world benchmarking simply means that the CPU is assigned to perform tasks like multitasking. The result is closer to what the CPU really have to process during normal usage.
Overclocking your CPU
Overclocking your CPU means that you are pushing your CPU over the default frequency it’s normally working. You are basically squeezing all you can get out of it. However, overclocking is very individual, some CPUs you can’t overclock at all, no matter how hard you try. This depends a lot on the CPU and you can’t really affect how your CPU will behave when overclocking it.
I don’t suggest overclocking if you are building your first PC because overclocking is safe only if you know what you are doing. When building your first PC, it’s better to concentrate on understanding how your computer works. When you have more knowledge about your system, how it performs and understanding of overclocking, then it might be something for you to try.
Overclocking might cause problems because the CPU is designed to be totally stable in the default clock rate it is working out of the box. The more you do overclock the more you understand the limits different CPUs have. Overclocking can make the CPU unstable if you don’t do proper testing before putting it to real use. And one thing you do want from your CPU is that it’s stable and won’t crash during gaming or working.
Overclocking will also affect the CPUs power consumption and make it produce more heat. If you are planning on overclocking your CPU you should have good cooling, in this case, water cooling is something to consider.