What Is a Computer Processor?
The Central Processing Unit, known and commonly referred to as the CPU, is the brain of a computer, as without it there wouldn’t be a center of operations and thus a PC couldn’t be used. When people are looking to purchase one they should be looking at three main aspects: the core count, the clock speed, and last but not least, the cache memory. These are just part of the processor’s speed equation, as the exact level of performance is a sum of all parts found inside a processor.
What Is Its Function?
The CPU is the heart of a computer, but do you know what its function is? Whenever the computer needs to perform an assignment, the required instructions to do so are temporarily stored in the random access memory (RAM). The processor then fetches these instructions, reading and executing them before it produces the result (or output).
Said instructions are stored to the CPU’s registers and can be browsed at any time by the processor, which can also output and compare the outcome of each function. If there isn’t enough free space in the registers, then the instructions are fetched back to the RAM unless the processor is done with its current workload and frees up space.
Clock Speed and IPC (Instructions Per Cycle)
You may think that in order for Intel, AMD, or any other manufacturer to create the fastest processor, they’d just need to find a way to increase the clock speed. But as mentioned in the introduction, this is far from the truth. The clock speed just refers to the processor’s clock’s cycles per second, which is measured in GHz (or MHz for the old years). The numbers have been steady for the past 7-8 years with only a few exceptions. What hasn’t been steady is the IPC (Instructions Per Cycle).
The IPC refers to the processor’s capacity to read and execute a number of instructions per cycle. This number, however, is hard to find, as companies don’t publicly distribute such information very often. This is because not many people are looking for that. So, for example, if you suppose that there are two processors of the same family with the exact same IPC, the one with the bigger clock speed would be the fastest processor.
A lower clock speed but a higher instruction rate would draw less power from the PSU and have reduced temperatures, as opposed to a CPU with higher clock speeds. This was the case with AMD’s FX series, as their performance was great but they would become hot and draw a lot of power under a heavy workload. There were multiple cores just to face this problem as the workload was equally distributed to many places.
Many times each second, the CPU requires very fast access to the computer’s memory: something that RAM isn’t always able to provide. This is where the cache memory comes in play. Each processor features different amounts of L1, L2, and L3 cache memory. Information is stored in them in ascending order. So when the CPU wants to fetch information fast, it will look for it in the L1 cache memory, and if it doesn’t find what it’s looking for then it will move to L2 and L3 accordingly.
The cache memory is usually very small. L3 cache memory is the most important out of them all, and when choosing a processor you should make sure that there is at least 1MB of L3 cache memory for each core. You wouldn’t need that much more, however, because the CPU has an average 80% success rate of predicting how much information and from where it will draw from.
What Is the Fastest Processor Today?
Starting with AMD we will be talking about three CPUs. These are the Ryzen 7 2700X, the Threadripper 1950X, and Threadripper 2990WX.
Ryzen 7 2700X
As far as average consumers go, the Ryzen 7 2700X is the fastest processor manufactured by AMD. This little beast has 8 cores and a base clock speed of 3.7GHz with a boost clock speed of 4.3GHz. Plus, there are 16 threads and 16MB of L3 cache memory. This model is sold alongside AMD’s own stock cooler, which is fairly good as it won’t reach temperatures greater than 85 degrees Celsius under maximum workload. It does, however, require a fairly high amount of power at 105W. It fits an AM4 socket.
Next up is the Threadripper 1950X which is designed to withstand extreme workloads such as professional video editing, 3D rendering, and multitasking. It features 16 cores and 32 threads with a base clock speed of 3.4GHz and a boost clock speed of 4GHz. There are 32MB of L3 and 8MB of L2 cache memory. It requires a whopping 180W to work, so you better have the budget to support that. It fits an sTR4 socket and can reach a maximum temperature of 68 degrees.
Just recently, AMD started selling the latest Threadripper: the 2990WX. This is currently the fastest processor the company has ever created, and we can’t imagine what a monster of a PC would rock that CPU. That being said it “only” has 32 cores and 64 threads, with a base clock speed of 3GHz and a boost clock speed of 4.2GHz. It has a combined cache memory of over 80MB and requires 250W to run! The maximum temperatures and the socket remain the same, however.
All three processors feature AMD’s SenseMI and CMT (hyper-threading) technologies for maximum efficiency and great performance. The Ryzen 7 features StoreMI, and the Threadripper 1950X has FuzeDrive, both of which are similar to Intel’s Optane Memory technology, which takes the workload of your HDDs and turning that into the performance of an SSD.
Moving on to AMD’s competitor, Intel has its fair share of options, but we will be talking about the i9-9900K, the i5-8086K, and last, the i7-7980XE. We decided to stay away from the Xeon series as those CPUs are mostly included in professional-grade servers rather than consumer-level computers.
i9-9900K and i5-8086K
Kicking things off with the i9-9900K, this CPU is equipped with 8 cores and 16 threads, featuring a base frequency of 3.6GHz and a maximum turbo frequency of 5GHz. Also, it has 16MB of SmartCache and a TDP of 95W. The i5-8086K, on the other hand, might be a bit cheaper as it has 6 cores and 12MB of SmartCache but its base frequency is 4GHz with the maximum turbo frequency reaching the 5GHz mark. This technically makes it the fastest processor that is commercially available on planet Earth. It has the same TDP though.
Despite the fact that this processor was first sold back in 2017, it still remains the best consumer-level Intel CPU, as it features 18 cores and 36 threads. Naturally, the base frequency has to be lower with so many cores. The i9-7980XE sits at 2.6GHz with a maximum turbo frequency of 4.2GHz which is quite the jump if we may say so. The frequency of some cores can even reach the 4.4GHz mark thanks to Intel’s own turbo boost max technology that provides increased performance to the best performing cores. Last, it has a total cache memory of 24.75MB and a TDP of 165W.
Just like before, Intel’s technologies, like their own Hyper-threading and Intel Optane Memory, will improve on the computer’s combined performance.
It’s clear that Intel processors require less power to perform compared to AMD, but this doesn’t prove that the latter has the fastest processor. Technically, the i5-8086K is still the fastest CPU you can buy, but this alone won’t cut it. When put to the test, the i7-7980XE performed slightly better than the brand new Threadripper 2990WX in gaming, but these CPUs aren’t designed for that. According to UserBenchmark’s website, the i7-7980XE is the 4th fastest processor; but the 2990WX is currently on the top.
The fact that the Threadripper 2990WX is brand new isn’t irrelevant, though as we have yet to find out whether AMD has finally succeeded in beating Intel by providing a better performing CPU at a lower price point. Chances are that there won’t be any problems though even if you take into account that all 32 cores couldn’t be easily used on the maximum level even from a professional user. The question remains though. Is the i5-8086K the fastest processor out there or has Intel been dethroned from the 2990WX? The answer is for you to decide!