When choosing a PC many people start from a kind of preconceived script that takes into account certain aspects that, theoretically, are very important to make the right decision. It is a good idea, of that there is no idea, but there is a problem, and it is that this kind of scripts often contain errorsand in many cases end up giving an inordinate importance to things that really should be more in the background.
Even for the most technically savvy people those less important details can be difficult to appreciatewe have therefore created this guide, in which we are going to explain in a clear and detailed way five things about a PC to which we attach too much importance, and which in reality are not only not so relevant, but can also lead us to make the wrong decision.
In many cases, this excessive importance is due to the fact that we tend to blinded by the figuresthat is, we believe that more always means that something is much better, when in fact the opposite may be true. Thus, for example, there are components that may seem much better than others, but in the end have a very poor price-performance value, so much so that they end up being a bad buy.
Continuing with the previous example, even in those cases where those components can make a difference in performance we find that, in the end, that difference is usually smalland can occur in applications or tools that we are not really going to use. I think the key idea behind this guide is clear, so let�s jump right in and look at those five things.
1.-The operating frequencies of a processor
It has been a long time since the MHz race has long since been supersededbut there are still people who believe that speed is still one of the most important aspects when choosing a new processor. This is not true, and can lead us to make very serious mistakes, since we will be omitting other much more important things to give priority to this one.
The frequencies of a processor job refer to refers to the rate at which a processor’s clock generator is capable of producing pulses. Such pulses are used to synchronize the operations of its components, and although they indicate the speed of the processor (clock cycles per second), they are only one more indicator of its performance, and are not the most important.
The performance of a processor depends primarily on the number of cores you have, of the IPC (instructions per clock cycle), of the number of processes and sub-processes that it can handle in total and, finally, the clock frequency it has. A processor with a low IPC and a high clock speed will offer significantly lower performance than a processor with a high IPC and a low clock speed.
Therefore, the most important parts of a processor are the IPC and the coresfollowed by the total threads (processes and threads) and finally by the operating frequency. This point is the least important of all, contrary to the popular belief that some people are still carried away by.
2.-Premium designs of graphics cards
This is a rather tricky subject, because it is true that a poorly designed graphics card, with a low-quality cooling system or a faulty assembly can end up causing serious problems. However, this is not an issue limited to the cheapest modelsi have seen graphics cards with premium designs that gave temperature problems due to poor contact between the radiator and the GPU and/or graphics memory, and I have also seen budget models offer excellent temperatures.
In the end, giving too much importance to the design of a graphics card can make us pay much more money for a model that will not offer a really big advantage over another seemingly inferior, and much cheaper, model. For example, a Radeon RX 6900 XT can cost us 1.049 euros with a good design (the Sapphire NITRO+ AMD Radeon RX 6900 XT SE OC GAMING) and perfectly functional that will not give us any problem of temperature, but it can also cost us 1.471.58 euros with the premium design of the PowerColor Red Devil AMD Radeon RX 6900 XT, which really won’t make any major difference despite the huge price difference between the two.
The two graphics cards we named in the previous paragraph will offer virtually identical performanceand they will work at completely safe temperatures, but the latter will have cost 400 euros more than the former. That extra expense is not justified in any way, so it is clear that such a graphics card would not have been a good buy.
The conclusion to be drawn from all this is very clear, and that is that the models with premium designs make no real difference to justify their high priceand it is therefore better to opt for more modest designs that are perfectly functional and much more economical.
3.-RAM latency
I know that some of our readers will have taken his hands to his head and will be thinking that this is very important, but the truth is that it is not, because today the differences in latency level between the RAM memory kits that are marketed in the market are usually minimal, and this means that in the end the difference in performance also is very small.
In most cases, one RAM kit running at the same frequency as another but with slightly lower latency will make very little difference in performance. Today, DDR4 kits running at a speed of between 3,200 MHz and 3,733 MHz, with latencies ranging between CL14 and CL18. Well, a kit at 3,200 MHz CL14 and another at 3,200 MHz with CL16 latencies will offer virtually identical performance.
Latency can be important if we are talking about very large differencesor if we are dealing with very high latencies in RAM kits running at very low speed, but this is not the norm in the kits currently on the market, as most of them have minimal differences in latency, and these will not result in a significant increase in performance, quite the opposite.
If we have two kits of 16 GB of DDR4 RAM memory at 3,600 MHz with latencies CL17 and CL16 with a price of 80 euros and 105 euros, respectively, the first one would be a good choice because is cheap and performs almost the same as the second one. Mind you, this is not to say that there are no performance differences between the two, but we are talking between nothing and two or three more FPS in games, at best.
4.-The sequential speeds of an SSD
You may have raised your eyebrow again, but this also has an explanation. The most important performance improvement that marks an SSD drive over an HDD drive is not in sequential read and write speeds, but rather in access times and latency. A hard disk has mechanical parts and suffers from the problem of fragmentation, which means that to perform something as simple as opening a game it has to:
- Activate the head and move it to locate the platter, and the area of the platter, where the file is located.
- Perform the read operation and, if necessary, find where the rest of the necessary files are located, which may be  “scattered” due to fragmentation.
The access times we have with an HDD are very slow, and latencies are high. With an SSD, the opposite is true, and this makes a huge difference, so much so that it is key to achieve that feeling of absolute fluidity and that the computer “flies”.
The sequential read and write speeds are not so important because any current SSD is already moving at values that are good enough to achieve a satisfactory experience. It is true that an SSD that reaches 7 GB/s will be faster than one that works at 3.5 GB/s, but in the end the SSD will be faster than another that works at 3.5 GB/she actual performance difference this will make in applications and games will be minimal. We told you about this in our guide to SSD myths. We also told you that there are very specific scenarios where these speeds can make a big difference, for example when we have to move huge files, but this is not a common occurrence.
5.-The size of a cooling system
Bigger is better, an expression that is not always true and, in fact, ends up being wrong on many occasions. We tend to think that mounting a large cooling system is essential because it will cool the equipment better, or the component to which it is attached, but the reality is that this actually is not as important as we think.
I have seen, and tested, cooling systems with huge radiators and two fans that in the end have recorded worse temperature values than others with a smaller radiator and with a single fan, and there are also a lot of fans made in China that look to attract the user starting from that ideabut which in the end have very poor thermal performance.
A good cooling system is not defined by the size of the radiator or fans, but by the heat transfer which it is capable of achieving, from the radiator fin density and the static pressure that the fan can generate. This explains why a huge cooling system can be worse than a more compact one, so we must be very careful when choosing, because as we have just seen, appearances can be deceiving.
The same applies to aIO liquid cooling kitsthe fact is that the quality of these kits has a great influence on performance, so much so that we can find 280mm kits that are able to offer almost identical or even better performance than other 360mm kits. The Corsair H115i RGB Pro XT kit is one of the best examples, as it offers fantastic performance, and has nothing to envy to other kits that, by size, seem superior, but really are not.
