Generally, when we talk about the speed of a processor we all know the magnitude of megahertz or gigahertz, and today it is easy to see processors that reach or exceed 5 GHz from the factory, or overclock records that approach 8 GHz in some cases. However, you will also have seen news that talk about processors that have reached 500 GHz , a figure that seems almost absurd but, to understand it and not be deceived with these figures, it is necessary that you know the concepts of Clocking and Switching , which is precisely what we are going to talk about in this article.
We are quite used to seeing overclock records, which are broken every so often by professional overclockers who manage to raise the operating frequency of a processor to levels much higher than the speed at which they operate from the factory; however, these magnitudes are usually of the order of less than ten gigahertz.
On the other hand, news of processors that have reached 350 GHz at room temperature and that have reached 500 GHz through cryogenization have also come to the fore, but how is this possible? To understand it, it is necessary to “go down in the mud” to understand the concepts of switching and clocking.
Clocking and Switching are not the same
When we see that an overclocker has broken a speed record in a processor, what it has done (explained in a basic way) is to increase its operating frequency, generally by increasing the voltage.
This produces a higher generation of heat and that is why they usually use advanced cooling systems, such as liquid nitrogen, thus avoiding problems with the system. In addition to increasing the operating frequency of the processor, the frequency of the base bus is also altered, so at the same time the speed of other components that depend on it, such as RAM, for example, is increased.
On the other hand, taking the example of the processor that works at 500 GHz (it was an IBM experiment), a cryogenization system was used for its cooling, but despite that at room temperature it already reached 350 GHz … we are talking about 3 , 5 THz. In the experiment they explained that they used a technology called BiCMOS SiGe for their transistors, but in any case, where is the catch? There is really no trap, but they refer to different things: in one case we speak of clocking , and in another of switching .
Clocking talks about the clock frequency at which the processor works, that is, the speed at which its transistors switch together. For example, when we say that a processor operates at 5 GHz we are actually saying that its transistors are capable of switching (switching between zeros and ones) 5,000,000,000 times every second.
On the other hand, switching is the speed at which a transistor can switch and go from one state to another. This is the case of the IBM example that we have given you before, since they referred to the fact that the transistors used with BiCMOS SiGe technology were capable of switching at a speed of 350 GHz at room temperature or 500 GHz with cooling by cryogenization; In other words, this means that the transistors used are capable of changing state 500,000,000,000 times in one second.
The difference is that when you talk about clocking, you are talking about the switching speed of all the transistors simultaneously , which is known as clock cycles. On the other hand, when talking about switching, it refers to the switching speed of the transistors individually , so one speed is not comparable with the other.
In a modern microprocessor there are many transistors working at the same time connected to each other. These interconnections create delays and therefore the clock frequency must be limited or else errors and instability will occur. In fact, you have to be quite careful about how the clock signal is distributed through the functional units of the chip so that it works homogeneously, since otherwise it is when instability occurs, calculation errors, ” hang up “, etc. that we see many times when overclocking (which is why it is considered unstable overclock).
Therefore, we must be certain “careful” when we see that they talk about a processor that works at a stratospheric speed like the IBM example we have been talking about, since in this case it refers to the switching speed but not even close to the clocking frequency of the processor.
Lithography is critical in determining switching speed
The size of the transistors is a fundamental aspect when determining their switching speed, and for this reason it is important that manufacturers update their manufacturing nodes using new structures such as FinFET, GAA, etc. in order to reduce the gate capacitance and thus improve the switching capacity.
In principle, the delay time depends on several factors such as the gate capacitance or the voltage and current used, but it also depends on the physical dimensions of the transistor since the switching speed also depends on the width, length and thickness of the logic gate. If these factors are reduced to moving from one lithograph to another (understanding that it is moving to smaller transistors), the switching speed could be increased at the same time.
In other words and explained simply, the smaller the lithography of the manufacturing node in which the transistors are made, the higher the switching speed can potentially be, which in turn can allow (although not necessarily) a higher operating frequency (clocking) of the processor.
As you can see, all the parameters influence and are related to each other when determining the speed, performance and power of a chip, but the physical lithography and the manufacturing node that determines the size of the transistors is one of the most important. for it.
