SoC vs. CPU – The battle for the future of computing

The CPU has always been called the 'brain of the computer'. Well, it is not really accurate. So for a computer to work, it needs to perform different tasks, like sending and receiving information, executing instructions, storing information and performing calculations. Different parts of a PC correspond to different operations. It is more like a complete computer than a CPU. The CPU is the component that performing processing tasks, like calculations and manipulating data. It performs basic arithmetic, logical and input/output (I/O) operations.

In short, the CPU fetches instructions on what to do and then follows them. Everything stored in computer memory is either data or instructions. All these instructions that the CPU performs and the data on which they are performed are taken from the memory. The operation or the instruction cycle of a CPU almost always has the steps fetch, decode and execute.

Now let us come to the thing around which is all this discussion from now, the SoC. Short for System-on-a-Chip, it is an integrated circuit which combines all or most of the components of a computer or the other electronic system. A typical SoC has CPU, memory, input/output ports and some secondary storage devices embedded into a single substrate or microchip. In some cases, it also integrates advanced peripherals like GPU, Wi-Fi Module, Mobile Network modules  ( like 5G networking chip) etc.

SoCs are typically fabricated using metal–oxide–semiconductor (MOS) technology and are commonly used in embedded systems and the Internet of Things. A typical SoC shall consist two major subsystems namely the functional unit that comprises the microprocessors for running code, memory, digital signal processors; and the inter-module communication system that contains the communication networks and topology (Network on Card) for communicating within between the modules/components of an SoC.

The number one advantage of an SoC is its size: An SoC is only a little bit larger than a CPU, and yet it contains a lot more functionality. If you use a CPU, it’s very hard to make a computer that’s smaller than 10cm (4 inches) squared, purely because of the number of individual chips that you need to squeeze in. Using SoCs, we can put complete computers in smartphones and tablets, and still have plenty of space for batteries.

Due to its very high level of integration and much shorter wiring, an SoC also uses considerably less power — again, this is a big bonus when it comes to mobile computing.

Cutting down on the number of physical chips also means that it’s much cheaper to build a computer using an SoC although; it may increase exponentially if we need it too compact.

SoCs do come with minute imperfections such as, due to its tight component packaging and no indentation, strenuous efforts are needed to fabricate it more and modify it. There is no way of repairing it even if it gets slightly damaged in the components.

Also, the major disadvantage is the inflexibility of an SoC. With your PC, you can put in a new CPU, GPU, or RAM at any time — you cannot do the same for your smartphone. In the future you might be able to buy SoCs that you can slot in, but because everything is integrated this will be wasteful and expensive if you only want to add more RAM.

Since there is always need for modifications, there is a high pressure on design engineers of SoC’s due to various market demands and achieving and designing new SoC’s will take time, as the design of the previous model of them is very difficult to be fabricated.

Back in 1965, Gordon Moore, co-founder of Intel, observed that the number of transistors on a one-inch computer chip double every year, while the costs halve, now that period is 18 months, and it's getting longer. Silicon is reaching the limit of its performance in a growing number of applications that require increased speed, reduced latency and light detection. The longer trend of humanity’s need for computation reaches back to the abacus, mechanical calculators and vacuum tubes, and will likely extend well beyond semiconductors to include superconductors and quantum mechanics.

While, back in 2019, NVIDIA stated that Moore’s law was no longer valid, Intel has come up with an alternative architecture based on ‘spin qubits’, which operate on silicon. A spin qubit uses microwave pulses to control the spin of a single electron on a silicon-based device, and Intel recently utilized them on its recent 'world's smallest quantum chip'.

With the future of tech at stake and the needs increasing exponentially, SoCs are the next step after CPUs. Eventually, SoCs will almost completely consume CPUs. We are already seeing this with AMD’s Llano and Intel’s Ivy Bridge CPUs, which integrate a memory controller, PCI Express, and a graphics processor onto the same chip. There will always be a market for general-purpose CPUs, especially where power and footprint are less of an issue. Mobile and wearable devices are the future of computers, though, and so are SoCs.

An Article By: ElecFest Team

---

DISCLAIMER

Shaastra TechShots’ publications contain information, opinions and data that Shaastra TechShots considers to be accurate based on the date of their creation and verified sources available at that time. It does not constitute either a personalized opinion or a general opinion of Shaastra or IIT Madras. The information provided comes from the best sources, however, Shaastra TechShots cannot be held responsible for any errors or omissions that may emerge.