Why Modern Programmers Rarely Consider the CPU: Reflections on Low‑Level Computing

On a weekend, the author saw an interesting question on Zhihu and began pondering why programmers focus on memory issues but rarely consider the CPU that actually executes programs.

While any program ultimately runs on a CPU (or occasionally on GPUs, FPGAs, etc.), most developers pay attention to heap vs. stack, memory allocation, pointers, and leaks, yet overlook the processor itself.

The author attributes this to the rise of higher‑level languages: decades ago, programmers wrote in assembly and had to understand CPU architecture directly.

Using the Intel x86 architecture as an example, the author poses several questions:

What is the difference between long jumps and short jumps? What are interrupt gates, task gates, and call gates, and how do they differ? What are the common x86 CPU registers and their purposes? What are virtual, logical, linear, and physical addresses, and how do they differ?

Most readers of the author's public account are C/C++ or low‑level developers, and few can answer all these questions, especially those coming from Java, Python, or JavaScript backgrounds.

Writing in assembly is mentally taxing and non‑portable across CPU platforms, prompting engineers to abstract away CPU details with higher‑level languages, compilers, and operating systems.

Later, the inconvenience of writing code tied to specific operating system APIs led to the emergence of virtual‑machine‑based languages that further hide OS and memory concerns.

The author observes that modern programmers rarely need to study computer organization, operating systems, or data structures, as many enter the field through short courses rather than formal computer‑science education.

With millions of new graduates and career‑switchers entering the software industry each year, competition is fierce, and relying solely on CRUD skills limits career growth.

Therefore, the author recommends that programmers supplement their daily work with knowledge of low‑level topics such as Linux internals, networking fundamentals, and database internals.