SESKStuff Everyone Should Know
SESK: Stuff Everyone Should Know · sesk.info/tech-gpu-transistors

Compared with a CPU, what does a GPU spend more of its transistors on?

Data processing units, while CPUs spend more on caches and control.

Start playing free

  1. Data processing units ✔
  2. Large caches
  3. Branch prediction and control

Why: A CPU runs a few threads as fast as possible, so it needs big caches and smart control. A GPU trades single-thread speed for many simple math units, great for doing the same math on lots of data.

Huh, didn't know that: NVIDIA's CUDA guide says a GPU is designed to run thousands of threads in parallel.

The dad joke
Why did the GPU never feel lonely? It always had thousands of threads to talk to.
Like this card?
Play the game

References

Answer: Data processing units. A CPU runs a few threads as fast as possible, so it needs big caches and smart control. A GPU trades single-thread speed for many simple math units, great for doing the same math on lots of data.

  1. NVIDIA CUDA Programming Guide: Introduction docs.nvidia.com/cuda/cuda-programming-guide/01-introduction/introduction.html
    GPUs are specialized for highly parallel computations and devote more transistors to data processing units
    Checked 2026-10-10.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Test yourself in the game

Tech & GadgetsAround 2004, why did chipmakers shift from faster clocks toward multicore processors?Tech & GadgetsWith Intel Hyper-Threading on, how does one physical core appear to the operating system?Tech & GadgetsA phone chip mixes fast, power-hungry cores with slow, efficient ones. What is that design called?

That's stuff everyone should know.