Why does lower latency of audio devices translate to higher CPU consumption?

Question

I'm wondering what the relation is between operating an audio device at a lower latency and it using more CPU power.

Can you explain it to me?

Answer 1

To achieve lower latency the CPU has to service interrupts from the device more often and so move data in smaller chunks - so more chunks will have to be moved. It turns out that the setup and teardown (the overhead) to move a chunk of data to or from a device is significant. So the fewer chunks/second you're handling, the less the load on the CPU.

However, modern CPUs are so powerful that the difference shouldn't be noticeable.