What is the differences between Ubuntu existing kernel vs rt kernel?

Question

What is the differences between Ubuntu existing general kernel of 11.10 vs so called linux rt kernel?

• does audio delay depends on general kernel vs rt kernel? Where some audio capture looks like it has always a few MS delay.

Example: other distro have them

ftp://ftp.nluug.nl/pub/metalab/distributions/tinycorelinux/4.x/x86/contrib/rt-kernel/

Follow up:

Choosing a Kernel for Your Use Case
These are some simple guidelines provided to help you understand which kernel, and in which order, you should test to fit your use case.

- If you do not require low latency for your system then please use the:
  -generic kernel.
- If you need a low latency system (e.g. for recording audio) then please use the:
  -preempt kernel as a first choice. 
  This reduces latency but doesn't sacrifice power saving features. 
  It is available only for 64 bit systems (also called amd64).
- If the -preempt kernel does not provide enough low latency for your needs 
  (or you have an 32 bit system) then you should try the:
  -lowlatency kernel.
- If the -lowlatency kernel isn't enough then you should try the:
  -rt kernel
- If the -rt kernel isn't enough stable for you then you should try the:
  -realtime kernel

Answer 1

The -rt kernel   vs   the -generic kernel

The -rt kernel is based on the Ubuntu kernel source tree with the PREEMPT_RT patch applied to it (and therefore can use the same features, patches, enabled hardware, security fixes and so on) and should offer the same services that the Ubuntu default -generic kernel offers.

e.g.,

It should be compatible with closed video drivers (nvidia and fglrx), any external DKMS drivers, and have available backport packages.

The goal of an -rt kernel is to obtain a real time variant of the Ubuntu kernel that is aligned with the -generic kernel.

As it can be read from the Control Systems Technology Group article "Realtime Linux":

The goal of -realtime and -rt (hard real-time) kernels is to achieve the lowest possible latency at every cost. In technical slang hard realtime systems should always meet their deadlines. Thus developers should use the most advanced programming techniques (sleeping spinlocks, PI Mutex, Full preemption, IRQ Threads and others) and sacrifice things like reliability, power-saving and throughput. Almost all parts of kernel should be analysed to guarantee that right behaviour always occurs.

The goal of -preempt and -lowlatency (soft real-time) kernels is to achieve good realtime characteristics, meanwhile offering welcomed features like rock-solid reliability, good power-saving features and also good throughput. But in this way it can't guarantee lowest latency under all conditions.

---

You audio Delay Issue

To you audio delay problem, the -realtime and -rt (hard real-time) kernels are your best option to achieve the lowest possible latency at every cost, thus reducing the sound delay.

Take note that other mitigating circumstances may be influencing your audio performance.

---

To answer your question directly

does audio delay depends on general kernel vs rt kernel? Where some audio capture looks like it has always a few MS delay.

Yes, the kernel type being used influences the hardware performance, thus influencing details like the audio delay.

Have it said, using -realtime and -rt (hard real-time) kernels, that have a more aggressive configuration, can further reduce latency, being this your best choice.