#437 Implemented a software carrier, is that sufficient?

With portable software carrier (#437) in place, having sane (probably, by necessity non-portable) implementations of hardware carrier sending is a good thing. The user can then select if he wants the hard- or soft implementation. This can be valuable in many cases.

@SaheblalBagwan : I suggest that you:

1. rewrite the patch on the top of Cleanup of hardware dependencies. Merge in SAM support.  #437 (branch beta?),
2. write some "documentation" on your suggested solution, possibly as comments in the code. What are the ledc*-functions, and why do they solve the problem?

having sane (probably, by necessity non-portable) implementations of hardware carrier sending is a good thing.

True.

I'll waiting on the updates suggested above then.

AFAIK 'ledc' is one standard method to generate PWM on the ESP32.
(ledc => LED Control Peripheral.....IR emitter is another LED)
There is also a dedicated IR RxTx peripheral on the ESP32, which would be the ultimate end game for this particular platform wrt IR (...and somewhat over-kill/over-engineered).

On most platforms, I would suggest that hardware based PWM for generating the carrier is a better approach than 'softPWM'. In other words, I would recommend 'hardPWM' over 'softPWM' in most cases, unless:
A - it is not available
B - specific pins/peripherals/interrupts/resources are not available for use.

Hello, what's about a compatible IR remote library using RMT peripheral?

Just to check, is there any code for RMT IR Receive on ESP32? I've only found IR send using RMT so far.

ESP32 send support is now included.