*  \brief  A variable controlling speed/quality tradeoff of audio resampling.

  *

  *  If available, SDL can use libsamplerate ( http://www.mega-nerd.com/SRC/ )

  *  to handle audio resampling. There are different resampling modes available

  *  that produce different levels of quality, possibly using more CPU.

  *

  *  If this hint isn't specified to a valid setting, or libsamplerate isn't

  *  available, SDL will act as if this hint was set to "fast".

  *

  *  Note that this is currently only applicable to resampling audio that is

  *  being written to a device for playback or audio being read from a device

  *  for capture. SDL_AudioCVT always uses the "fast" resampler (although this

  *  might change for SDL 2.1).

  *

  *  Most things can probably live with the "fast" resampler, but if quality

  *  is important or you can spare some CPU cycles, the other options are

  *  worth exploring!

  *

  *  libsamplerate's interpolators, that these hints map to, are explained here:

  *     http://www.mega-nerd.com/SRC/api_misc.html#Converters

  *

  *  This hint is only checked at audio subsystem init time and changes to it

  *  at other times are ignored.

  *

  *  This variable can be set to the following values:

  *

  *    "default"  - Use SDL's internal, resampler. (Default when not set. low quality, fast.)

  *    "linear" - Use libsamplerate's Linear interpolator (low quality, fast).

  *    "zero_order_hold" - Use libsamplerate's Zero Order Hold interpolator (low quality, fast).

  *    "sinc_fastest" - Use libsamplerate's fastest (lowest quality) sinc interpolator.

  *    "sinc_medium" - Use libsamplerate's medium quality sinc interpolator.

  *    "sinc_best" - Use libsamplerate's best quality sinc interpolator.

  */

 #define SDL_HINT_AUDIO_RESAMPLER_MODE   "SDL_AUDIO_RESAMPLER_MODE"

 /**