Technical Notes

Introduction

This section of the manual is dedicated to technical information about the internals of XSquawkBox to aid debugging and integration.

Warning

If you are not a software developer looking to solve problems with AFV-Native, or to develop a plugin that interfaces with XSquawkBox, the inforamtion in this section of the manual will not likely be of use to you.

Audio

There are two independant audio implementations inside XSquawkBox.

One handles aural notifications - that’s a simple wrapper onto OpenAL which uses OpenAL-soft on Windows, and uses the X-Plane11 OpenAL on all other platforms by inheriting the symbolspace from the X-Plane process.

The other implementation is the voice implementation that is part of AFV-Native.

AFV-Native itself is open source and more details about how the implementation works under the hood can be glened by reading it’s source code.

AFV-Native itself runs in a dedicated thread which uses a combination of mutexes & atomic variables to exchange the easily maintainable state between X-Plane and AFV-Native. A lock protected queue is used to pass major events (those normally raised by the callback hook on AFV-Native’s client class) from AFV-Native to X-Plane. For This reason, most datarefs that are directly linked to AFV-Native only update when the AFV-Native backend thread is running, and do so at a rate independant to the framerate - roughly linked to AFV-Native’s internal 20ms tick.

Certain details from AFV-Native are exposed as datarefs to assist with capture and real-time debugging of certain issues.

Dataref Name	Purpose
xsquawkbox/voice/input_overflows	Input Overflow counter. Increments whenever input device has more data than AFV-Native is able to process in a timely manner.
xsquawkbox/voice/output_underflows	Output Underflow counter. Increments when the output device requests more data than AFV-native has ready for playback
xsquawkbox/voice/active_channels	Number of independent voice streams currently being handled by AFV-Native. This is distinctly different to the number of audiable
xsquawkbox/voice/audiable_channels	Number of independent voice streams currently audiable - that is, they’re tuned, have audio data available, and are on a radio that is capable of playback, and has non-zero gain.
xsquawkbox/voice/iterations	Iteration counter. Increments whenever AFV-Native’s main thread loop runs
xsquawkbox/voice/input_peak	Input Peak in dB. Updates as long as audio is active.
xsquawkbox/voice/input_vu	Input VU in dB. Updates as long as audio is active.