Currently, processing of audio samples is called from AudioRenderer's Update method, using a fixed 4 buffers to process the given samples. Games call Update at variable rates, depending on framerate and/or sample count, which causes inconsistency in audio processing. From what I've seen, 60 FPS games update every ~0.004s, but 30 FPS/160 sample games update somewhere between 0.02 and 0.04, 5-10x slower. Not enough samples get fed to the backend, leading to a lot of audio skipping.
This PR seeks to address this by de-coupling the audio consumption and the audio update. Update remains the same without calling for buffer queuing, and the consume now schedules itself to run based on the sample rate and count.
Preliminary work for upmixing & general cleanup. Fixes basic issues in games such as Shovel Knight and slightly improves the LEGO games. Upmixing stitll needs to be implemented.
Audio levels in a few games will be fixed as we now use the downmix coefficients when possible instead of supplying our own
With all of the trivial parts of the memory interface moved over, we can
get right into moving over the bits that are used.
Note that this does require the use of GetInstance from the global
system instance to be used within hle_ipc.cpp and the gdbstub. This is
fine for the time being, as they both already rely on the global system
instance in other functions. These will be removed in a change directed
at both of these respectively.
For now, it's sufficient, as it still accomplishes the goal of
de-globalizing the memory code.
Amends a few interfaces to be able to handle the migration over to the
new Memory class by passing the class by reference as a function
parameter where necessary.
Notably, within the filesystem services, this eliminates two ReadBlock()
calls by using the helper functions of HLERequestContext to do that for
us.
Creating multiple "AudioRenderer" threads cause the previous thread to be overwritten. The thread will name be renamed to AudioRenderer-InstanceX, where X is the current instance number.
We shouldn't be incrementing if wave buffers are empty. They are considered invalid/unused wave buffers.
This fixes the issue of certain sounds looping when they shouldn't
Gets rid of the largest set of mutable global state within the core.
This also paves a way for eliminating usages of GetInstance() on the
System class as a follow-up.
Note that no behavioral changes have been made, and this simply extracts
the functionality into a class. This also has the benefit of making
dependencies on the core timing functionality explicit within the
relevant interfaces.
These two macros being used in tandem were used prior to the
introduction of UNIMPLEMENTED and UNIMPLEMENTED_MSG. This provides
equivalent behavior, just with less typing/reading involved.
Softlock explanation:
after effects are initialized in smo, nothing actually changes the state. It expects the state to always be initialized. With the previous testing, updating the states much like how we handle the memory pools continue to have the softlock(which is why I said it probably wasn't effects) after further examination it seems like effects need to be initialized but the state remains unchanged until further notice. For now, assertions are added for the aux buffers to see if they update, unable to check as I haven't gotten smo to actually update them yet.
Preserves the meaning/type-safetiness of the stream state instead of
making it an opaque u32. This makes it usable for other things outside
of the service HLE context.
Avoids including unnecessary headers within the audio_renderer.h header,
lessening the likelihood of needing to rebuild source files including
this header if they ever change.
Given std::vector allows forward declaring contained types, we can move
VoiceState to the cpp file and hide the implementation entirely.
GetAudioRendererSampleRate is set as a "STUB" as a game could check if the sample rate it sent and the sample rate it wants don't match. Just a thought of something which could happen so keeping it as stub for the mean time
