We want to resample before minimum phase reconstruction since that changes the
phase relationship of the sampled signal, introducing a slight bit of noise
from truncated sampling. It's not clear that the frequency domain resampling
method is accurate, so resampling prior to frequency analysis is an alternative
to ensure the resulting frequencies are given the proper phase for sampling.
This also cleans up some micro allocations in loops.
It's treated as 5.1 + 2 aux channels. This allows AL_DIRECT_CHANNELS_SOFT to
behave better, not forwarding rear left/right channel inputs to lower front and
upper rear, and allows reporting a more appropriate output mode to the app
instead of 7.1.
Creative's wrapper driver doesn't seem to handle EFX enums for alGetEnumValue,
let alone return values only for what's supported. The only way to check which
filter and effect types it supports is try to set them and check for errors.
The generated files won't play correctly if the player doesn't know they're 3-
and 4-channel UHJ (the third and fourth channels must be ignored, or decoded,
for playback). This is largely just for completion's sake, just in case someone
has a use for it.
The extended precision of the encoder's 0.6512*X term was guesswork, with no
real basis for it. Switch back to the original value until something better
actually comes up. Also updates the decoder to account for the change in the
encoder.
Classic B-Format uses scaling factors W=1, X=sqrt(2), Y=sqrt(2), and Z=sqrt(2),
which is +3dB louder than FuMa. The base factors are designed assuming classic
scaling, so encoding a 0dBFS FuMa signal without accounting for this would
result in the UHJ signal peaking at about -3dBFS. Similarly, decoding UHJ to
FuMa B-Format would be +3dB louder than intended.
So encoding needs to implicitly boost the signal by +3dB, and decoding needs to
attenuate by -3dB.
For 2-channel UHJ, two decoding equations are provided in the original paper.
The alternative one is most often referenced for 2-channel UHJ decoding, but
the original/general one can also be used by assuming T is fully attenuated
(which the format allows for, as T can be variably attenuated by a factor
between 0 and 1 to deal with an imperfect transmission medium).
Neither method can be perfect for 2-channel UHJ, it's irrevocably lossy to the
original source, but my subjective testing indicates the general equation
produces less audibly errant results.
There are no known file formats intended to support 3- and 4-channel UHJ, but
it is possible to store them in various audio files when a player/decoder is
aware of what it's dealing with. So there's no reason not to have it as an
option.
Currently only supports 2-channel UHJ, and the produced .amb files shouldn't be
played as normal B-Format (decoded 2-channel UHJ needs to use different shelf
filters).
This should help avoid destructive phase interference. The occlusion low-pass
filter is still applied in the time domain due to no clear topology (cutoff
frequency, slope, bandwidth, etc).
This should help improve memory use a bit since the SOFA file can be unloaded
before allocating some temp buffers for onset detection and FFT calculation.
This should produce far better results given it works directly on the frequency
response magnitudes prior to phase reconstruction, as it doesn't deal with a
linear phase filter on a truncated time-domain response (with the result also
getting truncated in both direction).
The in-library on-load HRTF resampler still uses the linear filter due to its
relative performance and simplicity benefits. It's good enough as a backup,
though users with custom HRTFs would benefit from resampling when creating the
mhr (adjusting its window size as appropriate/desired).
