[openal] OpenAL Soft 1.19.0 released!

[Chris Robinson]

On 09/07/2018 08:51 AM, Ian Reed wrote:
> Pretty awesome! Thanks for your hard work on this.
> 
> I see you added pitch and frequency shifting for EFX.
> Will that shift pitch for the direct sound, or only reverb?

It only shifts the send path the effect is connected to, not the direct 
sound. It also unfortunately only works in mono currently.

> I can adjust the pitch of a source without EFX, but that just alters the 
> frequency without preserving the duration.
> Does the new pitch shifting effect retain the original duration of the 
> sound?

Yes, it just raises and lowers the pitch without changing the speed or 
length of the sound. You can actually increase or decrease the pitch 
shifter's pitch by some amount, decrease or increase the source's pitch 
by the same relative amount (converting to/from semitones), and the 
source will play at a different speed while maintaining the original pitch.

> With your discerning ear, how does the 24-bit dual-ear HRTF data set 
> compare with the data set OpenAL Soft used in the past?
> Any thoughts on how it compares with the HRTF in Google Resonance, 
> Oculus, or SteamAudio?
> What does "dual-ear" mean in this context?

The difference with the new 24-bit data set should be rather minor. The 
original source data is 16-bit so this should only really help with 
avoiding yet another layer of 16-bit dither noise in the mix. However, 
with a higher quality HRTF data source, it would help avoid audible 
quantization noise.

I can't really say how it compares to the others, since it's highly 
dependent on both the data set, and how the data set "fits" to the 
person listening (a given data set may sound really good for one person, 
and not so good for another). The default set OpenAL Soft includes 
possibly isn't the best, but it should be fairly good for most people.

"Dual-ear" means the left and right ear responses can be different, 
instead of being perfect mirrors of each other. Though I suppose "side" 
may be more accurate terminology than "ear". Since the human head isn't 
perfectly symmetrical, the left and right responses shouldn't be exactly 
the same, and that allows data sets to have slight variations.

I should note that the default data set with OpenAL Soft isn't dual-ear. 
The source was only single-ear (or really, two sets of single-ear; one 
for large pinnae, one for small).

> Downloading the Windows binaries, I see that there are 3 HRTF def files, 
> with CIAIR.def being the largest.
> Which is best to use, and where do I need to place the file?

The HRTF def files are for the makehrtf utility, which can create 
external HRTF data sets for OpenAL Soft (mhr files). The CIAIR.def file 
would be used if you have the original CIAIR data set source, and want 
to make mhr files from them. You don't need to be concerned with them if 
you're not creating mhr files.

> When I run openal-info32.exe I see:
> Max auxiliary sends: 2
> Didn't that used to be 4?

Yes, I realized the EFX spec actually says the default should be 2. You 
need to request more through alcCreateContext's attribute list if you 
want more. It makes more sense this way, since an app that doesn't know 
about EFX (or is only expecting simple effect processing) isn't likely 
to explicitly request auxiliary sends, so it doesn't make sense to give 
more than necessary and waste processing time with unused sends (each 
send does incur a per-source cost, so it shouldn't be higher than you need).

> If you moved away from LGPL, what license would you use?

That's still an open question. Ideally it'd be something that encourages 
reciprocating back any changes, but doesn't require it in cases where 
NDAs and stuff may apply. Same for keeping the library open and/or 
replaceable, encouraging it but not requiring where it wouldn't be possible.

I may just go with something simple like MIT or BSD, but I'd like to see 
what other options are available.