Check you audio files guys

[RWY36R]

You really need to check and test all the audio files for the Saab 340. They could cause major issues.
Make sure they are in the proper wav format and test them. Not sure what you are using to create or test them.
But they have issues. I've edited and produced audio for over 30 years. So I do know a little bit of which I speak.
I own the Saab 340 and just don't fly much anymore. But I thought I would let you know in my latest scan of wav files to check for errors in the 
X-Plane directory your's are the only ones that spit out errors. And that's not a good sign.

[Ben Russell]

38 minutes ago, RWY36R said:

You really need to check and test all the audio files for the Saab 340. They could cause major issues.

Like what?

I have no open bugs against this issue.

Genuinely curious. Share your wisdom.

[Goran_M]

I'm not really sure what the point of this post is.  

Normally, when speculation is involved, it's common practise to send a private message asking for clarification on what we did and why we did it.  Not come into our forums, throwing your weight around with something about having 30 years sound experience, claiming we MIGHT have problems.  The Saab has been out for 5 years, and no one has had any issues with sound.  

You haven't listed your experience.  What company/companies you have worked with.  Your qualifications.  What software you used.  Anything that would corroborate your claims of experience.

Seeing as you have already made it public, how about itemizing everything you see that WILL cause problems, (not MIGHT), and listing exactly what problems they will cause.  

Apologies if I seem rather blunt, but I'm not used to people telling me I MIGHT have problems with my product, after 5 years in the community, because of something you might have found.  We're very receptive to feedback and criticism, but we need specific, factual details.  We can't follow up on "maybe".

 

Edited June 20, 2018 by Goran_M

[parsec71]

This post made be curious, because (amongst many things) I'm also an audio professional with my own home studio. So I just had to take a look at the files...

I will give one, friendly advice: Keep the files standardized. There's a lot of different bit rates and sampling rates in use, in addition to mono/stereo files. Depending on the audio engine, that will mean a resampling must be performed before the audio can be mixed and generated by a system's sound driver. This will take up processing time.

I guess the difference in files is done with space saving in mind, but all these files are so short in time that they can all be created in CD quality (16-bit, 44.1 kHz).

[JGregory]

@parsec71 Nothing you stated seems to align with the comment from the OP about these files causing "major problems".  And, since we have not had ANY reports of audio problems, I'm still curious just what the OP meant by that.

 

[parsec71]

Well, perhaps the OP will enlighten us...

...I most certainly could not find any major, show-stopping errors in the files I checked. And I've never had any sound issues with this great little bird.

[Ben Russell]

8 hours ago, parsec71 said:

Well, perhaps the OP will enlighten us...

...I most certainly could not find any major, show-stopping errors in the files I checked. And I've never had any sound issues with this great little bird.

I have my doubts that OP even understands the warning message he's interpreting.

I re-read the WAV file specs.

It could be as simple as a single byte missing from a number of different chunk types.

<WAVE-form> → RIFF('WAVE'
                   <fmt-ck>            // Format
                   [<fact-ck>]         // Fact chunk
                   [<cue-ck>]          // Cue points
                   [<playlist-ck>]     // Playlist
                   [<assoc-data-list>] // Associated data list
                   <wave-data> )       // Wave data

https://en.wikipedia.org/wiki/WAV#Specification

 

RIFF files consist entirely of "chunks". The overall format is identical to IFF, except for the endianness as previously stated, and the different meaning of the chunk names.

All chunks have the following format:

4 bytes: an ASCII identifier for this chunk (examples are "fmt " and "data"; note the space in "fmt ").
4 bytes: an unsigned, little-endian 32-bit integer with the length of this chunk (except this field itself and the chunk identifier).
variable-sized field: the chunk data itself, of the size given in the previous field.
a pad byte, if the chunk's length is not even.

https://en.wikipedia.org/wiki/Resource_Interchange_File_Format

 

Given that we have no bug reports on this issue in so long I'm going to error on the side of mundane and meaningless.

 

Had the OP named the chunk-type that was throwing the warning message it would add more weight to the post.

Had the OP named the tool he used to generate the warning scan it would've made investigations easier.

 

@parsec

Thanks for the notes regarding data-format matching and mixer overheads. Note worthy info that's easy to overlook.

In the grand scheme of things with 8 core machines and everything from crap on board sound to dedicated gaming sound chips it probably doesn't make a huge difference.

I'm honestly not sure if the OpenAL calls we use run in the game thread or their own. I suspect on Apple hardware at least it runs in a seperate thread.

(My brain is ringing bells from reading crash dumps where OpenAL is in a thread of its own...)

[Ben Russell]

Documentation for the OpenAL function that consumes the audio data after it's extracted from the WAV file container.

There is no mention of padding requirements.

There is a mention of data interleaving requirements for stereo files.

Perhaps (and this is pure guess work) the warning message being thrown is for stereo-format files where the interleave data is missing a trailing byte.

(Say: 30 samples for left data and 29 for right...)

 

If OP wants to chime in and provide more facts about what's actually wrong, what tool they were using or how to interpret the cropped screenshot they provided I could make more meaningful progress.

 

Bottom line: there are no errors logged against this for thousands of customers and varied hardware running it.