Bad AIF header info generated by WaveLab 10

Wavelab Pro 10 (WaveLab 10.0.20) appears to be generating wrong/incorrect AIF header info.

I compared the same AIF file data generated by Wavelab Pro ver. 9.5 and ver. 10, and compared the two w/ a hexadecimal editor:

  1. File generated by WL 9.5 shows the following header:
    FORM3ÅòAIFFCOMM+#@ªÄSSND3Åq…

  2. File generated by WL 10.0 shows the following header:
    RIFF≤Å3WAVEJUNKfmt ĪÄ2dataiÅ3…

The one generated by ver. 10 clearly shows a corrupt / wrong header. Besides… “RIFF (Resource Interchange File Format)” has to do with WAV file format; it shouldn’t be there at all for an AIF file.

Please correct this bug ASAP. Thank you!

What do you exactly do to produce the file #2 (which appears to be a WAV file, not a AIFF).
I did a quick test here (Save As), and no problem.

How to generate:

  1. Record an audio w/ WaveLab 10, then save as AIF (WL ver. 10)
  2. Next, open the generated file above w/ WaveLab 9.5, give a different name, then save as AIF

I’m using a Mac OS (if that matters)

I have been using the same file template (that creates an empty monaural AIF file, 24-bit 48000hz) inherited from WaveLab 9.5.

Let me see if creating a new template w/ WaveLab 10 matters.

OK here’s what I think what’s happening:

  1. The previously created WaveLab 9.5 templates somehow got mistranslated or corrupted when imported to WaveLab 10
  2. In the course of mistranslation, WaveLab 10 adds this “WAVEJUNK” header to each file since it cannot interpret the above corrupted instruction

I just created a new template within WaveLab 10, and now I see no issue.

So it may not be a major bug, but since all of my WaveLab 9.5 AIF templates are corrupted, I believe it should be noted.

Thank you very much for your time!!

The previously created WaveLab 9.5 templates somehow got mistranslated or corrupted when imported to WaveLab 10

If you have such a corrupted template, please send it to me.

  1. Record an audio w/ WaveLab 10, then save as AIF (WL ver. 10)
  2. Next, open the generated file above w/ WaveLab 9.5, give a different name, then save as AIF

Indeed, no problem here.

Philippe

Alas; I dragged them all out to the Trash bin and they are gone. We shall see; I bet someone else will bump into this issue sooner or later. Thank you.

I think I bumped into it a few weeks ago:
https://www.steinberg.net/forums/viewtopic.php?f=297&t=180117

I wouldn’t be surprised if these things are related. For some reason, WL wants to work with WAV files. I’ve even had a case where a Save of an AIF file wound up making an unreadable file because WL actually made it into a WAV file but retained the .aif extension, so it doesn’t surprise me that you found WAV file header info in an AIF file. In any case, something is strange about the way WL handles AIF headers.

@splunk
Your report is still under consideration, but on the WaveLab side, I could not find any problem yet.
I am waiting for some feedback from the HAlion programmer, to know why the sound is loud in HALion.

Philippe

Thanks for looking into this PG. Do you have access to a program called Myriad? It may have been discontinued, but I’m sure there are other applications that can read sample headers. I attached the report from Myriad with the two files I sent you. The top file is the one that was saved out of WL and the bottom one is the original file. You can see that in the Gain field of the header, WL has written something that is being interpreted as +115dB even though the original file has a gain of +0dB. Halion is simply reading this value, but I’m guessing it has a +96dB limit. Still, as you can imagine, having a file gained up 96dB unexpectedly is not pleasant. I’ve never seen Myriad and Halion disagree about gain settings before, but I’ve also never seen a gain setting that’s so high, so that’s the only explanation I have for the discrepancy between the two programs in this case.

(Hmm. I’ve tried attaching a screen shot to this post in a few different formats and I keep getting an “the uploaded file is empty” warning. I don’t recall having problems attaching things before, but maybe they weren’t images? Any suggestions?)

You sent me 2 files. Both of them exhibit the big gain you mention, in HAlion (latest version). Since you say the original file is ok, I don’t get it.
Anyway, the HAlion programmer has been contacted.

That’s not happening here. Are you sure you’re looking at the gain header parameter?

Here’s the original file in Halion:


Notice the Gain parameter directly above the selected sample. This shows a header gain value of 0.0.

Here’s what happens when I highlight the WL sample:


The Gain parameter now shows 96.0. These two files are definitely not the same and the only thing I did was to do some trimming or loop work on the original and save the file with a command-S.

Here’s what the file looks like in Myriad:


Are you actually not seeing any of these things over on your end? I have two different programs that agree that there’s been a big change in the header of the WL file, so it seems odd that these two files are identical for you.