MLoudnessAnalyzer wrong Integrated result

Official support for: meldaproduction.com
RELATED
PRODUCTS

Post

No, the target is-23 LUFS for broadcast.

Post

Are you saying the the problem occurs only after 90 minutes analysing? Or at lesser times too?
(My measurements were done over 3-4 minutes only).
DarkStar, ... Interesting, if true
Inspired by ...

Post

As longer the audio is, the more it diverges from the correct result.

Post

Hmm, and looking at your screenshots, several of the measurements differ by 0.8 to 0.9 dB

Image
Last edited by DarkStar on Tue Jan 18, 2022 3:07 pm, edited 1 time in total.
DarkStar, ... Interesting, if true
Inspired by ...

Post

Correct, I hope Vojtech takes a closer look and can fix it. Ticket is open.

Post

Some more info: I just ran a test over 40 minutes and the results were the same - no difference in the two meters
DarkStar, ... Interesting, if true
Inspired by ...

Post

Seems like it’s not a problem in general. I think Vojtech needs my audio file for further investigation. I’ll wait until he replies.

Post

Held wrote: Tue Jan 18, 2022 7:08 am I don't know much about LUFS, but I can see that the LevelNorm shows EBU R-128 whereas MLoudnessAnalyzer is set to EBU+18. Could that account for the difference?

And if your request is time-critical, you should probably contact Melda directly via their form/email.
From what I understand, R-128 is the technical standard and +18 is just a way to display the meters.

Post

Yes,
+18 means show a visible scale of -36 to +18 LU (for a target of -23, this is -49 to -5 LUFS)
+9 means show a visible scale of -18 to +9 LU (for a target of -23, this is -41 to -14 LUFS)
DarkStar, ... Interesting, if true
Inspired by ...

Post

I see. Thanks :)

Post

Fix is already done and will be released with the final 15.02.

Post Reply

Return to “MeldaProduction”