Event Volume Bug in Nuendo 7.1.35(64bit)

Hi All

I hope someone can help because this problem is becoming annoying and time consuming to constantly fix.

The problem started recently, I don’t know what the cause could be but random clips (that have an event volume value higher or lower than 0db)
would normalize back to 0db.

I thought I was pressing a button by mistake but that’s not the case because it happens to random events across the timeline sometimes, even when I’m not actively using Nuendo.

Has this happened to anyone else? does anyone have a fix or know which direction to point me?

Thanks in advance
Sizwe M

Yes, it’s a BUG and VERY annoying at that. I just got done with writing step-by step reproduction, when Chrome crashed upon posting it so I lost it. I’ll write a new one when I’ve got the time.



So, here we go again.

I must admit that I rarely visit Steinberg forums, so I’d be surprised if this post is the first time anyone actually addresses this issue. Given that the original post dates back three weeks and hasn’t been locked or moved because being a double topic, I have to assume it is.

As I understand, for an issue to be labelled a bug, it must be reproduceable. Here it is:

  1. Have two audio events (hereafter referred to as “left” and “right”) on the same track, both having event volume at anything below the possible maximum, that is +24 dB

  2. Set volume of the “right” event tomaximum, +24 dB.

  3. Select both events. The info bar will now show the “left” events volume (in yellow though, meaning multiple events are selected).

  4. Use mouse wheel to raise the event’s volume by a couple of dB from info bar.

  5. You’ll notice that visual volume, i.e. waveform height changes accordingly on BOTH events, even though volume of the “right” event was already at possible maximum, +24 dB. Give the events a listen - you’ll notice that the audible volume of both events has risen by the same amount.

  6. Save and close project.

  7. Reopen. Volume of the “right” event (that had the impossible volume “overshoot”) has defaulted to 0 dB.

  8. Here you go, there’s a BUG for you.

This behavior also appears with events on different tracks, the only difference is that the event on “upper” track acts as the “left” one in my reproduction.

This unintentional event volume overshoot is present in earlier versions, too (definitely N 6.5, maybe even earlier, but I can’t confirm that), but in these the overshot events never defaulted after reopening - so it was rather helpful, actually. However since upgrading to N7 I’ve regularly been sent back my mixes, because of having unintelligibly quiet bits of dialog in them - that were alright while mixing, which has been rather annoying, not to mention embarrassing.

I haven’t upgraded to N8 yet, however I have a strong ground to believe that the bug is no better there either. Namely I also own Cubase, in which the same issue presented itself with version 8 (I believe it’s the same generation as N7?). I can confirm that the issue is well present in C9 (same generation as N8?).

It might not seem like a major issue, but in my line of work (audio post for TV series) incoming audio regularly has rather random levels, for which the event based volume has been a wonderful tool. Now I’ve almost manically double-checking the hundreds of events’ volume for not acidentally overshooting again, which is rather time consuming and annoying.

So Sizwe M - sorry, I’ve got no fix for you. The ony workaround that has brought the risk of this happening to a minimum for me, has been to avoid raising event volumes to anything near +24 db - if needed, use offline processing, like “normalize” or “gain”. But as we both have concluded earlier, it’s time consuming and annoying.

I’m not entertaining the idea of this getting fixed in N7, but Steinberg, please look into it in N8. Thanks.

Oh, and it happens in both 32- and 64bit versions.

This is a terrible bug. Sending out mixes to clients with things being off is a terrible feeling.

I really wish Steinberg greatly increased their beta team or made released public betas. Because either beta/QC is the problem or the company knows about all these bugs and release versions anyway.

As I said, I’m happy I didn’t spend on version 8, because it was riddled with issues, and we’re still not left with a bug-free version 7 it seems. I’m just left with the feeling that I can’t trust the code and that any new version I’d have to test thoroughly for days to make sure it works properly , and when it doesn’t have to come up with workarounds…


Confirmed here on N7.1.40