Nuendo 7.1 available - VST MultiPanner/Dolby Atmos

Oh yeah baby! VCA issue is gone.

Finally all VCA links are now displayed correctly. I assume this was a long shot, thanks for the good work

VCA issue is fixed and working on my system.

Also, the new Atmos panner is amazing!

Thanks for pushing this out, Steinberg.

jeremiah

So you guys don’t have the trim issue with VCA? If you still have it, can you please specify just what “issue” is fixed and gone? We should be clear about this so that people don’t get the wrong impression.

Did your Repro begin with a blank project, as I would assume, or did you do the Repro with a project populated and already in process before your update?

Blank. Besides, bug fix B-17588 would make little sense if other issues remained broken. I will post a video shortly showing just how this isn’t working - assuming my way of thinking about this is correct.

https://youtu.be/_i50wPBRF4I

You should be able to see and understand what the issues are by just looking at the screen and follow the mouse clicks which are indicated. But either way, here are the items that seem to be just plain wrong to me:


— VCA Slave tracks with automation set to “global” never adjust faders to “+/-0dB” default when enabling TRIM in the automation panel.


— That breaks trim automation because those faders then appear to be writing a trim value based on the “automation read” value, which in turn is given by the VCA. This is in direct contrast to Pro Tools where the individual tracks are quasi-independent. We are able to change those tracks in PT independently of the VCA, including applying trim automation. The VCAs merely modify whatever the automation is on the tracks.


— Not only do these tracks (in my opinion) incorrectly write automation when it does write automation, it also does it in the wrong place. The automation written by me on the first two tracks were two short bursts of attenuation. In any logical implementation this would write automation where I touched the fader, not anywhere else. Instead the beginning of each track is now lower than the original 0dB. As you can also see clearly on the first track, despite me ONLY lowering the fader, there is a gradual ramp upwards before that short drop.


— It’s curious that the “TOUCH” track and the “TRIM” track behave entirely differently after automation has been combined. The first track no longer plays back automation after it has been written on it, whereas the second does.


— After “combine automation” has been undone, the first track (“TOUCH”) now plays back automation and appears to be behaving like the second. BUT, as you can see the “release” values are now entirely different. Despite all three tracks starting at 0dB, the first track now correctly ends at 0dB which is where no automation was ever written, whereas the second ends on roughly -8.5dB.


— Lastly, the third track, which had absolutely no automation written directly to it, and behaved as expected, is now at maximum gain at the beginning and end after having undone “combine automation”.


Let me know if the above behavior is all as planned. It would seem to me that if the “Version History” document specifically states that “Automation Trim mode can now be used on VCA Slaves” (B-12537) then it should be able to be used… in a predictable and logical manner.

So, am I misunderstanding what to expect here or what? I don’t have a problem being corrected here if you think that the above behavior is logical, intended, intuitive and more importantly industry standard.

Well if this doesn’t help Steinberg to reproduce it, I don’t know what would. Especially clear is the “after undo” part where automation and values were restored in the wrong way. This cannot be intended. And in case of that sudden and unexpected 12dB boost in the 3rd track potentially damaging to gear and ears, too.

Mattias, the small bug I came across has been fixed, but I must say that I am now too nervous to use the VCA system until the issues you describe have been clarified or resolved by Steinberg.

What a sorry saga.

It gets worse though:

I was saying earlier in the thread I started here that placing an automation point in the automation lane would solve the problem - however - during my last job on Nuendo I used a template in which I had created automation points by default to deal with this. I did something during the mix, and all of a sudden I saw a similar behavior but different. I couldn’t for the life of me figure out what I did wrong, because the automation points were clearly there.

Well, I just tried to break it again using automation points and I think I succeeded. This time I’m using a signal generator to illustrate part of the problem further. When watching playback keep an eye on both the fader and the meter showing the output of the track.

https://youtu.be/stW5ZAAGr0g

As you can see there are three different scenarios here;

  • After combining automation the fader moves, but the output level doesn’t move.

  • After then turning off trim neither moves.

  • Apparently combining after turning write off yields a different (correct?) result!



    Now let’s add to that what Guillermo has to say about it:


I had a talk with the some of the product planners about this, and you are not going to like it; In other words > they just mentioned it works the way it supposed to work and what you do is not the way you should do it… > So I am going to be brutally honest with you,> it is not going to be “fixed” because there is nothing there to be fixed.

I will try to get a more elaborate reply and translate it to english, but that’s it > either you re-adapt your workflow or maybe look for something else > as this relationship doesn’t seem to be going the way you think it should be going.

Best regards,
GN

So again:

The “Version History” document states clearly that trim on slave tracks now works. So, how is it supposed to work?

Thanks Mattias!
Now I understand the nature of unassorted anomalies with my automation. Had to fix it several times after projects were finished (as I thought). I also came to a trick with ‘starting points’ on automation lanes. Had no time to investigate it and report. Sad but true - now it’s clear that VCA based mixing is still something very unreliable. Personally, I would like to get reliable automation instead of Atmos futures (which are also cool, but I still know nothing about any Nuendo based Atmos studio)
But in general - I like Nuendo 7 more and more. A lot of great futures here and I believe Steinberg is able to find time to get it done!

Read what Guillermo wrote.

This looks quite bad.

Of course it’s frightening but I’m sure it isn’t official position.
Something tells me that I must go to realtime mixdown for a couple of months.

Crikey! - Those comments are not encouraging.

If it’s not the official position it would be nice to get an actual ‘official’ statement and clarification/explanation from Steinberg.

It’s possible that some automation modes work differently than in other software. There’s some room for interpretation I guess and one has to look what the manual says. Which in this case, says something that apparently doesn’t work the way it’s described.

Apart from that, the video clearly demonstrates bugs, stuff that really CANNOT be meant to work as intended. Faders not moving after combining automation? Faders moving but level not changing? Undo suddenly pops your volume up 12dB and thus is not an undo but does something weird? That’s the very stuff bugs are made of. It’s impossible to say this works “as intended”, sorry Steinberg.

I think this is a case of the official internal language badly interpreted.

“As designed” doesn’t mean it is meant to be wrongly designed.
It does mean that the developer has created the feature as planned/designed, regardless if it’s correclty or wrongly designed. I’m not going to discuss the severeness of the issue, I need to look into this much deeper, but if your bugreport is considered to be “exotic” (very unlikely to happen in “normal” working conditions - or better “designed to break the feature”) then indeed there is very little chance that they look into it.

But again, I haven’t looked into this.
And for the record, I always start a project with “create initial values”. There is so much that can go wrong if you don’t start your mix that way.

Fredo

That’s not the key phrase in what Guillermo wrote. The key phrase is “it works the way it supposed to work and what you do is not the way you should do it”. Fredo, there is zero ambiguity there. He’s straight up saying that either it’s user-error or incorrect expectations.

The videos are about 2 minutes each or so. So it’ll take you very little time to see what the problems are.

Further more, this isn’t necessarily solved by using initial values as I show in the second video. And lastly, because of the fundamental design here, regardless of workflow, we now run a much greater risk of severely screwing up our mixes by accident. I.e. not even misinterpreting the intent of the functionality, but simply touching a fader by mistake while it’s in touch-trim would screw things up. Normally one single node would be written and that’d be that, but here this could have severe consequences.

But I look forward to hearing your input on it.

At the risk of being extremely politically incorrect:

The wrong person or people are in charge of N7 VCA development, in my opinion. I first said this about 1 year ago. I’ve been mixing hit records using VCA style automation on Neve, SSL, Euphonix, etc, etc for over 30 years now. Even worked with Euphonix and a now-former (and fabulous) Steinberg employee on some of the initial Euphonix style automation features integrated into Nuendo. The current implementation is not how VCA Automation would be “intended” to work by any of the dozens of successful mixers I know and work/compete with.

Steinberg is going to have to relent on this sooner or later. This is just not working out. Some pain now or more pain later. A tough choice, and you have my empathy, but I think it’s going to become inevitable. But how can SB look at the posted videos and say “Yeah, we NAILED it guys…HI FIVE!” ?

I have no personal ill will against anyone at SB and believe best efforts are being made. I really do. But somewhere I think there is a bad match in work assignments. I will take no offense if this post is deleted, because I realize this situation is a tough pill to swallow. But to my point, making difficult (I would imagine) changes in this regard some time ago would have prevented this current scenario.

Do VCA’s work in Cubase?

Apparently the VCA implementation in Cubase is different from Nuendo. What’s the difference?

I believe Cubase’s system is different in that it relies on linked faders to carry out VCA operations. In other words, whereas PT and Nuendo creates a VCA-group/link, Cubase takes a regular “link” and has the VCA modify it. At least that’s what I’ve been led to believe.

It would have made more sense to develop the functionality once for both platforms. Instead we have twice the workload and twice the potential for errors (which we’ve now seen on both platforms).

I installed 7.1 on 2 of my workstations – the new panner shows up fine on one of them – the other workstation - the panner shows up as white block – no control – this is the same when I loaded up a fresh 5.1 template too – when I have a moment I’ll try uninstalling and re-installing – but maybe somebody knows a fix before I get to that – I vauguely recall having a similar problem years ago when pre 7.1 panner was introduced – I don’t recall what I had to do to fix – let me know if anybody has had to address this – thank you