@johnebaker @browj2

Hi John. (CB)

That doesn't happen for me. Nothing on the track on either side of an object, Curve does nothing. If there is an adjacent audio object, then it will affect that object, but not the track curve. Can you demonstrate this?

Yes. See below. Please bare with me as all this is new to me as well and I had to think about each step and selection to show that at least it could be done. To be frank, it's a bit of a mess. I'm doing much better now.

And I have now found this page on the Magix website
https://www.magix.com/gb/video-editing/how-to-adjust-your-video-volume/
Personally I think that information should be in every pdf manual and there is still a lot of information missing
such as John EBs tip for removing nodes on the timeline. although it can also be done with a rapid double left click of the mouse.

Speaking of mice. The higher the available sensitivity of a mouse used the better the experience seems to be. I tried three different mice, all with varying results.

There are some odd things I did notice while going through all of this.

It has made me rethink the whole audio mixing process.

First of all.

What is each adjustment available for?

Most analog and digital mixers both virtual and material normally have an input cut / boost trim pot at the top of each channel strip.
 
The Magix mixers for both MMS and VPX do not have them but instead I think that is what is located on the audio object itself.
This seems obvious to me as they have a very similar function and it is where the cut or boost signal options are applied if one right clicks the object and selects one of the cut / boost or normalize options. You would normally set this so the peak volume shown in the channel meter would reach the maximum of 0db (or less depending on preference or project demands)

This is by far more than enough headroom in a modern audio editing engine which should be doing its internal audio processing using 32 bit float calculations which would give the system a 1,528db of headroom to play with.
In practical terms this is only 700db of headroom but still an insane amount.

Most video files only produce 16 bit audio unless they can record in Dolby digital then they could be 24 bit.

MEP mixer also seems capable of excepting 32 bit files.

If the audio engine is worth its salt this should never result in internal audio clipping of a digital signal source
that isn't already clipped in the recording unless the master faders allow transient spikes to rise above 0db on the final output bus.

You can type a value of -0.01db into the base of the master volume sliders to prevent this as well as adding the Magix brick wall / soft knee limiter to the final output stage.(Magix does not state which is used but my guess is it is a brick wall type.)  

The audio key frame editor itself is a bit of a mystery. If you try to add individual nodes to the timeline as opposed to adding them individually in the Key Frame Editor, you seem to have to drag the mouse across the audio object to add node points.

Adding individual nodes at specific points along the timeline can only be done in the Key Frame Editor itself as there is no visible line ahead to the last node laid down to click onto to place or adjust the vertical position of the next node.
The maximum amount of nodes you can lay down on an audio track with a sampling frequency of 48kHz is 55 per second with the Key Frame Editor.
That is regardless of frame rate. All frame rates give the same maximum amount of node points over one second.
At 44.1 kHz the amount of nodes possible goes down to 50. Again regardless of frame rate.

You can add more nodes but only if you wish to change the nodes position vertically.

The amount of volume slider nodes that anyone can lay down on the timeline when expanded, only seems to be limited by the amount anyone can zoom into the timeline.

It seems infinitesimally more capable of fine adjustment. So much so I think I could take out a click or pop from a 78rpm recording just by lowering the volume during the duration of the click. Similar to what I could have done using a razor blade to cut out small chunks of audio tape in the analog days with quarter inch audio tape.
I will try that out later as the short experiment I did seemed to show promise.
Whether my ears could pick out that a small chunk of audio has gone missing or not I have no idea.
If one doesn't have a decent audio editor to take out such spikes, it could be an alternative.

narrow that view to see how short the reduction of volume is on the timeline.

You may well question whether the audio would be cut for such a small amount of time by using the volume mixer,
especially if the amount of nodes possible that are able to be created greatly outnumber those that can be created using the Key Frame Editor.

I think the answer to that is yes. Or possibly maybe, depending on the processing power of the system trying to do it.
I tried to reduce the volume by the typical time duration of a click from a 78rpm record and tried playing it back to see if I could catch seeing any slider movement.
That may be very silly to try to do as.
1: The duration of such a cut is much less than one 50th of a second so the slider movement may be faster than the video capture frame rate.
2: My graphics card nay bot be powerful enough to produce the movement quick enough to see or another part of the system may not be quick enough for the slider movement to show up on the screen let alone be able to record the evidence.  
Happily enough it seems to work. I did not remove a click or pop so nothing audible is heard. Keep your eye on the number 5 channel slider.

I had to slow down the resulting video to even catch the one frame each other pass caught of the slider movement.

It's promising that one would not hear the lack of audio for such a brief amount of time.
The duration of such an event is so short as the ear cannot normally notice a removal. Only if the click happens at the very beginning of a percussive sound would most ears detect the edit.
This potentially could be very handy as editing audio at less than frame level time-wise can get difficult if not normally impossible within MEP unless the file is externally edited and then if enough cuts were made it may result in an out of sync audio file. So could be useful for those pernickety about such matters.

Which leaves the Timeline Audio Key Frame Editor.
Should it be used to set audio levels?
I personally don't think so. I personally would use the volume slider on the object itself. The one that can be accessed and set by right clicking the audio object and selecting the Normalize option. I'm pretty certain that is its main purpose.
The Audio Key frame Editor does have some advantages over editing curves using the volume sliders via the timeline along with its additional controls but I think I personally would use it for any additional adjustments I needed after getting a reasonably accurate mix of what I wanted first. That way I could easily reset any adjustment if I didn't like the alteration without altering the basic mix.

As regarding the audio distortion on my video.

The distortion you can hear is mainly the background PA system. It is how that PA system sounded.

The audio in the file itself is not clipped.

The Music file is emulated 8 bit audio. Similar to what you would have got from an early Game Boy.

Subsequently some of the instruments sound sharp.

As for the LUFS meter.
Personally I find it is a tool for those that believe in it or need it because where they send their videos to demand it is done.
It is one of many checks that may be required including colour grading and other requirements.
Primarily that would be for TV broadcast where complaints of volume differences are taken seriously as a commercial TV channel does not like to lose viewer figure numbers. It's funny they never require feature films to follow the same rules now.
Some broadcasters now do not require LUFS or they have their own variation of what is acceptable. Not many broadcasters have the same requirements.
They also change their minds from time to time hence this site and others like it.
https://www.rtw.com/en/blog/worldwide-loudness-delivery-standards.html

To be clear. IMO LUFS will not cure all audio upload problems regarding relative volumes without additional checks.

 If anyone uploads to YouTube the sound will be re-assessed and dealt with according to their own criteria.
There aim it to make sure all uploaded videos play at the same average volume as people skip form one video to another.
You can see the alterations in the 'Stats for nerds' setting on each uploaded video. If a video fits the criteria it is generally left alone but if not YouTube will adjust it so it fits in with those that do.
People on YouTube tend to hop from one video to another. If anyone uploads a video with a low overall volume, YouTube will level it up anyway.
If the person listening has to turn their volume up to listen there would be a good chance that the next video they visit could blast their ears.
One of many reasons I prefer listening to audio via speakers over headphones for when that apparently doesn't work.
Other reasons are better stereo imaging, better sound pressure levels more evenly distributed, to the listener and cleaner base.
I understand that a lot of viewing now is done on phones and that creates another set of problems for creators wanting to please a larger audience. Some people will only use one earphone at a time when out for instance. I see that more and more.
Nothing is perfect, and you can't please all the people all of the time. No matter how hard you try.

There is no harm in trying but you won't beat the people in charge of the uploads.

Ray.