Puzzled about audio effects

Sebaz wrote on 1/19/2008, 6:27 AM
I haven't still read all the section in the manual about audio filters or effects, but for what I can tell so far, the way Vegas works in regards to audio effects applied to single events is pretty lousy. For example, in Final Cut Pro, Premiere, etc, you can simply apply an audio filter to a single event and it's not done in a destructive way, if you don't want it there anymore or if you want to change its parameters later you can.

But in Vegas, either you apply these effects in a non-destructive way to all the track, or you apply them to the events you need but then it forces you to save the modified event as a wav file, so it's destructive, because other than an immediate undo, if later you want to re-adjust the audio effect you can't, you have to find the source clip, trim it again and then apply the effect again.

I don't understand what's the purpose of this. Why didn't they program it in the same way as the video filters are applied, in a non-destructive way that you can later either delete or re-adjust? Am I missing something here?

Comments

Sebaz wrote on 1/19/2008, 9:49 AM
What I mean is that when you apply an effect to the video part of the segment you can open the effects box as many times as you want and keep changing its parameters. With audio, you can do that only in the track effects, but if you want to apply the effect only to the event, it seems your only choice is to right click and select "Apple Non-Real-Time FX" which opens first the plug-in chooser and then the box to adjust the effect parameters, but when you click OK it immediately opens a dialog to make you save the resulted rendered effect to a wav file. The original file may remain untouched, but now, in the timeline, where the audio part of the event was there's a new file, and other than doing undo right away, you can't keep adjusting the effect or either remove it completely and apply something else, unless you delete the event, search for the original file, open it in the trimmer, select the area that was originally in the even and drag it to the timeline again. This seems extremely bothersome for something that should be implemented in the same way it is for the video event, and as it is implemented in pretty much all other NLE.
musicvid10 wrote on 1/19/2008, 10:11 AM
All you are doing is creating a new take with the non-realtime effects to take the place the original event in your Project.

The reason they are not realtime is because certain effects are processor-intensive, and would not play back correctly from the timeline.

Vegas is a non-destructive editor. Your original files are still there, untouched and ready for you to use as you wish, unless you purposely changed or deleted them.

After applying the non-realtime audio effects, you can go back to a previous take by right-clicking on the event, and go to "Previous" or "Next" take, or "Choose Active Take." This can be done at any time, either before or after saving the project. No need to go through the process you described.

That being said, it's always a good idea to save your Project under a new name after applying a set of changes or edits. That way, you can go back easily to a previous save if you change your mind or make a mistake. Project files are small since they only contain a set of instructions, so I generally have 10-20 of them saved by the time I complete a project. Again, editing your Project has no effect whatsoever on your original files or your ability to revert to an earlier event take.
Sebaz wrote on 1/19/2008, 12:47 PM
You're missing my point. I know Vegas is non-destructive when it comes to the original captured file. However, I don't see why audio effects to events are applied in a different way as to video events. If the main reason were that certain effects are processor intensive, then with that reasoning, most video effects are way more processor intensive than audio effects, and they won't play in real time unless you do a RAM preview or pre-render them. So why the Vegas programmers implement adding effects to events differently to video than to audio, that's what I don't understand. And while you may be able to do a "Choose active take", then you're going back to the original audio, instead of the most logical which would be to open the effect that was applied and adjust its parameters in the same way you do with a video event. I just don't get it.
musicvid10 wrote on 1/19/2008, 1:53 PM
You're right, I am missing your point.

1) You complained about destructive editing, which is not the case. Then you said you weren't complaining about destructive editing.
2) You complained about having to go through a lengthy process to revert a take, which is not the case. Then you said the solution isn't good enough.
3) Now you're saying you want a button on the events that you can twiddle like the track fx or video fx. But if they won't play back in real time, wouldn't you need to pre-render them anyway? Now I don't get it. I actually find it easier to render a few takes and play them back to find the one I like best.

The reasoning you question was not mine, but came from the program's help files, so maybe you'll want to consult them first in the future.

If this thread is really about making a feature request, please start a new thread with the words "Feature Request" in the Subject. Someone from Sony might just see it.

This is a user forum, and I have neither the expertise nor the inclination to either explain or defend Sony's programming to you. So please accept my choice not to respond further.
Sebaz wrote on 1/19/2008, 2:29 PM
"1) You complained about destructive editing, which is not the case. Then you said you weren't complaining about destructive editing."

I meant destructive not as in destructive of the original file, but as in destructive of the event's audio.

"2) You complained about having to go through a lengthy process to revert a take, which is not the case. Then you said the solution isn't good enough."

The solution is good enough just to revert to the original audio of the event, but it's not good enough if I want to further adjust the effect's parameters. Say that you added a reverb and later on you want to check how much you added and add a little more, or subtract some. You can't. All you have is a wav file there, or the original audio, but the values of that effect that you applied are nowhere to be found, unless you wrote them down, in which case you would have to select the original audio as active take, then add the reverb effect again, then tweak the parameters again, and do that many times over until you like the way it sounds. It's beyond ridiculous. Now let's suppose you have to color correct an event. You add the filter, you color correct it and do a RAM preview or do a pre-render. You don't like the way it looks. Do you have to do the same annoying procedure as in audio events? No. You simply click on the FS symbol in the event which opens the color corrector filter and you keep tweaking it over and over as many times as you want. Well, why can't it be the same way for audio effects in events?

"3) Now you're saying you want a button on the events that you can twiddle like the track fx or video fx. But if they won't play back in real time, wouldn't you need to pre-render them anyway? Now I don't get it. I actually find it easier to render a few takes and play them back to find the one I like best.

If this thread is really about making a feature request, please start a new thread with the words "Feature Request" in the Subject. Someone from Sony might just see it."

I wasn't requesting a feature by posting my message, basically I was asking if this is really implemented in such an nonproductive way in Vegas or if I was missing something. It looks indeed that I will be posting a feature request, because it's beyond my reasoning how can something so obvious can be implemented in such an illogical way, instead of being the same way it's implemented like in video events, or audio events in Premiere, Final Cut Pro, etc, etc.
musicvid10 wrote on 1/19/2008, 4:23 PM
Have you tried automating Track FX envelopes to break in and out at the event borders? I mean you do have to set the points by hand, and not all effects are available, but they do "remember" their settings and play back in real time.
Sebaz wrote on 1/19/2008, 4:58 PM
I could try that, but still it seems like a lot of extra time setting up something that should be much simpler.
MarkWWW wrote on 1/20/2008, 4:41 AM
I think it's indicative of a different mindset between video and audio people.

People who start as video editing people always seem to want to use as few tracks as possible (perhaps they have a dim memory of assemble editing where you only had two tracks and you switched between them, A-roll and B-roll stylee). Consequently they feel that they need to have effects applied at the event level.

People who start as audio editors always seem to want to use as many tracks as they possibly can, ideally one track per item. (Have a look at the audio desks that are used in Hollywood film dubbing studios - literally hundreds of tracks/faders, needing several people to operate them, at least in the days before automation.) With this mental attitude there is no need to have effects applied at the event level - if an event needs to have a specific effect that no other event needs you just put it on its own track and apply the effect to that track. This approach would allow you to work in what I think is the way you want.

For example, you might have a track for the original audio recorded by the camera, a number of tracks for re-recorded dialog, one for each speaker/actor, a number of FX tracks, each containing a particular type of spot effect or background sounds (birds singing, door slams, footsteps, gunshots, etc, each on its own separate track) and perhaps several other kinds of audio tracks, each containg only a single kind of audio material which would be given its own particular collection of audio effects.

If you really want to have multiple audio events on the same track that require different audio treatments then you can achieve it by using effect envelopes and having each effect active only during the event you want, but I'd say this is more awkward than working in the usual one-track-per-item style of audio editing. It may be necessary on lesser systems where you have only a limited number of audio tracks available but on a system like Vegas where you can have as many tracks as you need you can work in the much nicer way that audio people prefer.

Mark
Geoff_Wood wrote on 1/20/2008, 3:06 PM
Just goes to show the danger of making assertions without reading the manual !

The track effects are non-destructive, realtime and , and fully tweakable at any time. Unless you choose to do the 'Apply non-realtime Effects" thing for a particular reason.

The reason that the effects operate on the whole track rather than per event on that track, is presumably because it would make a nonsense of the track automation.

If you want different effects chains on different events simply move each event to a new track, or use automation.

Either way, the standard FX mode is NON-DESTRUCTIVE. If you want to compare to the audio with/with the effect, simply tick/untick the effect box at the top of the effects chain window, or tug on an FX control to immediately hear a difference.

geoff
Sebaz wrote on 1/20/2008, 8:52 PM
Still I'm not hearing a reasoning good enough for event fx to be implemented in a different way than video fx and audio track fx. What if I want a reverb to just one event but I want to be able to tweak it over time? I have to add it as a track fx and then automate it only to that event? It's ridiculous. No matter how many times you try to make sense out of it, it still won't make sense to me. Now, if all the NLEs were designed this way then I would say well, maybe I'm wrong, but look at Premiere, Final Cut Pro, and I think Avid too, including Liquid, Canopus Edius, etc, have audio fx that can be applied at the event level and tweaked over time. So to me the way it's implemented in Vegas is flawed. The only plus I can see in Vegas is that it's easier to implement an audio fx to the whole track. Don't get me wrong, I do like Vegas for the most part, otherwise I wouldn't have spent money in it, but I think this is a definite flaw that hopefully will be addressed in future versions.
Chienworks wrote on 1/20/2008, 9:45 PM
I do recall a discussion in here a while back that one of the main reasons for not having event level real-time effects was that lots of effects add some sort of sustain which should carry over past the end of the event in order to sound correct. That makes sense ... just a little bit. I'm not sure why it wouldn't be possible to have these effects sustain past the end of the event anyway.

And on the video side, there have been many times i've wished for track level Pan/Crop and event motion.
Geoff_Wood wrote on 1/21/2008, 1:53 AM
No. You still haven't got it.

You either add a new track with the event with the so radically different FX chain OR you automate one existing track - you don't need to do both things.

There is no problem or inconvenience that anybody else has raised that I've noticed in this workflow in over 5 years, apart from me (!) , and I soon got my head around why not separate FX chain per event (as in CDA) , and how the alternatives were better.

Have you by any chance had a look at the manual or Help yet ? 'Overview' is a good place to start. Seriously.

geoff
drbam wrote on 1/21/2008, 6:42 AM
I would just echo the 1st sentence in MarkWWW's reply. Generally when audio folks begin seriously working with video, most will have similar comments as the original poster in that they simply cannot see or understand the reasoning behind many of the workflow features and functions in the the video editing software. And like here, they will typically start arguing that there must be some kind of design flaw because it doesn't make sense to them. From a brain science/learning perspective this can partly be explained under the theory of "state dependent learning." The brain learns and adapts to certain ways and methods of understanding and behaving via repetitive action, environmental factors and various states of consciousness. From a psychological perspective, it is seen as arrogance.
pwppch wrote on 1/21/2008, 8:30 PM
There are a lot of reasons we chose not to have track audio events to have FX like a video track event can. The number one reason is an audio DAW workflow model. The work flow is that if an event needs special treatment, just use another track. Tracks are cheap and the track then can be bussed in a number of ways. When the event has the FX aplied only to it, the bussing is still with the parent track and any other content that may be on it. This could present bussing issues that could not easily be solved later.

The general workflow is then to use tracks, sub buses, and automation to achieve a more flexable workflow for mixing purposes.

I am not saying this is the ideal workflow. It is however Vegas' audio roots approach and it is generally well understood by the audio centric community.

There are some technical issues that arise that some of the other forum members have mentioned. There are others. Far from impossible to overcome, but it would present unique workflow considerations. If you'd like I could enumerate some of these.

Having said this, we have heard your request before. We are very aware that other hosts - particularly in the Video NLE world - have this model of audio fx processing available. (Some DAWs also permit this approach.) We are not against this. We just have to weigh this request with the 1000s of other requests we get. We believe that the workaround by using more tracks is reasonable, if not ideal.

We will consider your workflow and request.

Peter


jbolley wrote on 1/22/2008, 2:36 PM
Welcome back Peter. Happy New Year!
Your contributions on this forum are greatly appreciated!!

Jesse
musicvid10 wrote on 1/23/2008, 5:45 PM
Thanks for the clarification, Peter.
Since most of my work is single event shoots with a continuous audio mix, the track fx have worked quite nicely, and when I've needed single -event control, setting the in/out points works just fine.
However, I can see how the multiple track approach works better for scene shoots, and assembly editing with audio. I've always been a bit stingy with tracks, tending to use as few as possible, but glad to see multiple tracks is an advantageous approach.
musicvid10 wrote on 1/23/2008, 5:46 PM
Thanks for the clarification, Peter.
Since most of my work is single event shoots with a continuous audio track (only the camera takes change), the track fx have worked quite nicely, and when I've needed single -event control, setting the in/out points works just fine.
However, I can see how the multiple track approach works better for scene shooting, and assembly editing with audio. I've always been a bit stingy with tracks, tending to use as few as possible, but glad to see multiple tracks is an advantageous approach.
Teetow wrote on 1/24/2008, 6:00 AM
The only problem I see with the current model / workaround ("tracks are cheap") is that it only works until you screen is full of tracks -- which usually pans out at about 15-20, if you want to see your waveforms clearly.

Event FX are an obvious solution to this. If all I want to do is adjust some EQ and / or stereo imaging, while retaining all the other track FX, using duplicate tracks and buses quickly becomes somewhat unsustainable. So, here's another vote for the "Event FX" feature =)

On a sidenote, I've been a strong advocate of the "fold to subproject" feature, which would create a subproject containing the selected tracks / events. I've created a script to do it, but it involves an awful lot of guesswork and assumptions which I believe can be avoided by making it a proper Vegas function. I also ran into problems trying to create an "unfold" script, which turned out to be almost impossible without access to effect parameters and a "merge into project" functionality. This would also reduce clutter, and retain the "overviewability" of a complex project.

Just my -2 dB.
Geoff_Wood wrote on 1/24/2008, 2:11 PM
"The only problem I see with the current model / workaround ("tracks are cheap") is that it only works until you screen is full of tracks -- which usually pans out at about 15-20, if you want to see your waveforms clearly."


Which is why I have being begging for 'folder tracks' since they were first conceived !

geoff