Audacity Forum

steve wrote:

Gale Andrews wrote:I will have a stereo track and Split Stereo Track. I will mute one track and work with Envelope Tool on the other track (so I can hear that channel on its own). Then I will select both tracks and Mix and Render. I do not want one channel silenced, obviously. What is going to happen in your scheme?

What if you had exported rather than mix and render? Isn't it confusing that mixing down to a file works differently from mixing down to a track?

It's not always ideal, as I already commented in this thread. But there is a simple rule that a track with Mute button down is not exported.

So when I want to export obviously I unmute the muted channel.

OR in the previously default "Standard" Solo Button mode (now called Multi-track) I would solo the channel I want to hear, leaving the other channel without Mute button down.

OR in "Simple" Solo Button mode I can SHIFT-click the Solo button.

These are all single operations. I don't have to keep doing it before every render.

Can we now get an answer to the question, please? Case 1: Left channel above Right, Left channel has Solo button down, Right has Solo and Mute up. Case 2: Left channel above Right, Left channel has Solo and Mute up, Right has Mute down.

What happens in Case 1 and 2 when I render under your scheme?


Gale

Gale Andrews wrote:Can we now get an answer to the question, please?

I thought i had already explained. The "mix down" is the same as the mix that you hear from the selected tracks.

Gale Andrews wrote:Case 1: Left channel above Right, Left channel has Solo button down, Right has Solo and Mute up.

In this case, if you press Play, then you will hear the left channel - is that right?
On "mix down" you will get a new stereo track which sounds the same as the selected tracks, which in the above case would be a bit pointless because it will just give you the left channel.
Assuming that what you really want is a mix of the two tracks, you would un-solo the first track. You can now hear both tracks. Select both tracks and mix down.

Gale Andrews wrote:Case 2: Left channel above Right, Left channel has Solo and Mute up, Right has Mute down.

In this case, if you press Play, hen you will hear the left channel - is that right?
On "mix down" you will get a new stereo track which sounds the same as the selected tracks, which in the above case would be a bit pointless because it will just give you the left channel.
Assuming that what you really want is a mix of the two tracks, you would un-mute the second track. You can now hear both tracks. Select both tracks and mix down.

The bumper sticker version: You get what you hear.

As we are talking about "workflow", why would you be mixing down these tracks?

steve wrote:

Gale Andrews wrote:Can we now get an answer to the question, please?

I thought i had already explained. The "mix down" is the same as the mix that you hear from the selected tracks.

I deliberately did not read some of the posts in this thread or look online. I don't know what a "mix down" is. Does that always imply what you hear?

Naive users won't know what "Mix Down" is - just like they expect Solo to work on one track only.

I know what Mix and Render is. It does what I would expect.

steve wrote:The bumper sticker version: You get what you hear.

Perhaps some term like that would be better than "Mix down"?

steve wrote:As we are talking about "workflow", why would you be mixing down these tracks?

Why shouldn't I? Because for now I want to start clean with no envelope points and I want to go back to hearing both channels and to selecting in both channels at once. I can do all those in one step: Mix and Render. It needs to remain in the menu, even if I definitely see the value of "Mix Down".


Gale

Gale Andrews wrote: Naive users won't know what "Mix Down" is - just like they expect Solo to work on one track only.

I know what Mix and Render is. It does what I would expect.

I had never come across the term "Mix and Render" until I saw it in Audacity.. Even now I can't think of any other application that uses that term.
On the other hand "mix down" is standard terminology and is an extremely common and well known term for DJs, home recordists, and just about anyone that is likely to do multi-track recording. Where there is standard terminology, I am very much in favour of using those standard terms rather than inventing new ones.

Searching on Google, there are two contexts in which "mix and render" occurs in the first page of search results. One is "Audacity", and the other is bricklaying. "Mix and Render" is not a commonly used standard term in audio production.

On the other hand, the term "mix down" shows a full page of links about audio mixing. The first hit is this Wikipedia page: Audio mixing (recorded music). "Mix down" is a commonly used standard term in audio production.

Gale Andrews wrote: Why shouldn't I? Because for now I want to start clean with no envelope points and I want to go back to hearing both channels and to selecting in both channels at once.

Well you probably don't "need" to mix and render straight away. You could "make stereo track" and defer mixing down until when and if you need the envelope applied. For a long track, "make stereo track" is much quicker than mix and render.

On the other hand, considering the original poster's question, they "need" to go through every track, selecting / deselecting as necessary in order to get the required mix. This is much more inconvenient and prone to user error than the very minor inconvenience of your use case in which you only need to unmute one track (which personally I would probably unmute before mix and render anyway).

The reason I raised this issue was because:
1) It is not really logical or intuitive that "mixing" should ignore the mute status when mixing to a track, especially as the mute status is not ignored when mixing to a file.
2) The more logical approach of the "mix and render" track being the same as the mix that you hear, would provide a convenient solution to the original poster's question. Rather than laboriously going through each track, selecting and deselecting track as required to make a mix down that sounds the same as the mix that he is listening to, all he would need to do is to "select all", then "Mix and Render" ("mix down").

steve wrote:

Gale Andrews wrote: Naive users won't know what "Mix Down" is - just like they expect Solo to work on one track only.

I know what Mix and Render is. It does what I would expect.

I had never come across the term "Mix and Render" until I saw it in Audacity.. Even now I can't think of any other application that uses that term.
On the other hand "mix down" is standard terminology and is an extremely common and well known term for DJs, home recordists, and just about anyone that is likely to do multi-track recording.

Where there is standard terminology, I am very much in favour of using those standard trms rather than inventing new ones.

Nonetheless most of our users do not do multi-track recording. And if they do, they then complain that the audio "sounds together".

steve wrote:Searching on Google, there are two contexts in which "mix and render" occurs in the first page of search results. One is "Audacity", and the other is bricklaying. "Mix and Render" is not a commonly used standard term in audio production.

I don't see what "and Render" adds to the understanding, because it seems obvious that waveforms are going to change. So if we have new "Mix" functions perhaps we can revisit the "and Render".

Despite that the current "mix" does do what I would expect it to. None of the other Tracks menu items take into account audibility of the track.

Aside: Assuming the export behaviour probably won't change without a lot of argument I do think we should warn whenever we export with a Mute button on, not just when all tracks are muted. This prevents nasty surprises. Then add "Exporting tracks with Mute on" to Warnings Preferences. I propose adding this to Bugzilla (as a bug not an enhancement). Anyone disagree?

steve wrote:On the other hand, the term "mix down" shows a full page of links about audio mixing. The first hit is this Wikipedia page: Audio mixing (recorded music). "Mix down" is a commonly used standard term in audio production.

I already saw there were a lot of results. That does not mean the term means anything specific to most of our users, beyond reducing the number of tracks to one mono or one stereo track.

As I asked, does "mixing down" necessarily mean the mix is what you heard before the mix? This matters when we choose the name. Note that Warnings Preferences refers to "Mixing down... during export" but that "mix down" is not necessarily what you hear before export.

steve wrote:

Gale Andrews wrote: Why shouldn't I? Because for now I want to start clean with no envelope points and I want to go back to hearing both channels and to selecting in both channels at once.

Well you probably don't "need" to mix and render straight away. You could "make stereo track" and defer mixing down until when and if you need the envelope applied.

Sorry but I like to clear the envelope points as soon as possible so if I use Envelope Tool again I start with no points.

steve wrote:On the other hand, considering the original poster's question, they "need" to go through every track, selecting / deselecting as necessary in order to get the required mix.

That's fine. Let's address that, and I agree a Mix menu item is a good way to do it, but please don't inconvenience other people assuming you will only cause them "minor" inconvenience. Either make a new pair of mix <some term> menu items, or add a Tracks preference for mix behaviour.


Gale

Gale Andrews wrote:Aside: Assuming the export behaviour probably won't change without a lot of argument I do think we should warn whenever we export with a Mute button on, not just when all tracks are muted. This prevents nasty surprises. Then add "Exporting tracks with Mute on" to Warnings Preferences. I propose adding this to Bugzilla (as a bug not an enhancement). Anyone disagree?

I think that's a reasonable enhancement which may help some novice users. It's not a "bug" because applying the mute status is "doing the right thing".
I think it would need to be in addition to the current warning about all tracks being muted. The first thing that I would do with the proposed new warning is to disable it, but I'd still like to keep the warning about all tracks being muted.

Gale Andrews wrote:I don't see what "and Render" adds to the understanding,

I think it's because it may be applied to a single track and not just mixing multiple tracks.
The term "mix down" is generally thought of as synonymous with "rendering" one or more tracks. The term "render" is becoming more common in semi-pro audio software, but I'd guess that the term "mix down" is much more widely known and more understandable for novice users.

Gale Andrews wrote:As I asked, does "mixing down" necessarily mean the mix is what you heard before the mix?

That is the usual meaning.
As an example, from the Ardour manual:

You can mute tracks, pan tracks, turn effects on or off, change the output level. Whatever you do to adjust the sound will show up in the two mixed down tracks.

Gale Andrews wrote:but please don't inconvenience other people assuming you will only cause them "minor" inconvenience.

Perhaps I didn't express it very well. The current behaviour assumes that the user will want to disregard the mute status of the track, even though this is inconsistent with how mixing works when exporting and is a big inconvenience when working with multi-track projects.

I agree that we could have a preference to ignore track mute and solo when performing "Mix and Render", but if we go down that route, should we also have separate options to ignore time tracks, envelopes, track pan, track gain and track channel?

"Mix and render" will be less incomprehensible to the newcomer. But I think Steve is right that "mix" and "render" both are mostly the same.

Iirc, there is at least one DAW that has it formulated "mix and render" in the menu.

It's one area where some DAW's go wrong, imho, that's in calling it "quantizing". The only purpose it server is to shield the program from newbies.

There also isn't a standard behaviour anymore for mute buttons. On the analog tables this usually didn't affect aux sends. On some it only affected monitoring. Some had jumpers to set the behaviour.

In today's software, I can't think of an example of setting mute behaviour in prefs. In most DAW's, it can be set through some form of automation. This is an example how mute is handled in Reaper:

There is a law that says if you change a behaviour without giving a way to access the old behaviour you are bound to inconvenience someone else who isn't currently complaining.

steve wrote:

Gale Andrews wrote:but please don't inconvenience other people assuming you will only cause them "minor" inconvenience.

Perhaps I didn't express it very well. The current behaviour assumes that the user will want to disregard the mute status of the track, even though this is inconsistent with how mixing works when exporting and is a big inconvenience when working with multi-track projects.

First, as I understand your proposal, you want Mix Down to ignore tracks with greyed out waveform. I still think that is a good option to have but that is not consistent either with how mixing currently works when exporting, which is to ignore tracks that have Mute button on.

So if your Mix Down is a preference I don't think it should be default while the export behaves differently, because it would encourage a perception that export will ignore greyed out tracks.

Second, look at the case where you have some tracks audible and some not. You want to add more tracks and to make space for those, you want to "Mix and Render". With your Mix Down that user has to un-necessarily go through the tracks and make them all audible. Another reason any Mix Down preference should not be default.

Finally I worry novices would find your Mix Down behaviour unexpected, because it does not act like other Tracks menu items. Another reason Mix Down should not be default.

steve wrote:I agree that we could have a preference to ignore track mute and solo when performing "Mix and Render", but if we go down that route, should we also have separate options to ignore time tracks, envelopes, track pan, track gain and track channel?

Not in my opinion. What benefit do you see in ignoring track channel?

Perhaps have a separate option to ignore all the others including Mute and Solo?


Gale

Gale Andrews wrote:What benefit do you see in ignoring track channel?

A user may have set two mono tracks to left and right channels so that they have "separation" in their headphones while monitoring them, with the intention of mixing them down to one mono track. Setting the mix to "pre-channel" would save them from having to manually set the channels to mono.

However, this is just getting caught up in details.

There is an underlying problem / limitation that Audacity does not have a properly defined system for "stacking" effects (including commuting operators). The order in which "effects" (including Time Track, Mute, Solo, Track Gain, Envelopes et al) are performed is defined on an ad hoc basis. This gives us inconsistency in use and excessive repetition in the code. It also gets in the way of further development, such as Leland's experimental effect rack. This is the issue that really needs to be addressed, then consistent behaviour should follow naturally from the defined structure.