If we use the spectrum module, we get alternating emphasized and de-emphasized bands added/subtracted to the mono signal to create the respective left and right signals. Now, if we want to place the resultant stereo track in a sound stage position that occupies, for example, center to right, we cannot do that with the DAW mixer's pan control or the emphasis/de-emphasis will get lopsided.
Is this what the Panorama control in the top row of controls in MStereoSpread will do for us? That is, will it shift the resultant stereo in a way that reduces the width and keeps the L/R balance sounding correct before doing the pan? Or would we be better off using an external tool such as Waves S1 Stereo Imager to get the job done right?
Quick Question Re. MStereoSpread
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122 - KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hmmm, I'm a little lost here - if you use panorama or any other tool (stereo imager is just more complicated panorama...), you will loose some stereo width of course, that's inevitable and logical. So what is the problem exactly? The panorama control does normal panorama, pretty much the same thing as in your host probably.
-
- KVRAF
- Topic Starter
- 1759 posts since 11 Nov, 2009 from Northern CA
OK, let me try to explain this better. With the spectrum trick used in MStereoSpread, I assume that you create a signal that you add to the mono signal to make one channel and subtract from the mono signal to make the other. Result: two different channels that become L and R. Anything that claims mono-compatibility has to have the stereo-ization changes cancel out if taking the stereo back to mono, and the add/subtract operation would make for mono-compatibility.
So, let's say we pan the newly generated stereo from center to halfway right, just as an example. When centered, the changes cancel each other out, but after the pan, one side gets emphasized. The stereo-zation changes would start to become lopsided. The added information would become more prominent than the subtracted (or vice versa). In this example, the right channel changes would become more predominant than the left.
To get some panning like this correct - i.e., no lopsidedness - a tool like Waves S1 takes care of the problem for you. I suppose that something like Voxengo MSED could be used as well, but there'd be more guesswork involved with no graphics to confirm how you are the distributing of sounds. But just straight panning in your DAW won't do it correctly - you get the lopsidedness. So, my question was whether or not the Panorama control in MStereoSpread just did basic panning or did something more sophisticated.
Does this make more sense?
So, let's say we pan the newly generated stereo from center to halfway right, just as an example. When centered, the changes cancel each other out, but after the pan, one side gets emphasized. The stereo-zation changes would start to become lopsided. The added information would become more prominent than the subtracted (or vice versa). In this example, the right channel changes would become more predominant than the left.
To get some panning like this correct - i.e., no lopsidedness - a tool like Waves S1 takes care of the problem for you. I suppose that something like Voxengo MSED could be used as well, but there'd be more guesswork involved with no graphics to confirm how you are the distributing of sounds. But just straight panning in your DAW won't do it correctly - you get the lopsidedness. So, my question was whether or not the Panorama control in MStereoSpread just did basic panning or did something more sophisticated.
Does this make more sense?
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hmmm, quite frankly it doesn't make sense to me. Basically, panorama isn't "some signal added to one channel and another to the other", it is the difference between 2 channels. Thank of it this way: the biggest stereo you could ever imagine is phase inverted mono. This way +1 in R results in -1 in L for example. But that's not mono compatible. So the biggest stereo while being mono incompatible is fully panned to one side.
A little side track: MStereoSpread does is way more complex, but for simplicity you could imagine that it selects specific frequencies and attenuates them on one side and amplifies on another, so it's like it pans individual frequencies to different sides. The relationships between the frequencies, that's the key for making it natural sounding.
Now back to the original simple case - you have a fully panned signal, and you start panning it back to the center, well, there's no loss, you just, again, change the difference between the channels, no big deal.
What pretty much all "stereo imagers" do (I didn't check S1, but I'm pretty sure it's just another one) is manipulate the mid & side signals, which is just another approach for panorama.
A little side track: MStereoSpread does is way more complex, but for simplicity you could imagine that it selects specific frequencies and attenuates them on one side and amplifies on another, so it's like it pans individual frequencies to different sides. The relationships between the frequencies, that's the key for making it natural sounding.
Now back to the original simple case - you have a fully panned signal, and you start panning it back to the center, well, there's no loss, you just, again, change the difference between the channels, no big deal.
What pretty much all "stereo imagers" do (I didn't check S1, but I'm pretty sure it's just another one) is manipulate the mid & side signals, which is just another approach for panorama.
-
- KVRAF
- Topic Starter
- 1759 posts since 11 Nov, 2009 from Northern CA
I don't know why I didn't think to ask my question in a *much* simpler fashion. Here it is: if MStereoSpread is the last insert in a track, does the Panorama control do anything different from what the DAW's track Pan control will do?
By the way, yes, stereo imagers are all based on M/S manipulation. What distinguishes Waves S1 is a nice graphic that helps you visualize the result you're after. It doesn't do anything much less expensive or free plug-ins do, but the visuals are actually pretty elegant.
Thanks for your patience, Vojtech.
By the way, yes, stereo imagers are all based on M/S manipulation. What distinguishes Waves S1 is a nice graphic that helps you visualize the result you're after. It doesn't do anything much less expensive or free plug-ins do, but the visuals are actually pretty elegant.
Thanks for your patience, Vojtech.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Nope, it does the exact same thing 
. You really like the graphics on S1? 
It's just a rectangle . And does it at least analyze the input? Because from what I remember it doesn't, which makes even the rectangle completely pointless
. You really like the graphics on S1? It's just a rectangle . And does it at least analyze the input? Because from what I remember it doesn't, which makes even the rectangle completely pointless -
- KVRAF
- Topic Starter
- 1759 posts since 11 Nov, 2009 from Northern CA
What makes S1 appealing is the combination of Rotation and Asymmetry controls. If you started with a full stereo track and wanted to position it so that it sounded like it occupied the panorama between, say, center and fully right, you need to adjust both the Rotation and Asymmetry S1 controls (adjust so that the top of the triangle is horizontal in the graphic representation). The graphic makes that quick and easy - faster than tweaking just by ear. But with that said, a decent M/S channel manipulation tool like the free Voxengo MSED plug-in could get the job done with similar audio results.
I don't believe analysis of input is required under any circumstances for stereo imaging of this sort. The math for M/S manipulation is pretty basic and is independent of audio content in all cases, as far as I know.
I don't believe analysis of input is required under any circumstances for stereo imaging of this sort. The math for M/S manipulation is pretty basic and is independent of audio content in all cases, as far as I know.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hmmm, that doesn't seem to be logical to me. It basically implies that your tract you are processing has the exact properties of the "typical stereo track they used", which is, well, never true! Without analysing the actual input, you just never know...
-
- KVRAF
- Topic Starter
- 1759 posts since 11 Nov, 2009 from Northern CA
The stereo imaging plug-ins that I'm familiar with do make the assumption that you are starting with a balanced stereo signal that is perfectly stretched left to right in the soundstage, in which case no special analysis is needed. But I can see your point. If you wanted to take a signal that was not balanced stereo and "fix" it, then some analysis would certainly be required - and probably a tricky business at that.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Well, now what is "balanced stereo"? Quite frankly I have never seen 2 signals with identical stereo spectrum 
. That's what MStereoScope is for - it analyzes the width. Most plugins use the primitive correlation meter, this one is based on some statistics and its actually not that simple and definitely every track has different properties...
. That's what MStereoScope is for - it analyzes the width. Most plugins use the primitive correlation meter, this one is based on some statistics and its actually not that simple and definitely every track has different properties...