The separate PA thing means that you lose all phase related issues (providing the vocal PA is a mono single effective source, rather than a pair of speakers...).
You hardly ever pan main vocals anyway, so you're normally using your stereo pair to produce a mono sound. In other words, both speakers are producing the same sound at the same time. Since hardly anyone in the audience is an equal distance from both speakers, they'll hear a messy comb-filtered version of the main vocals, rather than them cleanly.
You can make a very good argument for not bothering with stereo at all for live sound (except for effect stuff, but then we're back into separate-pa-for-each-instrument grounds...). The trouble with that is how much harder that is to rig. If you're stacking stuff on the ground at all, then you can't really stack center stage, you need to stack to the sides. (Or behind the stage, with all the feedback issues that involves). If you do fly your PA, then you still need legs, usually. And a big PA is stinking heavy, and flying it near the legs is a lot easier than near the middle, due to lengths of truss, loading points, etc.
I've never tried it personally, but people who's ears I trust all say that besides all of that, just using a secondary sound system for vocals still makes them come through a lot clearer and everything sound better.
One possible reason for that is that a speaker is essentially just a paper cone with an electromagnet on it. There are physical limits on what that one bit of paper can do in terms of vibrations. We like to pretend the world is a lot more digital and measurable than it really is. By separating out the vocals from the other instruments, you allow the speakers to only move when the vocalist is singing. I can see why it could be cleaner. I believe the Chilli Peppers may be doing 2 PAs at the moment splitting off the vocals to their own mix...
You'll still always need a good sound guy, and you'll still need somehow to control the (vocals:instruments) ratio in the audience.
Ideally, bands would think of the sound guy as a member of the band. Kind of like the conductor of an orchestra is a vital member of the orchestra. A good sound guy is very similar. A bad sound guy has all of the same control, but none of the feeling.
For me one of the most impressive parts of this article where how they managed to eliminate feedback on the vocal microphones.
They did this by using two microphones, using one of the microphones for the singer, the other for the background noise. By inverting the phase on one of the microphones and then summing these two signals they filtered out the background noise. At least this was my understanding.
Sometimes it's really great how far we have come with technology and what tricks they've used in the past.
@good sound guy:
you're absolutely right - but most of the good and / or big bands have their sound guy anyways.
I always wondered if it still sounded natural, or if there was some kind of comb-filtering to the vocals that just didn't matter due to the style and volume of the music.
I've played with multiple out of phase mics myself (to try and reduce drum bleed from the main vocalist mics) but never found it as effective as I'd hoped.
I think the problem was just how stinking loud the ambient drums just in general were to the room. Maybe outdoors with a bigger rig it would have worked.
Still pretty fun to play with though.
There's so much mad stuff you can do when you start playing with out of phase mics, delay, etc. As a party trick I occasionally pull a mic up to the point of feedback, whistle until I have a stable loop, and then can whistle in harmony with myself.
I think one of the guys in the article says that the sound was not too good. I think I've read the term "comically". Another problem was that it didn't work well if the singer was not singing close to the mic, so the ambient mic direct vocals on it too.
But I think it was a big improvement to what they had before.
IIRC they had the double microphone configuration in The Grateful Dead Movie (it's been a while since I have seen it). If I remember correctly, it sounds a bit different indeed.
It's probably on Youtube, so you can listen and judge yourself ;).
Odds are pretty good that your cellphone uses a similar noise-cancellation system, with a mic on the top sampling the ambient noise to subtract it from the mic by your mouth.
You hardly ever pan main vocals anyway, so you're normally using your stereo pair to produce a mono sound. In other words, both speakers are producing the same sound at the same time. Since hardly anyone in the audience is an equal distance from both speakers, they'll hear a messy comb-filtered version of the main vocals, rather than them cleanly.
You can make a very good argument for not bothering with stereo at all for live sound (except for effect stuff, but then we're back into separate-pa-for-each-instrument grounds...). The trouble with that is how much harder that is to rig. If you're stacking stuff on the ground at all, then you can't really stack center stage, you need to stack to the sides. (Or behind the stage, with all the feedback issues that involves). If you do fly your PA, then you still need legs, usually. And a big PA is stinking heavy, and flying it near the legs is a lot easier than near the middle, due to lengths of truss, loading points, etc.
I've never tried it personally, but people who's ears I trust all say that besides all of that, just using a secondary sound system for vocals still makes them come through a lot clearer and everything sound better.
One possible reason for that is that a speaker is essentially just a paper cone with an electromagnet on it. There are physical limits on what that one bit of paper can do in terms of vibrations. We like to pretend the world is a lot more digital and measurable than it really is. By separating out the vocals from the other instruments, you allow the speakers to only move when the vocalist is singing. I can see why it could be cleaner. I believe the Chilli Peppers may be doing 2 PAs at the moment splitting off the vocals to their own mix...
You'll still always need a good sound guy, and you'll still need somehow to control the (vocals:instruments) ratio in the audience.
Ideally, bands would think of the sound guy as a member of the band. Kind of like the conductor of an orchestra is a vital member of the orchestra. A good sound guy is very similar. A bad sound guy has all of the same control, but none of the feeling.