QUOTE(Lyx @ Aug 29 2005, 05:40 PM)
Actually, i have done some diagrams today(will post them in a few minutes). 40° vertical dispersion is okay if one stacks two speakers horizontally from a single point per channel(facing outwards from each other in the correct angle to avoid distortion due to overlapping) - however, 60° of course wouldn't hurt, because then there would be no need to place them an their side(why did you mention that one cannot easily do that? Just asking because the MSR100 are built so that they can also be placed on their side, so that they can also be used as stage-monitors. Their shape even is made for just that).
If you place MSR's on their side, then you'll have 90deg vertical dispersion and 40degs horisontal (applies to horn mostly I guess) - its sorta beaming, you'll get peaks and valleys of spl in the room and there'll be more places where it sounds bad. And due to unavoidable ceiling/floor reflections then (90deg vertical dispersion) you'll get comb filtering effects that makes it only worse. On stage its a completely different game. You can try blending 2 speakers together, but imho, that would not work too well, you'd face bad lobeing distortions, comb filtering, early reflections, etc. Imo you'd waste too much time on fighting room acoustics and still won't be happy. But I can't say I'm right, I've never been into serious acoustics practically. Its just seems like you'd first create the problem and then try to fix it instead of avoiding trouble in the first place..
I would think that if you can place two speakers into same source spot at angles, then you can place one speaker at normal vertical position. Its not that bad is it. These MSR's are pretty small.
http://www.lenardaudio.com/education/08_live.htmlReally, every bit on these edu pages is packed with useful things. Check on the historic perspective, where he shows errors made. It also touches the lobing and comb distortions, also some problems with small club venues (closest to your case).
Imo, small vertical dispersion is a good thing. They use line arrays to reduce vertical dispersion. That helps to equalize sound level perceived over greater distance, reduces reverberant effects and gets more people into near field.
Of course, reverberant rooms reduce critical distance, and I'd think that you'd want to focus on-axis (vertical) response so that hot-spot of higher frequencies due to beaming is where lower frequencies are already in far field. Goal being to equalize perceived frequency response for those off-axis (closer and below speakers axis) and those on-axis but further away. From here imo arises need for optimal vertical dispersion. I don't have idea what it should be or how to determine it, I just imagine that there is some optimal where you can achieve flattest FR over largest area, by combining speaker placement, aiming, their dispersion and DSP equalizing.
I guess you want larger vertical dispersion when you put speakers higher up from the ground and direct at sharper angle towards people, and lower vertical dispersion when speakers are closer and thus more parallel to the ground.
I'd also think of placing pair of speaker into far end of the room, and 2nd pair to the half way of the room, both pairs directed into same direction and covering half of the room each. You'd need delay correction between pairs to blend them together, but my gut feeling tells that this *might* give good results. Afaik such arrangement is used in modern cinemas, larger venues with multiple sound reinforcement systems to keep sound field uniform.
QUOTE
One thing which i ran across when doing the diagrams is: how evil is it to place the two channels at opposite sides of the room(creating a widened stereo-image) and then narrowing the stereofield a bit with a DSP?
Imo, this is most evil thing to do. Afaik there is zero stereo effect to be listening at in pub anyway, unless you want to focus sound source to the stage, but what you'd get is ruined intelligibility (destroy near field) and wild frequency response fluctuations (excite room modes more), with people halfway between the speakers overwhelmed by spl. Besides, sounds would feel like jumping all over the place around people and actually distract. The only DSP for narrowing the stereofield a bit I would think of is stereo->mono DSP.
"The golden rule - intelligibility improves if sound comes from a single point source". Note that it isn't always best to have 2 spots for speakers - many rooms would benefit if the speakers are actually in single spot together. For eg. in the middle of the side wall, facing away and covering 90degs each.
Of course this all is only my opinion, not even professional one.