Andrew B. wrote: > As I'm sure Pat realises, it is all more complicated than this with > respect to delay etc.
You bet, this stuff gets complicated fast. > 10ms is a very tiny delay which may slightly > thicken the music if you were doing it with two sets of speakers > pointing at you, one delayed and the other not. True, but the key point is that humans need delay to clearly tell spacial clues. Without that delay, it is hard to pick up location clues. And the 10ms number is only a rule of thumb, Different people, music, rooms, speakers and zillions of other things directly and indirectly impact it. But even decent, non-audiophile speakers can sound a ton better 8 or so feet from walls and corners. The bad thing is that most of us don't like in rooms with more than 8 feet ceilings, and it is really hard to get speakers more than 3 feet off the floor. This is where good WAF things like rugs do wonders. You also need to move your listening chair away from the walls (and corners). > Firstly, the sound radiating backwards from the speakers will contain > very little of the higher frequency information since this is all > beamed forwards from the speaker. In the tweeter range (say above > 3.5kHz) you will probably be at least 25dB down if you are 90 degrees > off axis. Very true. But on some systems, the tweeters come in a lot lower, say 1200 Hz. Above that is stuff important to 'air' and 'shimmer' and all the critical overtones. Huge amounts of location information are in the 500 Hz to 1200Hz area. This is the sweet spot for fundamentals of huge numbers of things from guitars to human voices, the middle of a piano keyboard, etc. > [snip], whereas the problem > in the treble and upper mid frequencies (which affect imaging) is more > often the result of side wall (or side cupboard!) reflections since > they are receiving sound at perhaps 45 degrees off axis. Quite true. imaging is all about 200 to 1200 or so, and not all that much can be done as the lower frequencies are omnidirectional. > This directional higher frequency effect is another reason why toeing > in speakers can often improve imaging/ treble response. By toeing in > the speakers you are making the side walls more off-axis to the > speakers than they were previously and therefore you are reducing the > amount of higher frequency sound that is hitting the walls and able to > refract from them. The same effect can be achieved by using > acoustically absorbent materials on the walls, of course. Again these > will have differing effects at different frequencies. One big problem is that commonly available acoustic treatment materials, from companies like Aurlex, etc. are usually installed in thicknesses that are too small to do much at the important frequencies. Something an inch thick is mostly effective at frequencies with a wavelength in the same range or at least same orders of magnitude. Using 1000 f/s as the speed of sound, a 1000 Hz tone has a wavelength of about an foot. An inch is about 10kHz. So absorbing frequencies with an inch or so of material (which is already too much to have a good WAF) is problematic. Quadratic residue diffusers do a lot better, which is why the 'skyline diffusers' are so popular. A bookshelf full of books of differing sizes that are not lined up nearly can be a decent approximation of a QRD while still having a decent WAF. The really cool thing about trying differing speaker placement is that it is often free to experiment. Even if you have to buy longer speaker cables, you can start with some Home Depot cables for $20 to do testing. If it doesn't work, fine. If it does, then you can look for cables that have a better WAF. -- Pat http://www.pfarrell.com/music/slimserver/slimsoftware.html _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/lists/listinfo/audiophiles
