Imaging control

[fragman56]

Just a quick question.

What characteristic of a loudspeaker affects or controls the imaging? Holographic and pinpoint imaging is what I refer to here. Is it the "beaminess" of the mid and high frequency transducers?

I have auditioned many speaker systems and prefer those with exceptional stereo imaging.

Is there a simple answer?

Thanks,

Rick

[SY]

No simple answers, but two observations born of hard experience:

1. L-R matching. This includes both the speakers and the environs. This is the sine qua non of imaging.

2. Polar patterns. Get them even and non-lumpy. I've heard beamy speakers that image well and more-or-less omnis that image well. In both cases, there were no odd off axis horns or lobes. You can partially compensate for those flaws with acoustic treatment, but that's like putting sugar into urine to make it taste better.

[Ilianh]

Well how's the imaging on esl's then? Their polar pattern is very, very bad.

[mwmkravchenko]

Imaging

Here's something to chew on. Your key areas in hearing the location of objects in real life or in a recording are in the midrange from around 500 hz to just around 2500 hz. This is the area of most importance. Clean reproduction is one thing. Good matching of volume from one speaker to another is important. The channel matching is hard to keep up unless you have a stepped attenuator. Volume pots just don't hack it. The wierdest thing is which speakers provide good imaging. Your placement in the room and the placement of the speakers in ypur room effect this as well. Listen to decent computer speakers and you may get a bit of a surprise too!! If you really want to learn the nitty gritty of it all get ahold of Floyd Tooles JAES papers and read up on his psychoacoustics studies. I have them but I would have to find them, and re read the papers. The balls in your court bud.

Mark

[SY]

Originally posted by Ilianh
Well how's the imaging on esl's then? Their polar pattern is very, very bad.

Depends on the ESL- but my response was directed toward point source radiators, not panels or line sources.

The polar patterns of ESLs with a good reputation for imaging (e.g., Quad 63) are actually pretty excellent.

[7V]

Some other things to throw into this discussion ...

Crossovers, particularly in the midrange, can play havoc with imaging - something to do with phase errors at crossover frequencies.

Slim baffles are widely held to be good at imaging.

Noisy cabinets (vibrations, etc.) will mess up the imaging.

Also, linear array type designs can image beautifully.

Steve