So far I haven't found any negatives to fixing up SQ, prior to the acoustic transducers doing their thing ...
An 'extreme' example of what can be accomplished is Robert Johnson original blues recordings -- everything is wrong that can be wrong: 'terrible' recording technology, rough and ready job done capturing the session, the available versions of songs are on poorly produced, poor condition shellac copies - this is about as bad as it can get. Yet ... on a system that is totally on song, a miracle happens - there's a real musician there, in a space, all sorts of little subtleties of the playing come through -- you're in the room with the man himself, it's a real event hearing him play - not a listening torture session ...
Who said anything about ideal?
The listening tests were from several publications from back in the day, say some 25 years ago or so. I couldn't tell you which. Tho I've done and had many of my friends sit in on my various auditions, tests and just general musing around.
Flip a Snell Type One upside down with the floor baffle against the wall, but midwofer canted more like at a 35 degrees. I made a wedge out of wool felt and carved it into shape with electric carving knife. The attempt was to even out the acute angle from the baffle face and wall baffle. The attempt was to kill off as much of the sound wave directly behind the tweeter. Shaped so that somewhere between ~40% of the tweeters dome (1" audax softdome) facing the wall was blocked. Later I had some paneling scraps that were 35" (if mem serves me correct). I wanted a bit wider, for the best coupling/ freq. this was butted against the wall baffle (for the lack of a better description) and arced upto the ceiling being roughly 30" out from the wall. This improved imaging and smoothed things out a bit.
The midwoofer was a Realistic 5" full range P#40-1909B's. I had multiples of the A's, B's and C's. Nice little driver for cheap
Shame to, only driver they had I ever liked and have no more
Flipside: they sounded better when laying flat on your back on the floor. Pillows directly behind your head is not the best situation for listening. Next on a Couch, but the back has a suckout factor one side or the other. Perhaps in a bedroom would have been better. Tried about every listening position/placement possible.
Today I'm messing with the thought of this, but more quasi. Think FCUFS and then add a small mtm with dual 3-4" drivers as a direct radiator. But it hit me the other day that a single FR88 or 89 might be a good candidate for this test/task. Time delay'd to coalesce with the floor wave.
Who said anything about ideal?
The listening tests were from several publications from back in the day, say some 25 years ago or so. I couldn't tell you which. Tho I've done and had many of my friends sit in on my various auditions, tests and just general musing around.
Flip a Snell Type One upside down with the floor baffle against the wall, but midwofer canted more like at a 35 degrees. I made a wedge out of wool felt and carved it into shape with electric carving knife. The attempt was to even out the acute angle from the baffle face and wall baffle. The attempt was to kill off as much of the sound wave directly behind the tweeter. Shaped so that somewhere between ~40% of the tweeters dome (1" audax softdome) facing the wall was blocked. Later I had some paneling scraps that were 35" (if mem serves me correct). I wanted a bit wider, for the best coupling/ freq. this was butted against the wall baffle (for the lack of a better description) and arced upto the ceiling being roughly 30" out from the wall. This improved imaging and smoothed things out a bit.
The midwoofer was a Realistic 5" full range P#40-1909B's. I had multiples of the A's, B's and C's. Nice little driver for cheap
Shame to, only driver they had I ever liked and have no more
Flipside: they sounded better when laying flat on your back on the floor. Pillows directly behind your head is not the best situation for listening. Next on a Couch, but the back has a suckout factor one side or the other. Perhaps in a bedroom would have been better. Tried about every listening position/placement possible.
Today I'm messing with the thought of this, but more quasi. Think FCUFS and then add a small mtm with dual 3-4" drivers as a direct radiator. But it hit me the other day that a single FR88 or 89 might be a good candidate for this test/task. Time delay'd to coalesce with the floor wave.
Laying on back, speakers on floor basically pointing at ceiling, sound comes from ceiling. Speakers standing up slightly in front of listening position, pointing away with tweeters aimed at front wall, sound comes from front wall. This is a precedence effect, coupled with faster processing of shorter wavelengths v longer wavelengths within directional cues formed as transient events.
yes, a correct proportion between HRTF freqs coming from below and from above is necessary, and perhaps each time adjustment is needed for the particular user in the particular room
but even then it is still less troublesome than precision toeing in and out in a conventional setup
actually no, for example You didn't, You specifically denied that You had experienced the sound coming from the floor phenomenon
The thing is - dumping a standard loudspeaker on the floor facing up introduces a lot of problems. It was never designed to be used that way. Putting a coaxial on the floor? Fewer problems. Putting a coaxial that has an integrated waveguide on the floor? Fewer still. A coaxial with an integrated waveguide and flush (inset) to the floor? Now we are getting somewhere..
yes, I understand Your point
however I don't think that flush mounting to the floor is necessary and I believe that with a correct type (like a coincident UniQ) and size of the coaxial a waveguide isn't necessary either
here is an idea of a simple OB adjustable FCUFS (with a coaxial mounted):
Attachments
Last edited:
Then what about "working" instead of "ideal"? I did a lot of tests with a wide range of directivities, angling and equalization. It didn't work. If it works only on an individual basis, there has to be a reason for that.
Interesting hypothesis. This would explain why sound sources were all over the place when I used a 8" fullrange driver aimed at the ceiling.
The "processing delay" would have to be rather large because the ceiling reflection arrives at about 5ms.
Last edited:
Can You please give any reference on the precedence effect working in the vertical localisation?
Last edited:
Psychophysical and physiological evidence for a precedence effect in the median sagittal plane.
Litovsky RY, Rakerd B, Yin TC, Hartmann WM.
"models that attribute the precedence effect entirely to processes that involve binaural differences are no longer viable."
again - LOCALISATION
question is not about precedence effect as an echo suppression mechanism. It is but about precedence effect as an echo suppression mechanism working IN THE VERTICAL LOCALISATION
Last edited:
Cherry pick much? I also said it smeared details, stretch images, blunted transients, had no punch, was not immersive, and was not accurate.
You have a habit of patially quoting information in the name of framing your argument. That does not work with me.
So true
I'm particularly sensitive to phase shifts and can pin down a source
13-14ft ceilings = the typical width of a living room, hardly an apt comparison. Btw I dont't listen laying flat on my back, it was just an interesting footnote made in testing. Also I never mentioned what my head orientation was in relation to the speakers. Where the speakers effectively above or below ie at your feet or north thereof. The balance between direct and reflected were ~1.5:1 ratio (rough guess). Distance from the source and reflected were approx the same, 8' ceilings, minus the living room which had a cathedral ceiling, peak at 11'. Delay from direct to reflected was mared by ~0.3ms ( a few inches or so)
Would not suggest this for large or high ceiling areas. Smaller seem a better fit. Subjectively this wasn't a president effect as the amplitude of the ceiling / wall reflection was lower in level using a simple SPL meter (above).
-------
Please remember this was some 25 years ago and only a spl meter for measurments. My Tek O'scope / Fluke meter / AC/DC ammeter / B&K function generator or LCR meter to name a few was of little help here. Wrote my own Ti59/HP41C calculator programs for modeling drivers, transferred these to a Ti99/4a in '84 so I could print plots than play connect the dot. Alot of info has been derived in the last 20y since I did anything audio. Not that I'm wholy unaware of this but simply cannot quote this body of work like most of you can. Kuddos to you all -- trust me, even if we disagree, I have an immense amount of respect.
What I do know is what I have discovered mostly in isolation albiet reading between the lines of every magazine, tech manuals, books and the ramblings of highend salesmen
After dropping 20k on my stereo by the time I was 25 felt that most of what I'd read by the establishment was full of malarky. Reading Lynn Olson's thread Beyond the Ariel recently mentioned something I'd noticed about what he (WE) expected to see happen with driver design didn't occur after all the lessons learned in the 70-80's and that's controlled breakup. Alot of the drivers we have today are in many ways far inferior to the drivers of days past. Rigid cones with hellish breakup modes that ruin off axis dispersion. This was what I believed was in part of the reason for floor localization. The other part, oddly enough from reading jmmlc's insight on horn design. It's the difference between proper wavefront generation and propagation vs the typical baffle with at least 3/4" round overs and flush mounted drivers.
BTW I'd knock heads with the Pro Audio guys back in the day when I maintained the Charolette Colisium, Peace Center for the Arts, Glaxo, UNC Chapel Hill to name a few. Oddly enough the consultants always snatched me up first to help them. Perhaps I didn't give dumb looks at the wrong time or gave dumb looks at the right time, who knows
That was back when everything was fresh in my head, before the significant other, before the kid and the hospital and the kid and the hospital and out of the loop twenty years later.Cheers,
Mike
Can anyone?
Anything?
Please
On the the precedence effect working in the vertical localisation? (NOT just as an echo suppression mechanism as in Elias' reference)
I noticed a mistake in the explaining notes - of course it should be written just "first order front wall reflection path" not "most first order..."