And also that there are reflections in the recording that are convolved with the reflections in the listening environment. The use of measures from architectural acoustics are not readily applicable in sound reproduction.
This is a strong point if you are listening to a orchestran recorded in auditorium. I like to reproduce this in an environment with little reflections. The registered environment is good.
If I listen to one or a few instruments (which I would listen live from a small distance, in a non-anechoic environment) with a high absorption of reflections, I find a poor music. A sound that comes out of the speakers, unnatural, even if it contains "his reflections".
If I listen to one or a few instruments (which I would listen live from a small distance, in a non-anechoic environment) with a high absorption of reflections, I find a poor music. A sound that comes out of the speakers, unnatural, even if it contains "his reflections".
That is your preference, some say you should be content with the mix ("his reflections"?) regardless
Well, random anyway. Spacious maybe. Having had a listening room that was all stone with large piles of rock, the effect can be drier than you might expect.
Just to dive in, throw a spanner in the works and then vacate I had an interesting experience recently with toed behind the listener vs toed in front of the listener, and the amount of damping in the room.
Those that have read my waffle before know that I have reasonably directional speakers. (Full range driver and wave-guide ribbon on a reasonably wide baffle, at least by today's anorexic speaker standards
)I have always used them toed behind the listener, as is more typical of speakers and while I've tried to toe them in front of the listener before I've always found it a bit too dry and the width compressed, despite having leanings towards liking a drier sounding room.
So I've been using them toed behind the listener in a room that I felt was was OK, but on the verge of being too live, and the result has been good.
We recently changed sofas from some big plush well padded and curved fabric covered ones to some bare leather much flatter/squarer design sofas, and as the sofas are the second biggest absorptive item in the room (after the carpet) the acoustics of the room changed quite dramatically for the worse, and by much more than even I had expected.
It's hard to describe exactly what it did, but certainly a lot of vocal clarity was lost on TV/Movies, the image became a bit indistinct and stretched apart, and rather than becoming bright if anything I found it was a bit too strong in the lower midrange so a bit "woolly" sounding.
This was primarily due to the second sofa on the right hand side wall which is quite close to the right speaker, so that now flat, smooth, hard leather surface was acting as a good specular reflector all the way from the low midrange to the treble. If I threw a blanket over that sofa some of the old sound of the room returned but not all, and of course that blanket has zero WAF.
I actually ended up fiddling with the EQ and raising everything above about 1.5Khz by 0.5dB which improved the clarity a bit for TV watching but didn't really solve the problem properly.
I've been unhappy with the way it has sounded since then, then a couple of days ago I thought fine, I'll try toeing the speakers in in front of the listener the way Earl always recommends to see how it sounds, on the theory that more toe in might be beneficial for a room that is now "too live" and has an unwanted source of specular reflection (instead of absorption that was there before) on the right hand side wall, since the speaker should be sufficiently directional that this extra toe in will cut a lot of that early sidewall reflection.
Wow is all I can say. Sound completely transformed.
I used a fair bit of toe in - they intersect at least a metre in front of the listener.The somewhat indeterminate phantom image that I've had since the sofas changed has gone and been replaced by a rock solid phantom image. I've had to remove the 0.5dB high frequency lift shelf because it now sounded too bright (go figure!) and go back to my original frequency response.
Now it sounds well balanced, no longer "woolly" in the lower midrange, (although I'm not sure why!) and dialogue clarity is better than it has ever been.
Unlike the last time I tried more toe in with a more dead room, the stereo spread is perfectly acceptable, even though my speakers aren't nearly as far apart as they ought to be (Kids toy boxes tend to get in the way of where I'd like the speakers to be
) and the room doesn't sound too "dead" this time around. It has a good balance of liveness yet still has good stereo imaging and sharp focus as well.And of course, the sweet spot is much wider, as is usually the case with speakers toed in front of the listener due to time/amplitude trading as you move yourself laterally. All it's really missing is for the speakers to be a little bit further apart but I can't do anything about that at the moment.
This is a very interesting result to me - a lot of discussion goes on about toeing behind the listener vs toeing in front of the listener and I had always put that down to people debating personal preferences.
But it now looks to me that toeing behind the listener might work best in a room that is fairly acoustically dead (especially if it has absorption on the side walls like a plush sofa) while toeing in front of the listener might work best in a room that is fairly acoustically live and un-damped, as my room now is.
It's literally the same room, same speakers, all that's changed is the sofa's and therefore the damping and reflectivity characteristics of the room, and my preference has changed completely from toeing behind to toeing in front...
Last edited:
This depends on the timing. All lateral reflections add to spaciousness, but the early ones also degrade image. So if you want good imaging and spaciousness then you need to minimize the early reflections and enhance the later ones, which has always been my goal.
Sure. I fell into the trap of "good distortion".
But if a violin or a guitar sounded in my room, I would feel a lot from the ceiling. The ceiling that comes out of the speakers is quite low.
Distortion?
Sounds like a "man cave". I would think a literal cave, assuming its large enough, would be an ideal room because there is so much randomness and so few parallel surfaces.
It seems that people prefer a sound that is similar to something like an amphitheater. You have a stage, and then open space to the right and left. Also with no ceiling but sidewalls to reflect laterally.
So then you have some random early reflections from the people on the stage itself, but no early side reflections, only later ones. You also don't have the usual ceiling reflections which seem to be universally disliked. One could actually design a speaker with the right amount of vertical directivity to match this.
Extreme toe in can help with this in a smaller room, and in some cases the speakers will begin reflecting early off each other instead of the side walls, which might be a consideration in the shape of the speaker. I suspect most typical rooms have sidewalls too close.
Last edited:
At what time interval difference are reflections detrimental and at what time intervals are they preferred?
That is one thing I try to figure out. It is likely the reason why there are different opinions about lateral early reflections. I was reading that the brain merges sounds if the time delay is under 2ms but that was measured with simple clicks and not music. So too early reflection will be just distorting the original signal but what is the real threshold for music (in time and consequently distance of reflective surface). Also how much is that frequency dependent?
I can´t confirm if measured data for loudspeakers over the years would corrobate this theory (compensation for the more common losses at low frequencies) but it could be.
The higher leakage of drywall constructions was the reason for my post, as sound transmission attenuation is considerably lower in comparison to stone walls (difference depending on the specific construction details though).
Typical difference in attenuation can be as high as 10-16 dB for a wall.
Attachments
I disagree. The reason for many of the issues in this discussion is because most of the work concerning early reflections pertains to architectural design and therefore to a sound source in its primary acoustic environment. The extrapolation of this work to the reproduction of recorded material in a second listening environment is not trivial and the reason for much of the "mystery".
A friend from California sent me a picture of her refurbished room.
My envy led me to recommend a smoke detector.
She answered with something about the anti-seismic practices of my house.
Attachments
This last point is the principle one. If one looks at the gammatone filter model of hearing it is clear that the highest audible frequencies are detected and resolved in under a ms. while the lowest can take as long as 20 ms to resolve. Hence a 10 ms reflection will affect only those frequencies below about 700-800 Hz. The bottom line is the longer the delay and the lower the level the less of an effect the early reflections will have on imaging.
This is why I recommend high DI in small rooms, because one can prove that the higher the DI the slower the reflections/reverb will build and the lower the level of these earliest reflections. For longer time high DI and low DI will both have about the same amount of reflections - high DI just takes longer.
Hi,
About ER time delay and level there is a relatively simple test everyone can make ( if you have a software which allow use of (serious) reverb plug ins): you disable tail to only have ER and you play with time delay and level.
I suggest to try 10ms/-10db, 15ms/-15db and 20ms/-20db.
Very revealing of the difference of 'presentation' you can have and will give you an idea of what YOU may like or not ( the target you prefer).
Then reactivate tail once you have done this and compare them too. You will have an idea of overall target you prefer (ER target and RT).
For my own taste i like 15ms/-15db and 20ms/-20db, the former giving an 'in your face' kind of presentation, the latter a kind of 'step back' presentation. Or described differently, front row seat ( wide stereo image but you loose a bit of depth of projected image), last row seat (not as wide stereo as previous but more feeling of depth of rendering ( but not nescessarely of the message reproduced*).
Try it it's fun.
Leather couch. Yeah it does change sound dramatically! Mine introduce some weird things around 1k/3k when i'm deeply seating in it... to the point it is almost like some phase manipulation fx with larger than life width ( but i loose depth). For critical listening better not be deep in it! Next one will be fabric for me, or better a cinema/theater seat if i can find one.
*: depth may be a poor choice of word as it may be interpreted as depth within the message reproduced, perspective could be better choice as it is more relative to presentation in my view.
About ER time delay and level there is a relatively simple test everyone can make ( if you have a software which allow use of (serious) reverb plug ins): you disable tail to only have ER and you play with time delay and level.
I suggest to try 10ms/-10db, 15ms/-15db and 20ms/-20db.
Very revealing of the difference of 'presentation' you can have and will give you an idea of what YOU may like or not ( the target you prefer).
Then reactivate tail once you have done this and compare them too. You will have an idea of overall target you prefer (ER target and RT).
For my own taste i like 15ms/-15db and 20ms/-20db, the former giving an 'in your face' kind of presentation, the latter a kind of 'step back' presentation. Or described differently, front row seat ( wide stereo image but you loose a bit of depth of projected image), last row seat (not as wide stereo as previous but more feeling of depth of rendering ( but not nescessarely of the message reproduced*).
Try it it's fun.
Leather couch. Yeah it does change sound dramatically! Mine introduce some weird things around 1k/3k when i'm deeply seating in it... to the point it is almost like some phase manipulation fx with larger than life width ( but i loose depth). For critical listening better not be deep in it! Next one will be fabric for me, or better a cinema/theater seat if i can find one.
*: depth may be a poor choice of word as it may be interpreted as depth within the message reproduced, perspective could be better choice as it is more relative to presentation in my view.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.