mhardy6647
Grand Contributor
- Joined
- Dec 12, 2019
- Messages
- 11,482
- Likes
- 24,971
I chuff. Big time.
Don't make me trot out the Port Noise Complaint joke.
I chuff. Big time.
I'm not convinced enough about the validity of the hypothesis to be comfortable with the seeming shift of consensus in ASR to wide directivity = good. It may or may not be true.
Don't forget that (human) evolution has selected for acuity in the midrange -- to wit, in the range of frequencies spanned by the human voice.
OK, probably the noises our predatorsmademake, too!
On a more serious note, I have a feeling something essential is missing in the preference scoring system.
I think the wide dispersion hypothesis needs to be demonstrated reliably for me to accept it at face value. Especially considering the mono listening test as done here. IIRC, there's quite the gap between listening in mono and listening in stereo when evaluating the spatial qualities of speakers, which is not surprising.
It's quite popular to use the blind testing of M2 vs Salon over at AVR as proof of that hypothesis, but the results could be because of totally unrelated stuff. Interestingly lots of people have written that they vastly prefer the Beolab 90 in Narrow Mode vs Wide or Omni, which is contradicting the hypothesis.
I'm not convinced enough about the validity of the hypothesis to be comfortable with the seeming shift of consensus in ASR to wide directivity = good. It may or may not be true.
More listeners
100% agree .Actually my experience is that people like what they are used to in speakers, that is why some believe in break in, it isn't the speakers breaking in but them becoming accustomed to the difference.
I would not expect Amir to immediately like something different to what he is used to any more than anybody else would and given expectation bias, which we all have, the subjective result is unsurprising.
I think the measurements are fantastic.
I think the subjective opinion is more or less what one would expect and of no value to me personally whatsoever, so I take it with a pinch of salt.
Is there a reason the estimated typical room response does not include room gain?Putting all of this together, this is our predicted response in a typical room:
Because there is no way to calculate that based on the graphs given. Room gain is very hard to predict, as it’s not just the room but the placement as well. I would wager the graph is similar to what you would get if you placed it in the center of the room.Is there a reason the estimated typical room response does not include room gain?
I think the measurements are fantastic.
I think the subjective opinion is more or less what one would expect and of no value to me personally whatsoever, so I take it with a pinch of salt.
I'm finding that hard to accept given it's for some mythical typical room, and the whole response is dependant on size, position and absorption of the room.Because there is no way to calculate that based on the graphs given. Room gain is very hard to predict, as it’s not just the room but the placement as well. I would wager the graph is similar to what you would get if you placed it in the center of the room.
It’s better to under-predict than over-predict.
I don't doubt for a second that wide dispersion is universally a good thing when listening to stuff in mono, but Toole's research shows quite clearly that those differences attributed to mono almost vanishes in stereo.Totally fine to be skeptical about that, though I'm not sure that there's really a shifting hypothesis as much as it is just following what Toole says on the matter, and that there seems to be a trend with Amir's preferences (which for this review he specifically noted the identifiable source as potential issue). Toole does spend like half the book espousing the benefits of wider directivity/louder sidewall reflections after all, while acknowledging it's also a matter of preference.
Agreed that we could all use more research there. For me -- perhaps unsurprisingly, having gotten so much of my knowledge from Toole's book and the research included in it -- it's more that I've seen less evidence to support narrow directivity being preferred than wide. And on a personal level, I very much know this to be a subtle but clear preference.
On the other hand, as I mentioned elsewhere, narrower directivity speakers do give you more flexibility with playing around with sidewall reflections, especially if you always sit in the same spot.
More (diverse) speakers should be a higher priority IMO.
We already quite strong evidence already that flat direct sound is preferred on average and that smooth dispersion also tends to be preferred. The biggest unanswered question's are therefore about what pattern that smooth dispersion should have.
So in my view, the next round of research should focus on determining listener preferences between only excellent examples of speakers of varying dispersion patterns.
I'm finding that hard to accept given it's for some mythical typical room, and the whole response is dependant on size, position and absorption of the room.
I don't doubt for a second that wide dispersion is universally a good thing when listening to stuff in mono, but Toole's research shows quite clearly that those differences attributed to mono almost vanishes in stereo.
View attachment 64538
This graph shows the importance of not assuming too much when it comes to any one particular trait because one factor may not be that important when you introduce other factors. I agree completely with @andreasmaaan that we need to evaluate only high quality speakers with different dispersion to get any clear answer.
In order to do that with any relevancy to real-life situations, we also need to do that with a wider room and a smaller room to see if any preference between dispersion width is universal or situation-based.
We are in a situation where the absolute best "Toole-style" speakers almost by default will be coaxials and, because of that, the dispersion pattern in the theoretical "best" will be somewhat limited by the necessary size/shape of the coaxial drivers.
It makes no sense whatsoever to have the "perfect" speaker with regards to the preference score if it's nowhere near the top of the subjective listening evaluations due to limited dispersion.
You don't see any room gain in the low frequencies compared to the anechoic measurement?In my home the PIR correlates very closely with the actual in room response for a single speaker at 2 to 3m. The top octave may vary a bit because the direct sound has a great influence in this region, so it can vary with toe in and such, and if you take a stereo measurement it tends to drop down a bit further.
True, I was just making the point about the tests here specifically; that the preference score was never meant to define a single user's preference. While it does seem most people prefer the same speakers in blind tests, I'm not sure that can be totally extrapolated to correlation with a single listener over a large span of speakers.
You don't see any room gain in the low frequencies compared to the anechoic measurement?
I don't doubt for a second that wide dispersion is universally a good thing when listening to stuff in mono, but Toole's research shows quite clearly that those differences attributed to mono almost vanishes in stereo.
View attachment 64538
This graph shows the importance of not assuming too much when it comes to any one particular trait because one factor may not be that important when you introduce other factors. I agree completely with @andreasmaaan that we need to evaluate only high quality speakers with different dispersion to get any clear answer.
In order to do that with any relevancy to real-life situations, we also need to do that with a wider room and a smaller room to see if any preference between dispersion width is universal or situation-based.
We are in a situation where the absolute best "Toole-style" speakers almost by default will be coaxials and, because of that, the dispersion pattern in the theoretical "best" will be somewhat limited by the necessary size/shape of the coaxial drivers.
It makes no sense whatsoever to have the "perfect" speaker with regards to the preference score if it's nowhere near the top of the subjective listening evaluations due to limited dispersion.
It would be better if it didn't plot frequencies it cannot estimate then. Where is the cutoff?Oh absolutely. MZKM's point earlier, I believe, was that the PIR isn't meant to be very useful for low frequencies. It does okay in the center of a room though.
Yeah, I agree with you completely. I mean I think the point of that study was that trends remain similar, but I also agree narrower dispersion speakers are at an unusual disadvantage in mono, especially as you can adjust sidewall reflectins with toe in and such.
My hunch is that wide directivity/louder sidewall reflections will still be preferred by a majority of listeners in a majority of rooms because we adapt to rooms. But yeah, we don't know for sure.
I totally agree with you. I understand the appeal of applying preference scores to speakers, but I don't think the system is nuanced or refined enough at this point for the scores to be of much specific value.
I also think that applying preference ratings can be counterproductive at times, as a numerical value tends to give the illusion of certainty, and to obscure any uncertainties underlying it.
It reminds me a bit of the practice common in the German hifi press of giving subjective ratings to speakers that are broken down into categories like "imaging", "tonality", etc, with a numerical mark assigned for each and then an overall mark derived from those. On one hand, it may help reviewers be more rigorous in their assessments, but on the other hand, it tends to obscure the underlying fact that the review is a sighted, subjective one, giving a false veneer of rigour, precision, and objectivity.
It depends on where the transition happens in each room. A moving target. The PIR is still useful for determining relative bass amounts. A speaker with good bass extension in the PIR will still sound bassier than a speaker with less extension.It would be better if it didn't plot frequencies it cannot estimate then. Where is the cutoff?
I'm also intrigued by the directivity discussion and I also much prefer a wide soundstage to a more focused image but oddly I use LS50's and don't consider them constrained at all. I have them in a smaller room about 3 feet away from the sidewalls with no toe-in, that probably helps but I also think accurate vertical reflections play a part in that as well. For some reason the LS50 sound "bigger" than any other speaker I've compared them to including speakers with a 64-10 RAAL and even the KEF R3, and these were blind evaluations. I was thinking the coaxial was the reason for what I was hearing but the R3 put that to rest, now the only other explanation I can think of is maybe the rounded cabinet is playing a part in that.
I do agree that room size plays a part because listening to the LS50 at my dealer, they were underwhelming and I didn't feel any spaciousness at all. Those were in a large room further away from the sidewalls. But then again I also listened to some Reference ones in an even larger room away from sidewalls and they sounded huge so I'm not sure what to think...