• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Binaural blind comparison test of 4 loudspeakers - II

Which speaker comes closest to the original recording?

  • Speaker A

    Votes: 7 25.9%
  • Speaker B

    Votes: 4 14.8%
  • Speaker C

    Votes: 3 11.1%
  • Speaker D

    Votes: 13 48.1%

  • Total voters
    27
  • Poll closed .

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
if you check the spectra graphs that @ctrl made the deviations are barely enough to make the dramatic effect that you're mentioning. the direct sound ofcourse is still flat after all.
So this is what I don't understand. Is the listening position judged to be in nearfield relative to the room size? What is the direct/reflective ratio here? If the direct sound is dominating here, I can understand better.
 

steve59

Major Contributor
Joined
Dec 18, 2019
Messages
1,018
Likes
727
I’m posting right after reading up to the big reveal. can someone explain how the speakers we listen thru will effect the result? If I listen thru B&W wouldn’t that treble rise double making my favorite speaker unlistenable? Curious to what I’ll learn here and now knowing which speakers are used would there be a way to get more accurate results
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
When I read about their new methods, they seem to capture the impulse response which is then processed on any music of choice. Is that correct? So does that solve the room issues?
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
I’m posting right after reading up to the big reveal. can someone explain how the speakers we listen thru will effect the result? If I listen thru B&W wouldn’t that treble rise double making my favorite speaker unlistenable? Curious to what I’ll learn here and now knowing which speakers are used would there be a way to get more accurate results
Headphones needs to be used. Neutral ones.
 

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,081
Likes
8,917
An interesting thing for the DIYer is that GRIMM is very open how and why they have done things. The drivers used can be bought.
A skilled DIY:er can make a clone very easy.
DIY, I'm lucky if I can do basic electrical and plumbing repairs. Woodwork is for the lucky people who know how to do it.

Does anyone sell the Heco Direct in the US?
 
Last edited:

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,763
Likes
16,237
I agree but that doesn’t mean that the design is flawed, it’s very difficult to match the drivers so well like they did. It might be the most linear non-coaxial speaker in this forum.
It has impeccable engineering but as said that its easier to get the vertical directivity good with a 4" woofer than with the usual 5"-8" ones and the DSP crossover is an additional help to get very smooth responses.
 

Nicolaas

Active Member
Joined
Feb 2, 2020
Messages
132
Likes
114
Can you really find them at that price locally? Here they're quite a bit more expensive (~3000€) but you can find the Einklang for about 1600€. Looking at it for a millisecond, I assume that this is a wildly different design though.

Still though, even for the ~3000€ that the Direkt goes for locally - it's dirt cheap compared to the competition!
Look at MD-sound.de
 

fffffgggg54

Member
Joined
Sep 22, 2021
Messages
18
Likes
30
Location
Greater Seattle Area
Preference for D.
Looking at the graphs makes sense why. Final decision based on the piano track and tonality relative to original. D sounded the least recessed (quieter, not as “full” a sound) in the higher parts of the clip, especially compared to B and C. A sounded ever so slightly recessed compared to D in this region as well. Conclusion matches graphs, as the responses of A, B, and C are all in some way recessed around 1-2 kHz, corresponding roughly to C6-C7 on the piano.

Might have heard the lack of bass roll off in D as well, but less noticeable than the upper mid dip.

Perhaps some other people not noticing/mentioning recessed piano notes may be because of different focus in listening or due to age and hearing loss (I’m <20 and can hear up to around 18-19 kHz).
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
Preference for D.
Looking at the graphs makes sense why. Final decision based on the piano track and tonality relative to original. D sounded the least recessed (quieter, not as “full” a sound) in the higher parts of the clip, especially compared to B and C. A sounded ever so slightly recessed compared to D in this region as well. Conclusion matches graphs, as the responses of A, B, and C are all in some way recessed around 1-2 kHz, corresponding roughly to C6-C7 on the piano.

Might have heard the lack of bass roll off in D as well, but less noticeable than the upper mid dip.

Perhaps some other people not noticing/mentioning recessed piano notes may be because of different focus in listening or due to age and hearing loss (I’m <20 and can hear up to around 18-19 kHz).

I would say it depends on many things. The main things IMO are how the speakers dealt with the stereo system errors and also how the recording was made, taking into account the ratio of direct to reflected sound. I preferred B for piano because in my mind of the more natural timbre.
 

BostonJack

Active Member
Editor
Joined
Jul 2, 2019
Messages
288
Likes
350
Location
Boston area, Cambridge, MA
Is it possible to do this test in a recording format other than binaural. The original test recording was really strange, like strangled somehow.
I've heard some recordings using ellipsoid (<- I think, not certain) microphones that capture room sounds.
 

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,437
Likes
4,686
Is it possible to do this test in a recording format other than binaural. The original test recording was really strange, like strangled somehow.
I've heard some recordings using ellipsoid (<- I think, not certain) microphones that capture room sounds.

You have to come to terms with that feeling. I had the same reaction on the first test as in "people paid real money for those crappy speakers?"
Once you've heard a few, you can put things in perspective, none of those sound "good" as speakers in the room with us would sound. But, so far, it seems that despite the recording hurdle, what the majority of people chose correlates decently with what the preference research said they would (on average).

I wouldn't draw huge conclusions from such tests though. Although most people made the expected choice, one could argue they shouldn't actually have done that in theory, given that their devices weren't calibrated or at least standardized. But maybe they are to some point (selection bias is running at full speed when you get voluntary participation, on a site called audiosciencereview, by a minority of active users. It's confusion all the way.

But is is fun.
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,789
Location
Sweden
I would say it depends on many things. The main things IMO are how the speakers dealt with the stereo system errors and also how the recording was made, taking into account the ratio of direct to reflected sound. I preferred B for piano because in my mind of the more natural timbre.
There is, like you say, many variables but nevertheless very interesting with such a test.

Both GRIMM and Revel ( both winners in this comparisons, part 1 and II. ) has :

1. Excelent directivity.
2. Low distortion
3. A straight, smooth frequency response, with no stereo-system compensation.

Its very tempting to say that Toole and Kevin Voecks is right in what makes a loudspeaker sound good and natural.
That, in turn, would mean that some other loudspeaker experts around the world, is wrong about certain things.
Maybe the need for stereo-system compensation is just a myth, and maybe even a salespoint?;)
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
There is, like you say, many variables but nevertheless very interesting with such a test.

Both GRIMM and Revel ( both winners in this comparisons, part 1 and II. ) has :

1. Excelent directivity.
2. Low distortion
3. A straight, smooth frequency response, with no stereo-system compensation.

Its very tempting to say that Toole and Kevin Voecks is right in what makes a loudspeaker sound good and natural.
That, in turn, would mean that some other loudspeaker experts around the world, is wrong about certain things.
Maybe the need for stereo-system compensation is just a myth, and maybe even a salespoint?;)
The F52 has a dip around 3 khz and several smaller Revels are voiced with slight dips 3-4 khz and even 7-8 kHz and are highly praised:


As Toole and others already have said several times:
A: A speaker with a flat on-axis and smooth off-axis is the most preferred, when evaluated in mono with music.
B: The same speaker sound different when in stereo setup playing the same music in mono, because of the stereo system errors (comb filtering).

So if A is true, the same preference for B cannot at the same time be true. You need another frequency response to mimic the mono speaker setup.

Tooles solution is to add a center channel.
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
Wow. More interesting results. The Grimm LS1Be definitely has the best measurements, and once again, the "best" measurements win the poll, despite the recording flaws. The KEF is really interesting to me. I bet it would do much better with some EQ to reduce the bass. I think KEF + EQ could probably rival Grimm + EQ, though I can see why the Grimm won without EQ. Why does KEF "voice" it that way(with 5dB bass boost)? Maybe to raise the sensitivity? The bookshelf models seem to be voiced more neutrally.

Anyways, I'd like to keep this going. I just purchased this. What else do I need? I may need some coaching as to how to appropriately measure and apply compensation(if needed).
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,789
Location
Sweden
The F52 has a dip around 3 khz and several smaller Revels are voiced with slight dips 3-4 khz and even 7-8 kHz and are highly praised:


As Toole and others already have said several times:
A: A speaker with a flat on-axis and smooth off-axis is the most preferred, when evaluated in mono with music.
B: The same speaker sound different when in stereo setup playing the same music in mono, because of the stereo system errors (comb filtering).

So if A is true, the same preference for B cannot at the same time be true. You need another frequency response to mimic the mono speaker setup.

Tooles solution is to add a center channel.
Please dont mix up 5.1 sound with 2 channel sound. Toole says that stereo listening is flawed, but he dont rekommend a stereosystem-compensation in the loudspeaker. If a loudspeaker measure flat, and has excellent directivity, then even in stereo, the sound will be good.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
Please dont mix up 5.1 sound with 2 channel sound. Toole says that stereo listening is flawed, but he dont rekommend a stereosystem-compensation in the loudspeaker. If a loudspeaker measure flat, and has excellent directivity, then even in stereo, the sound will be good.
But he admits that the 2 channel speakers sounds different from the ideal speaker in mono. He also says that don't use 2-channel, because it is flawed. If you are using 2-channel, the timbre will be different from the same speaker in mono. So the ideal speaker in mono is not the ideal speaker in stereo. It is an impossible equation to prefer one timbre in mono but another in stereo, both being the ideal speaker.

And the Revels listed are voiced different from linear on-axis, it is quite clear..
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
Thomas A wrote:
"As Toole and others already have said several times:
A: A speaker with a flat on-axis and smooth off-axis is the most preferred, when evaluated in mono with music.
B: The same speaker sound different when in stereo setup playing the same music in mono, because of the stereo system errors (comb filtering).

So if A is true, the same preference for B cannot at the same time be true. You need another frequency response to mimic the mono speaker setup.

Tooles solution is to add a center channel."


Isn't it a bit strange to use monophonic material to evaluate a stereo pair of speakers dedicated to stereophonic music material?
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,789
Location
Sweden
But he admits that the 2 channel speakers sounds different from the ideal speaker in mono. He also says that don't use 2-channel, because it is flawed. If you are using 2-channel, the timbre will be different from the same speaker in mono. So the ideal speaker in mono is not the ideal speaker in stereo. It is an impossible equation to prefer one timbre in mono but another in stereo, both being the ideal speaker.

And the Revels listed are voiced different from linear on-axis, it is quite clear..
The test result ( where two loudspeakers are used ! ) shows that GRIMM ls1 is the best sounding loudspeaker in the blind test II. Its ruler flat with no compensation whatsoever .

How do you explain that ?
24D3C851-B6B3-407B-B77A-61F95CDB1976.jpeg
 
Last edited:

TimVG

Major Contributor
Joined
Sep 16, 2019
Messages
1,181
Likes
2,573
The test result ( where two loudspeakers are used ! ) shows that GRIMM ls1 is the best sounding loudspeaker in the blind test II. Its ruler flat with no compensation whatsoever .

How do you explain that ?


Well it's not very complicated. The recordings here were stereo, and two loudspeakers were used. Any stereo recording, mixed and mastered in stereo is bound to have any needed corrections, be it loudness, presence crosstalk dip, etc, done so in the recording itself.

The result won't change if we switch to one channel (mono loudspeaker) because we are looking for the least flawed loudspeaker. In this case each of the candidates would have the same flaw, and we will adjust and unconsciously look for the next flaw.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,423
Likes
2,408
Location
Sweden
The test result ( where two loudspeakers are used ! ) shows that GRIMM ls1 is the best sounding loudspeaker in the blind test II. Its ruler flat with no compensation whatsoever .

How do you explain that ?View attachment 160418

If you put a speaker with a linear response +/-1.5 dB against speakers with +/- 5 dB it will be quite certain which will win.

With respect to stereo system corrections you need to test speakers that are within +/- 1.5 dB. Such a test has not been performed in the Scientific literature. Yet.
 
Top Bottom