Best Sounding Speakers According to Scientific Tests

[largedoor]

Bad tests, not remotely awesome!

I know this is an old thread, and I didn't even read more than the first 4 pages. After that, I just skimmed a bit.

But, wow..what a mess. It really doesn't sound like good tests to me.

To begin with, each speaker needs to be placed properly to interact with the room in it's ideal fashion. To do this, it takes a trained set of human ears, with good knowledge of what they are doing and why they are doing it, and a hour or so. I simply don't believe that this was done with every speaker.

Next, arguments about very important lines of speakers being omitted are very valid. Some of the better speakers in the world weren't mentioned. Bowers and Wilkins, Paradigm, Sonus Faber, MacIntosh, JM Labs, etc...etc...? Without these, why bother?

Let's talk about the panel. They had their hearing tested, and performed well. Tested for what? Frequency response?
Was any testing done for the ability to recognize imaging, soundstage size and depth, phase issues, decay and all of the other things that audiophiles understand but the general public generally doesn't? Many people with "good" hearing couldn't tell a world-class speaker from a Bose-wave radio if it bit them on the butt. Many people with poor hearing, but knowledgeable about high-end audio do great at evaluating these things. It takes a lot more to create a quality speaker than just a pleasing tonal response. Most people have no clue what to listen for.

During testing 8 people were placed in a room at the same time? Seriously? Eight human bodies alone affect the perceived acoustics of a listener. Was there a front row and a back row? If there was, both the front and back rows heard different things, based on the interference from the other listeners. No? They were in a line? So the guy on the far right was expected to evaluate sound stage depth and imaging? There is no way this would work.

The speakers were hidden from the panel? How were they hidden? How did this affect acoustics and imaging?

The cable findings were interesting. I've done many repeated controlled blind listening tests when I worked with a large high-end cable manufacturer for many years. I've repeatedly had results where the panels could clearly distinguish differences in audio through different cables. This occurred predictably and repeatedly from panels testing cables from inexpensive $50 cables to $10,000+ cables.

Anyway..I'm with the guy who said he didn't believe these tests occurred. It sounds like BS to me. More-so, if they did occur, it sounds like they were so sloppily conducted as to be useless.

A lot of money is going towards testing that will produce no valid use. OR...a lot of words are being spent on a useless gag on a forum. The whole premise of this thread sounds like troll-bait.

[rlpiii]

Originally Posted by PalmHarbor

I have been involved in high fidelity since 1969 and I finally got Martin Logan ESL and my search for the best speaker is over....its Martin Logan. Period

lolz.

[Celt16]

Yep, I applaud the effort conceptually, but have to join the growing din on the approach. Add fine quality products like Bowers & Wilkins (B&W) speakers (they're good enough for Abbey Road studios) and others to your test, and leave Bose to the folks who want to over-pay for home-theater-in-a-radio-foorprint.

[ChiefTK]

I don't see Revel included in the testing.

The original Ultima line Studios and Salons are exceptional loudspeakers and the Ultima2's are even better.

For the price, there isn't anything out there that sounds as good as my Salon2's.

The Vandersteen 7 at $45K is better.

[PapaSquillo]

Let's take the Way-Back machine and see how little this test really means. And I say this with great respect for the long hours of work and planning that went into it. I am delighted that people are finally getting around to listening again!

Way back in the '70's, my friend Gregg and I used to stay till almost dawn at Diener Stereo in Champaign, IL, trying to use the Advent speaker "comparator" switcher to give our clients a chance to listen between all of our speakers at equal volumes, so they'd pick the right speaker for their own personal taste. Trouble is, you can't even do an absolute volume match. You could play pink noise and balance the volumes with test equipment, and the total SPL could read exactly the same for two speakers. Then you play organ music on it and the speaker that has deeper bass plays louder, and louder almost always wins in a demo. Play another recording that favors other frequencies and the other speaker is louder. Take two identical pairs of speakers and adjust one pair slightly louder and it will win, or the slight difference in room placement will turn the two identical pairs into entirely different speakers...

Now try to compensate for the dynamics differences, dispersion, and the 20 other major qualities, and you still haven't touched the interaction of the amp or the room acoustics or the quality and balance of the recording.

All this being said, listen to enough speakers on your favorite, well-recorded music, and you'll find your own preferences. Just stay open, and don't fall in love with a speaker, or you'll never be objective again!

The first thing you need to do is to find out what qualities of music turn you on the most and make you fall into the music the quickest. Go to lots of acoustic live events and learn what your favorite music sounds like for real.

Then go back to the showroom and flip the switch in your head that says that it's OK to expect a pair of speakers to sound like you're listening to the real thing. Suddenly, the speakers you like most will be much more obvious.

Good luck, all of you.

PS

PS- If you find a speaker that does everything well, it will appeal to a wider range of listeners. If you pick a speaker for yourself because someone else says it's the best, you're due for a long spell of listening frustration, unless you listen exactly the way your mentor does. -PS

[cigarjohn]

Originally Posted by Lotus

I was asked by Wes to post some results and ratings from a scientific study I'm a part of.

What we've done:

2,000 "Test Subjects," had their hearing checked and passed. We decided they had "above average to good," hearing for their age and sex.

Then we tested different parts of the audio chain: cables, amp, pre-pro, speakers, and CD players. Only 8 people could be tested at a time due to the limited "sweet spot," for sound listening for the tests. For speakers our reference rig consisted of a complete Arcam FMJ setup: amp, cd player, pre-pro. The wiring was all mid-priced Cardas. They listened to a track of Beethoven, Pink Floyd, Waylon Jennings, and Public Enemy (just so we could affect the political race, and get the phrase "corrupt like a senator," ingrained in their brains). 5 identical listening rooms were used. Each had 2 sets of speakers properly set up. They'd listen to one track and move to another room. Speakers would randomly be switched from "A," and "B" for example in room 1. So say a PSB and a Def Tech were in that room, one was A and one was B, this could be switched without the listener knowing we had changed designation for a different track.

The test was completely blind. The test subjects were never told what speaker they were listening to, nor could they see them.

We tested hundreds of different speakers at first (pretty much every brand you could think of). After many tests some speakers were removed because it was apparent they were "disliked," from results. Basically in one day 40 people would test 40 different speakers. We did this by price point. The first tests were for $0-$1000 pair of speakers. They weren't asked to "Rate," just pick the best 5 from a group (of 10). It was never the same 10 vs the same 10 more than per 40 people. We mixed it all up. Still we were able to identify speakers that were "preferred," to "disliked." Interestingly this process of elimination only resulted in nearly 30% of the speakers priced under $1,000 to be eliminated. The entire process was repeated multiple times until we had valid data showing the 10 most preferred speakers. Then those 10 speakers were "rated," by the test subjects.

Then we took the top 5 rated speakers under $1000 and added them to the speakers under $2,000 and repeated the tests with the speakers priced $1000-$2000. We again reached a point where we had the top 10 speakers. These were rated and the top 5 were then compared to $2001 to $5000 speakers. Ten speakers were preferred, then rated. Then $5001 to $10,000. Then $10,000+.

These tests are not finished. So my posting results should be taken as a work in progress. So here are some incredibly interesting findings:

The Top Rated Speaker for under $1,000:

Magnepan MMG, $550 direct from Magnepan. This speaker was the overwhelming choice in the category.

The Top Rated Speaker for under $2,000:

Magnepan 1.6 for $1,800 from Magnepan Dealers. Interesting note: The Magnepan MG12 and MMG (the top under $1,000) also made the top 5 highest rated under $2,000 pair of speakers. The Dali Ikon 6 and Martin Logan Source also made the top 5 rated. (Maggie 1.6, Martin Logan Source, Maggie MG12, Dali Ikon 6, and Maggie MMG in that order for those wondering).

The Top Rated Speaker for under $5,000:

Martin Logan Vantage, Magnepan 3.6, Anthony Gallo Reference 3.1, Def Tech Mythos Super Towers, Martin Logan Purity

The Top Rated Speaker for under $10,000:

Martin Logan Vantage, Martin Logan Summit (technically more expensive than the Vantage)

Top Rated Speaker for over $10,000:

Martin Logan Vantage

We're not done with our testing. We're about 60% finished, so the numbers could change. We got most of our gear from donations, the rest we bought or have on loan (the 7 speakers we have over $10,000 for instance). We had to skip certain models and generally chose the most expensive of a line or 2nd highest in a line.

Speaker Brands we tested: Bose, Klipsch, Polk Audio, Martin Logan, Magnepan, Def Tech, PSB, Mirage, Wilson, Anthony Gallo, Sony, KEF, Canton, and Krell just to mention some of them. Like I said we tried to test as many cheap speakers as possible and removed complete lines that didn't test well. So take Bose for an example. They tested horribly, and by the time we eliminated their 101s and 201s, we only tested one of the Life series. Polk Audio, Klipsch, and many others suffered the same fate, with us just picking one of their line from a price range.

I'll let everyone here know when we're finished, because we'll be selling the gear.

Those Magnepan 1.7's look to be a great buy for the money. It has gotten great reviews. But I don't understand why no one has reviewd the new B&W 805 Diamonds. Isn't B&W a good, high quality compnay too?

Peace!