Magico Ultimate vs Magico M6

[paddycrow]

Harman Kardon

It's a legitimate beef. Tests should be blind to eliminate bias.

 

[Beave]

Isn't Harman Kardon is a different division of Harman? Did they really use employees from the Harman Kardon division?


(My previous post was tongue-in-cheek pointing out the all-too-common misuse of HK when referring to Harman.)

 

[paddycrow]

Isn't Harman Kardon is a different division of Harman? Did they really use employees from the Harman Kardon division?


(My previous post was tongue-in-cheek pointing out the all-too-common misuse of HK when referring to Harman.)

I prefer to use Revel, it's more specific. And HK is a firearm where I'm from.

 

[Snoochers]

Spinorama is not the end-all.
The Spinorama doesn’t show you anything about nonlinear distortion, which is a significant part of the overall SQ (what happens when SPL increases, etc.). The same speaker in MDF enclosure would sound completely different than if it was made out of fiberglass or aluminum, yet it would look the same on the Spinorama. And, yes, it is objective, you can measure it, if you only choose to.

Yes of course spin isn't everything. I'm just saying that all this nice "stuff" doesn't mean anything if it doesn't actually improve sound objectively, regardless of what that objective measure is.

 

[vavan]

The Revel also sound more strained and grainy as SPL increases

sounds somewhat puzzling given all praises and recent f226be review/test from Erin. I only have f208 and have had no chance to audition neither BE version nor any Magico so cannot compare unfortunately

 

[ttimer]

When you say "biased", do you mean that because the subjects have been trained to prefer neutrality, they will be biased against a speaker like the A3 that has a less-neutral presentation? This seems a fair criticism IMO.

Harman Kardon
It's a legitimate beef. Tests should be blind to eliminate bias.

It is possible that a lot of Revel employees are able to recognize their own product in a blind test. Which defeats the purpose of a blind test.

And then there is sample selectivity. Testing such a very specific subgroup is probably not very informative for predicting general population preferences (whether or not that is the right target for high-end speakers is another question).

 

[napilopez]

Yeh, I also would have expected the 228 to be more preferred. I don't see much of a difference in DI around the midrange though, maybe just a slight difference around 200-400Hz...? My guess is that it has more to do with bass extension.

Assuming you're looking at the spins, unfortunately they're unreliable when it comes to directivity as we usually mean it when discussing soundstage. The ER breakdown is much more useful for this, but I've never seen Harman share that, though for the L82, they did share a graph with a horizontal DI curve, which I appreciated.

Stereophile's measurements of the F228Be and Salon2 make it very easy to the Salon2 would appear to have a significantly louder and, probably, more precise-sounding soundstage.

F228Be

Salon2

For me roughly 2K to 10K is where the soundstage magic happens. The salon 2 appears to be 2-8dB in louder in this region, while also being better controlled.

I would imagine both speakers are flat enough that such a significant difference in horizontal directivity plays a larger role in preference.

 

[napilopez]

It is possible that a lot of Revel employees are able to recognize their own product in a blind test. Which defeats the purpose of a blind test.

Not really, from what I've read. I mean, Im sure it happens occasionally, but once three speakers are in the mix, things get more complicated. I'm pretty sure I've heard about about Harman engineers picking bookshelf speakers over their more expensive towers sometimes too and swearing it was a tower they were listening to.

 

[thewas]

Not really, from what I've read. I mean, Im sure it happens occasionally, but once three speakers are in the mix, things get more complicated. I'm pretty sure I've heard about about Harman engineers picking bookshelf speakers over their more expensive towers sometimes too and swearing it was a tower they were listening to.

Here is a similar case where they even graded their own product worse at a sighted test http://seanolive.blogspot.com/2009/04/dishonesty-of-sighted-audio-product.html

 

[andreasmaaan]

It is possible that a lot of Revel employees are able to recognize their own product in a blind test. Which defeats the purpose of a blind test.

And then there is sample selectivity. Testing such a very specific subgroup is probably not very informative for predicting general population preferences (whether or not that is the right target for high-end speakers is another question).

Re: sample selection, according to Olive, the difference between trained and untrained listeners is that trained listener preferences are more consistent but not different, which I presume is another way of saying that the mean is similar for both trained and untrained listeners, but that the standard deviation is smaller for trained listeners.

For this reason, he (or was it Toole in his discsussion of OIive's work?) argues that there can be no basis for the argument that the preferences of trained listeners are not representative of the general population. Trained listeners are not different; they are just more reliable. I have to say, it's been a while since I read through the material on this, so I don't know how strong this argument is. Will revisit it at some point, but perhaps others can clarify whether I've done the argument justice, and/or whether there might be any holes in it that I may have missed?

 

[andreasmaaan]

Stereophile's measurements of the F228Be and Salon2 make it very easy to the Salon2 would appear to have a significantly louder and, probably, more precise-sounding soundstage.

Very informative - thanks!

It seems that the 228Be achieves its smoother DI/PIR by counterbalancing the vertical off-axis crossover-region dip with a horizontal off-axis boost in the same range of frequencies:

So in the case of the 228Be, it seems that, as far as the Spin is concerned, two wrongs are making a right.

Thus the difference in preference may be explained by beamwidth, as you suggested, or it may be to do with the fact that the Salon2's horizontal off-axis performance is smoother (or a bit of both ofc).

 

[HenryB]

That's because it's a--let's say exaggerated--statement. Either he is listening to them at above 110 SPL at 5 meters or more (in which case a PA system is more suitable--or maybe it's his ears giving out at such a high output) or it's a declaration that he has "magico" ears.

No Sir, not exaggerated at all. I am sure your "Revel" ears will hear it as well. Check back with us once you actually make a comparison, like I, and all my audiophile friends did
BTW, some dealers have both brands, should not be that difficult to do.

 

[andreasmaaan]

Is it really so inconceivable that he prefers the Magicos partly/mostly because of how they sound? I agree that psychology could be - must be - playing some role here (it always does), but it also seems, based on the measurements we have, that the Magicos have a more laid-back presentation with a slightly recessed upper-midrange/low-treble (@HenryB, we await your measurements!). And perhaps they have better bass extension and/or their sealed bass alignment better complements Henry's room. It's not as though there appears to be anything drastically wrong with them.

I just don't find it so inconceivable that someone couldn't possibly favour this kind of a tonal balance. And I also think it might explain why the speakers seem easier to listen to at high SPLs.

 

[HenryB]

Is it really so inconceivable that he prefers the Magicos partly/mostly because of how they sound? I agree that psychology could be - must be - playing some role here (it always does), but it also seems, based on the measurements we have, that the Magicos have a more laid-back presentation with a slightly recessed upper-midrange/low-treble (@HenryB, we await your measurements!). And perhaps they have better bass extension and/or their sealed bass alignment better complements Henry's room. It's not as though there appears to be anything drastically wrong with them.

I just don't find it so inconceivable that someone couldn't possibly favour this kind of a tonal balance. And I also think it might explain why the speakers seem easier to listen to at high SPLs.

Indeed possible, but I believe it is it's more than that. Distortion measurements could have settled that. I am surprised a capable compony like Revel does not go there. Also, I will add that I don't necessarily "prefer" the sound of the A3, I am well aware of the somewhat laid back presentation, and indeed wish it will have more presence, but without the strain that comes with it.

 

[andreasmaaan]

No, not at all. It doesn't matter that he likes it more. Who cares. But I was very precise about what I found troubling. Please re-read the post.

@franspambot I intentionally didn't quote you because I wasn't directing my comment specifically at you

What you found troubling was the idea that distortion was the explanation. I tend to agree with you on that count. Hence my suggestion that the differing tonal balance and/or bass alignment were more likely explanations.

Indeed possible, but I believe it is its more then that. Distortion measurements could have settle that. I am surprised a capable compony like Revel does not go there.

Yes, also a possibility. The reason I tend to think this is less likely is simply based on the research on distortion audibility, which shows that it is not strongly correlated with listener preference unless extreme. Will be interesting to see the results when your friend visits with the microphone.

 

[napilopez]

Indeed possible, but I believe it is it's more than that. Distortion measurements could have settled that. I am surprised a capable compony like Revel does not go there. Also, I will add that I don't necessarily "prefer" the sound of the A3, I am well aware of the somewhat laid back presentation, and indeed wish it will have more presence, but without the strain that comes with it.

@HenryB, as mentioned by others, I think the difficullty some of us are having is the inclination towards assuming distortion having all that much to do with it when FR, directivity/room interaction, and individual psychological factors including personal preferences appear much more likely. It doesn't seem to jive much with the science. For example, you mentioned earlier that no one would hear the differences from 100-300Hz, which researchs suggests should certainly be audible, but are pointing to (in my view) relatively small differences in distortion as defining the difference. We know people can detect tiny changes in frequency response, and it having an effect on preference.

With distortion, things are a lot less clear. Even when we have a rough idea of audibility thresholds, we have a far muddier picture of how much distortion is needed for something to sound bad.

BUT

As a counterpoint to myself, it's worth noting that at least some at Revel themselves have also pointed to distortion mattering at the top level. For example, when sound and vision compared three revel speakers in Harman's double-blind listening lab, the Revel F208 beat the Ultima Studio2 (not the salon2), by a small margin, and both beat the F206. When asked why the F208 beat the studio2, one possible explanation was that the F208 simply had a little more bass, but Kevin Voecks also had this to say:

"As you know, many people argue about double-blind tests. Most of their arguments are without merit, but not all. One of the most important is that in my opinion and observation, it does indeed take extended listening sessions to hear the more subtle differences. The important thing is that these more subtle differences can indeed become more evident over time. Having listened to the Performa3 series and Ultima2 series both for very long periods of time, the difference at high frequencies especially is dramatic. The Ultima2 tweeter is so much "cleaner," with vastly lower distortion (even though the Performa3 distortion is far below most speakers) that it is much easier to listen to without fatigue. Combined with the advantages of low diffraction, it is the high frequency range that causes the Ultima2 series to win in long-term listening tests. Getting back to the blind testing, that kind of difference is best heard with longer sessions. There must be breaks between long sessions, as fatigue sets in, but that is where differences that audiophiles live for become apparent."

So, if Kevin Voecks can say that the difference between the Performa3 and Ultima2 series at high frequencies is "dramatic," seems fair that HenryB can say he thinks the difference between the Magicos and Revels is dramatic in extended listening as well

....

Now counterpointing my counterpoint, I'll note that hasn't personally been my experience. As a reviewer, I've been doing an average extended listening of about 1 new pair of speakers a month for the past couple of years, and not a single time have I felt my opinions change with extended, eyes-closed, all-in listening. Then again, I've also never experienced listening fatigue other than from obvious high SPLs so *shrugs.* Perhaps my ears are just silver, not golden.

 

[richard12511]

Yeah, the distortion figures I've seen for both speakers are really good, and I honestly doubt the differences between them are audible. I would think the frequency response differences are going to be far more audible, and far more likely to be responsible for differences in preferences. @HenryB may just prefer the more laid back presentation of the A3, and I don't think that necessarily disagrees with the science. All the science can really tell is that "most" people should prefer the more neutral 228be over the A3, but "most" is not everyone. It's all but impossible to account for individual preference.

 

[Zvu]

I would not try to compare the distortion graphs between those two based on NRC measurements. Those measurements show a line and nothing more. Its harmonic content however is hidden - does the third, fifth or second harmonic dominate, are they equally represented in that line and so on. All the research on distortion relies upon something called statistical significance. There may well be that 5 percent could hear higher distortion every time but in gross sense of the whole research they represent anomaly. I never found results about percentages of tested people that could hear or could not hear the differences - we see just results.

Decent test was JBL M2 vs Revel Salon 2. Only there it is mentioned that while for the whole bunch Salon 2 wins by a margin, there were also people on both sides that always preferred M2 and other that always preferred Revels. For those guys the M2 isn't just 5 percent better than Salon 2 - it is 100% better as they always preferred that sound and could recognize it. Even bigger thing for me is that it is not their familiar listening enviroment.

Because of that, in my opinion, loudspeaker sound will never be completely described with few measurements. Maybe it can be described with relations of measurements that we know: how low distortion* is low enough that no one on planet can perceive it, how smooth response is smooth enough that no one on planet can perceive the differences, how tilted sound power response is tilted enough for a certain enviroment... Measurements of today gives us very good clues and only if they are done with great resolution, but you still have to go listen to see if you will like it.

*By distortion i mean resonances, harmonic distortion, diffraction and everything else that is not fed in the loudspeaker but is a product of the loudspeaker itself.

 

[o2so]

perhaps MAGICO A3 VS. REVEL 228 BE conducted by harman where "All listeners rated Magico as their least favorite choice"

Very helpful. One has to hope the listeners weren't all already very familiar with how the revels sounded in that room.

 

[Frank Dernie]

I'm looking at houses that don't cost that much!

You are not their target market then