Blind Listening Test Results (N=4) : Revel F206 vs Ascend Sierra RAAL Towers

[echopraxia]

Ja, I feel the same regarding the so called classical music. A term that is more often than right used wrong. I was kind of a fanboy when listening to "it" ( wrong term, that means ;-) through the 5 valve radio of my parents. But then electronica chimed in, again through the (German) Dampfradio. PGP sequencer and stuff like that, and / or the early Stockhausen, anybody? Steampunk by todays standards.

On the other hand, Stairways To Heaven is fun to listen to, but as a testprogram, I don't know. Wasn't it the last piece on a vinyl record? Last piece means least quality due to slowing down of the linear speed of the groove. All the sounds gets so congested. It starts clear with few instruments and becomes thicker and thicker, without actually changing loudness. Not to forget the accumulated dust at the needle, and accumulated wear out there. To me it more or less represents a practical joke of the sound engineer. Super drama, so close to going kitsch, good art alltogether, but as a test?

I agree in general, but for this test I decided I will not influence the participants' music choice in any way. Maybe it's true that this makes the test less widely applicable in some ways, but in other ways it's definitively more useful in its conclusions to those actually listening who would rather investigate the question of "what speakers do my favorite music sound best on?" (some of which include music the participants heard live at concerts in person) without some stuck-up "audiophile" telling them how poorly recorded/mastered some songs might be, or how their genre might not span enough of the frequency spectrum to be a reliable indicator of all the speakers' tradeoffs.

Aside from the general annoyance such a thing would cause, my worry with forcing a set of 'approved' test tracks on participants is that it puts them in a position to choose which speaker sounds best for a song they may not like at all, or may have no frame of reference of what it's 'supposed' to sound like otherwise.

Anyway, this is why in the full log I record the song being played (and I think there are enough genuinely high fidelity songs tested here to be useful), and reasonably detailed notes of the impressions of each song -- so you can select from the full list whatever songs you choose to be relevant, and take from the impressions whatever conclusions you want. Notes from "Participant 2" were sparse since that was me, since I've already written enough at length about my impressions of these speakers (and in this case my blind impressions matched exactly my expectations).

P.S. Regarding older songs, I was so impressed recently with this song from 1965 that is incredibly well mastered/recorded, given the era, that I used it as one of my test songs. It's really amazing the range of recording/mastering qualities that exist from just about any era.

 

[GelbeMusik]

... I was so impressed recently with this song from 1965 that is incredibly well mastered/recorded ... .

Listen, I think You simply love good music ;-)

With Curtis' song, how much forward, and bodied at the same time was his voice? Not the least the rather subdued bass, a killer groove. I think I would love this piece regardless of the speakers anytime.

Now, the "classical" violin that was fake. First I read John Williams as Aaron Copland. But the piece was so devoid of any ... what I would be after with music. And I go to the concert quite often, if You accept the reference. So hifi-ish it is presented.

Better go with Aaron. He has this famous Fanfare For The Comman Man.

 

[Feanor]

@echopraxia, did you listen to any real music, i.e. classical? You don't really have to like classical music but good classical selections are a much better basis for judging the quality of a speaker.

I apologize, @echopraxia, I was being a silly troll.

I stick with Classical for my own evaluation purposes but concede that some types, e.g. chamber music are not so useful in general. Personally I use orchestral and large-scale choral almost exclusively for that purpose and those forms I find particularly useful when it comes to detail, transparency and soundstage.

Other types of music are, to be sure, at least equally appropriate for people who mostly listen to those types.

In any case you blind preference testing is very commendable and interesting.

 

[Kal Rubinson]

This weekend, I finally got around to doing a proper blind listening test between the following two tower speakers:

Can you tell us in detail how the "proper blind listening test" was done?

It’s a perfect tie!!

Or a the outcome of random results.

Excellent music choices!

If you say so. I don't recognize all but one.

In terms of genre correlation to speaker selection reliability, we have this from research:

I think that mostly reflects the general familiarity of listeners with a particular genre. It would be interesting to do a meta-analysis grouped by genre preference.

 

[echopraxia]

I apologize, @echopraxia, I was being a silly troll.

I stick with Classical for my own evaluation purposes but concede that some types, e.g. chamber music are not so useful in general. Personally I use orchestral and large-scale choral almost exclusively for that purpose and those forms I find particularly useful when it comes to detail, transparency and soundstage.

Other types of music are, to be sure, at least equally appropriate for people who mostly listen to those types.

In any case you blind preference testing is very commendable and interesting.

Sorry, I just couldn't resist replying the way I did -- no hard feelings I assure you, it was just fun to pick on you picking on non-classical -- all in good fun over here

BTW I'd be curious to see some links to good orchestral recordings you'd recommend to use for comparisons. For some reason, it seems rather difficult to find good recordings in these genres.

 

[echopraxia]

Can you tell us in detail how the "proper blind listening test" was done?

Yes: Without bias or conflicts of interest, and published for the good of audiophiles everywhere. Interpret the data however you wish.

Ask a snarky question, expect a snarky answer In seriousness, if you follow the links you'll see more in-depth explanation of the procedure. I'm sure I didn't explain everything, so feel free to ask any specific questions you may have. That said, my use of the word "proper" here was mostly used to colloquially to distinguish this from less carefully controlled experiments such as sighted subjective comparisons or impressions.

Or a the outcome of random results.

Yes, the perfect tie is most certainly just a rare random outcome here. The impressions on treble/bass however were almost certainly not random, if you observe the consistency of the data there.

But yes, we are always in need of more data. Too few people actually perform and post blind listening tests these days

In controlled blind tests you hosted, how many participants did you have per test, and song tracks did you use? Forgive me for my ignorance, but could you share a link to the data from blind listening tests that you performed with N>1? I'd love to read them.

If you say so. I don't recognize all but one.

Rumor has it there might be more good music tracks than those we've already heard

And I mean this as a general rule -- not a comment about any single person.

 

[ernestcarl]

I don't recognize all but one.

Just one? LOL

I've heard 10 of the tracks in total myself, but do know probably half of the artists. None of them I'd probably include as part of my own "test tracks" playlist though... Since I just stick with test tracks I'm already most intimately familiar with -- wouldn't even call the stuff in that playlist of mine all "excellent" music. It does inlcude a wide variety of genres, though; highlighting the sub bass, bass, mids, and treble frequencies as well as some extreme 'stereo field' test tracks for good measure.

 

[krabapple]

It is a typical use case for an EQ. It may be set to a grand average, weighing in the most preferable records more, or one would even go for an EQ setting specific to a particular genre, listening volume and such--again Dr. Toole.

Dr. Toole advocates for old fashioned things like broadband bass/treble pots, to adjust for EQ problems in individual recordings (rather than genres)....some of which are attributable to the unfortunate 'circle of confusion' that exists between the music recording and playback spheres.

 

[KaiserSoze]

Another thing that would have been interesting is if I EQ’ed a slight bass boost into the Ascends for this test. This is actually how I usually run them. I find that with a slight midbass boost, it mostly “fixes” what is lacking on the Ascends on some tracks. However, even when doing this, I still find the Revel bass to be preferred (not sure why), which is why I still really like the Revels, and partly why I plan to upgrade to the PerformaBe series.

I did not EQ either speaker for this test. It would be really interesting to compare with both equalized, but it seems even Harman’s blind tests don’t do this, so I didn’t want to diverge too much from a pure test of each speaker pair.

Since simple equalization via a tone control is ubiquitous, the way that speakers really ought to compared is with each equalized to the extent that is possible using simple bass and treble controls. Then it becomes a question of which one plays bass the cleanest, which is the more meaningful question given the ease with which simple equalization may be applied. The problem, though, is that isn't necessary fun to listen to speakers that are driven to distortion. An alternative approach would be to apply equalization to one or both so that they both exhibit similar levels of non-linear distortion in the bass, and then let the listeners judge which has the more capable bass. This approach would probably work in the great majority of cases where both speakers are naturally bass-shy. In cases where one speaker ends up with exaggerated bass, then it would be necessary to lower the bass level of both speakers, keeping the distortion profiles similar to each other while doing so. In any case, it has always seemed silly to me to judge speakers based on their natural bass loudness with indifference to distortion level.

 

[stevenswall]

The RAAL tweeter is quite special. What has been debunked as myth for long now is the low-mass / fast transient confusion.

Didn't know it had been debunked, but I found this a year ago or so. Looks like less ringing, and the same transient response. See page seven here:

http://www.ascendacoustics.com/pages/products/speakers/SRT/Ascend Sierra Ribbon Tower.pdf

I wish the paper included beryllium tweeters.

 

[Kal Rubinson]

In seriousness, if you follow the links you'll see more in-depth explanation of the procedure.

Sorry, I missed that link. Weren't the speakers visible? How symmetrical is the room? Were there relative positions varied?

In controlled blind tests you hosted, how many participants did you have per test, and song tracks did you use? Forgive me for my ignorance, but could you share a link to the data from blind listening tests that you performed with N>1? I'd love to read them.

I have never conducted double-blind speaker tests although I have done innumerable double-blind tests in my scientific work which has nothing to do with this. I have be a subject in a number of blind listening tests.

Rumor has it there might be more good music tracks than those we've already heard

No doubt. It is just that I cannot directly relate to the descriptions of the listeners except very generally.

 

[echopraxia]

I have never conducted double-blind speaker tests although I have done innumerable double-blind tests in my scientific work which has nothing to do with this. I have be a subject in a number of blind listening tests.

I recommend it It's very useful when comparing two speakers -- so much more than any sighted test. Of course, no matter what you do, your results will be questioned. I don't worry about that though except to the extent that I can directly improve future tests.

Sorry, I missed that link. Weren't the speakers visible? How symmetrical is the room? Were there relative positions varied?

The speakers were visible, but which speaker was playing is not known to the listener (and the assignment of "A" vs "B" from the listener's perspective is randomized between tracks), and the listener is invited to close their eyes. Speakers were placed symmetrically in the room except that the speakers were laid out as (AB ----- AB) such that the width between each speaker pair is identical, rather than (AB ----- BA) which can cause soundstage width differences.

The downside of this approach is that it means there is still a position difference between two speakers. This concern was mildly mitigated by randomly shifting the speakers a bit between test sessions and mirroring (AB-AB <-> BA-BA) as in my previous blind comparison. This is obviously not as good as having a robot rig like Harman does, but it seems to work okay.

What is not well controlled for is listener preference and room effects, but for that we do need many more people to contribute by participating in and performing their own such tests.

No doubt. It is just that I cannot directly relate to the descriptions of the listeners except very generally.

Definitely agree. It would be nice if there were a standardized set of tracks we could all use. I suppose I could use the list of songs from here:
https://www.audiosciencereview.com/...sic-tracks-for-speaker-and-room-eq-testing.6/

I still worry a bit how this would impact the results of a test if the listeners don't like or listen to these genres, though. It's not a problem for me as I like these tracks and listen to a wide range of genres, but for participants who donate their time it's a bit more ambiguous how well it would work.

 

[GelbeMusik]

Dr. Toole advocates for old fashioned things like ...

Yeah! The most valuable element of kind of a reasonable hifi-ish stereo equipment is beyond any question a measurement microphone. And it costs the least of all the devices involved. Panasonic XYZ for 2$ each. O/k, with some inventive additions, not the Linkwitz cut, but that's for fun only.

Hifi without an individual microphone for cross check is like fishing without water ...

Since simple equalization via a tone control is ...

Utter confirmation. Toole is so much right in his goundbreaking answer to the ever burning questions, which parameters actually define the perceived quality of re-production of auditory sensations. Spectral balance is a first. But then comes sheer volume, with maintained spectral balance, including harmonic and, sad to say, intermodulation distortion.

Is it fair to compare speakers without equalization for let's say in-room response? In past time, what would You do? But today?

My latetst attempt to design a speaker is to incorporate a passive xo'ed directivity optimized main speaker, equalized by miniDSP, plus the opportunity to enhance its level capability by a pair of subs, again individually equalized. Four channels, two 2-channel power amps. Cheap but, as I hope, effective. She can keep her Braun A1 (Dieter Rams) for the looks ;-)

 

[echopraxia]

Is it fair to compare speakers without equalization for let's say in-room response? In past time, what would You do? But today?

I don't think it's fair to assume the average listener will spend the time and effort to use miniDSP and REW. Even among serious audiophiles outside of the ameture (and some pro) 'speaker designer' crowd here, I think very few venture this far into the art/science of room correction.

Most products have barely more than bass/treble EQ controls (e.g. Sonos Amp), and many AV Receivers use Audessy which makes everything sound worse by default because it tries to equalize above the Schroeder frequency, so everyone I know (including myself) who tries it often ends up turning it off until you extensively research online to understand what you can do via an extra app you have to go find and pay an extra $10 for, and learn how to operate and use to avoid aggresive EQ of the mids and treble.

 

[Wes]

we need a revscend speaker...

 

[GelbeMusik]

I don't think it's fair to assume the average listener will spend the time and effort to use miniDSP and REW.

I understand Your argument. My caveat addresses the more demanding enthusiast. If there is some desire to have it all right, why not ask the consumer, or his service man, for some money of course, to do the job of EQ? Or teach people to EQ, for some reasonable money. But then they are free to adjust every record to their home curve ...

If spectral balance is so important as Toole says, why not cross check its realization in the actual setup. Don't we talk about investments exceeding the thousends easily?

 

[echopraxia]

I understand Your argument. My caveat addresses the more demanding enthusiast. If there is some desire to have it all right, why not ask the consumer, or his service man, for some money of course, to do the job of EQ or teach people to EQ?

If spectral balance is so important as Toole says, why not cross check its realization in the actual setup. Don't we talk about investments exceeding the thousends easily?

Ok, so aside from the calibration effort itself, there is still a major catch: How do we achieve the DSP capabilities needEd without adding more really expensive equipment to the stack, or without compromising a lot of convenience

For example, in a few rooms I use the Sonos Amp for the convenience it offers. Not only does it make streaming music to speakers extremely easy, but allows multiple rooms to play in sync when you want, AND supports HDMI ARC + CNC so you can use it connected to a TV as well.

But there is no way (as far as I know) you can perform room correction through a Sonos Amp while retaining all these convenient features.

So maybe I should phrase it as a question: How else can I achieve all these features with a equal or greater level of convenience, with 200wpc, in a small form factor, while also enabling room correction?

I do expect some compromise of convenience in optimizing/calibrating my primary system, but it sure would be nice if I didn’t have to compromise. And beyond that, my impression is that the vast majority of people will favor convenience and simplicity of setup most of the time.

 

[Senior NEET Engineer]

I'm curious what the double blind forum reactions would be if the Revel scored significantly better.

 

[echopraxia]

I'm curious what the double blind forum reactions would be if the Revel scored significantly better.

Probably something like "Wow, well would you look at that -- this is science in action, folks! Thank you Harman International for showing us how it's done -- yet another confirmation the true infallibility of this wholly owned subsidiary of Samsung Electronics! Go out and buy Revel speakers ASAP, everyone!"

Just kidding (mostly) But yes, the Revel/Harman fan-ism here is strong. I can't blame anyone though, they make fantastic speakers based on solid science, and many of the people behind the brand seem worthy of our trust. But speaking from experience, Ascend makes some pretty fantastic stuff too (of a different flavor).

The Revel (of Harman, of Samsung) brand makes great speakers. I am undoubtedly a fan. So much so that I ended up placing an order for Revel Salon2's today instead of my originally planned F228Be's!

 

[Senior NEET Engineer]

Probably something like "Wow, well would you look at that -- this is science in action, folks! Thank you Harman International for showing us how it's done -- yet another confirmation the true infallibility of this wholly owned subsidiary of Samsung Electronics! Go out and buy Revel speakers ASAP, everyone!"

Just kidding (mostly) But yes, the Revel/Harman fan-ism here is strong. I can't blame anyone though, they make fantastic speakers based on solid science, and the brand seems worth trusting. But speaking from experience, Ascend makes some pretty fantastic stuff too (of a different flavor).

The Revel (of Harman, of Samsung) brand makes great speakers. I am undoubtedly a fan. So much so that I ended up placing an order for Revel Salon2's today instead of my originally planned F228Be's!

You are truly a madman just like Amir. I look forward to the impressions.