Revel Salon2 vs Genelec 8351B - Blind Test Preparations

[LTig]

Roger Waters 'Amused to death' the song Three wishes has a woman speaking from far left, almost behind the listening chair...well the salon 2 would place the phantom image near center.

Nice track, nice recording. The woman speaking however sounds "phasey" and a bit left of my left speakers (K&H O300D). The "good night" from Stings track When the Angels Fall from Soul Cages is more stable and almost 90 degree to the left.

 

[LTig]

Damn, i just listened to that track on my kef r3 and its so creepy. It felt like the woman was speaking into my ear!

and i thought the growling sound slightly later in the track was coming from the street downstairs. Some fantastic mixing there

Yep, very good, very dynamic. Here is the spectrogram:

 

[preload]

I now wonder if my kimber 4tc wasn't a heavy enough gauge wire for them as the bass or lack of bass in my case was some of the disconnect with them.f

4TC is 13GA equivalent. That should be plenty. Unless they're super long. And the kimber design is nice and low capacitance so they should be very transparent.

Have you taken FR measurements with REW? I bet they'll shed some light on your bass issue. How far from the back walls are you placing them?

 

[LTig]

I had them about 9 months and tried them along the long wall in my 17x27x7' room and the short wall with varying amounts of toe in. I now wonder if my kimber 4tc wasn't a heavy enough gauge wire for them as the bass or lack of bass in my case was some of the disconnect with them. I know I bought the exact pair I listened to and they sounded great in the demo.

Bass performance varies a lot between rooms, and also between different positions of speaker and the person listening in the same room. Both geometry and wall substance play a role.

 

[ROOSKIE]

I am going to be doing a controlled blind test of these speakers soon. Opinions on best test procedure are welcome. Edit: It seems there is nothing remotely approaching consensus here on what is even an “acceptable” test procedure in this thread so far. I’ve canceled any plans of doing a test in the near term as a result, but I definitely still plan to test these — once we better understand consensus for what methodology would be required in order for a victory of one speaker over the other to be accepted as such.

Need to determine: best test track list, mono vs stereo test, with vs without subwoofer, best test procedure for placement, and room correction procedure (if any) for bass, etc.

Current in-room measurements (prior to any EQ) with Rythmik F18 integrated with 80hz crossover shown below, from a single identical measurement position. The Genelec seems to have a flatter in-room response, but as a coaxial this makes sense as I am not averaging multiple measurements for the Salon2’s. Still, the Genelecs in room performance is amazingly good.

Question is, other than bass EQ (which I will be doing to flatten that out), should I attempt to EQ the Salon2's in-room response closer to that of the Genelec, or is it more fair to just leave it alone (other than bass EQ)? Edit: Current plan is not to EQ anything above 200hz and realistically probably not above 100hz.

Genelec 8351B + Rythmik F18 (10hz - 22khz) — no room correction yet:

View attachment 79237

Revel Salon2 + Rythmik F18 (10hz - 22khz) — no room correction yet:

View attachment 79234

(Boomy subwoofer response from room modes as you can see, but this is good since it will provide headroom for bass room correction.)

Vertically stacked orientation shown here. Edit: Note: This is probably not good enough, so will likely have to end up using a turn table behind a acoustically transparent curtain.

View attachment 79236

Hi, great thread, just found it - read about 7 or 8 pages so sorry if something I say is redundant and not additive - I had to skip some of this thread due to time.

IMHO

• don't try to get it perfect, this is your first time. Be reasonable with your expectations.
• This sounds fun and totally worth it, vastly superior to not doing it from a hobbyist viewpoint.
• do not shut the whole thing down because there is no unanimous agreement or even a general consensus. Most of the folks here have never done this type of thing either. Just take the best path you can that still allows something to happen. If it presents as a bust then redo it. But get it out of heads and into the room.
• learn
• have fun, this is only a hobby for 99.9% of us and for the .1% who's job is impacted by this they really ought to stay silent for now.
• publish what you find & be open to constructive criticism and new ideas for next time.
• Don't overestimate the value of your results by declaring a "winner". Focus more on the insight you have gained by the process of experimenting in real time/life.
• You never going to be able to tell me which speaker I will prefer while testing 2 speakers with 4 or 5 folks. So don't try. You might find your favorite pair though!
• Know why you are doing this test, and mean really get to know why. It most certainly is not to simply find the best speaker between the two- I guarantee you can not do that this way.

1. If it was me I would definitely want to use the speakers as I plan to in real life. Nobody on this site can tell they get better sound by NOT using room correction and PEQ and other tools. Personally I literally could care less how a speaker sounds in anechoic chamber. That chamber serves a design purpose for measurements but not for the consumption of the music played back.
2. Your in room responses ought to be corrected exactly as you would for personal use. This will be argued. So is the value of $500 a foot speaker wire - everything can be argued. This is my view though. I will never again not be willing and able to correct many room influences, I will also never again NOT be able to shape the response to a reasonable degree toward my preferred balance/curves.
3. Use a sub if you will be using one. It is a speaker system after all. If subs will be in the system then they better be in the test.
4. Listen at different volumes. In real life I listen at a variety of volumes. Listen only at 80db or whatever doesn't tell me much about how they sound at 65db nor 100db. I'd run the testing 3 times at 3 volumes to account for this in a meaningful way.
5. Set the levels by using pink noise playing 500-2000hrz
6. After the blind testing run a series of sighted testing's. This actually gives you a reference and will be very interesting to see how/if the results change.
7. make sure speaker A is not always the same speaker, make sure the order is very randomized. The second playback can have a slight advantage.
8. It is possible to build a turntable that spins the speakers so they play on the exact same spot. Realizing that they are heavy/huge in the case of the Salon's, I still point this out for future match-ups.

Regarding the Salon's specs. I think revel has published fantastic material. Power handling is a "fake" spec. Really stupid. The speaker deals with so many variables with musical content and other variables. No high-end speaker benefits from such a spec. Thermally the drivers are likely to handle about as much power as any quality driver on the market and the sensitivity is 86ish so the math is there if need be.
They are know for handing lots of power and playing very loudly, like any speaker they have limits. Best to use a very powerful amp and thus assure little chance of clipping distortion. That is the tweeter killer.
I have never blown a tweeter in my life and have played music very loudly. It does sound like your room is huge if you are 20 feet away. Luckily that isn't that different from my 10/11ft distances. I bet you can rock those Revel's hard. Do it.

The Genelec listed maximum SPL, might be theoretical. I don't know how they ascertain it, it could be in a variety of ways & therefore means very little beyond Genelec being confident it gets pretty loud before it craps out. I am sure it is only in ideal situations. 113db is quite impressive for such a speaker and therefore a bit dubious - even given Genelec's usual honesty. Again I don't actually know this.
At 20 feet it is about 97db per single speaker maximum(there may be room gain as well, but we don't know exactly how to value the maximum spl spec). I'd be surprised if the big Revel's can't match/exceed this. (obviously with a full tank of big amps)

Anyway have a blast!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Just do it right?

 

[ROOSKIE]

4TC is 13GA equivalent. That should be plenty. Unless they're super long. And the kimber design is nice and low capacitance so they should be very transparent.

Have you taken FR measurements with REW? I bet they'll shed some light on your bass issue. How far from the back walls are you placing them?

One test that could be run here very easily is a blind wire test!!! & these big Revel's are speakers to test such wares with.
Expensive wire vs "normal" vs inexpensive/Monoprice. Plus maybe even different gauges.
Really want to see those results - even more than the speaker shoot-out.

 

[daftcombo]

I wanted to add my experience comparing the old vs new style waveguide and how a speakers soundstaging capabilities can't be compared the harman way.

I have also experienced that speakers with no waveguide can have a great stereo image. I don't know why. A waveguide is usually better for tonality though.

 

[Chromatischism]

I've found a waveguide helps with both. Consistent dispersion ensures reflections are of the same tonal makeup as the direct sound, which prevents reflections from coloring the sound tonally and skewing/softening imaging.

 

[thewas]

I've found a waveguide helps with both. Consistent dispersion ensures reflections are of the same tonal makeup as the direct sound, which prevents reflections from coloring the sound tonally and skewing/softening imaging.

Same in my current quite reflective music room, smoother and higher directivity loudspeakers image more precise.

 

[Frank Dernie]

Damn, i just listened to that track on my kef r3 and its so creepy. It felt like the woman was speaking into my ear!

and i thought the growling sound slightly later in the track was coming from the street downstairs. Some fantastic mixing there

It uses "Q"-sound which I think, whilst it gives remarkable stereo positioning effects, is super sensitive to the system.
I use "Amused to Death" as my main evaluator of stereo imaging and find it very revealing.

 

[Frank Dernie]

Same in my current quite reflective music room, smoother and higher directivity loudspeakers image more precise.

I have found the opposite.
Wide directivity gives a euphonic stereo and narrow directivity more precise imaging IME.
Mind you it seems a majority prefer euphony to precision, given the Harman test preference, and a wide directivity with poor off-axis FR is neither euphonic nor precise.

 

[thewas]

I have found the opposite.
Wide directivity gives a euphonic stereo and narrow directivity more precise imaging IME.

Opposite? You also write the same, namely that higher (=> narrower) directivity loudspeaker image more precise.

Mind you it seems a majority prefer euphony to precision, given the Harman test preference

But as discussed in several threads those tests were done in mono and in stereo or multichannel the preferences seem to change.

and a wide directivity with poor off-axis FR is neither euphonic nor precise.

Of course, I personally still prefer though rather more narrow smooth vs very wide smooth as I like "diving" into the recording (of course only if it has the according qualities and its not just panpot mixed).

 

[andreasmaaan]

I have found the opposite.
Wide directivity gives a euphonic stereo and narrow directivity more precise imaging IME.
Mind you it seems a majority prefer euphony to precision, given the Harman test preference, and a wide directivity with poor off-axis FR is neither euphonic nor precise.

By Harman test preference, do you mean the studies and model developed by Olive and Welti?

In fact, these studies did not find any specific preference with respect to directivity. Olive summarises this as follows (my bold):

The ideal target slope for the on-axis and listening window curves...is identical for both test samples, which indicates that the on-axis curve should be flat, while the off-axis curves should tilt gently downwards. The degree of tilt varies among curves for Test One and the larger sample. Test One includes mostly 2-way designs whereas the larger sample includes several 3-way and 4-way designs that tend to have wider dispersion (hence smaller negative target slopes) at mid and high frequencies. This suggests that the ideal target slope may depend on the loudspeaker’s directivity.

I'm not aware of any other Harman studies in which wide-directivity designs were found to be preferred, although ofc Toole's view (based on other studies) is that they are preferred.

 

[thewas]

I'm not aware of any other Harman studies in which wide-directivity designs were found to be preferred, although ofc Toole's view (based on other studies) is that they are preferred.

I think the 1985 Toole AES paper is meant which had compared

■ Loudspeaker AA was a two-way design, 8 in. (200 mm) woofer and 1 in. (25 mm) tweeter (Rega model 3).
■ Loudspeaker E was a three-way design: 12 in. (300 mm) woofer, 5 in. (110 mm) midrange, and a 2 in. (50 mm) tweeter (KEF 105.2).
■ Loudspeaker BB was a full-range electrostatic dipole, employing a diaphragm subdivided into areas driven in a manner to approximate a spherically expanding wavefront. The center circle, the “tweeter,” was about 3 in. (76 mm) diameter (Quad ESL-63).

where AA was prefered in the spatial rating when listening in mono, but BB which had a poor rating in mono got an almost as good rating as AA in stereo listening. The ratings plot can be found in his book and were also posted here some months ago but can't find the thread right now

 

[andreasmaaan]

I think the 1985 Toole AES paper is meant which had compared

■ Loudspeaker AA was a two-way design, 8 in. (200 mm) woofer and 1 in. (25 mm) tweeter (Rega model 3).
■ Loudspeaker E was a three-way design: 12 in. (300 mm) woofer, 5 in. (110 mm) midrange, and a 2 in. (50 mm) tweeter (KEF 105.2).
■ Loudspeaker BB was a full-range electrostatic dipole, employing a diaphragm subdivided into areas driven in a manner to approximate a spherically expanding wavefront. The center circle, the “tweeter,” was about 3 in. (76 mm) diameter (Quad ESL-63).

where AA was prefered in the spatial rating when listening in mono, but BB which had a poor rating in mono got an almost as good rating as AA in stereo listening. The ratings plot can be found in his book and were also posted here some months ago but can't find the thread right now

Yeh, I know the study you're talking about I just wanted to point out that Toole's view is not based on the Harman research findings, and that the Harman studies do not reach any conclusion regarding narrow vs wide directivity.

There seems to be a common misconception that Harman's research found that people prefer wider-directivity speakers. It did not actually make any such finding.

 

[Frank Dernie]

Yeh, I know the study you're talking about I just wanted to point out that Toole's view is not based on the Harman research findings, and that the Harman studies do not reach any conclusion regarding narrow vs wide directivity.

There seems to be a common misconception that Harman's research found that people prefer wider-directivity speakers. It did not actually make any such finding.

In my head it is hard for me not to see Toole research = Harman starting point.

 

[LTig]

It uses "Q"-sound which I think, whilst it gives remarkable stereo positioning effects, is super sensitive to the system.
I use "Amused to Death" as my main evaluator of stereo imaging and find it very revealing.

Same as I do with Sting's Soul Cages - also using Q-Sound.

 

[Frank Dernie]

Same as I do with Sting's Soul Cages - also using Q-Sound.

Didn't realise that, I've got it so will have a listen when in the mood.

 

[steve59]

By Harman test preference, do you mean the studies and model developed by Olive and Welti?

In fact, these studies did not find any specific preference with respect to directivity. Olive summarises this as follows (my bold):


I'm not aware of any other Harman studies in which wide-directivity designs were found to be preferred, although ofc Toole's view (based on other studies) is that they are preferred.

That the revel waveguide designs started deep and narrow in the performa 2 and have evolved to wide shallow in the latest models leaves one to conclude their studies, which are performed in single speaker comparisons have taken them in the direction.

 

[andreasmaaan]

That the revel waveguide designs started deep and narrow in the performa 2 and have evolved to wide shallow in the latest models leaves one to conclude their studies, which are performed in single speaker comparisons have taken them in the direction.

That seems like a lot of speculation IMO. We don't know the results of the studies you're speculating about, let alone that they even exist in the first place.

Also, of all the Harman brands, Revel is their only one that produces speakers with wider than average directivity. That could very easily be down to the personal views of Kevin Voecks, or some other (unknown) factor.