Revel Salon2 vs Genelec 8351B - Blind Test Preparations

[echopraxia]

I would test in stereo, otherwise the Salon's wide dispersion will be dramatically advantageous.

Hmm. Maybe I should then. The problem is the subwoofer issue mentioned above via my 4-channel-out miniDSP SHD.

Maybe I can test without subs, but with DSP carefully crafted to make the bass response as exactly similar as possible. I don’t see why this can’t work in theory, and it might even be mire accurate than trying to integrate a sub equally well with either.

 

[echopraxia]

You'd have to switch manually, but I'm afraid that both stereo and a sub are needed to make this useful: there's no "halfway useful" here.

What about EQ’ing the bass response of each stereo pair to match almost exactly? What would a sub be adding beyond that?

 

[q3cpma]

What about EQ’ing the bass response of each stereo pair to match almost exactly? What would a sub be adding beyond that?

Lower bass distorsion. The Genelec is simply not made to be used without a subwoofer in another situation than nearfield.

 

[echopraxia]

Lower bass distorsion.

I’ll do some no-sub bass distortion tests with the Genelec and Revel at 90-100db tomorrow and see what I find.

Worst case, I certainly could do a virtual crossover with just the 80hz high pass (with just no sub to fill in the rest). Of course it wouldn’t sound very pleasant to listen to, but I would think it would be fair. But is there anything else this would be missing, technically?

 

[q3cpma]

Don't know, honestly.

 

[Beave]

Even if they are clearly distinguishable, I still think it’s best keeping it as blind as possible. Because at least we won’t know which one is playing, and this will prevent brand/appearance/size/price or other biases from entering the picture and distorting impressions.

Also, if what you’re saying is true (+-1db making blind tests pointless), then all blind tests are useless because no two speakers are identical within 1db across all frequencies.

No, not at all. Remember, in this situation, you're comparing two speakers that you own, that you have heard in advance, that you are familiar with. And you know the brand, model, measurements, etc, for each.

Now picture walking into a facility, say Harman's for example, and they sit you down and have you listen blinded to four different speakers. You don't know what brand or model ANY of them are. You have four complete unknowns.

Those are two totally different situations - one in which blind comparisons are appropriate and very useful, one in which it's kind of a stretch to call it a blind comparison.

 

[echopraxia]

No, not at all. Remember, in this situation, you're comparing two speakers that you own, that you have heard in advance, that you are familiar with. And you know the brand, model, measurements, etc, for each.

Now picture walking into a facility, say Harman's for example, and they sit you down and have you listen blinded to four different speakers. You don't know what brand or model ANY of them are. You have four complete unknowns.

Those are two totally different situations - one in which blind comparisons are appropriate and very useful, one in which it's kind of a stretch to call it a blind comparison.

I am not going to be a primary listener here. I expect at least 3 untrained listeners, and hopefully more, none of which include myself. I will participate but only out of curiosity, and will identify the logged results that I produce as belonging to my test session so anyone reading the results can filter the data however you like.

 

[Beave]

How will you hide the identity of the speakers? Acoustically transparent screen? Actual blindfold on the listeners? Just curious.

 

[echopraxia]

How will you hide the identity of the speakers? Acoustically transparent screen? Actual blindfold on the listeners? Just curious.

Current plan is for no blindfold or curtain, since the listeners already know what speakers are being tested (though with little familiarity of them). Listeners obviously will not know which speaker they are listening to during each test. I will be the only listener with significant time already spent prior listening to both speakers, so other listeners will not be able to reverse correlate sonic signature.

I already cannot change the fact that the participants will know what speakers these are, so I have to work with that as best I can. I know it’s not ideal, but I think it shouldn’t affect which (if any) speaker wins, only the perceived magnitude of difference.

 

[echopraxia]

I plan to use a random number generator for ABX testing.

I will probably be writing a simple Python script or maybe a very basic UI app to manage mapping the (per track shuffled) A B X identifies to miniDSP profile buttons, and help record participant ID, song ID, conclusions per track and misc. notes.

 

[q3cpma]

I plan to use a random number generator for ABX testing.

I will probably be writing a simple Python script or maybe a very basic UI app to manage mapping the (per track shuffled) A B X identifies to miniDSP profile buttons, and help record participant ID, song ID, conclusions per track and misc. notes.

Could just use a dice, though.

 

[Juhazi]

Many good points covered in replys.
I question too long listening distance and eq of DUTs.

 

[echopraxia]

Many good points covered in replys.
I question too long listening distance and eq of DUTs.

Can you clarify your acronym DUT, and why and what specifically you question about long listening distance in a large room? I welcome questioning here, but am hoping for a bit more insight than just the statement that you question something

 

[Chrise36]

I tried, but I cannot tell the difference in height. This is at a 20 foot listening distance, and the tweeter height difference here is maybe a foot or two when stacking.

Have you tried distinguishing height differences like this? I find it vastly more difficult than horizontal differences, which are quite obvious.

Building a robot capable of shuffling a speaker as heavy as the Salon2’s is just not practical for me, so an alternate needs to be chosen. This vertical stack seems the best candidate so far.

Anyway, I will never be able to match the positions perfectly even if I could use a robot, because this is comparing a 3-way coaxial standmount with a 4-way non-coaxial floorstander. I suspect Revel’s robot does not normalize height differences between towers (though I don’t know this), because if you did you would have to choose a reference axis, and this wouldn’t really even make sense for towers.

For example, how would Revel make the Salon2’s tweeter as low as the F208? You just can’t do this unless you either raise the F208 way up (not how it was designed to be used), or dig a hole and sink the Salon2 into it (not how it was designed to be used). They were designed to be floor standing. So I would be surprised and skeptical of any test procedure that tries to mess with that.

I tried, but I cannot tell the difference in height. This is at a 20 foot listening distance, and the tweeter height difference here is maybe a foot or two when stacking.

Have you tried distinguishing height differences like this? I find it vastly more difficult than horizontal differences, which are quite obvious.

Building a robot capable of shuffling a speaker as heavy as the Salon2’s is just not practical for me, so an alternate needs to be chosen. This vertical stack seems the best candidate so far.

Anyway, I will never be able to match the positions perfectly even if I could use a robot, because this is comparing a 3-way coaxial standmount with a 4-way non-coaxial floorstander. I suspect Revel’s robot does not normalize height differences between towers (though I don’t know this), because if you did you would have to choose a reference axis, and this wouldn’t really even make sense for towers.

For example, how would Revel make the Salon2’s tweeter as low as the F208? You just can’t do this unless you either raise the F208 way up (not how it was designed to be used), or dig a hole and sink the Salon2 into it (not how it was designed to be used). They were designed to be floor standing. So I would be surprised and skeptical of any test procedure that tries to mess with that.

You need a large and stable turntable for these tests or some kind of rolling platforms they could be mounted.

 

[echopraxia]

You need a large and stable turntable for these tests or some kind of rolling platforms they could be mounted.

I’m open for donations of such apparatus (Complete with the necessary acoustically transparent but visually opaque curtains, electromechanical automation, controls, and end-to-end installation work, etc. necessary for it to be practical.)

Otherwise I will have to find the least bad compromise.

 

[echopraxia]

Could just use a dice, though.

Yeah but that requires too much work. Writing a script is so much easier

 

[echopraxia]

Hey echo,

Thanks for doing this shootout!

One thing to consider with the speaker stacked so closely above each other is the baffle step. The frequencies that would normally radiate up and over the top are now being reflected forward. It's essentially going to change the vertical response of both speakers. Side by side with some space between them may give you a better result.

I would guess the genelec would fare better in this regard as its designed to have stands or something below it. The revel definitely wasn't designed to have a flat surface above it.

I think this point is the main remaining unresolved question.

Assuming that nobody here is going to donate and install for me a fully automated powered belt or dolly system to swap out speakers in the same position in the horizontal plane, the next best way to keep the position identical in the horizontal plane is to stack them.

But as you point out, that will affect the vertical response.

My question then is, would it be sufficient to compare the Salon2’s frequency response with vs without the Genelec 8351B stacked on top, in order to determine whether this is a problem or not? Or, is this not something that can realistically be confirmed to be a problem or not with the basic measurement equipment I have (UMIK1 and REW)?

Do we expect the influence of the vertical stacking to be worse than the downside of placing the Genelecs on their own stands beside the Salon2’s (the downside there being the very audible horizontal placement difference, especially in stereo one pair of speakers is wider apart than the other)?

 

[Doodski]

especially in stereo one pair of speakers is wider apart than the other)?

Stagger them so the left Genelec speaker is to the right of the Revel and then at the right speaker the Genelec is to the right of the Revel. Each is the same distance apart and hopefully near the same height for the tweets. That's how it's done in store sound rooms.

 

[echopraxia]

Stagger them so the left Genelec speaker is to the right of the Revel and then at the right speaker the Genelec is to the right of the Revel. Each is the same distance apart and hopefully near the same height for the tweets. That's how it's done in store sound rooms.

Yeah and that’s how I’ve done it in the last two blind tests I’ve done as well. It works well I think, but the only problem is that it creates an audible rotation/shift of the soundstage to the right or left when switching speakers.

I’ve been able to mitigate that before in a few ways but nothing is quite perfect. A pause when switching helps make the shift less obvious, for example. But if you keep your head completely still while the shift occurs, you can still hear which direction it has shifted. You kind of have to trust the participants won’t try to listen for this, but needless to say this is far from ideal vs a setup that is truly blind.

Brainstorming, I suppose I could configure the miniDSP to send a mono signal to each speaker, but also apply a slight “pan” effect to each configuration, to compensate for the left right shift so the center image remains perceived in the same spot. But, I don’t know if the illusion will really be good enough, and I would worry this might risk other side effects. As long as it is symmetric maybe it won’t hurt? I don’t know.

 

[Doodski]

Yeah and that’s how I’ve done it in the last two blind tests I’ve done as well. It works well I think, but the only problem is that it creates an audible rotation/shift of the soundstage to the right or left when switching speakers.

I’ve been able to mitigate that before in a few ways but nothing is quite perfect. A pause when switching helps make the shift less obvious, for example. But if you keep your head completely still while the shift occurs, you can still hear which direction it has shifted. You kind of have to trust the participants won’t try to listen for this, but needless to say this is far from ideal vs a setup that is truly blind.

Brainstorming, I suppose I could configure the miniDSP to send a mono signal to each speaker, but also apply a slight “pan” effect to each configuration, to compensate for the left right shift so the center image remains perceived in the same spot. But, I don’t know if the illusion will really be good enough, and I would worry this might risk other side effects. As long as it is symmetric maybe it won’t hurt? I don’t know.

I would run with a simple staggered left to right approach.