AsciLab F6Bs Speaker Review

[mj30250]

The DBR62 review is over 5 years old. Amir has reviewed and provided subjective impressions for, what, 100+ speakers since then? How could he possibly answer that question with a large degree of accuracy, especially given acoustic memory begins to degrade after only a few seconds?

You can find multiple differences between the speakers in the measurements and extrapolate from there, no memory required.

 

[somebodyelse]

Neither CTA2034 nor Olive/Toole's Preference model can predict listener preference with 99% confidence. That's nonsense.

Anyone who puts that much trust into preference scores is misguided.

I wish we had a list of Toole's posts covering why the score isn't the be-all and end-all, and why you need to understand the various spinorama plots and make a judgement. Here's one among the many that I can't find immediately. From memory you need a score difference of >1 to be somewhat reliably predictive.

 

[a4eaudio]

Why don't we go to What Hi-Fi and ask there too if we just want subjective opinions? Or ask AI to compare Amazon reviews?

The Spinorama measurements are valuable, just not very "predictive"? Instead of pretending to predict something with 99% accuracy, why not point out the objective attributes of the speakers and comment on the implications.

For example: The Ascilab is more linear than the Elac and in some research greater linearity has been shown to be more preferred. Additionally, the Ascilab has more bass extension which has been found to be one of the most important indicators of speaker preference.

More facts, less hyperbole.

 

[nathan]

I wish we had a list of Toole's posts covering why the score isn't the be-all and end-all, and why you need to understand the various spinorama plots and make a judgement. Here's one among the many that I can't find immediately. From memory you need a score difference of >1 to be somewhat reliably predictive.

Good point.

In the post you link to, Toole says "The 13 bookshelf loudspeakers in the well-controlled test were those evaluated by CR in their review. Sean’s calculated sound-quality ratings correlated with double-blind subjective ratings with a correlation coefficient of 0.995 (perfection!) with a high significance p < 0.0001." IOW, Sean's scoring predicted the way the speakers in question were ranked by blind listeners perfectly. This is also the discussion I link to in my post about "99%" where I guess I should have said in this pool of speakers it was 100% predictive

But yes, this is not going to always be the case.

• For one, bass extension will trump many of the smaller nuances (though in this particular case that doesn't seem to be the case?)
• Listening in different rooms, with different modal characteristics due to different placement of speakers and listener can trump smaller nuances.
• And other rooms factors and use cases will matter, as well.

And yes, at some point the preference scores are close enough that the process is more complicated. 7.8 versus 8.5 for the two speakers in question (bass managed) may be within the margin of error. I thought the margin of error for prediction falling apart was smaller, but I can't find that number, either, at present.

 

[a4eaudio]

...with a correlation coefficient of 0.995 (perfection!) with a high significance p < 0.0001." IOW, Sean's scoring predicted the way the speakers in question were ranked by blind listeners perfectly. This is also the discussion I link to in my post about "99%" where I guess I should have said in this pool of speakers it was 100% predictive

There are many posts about the correlation coefficient not being predictive and some other potential statistical errors. I think the fact that Toole and Olive are not statisticians is fine and we should appreciate the research they have done without overstating the implications.

 

[nathan]

There are many posts about the correlation coefficient not being predictive and some other potential statistical errors. I think the fact that Toole and Olive are not statisticians is fine and we should appreciate the research they have done without overstating the implications.

Fair enough. I don't disagree per se.

I'd still bet money on a person with good hearing being more likely to prefer the AsciLab in this matchup

 

[nathan]

Fair enough. I don't disagree per se.

I'd still bet money on a person with good hearing being more likely to prefer the AsciLab in this matchup (yes, that's not even close to 100%).

 

[Karmacoma]

From my little and very subjective experience with different speakers that have been measured, including the ruler-flat on-axis Neumann 150, the differences between these speakers would shine in the mediums, listenning to choirs and philarmonic orchestras with a flatter on axis response, the same notes played with the slighest differences of tones would turn the garden into a forrest, and based on measurements only I'd pick the Ascilab's hands down. My preference would certainly be less obvious with jazz and more simple compositions where I could fall for a voicing and extra bass more easily.

 

[Purité Audio]

As always the only valid method is to hear both in your own space at the same time, and even then problematic.
Keith

 

[MarkS]

As always the only valid method is to hear both in your own space at the same time

Yes

and even then problematic.

Can you clarify what you mean by this?

 

[Purité Audio]

Only having to level match, ( generator and an spl meter) and then the positioning, side by side, on top of each other, interleaved unless you have a Hartman ‘shuffler’ in your front room.
Keith

 

[MarkS]

Only having to level match, ( generator and an spl meter)

There is an alternative procedure that makes this unnecessary: do not switch instantly. Instead, start at zero volume after each switch, and adjust the sound level by ear only.

For anyone who says this isn't fast enough, please note that it's faster than the Harman speaker shuffler.

 

[amirm]

@amirm what do you think?
Is DBR62 enjoyable over F6Bs during music listening?

I have no memory of DBR62. Heck, I don't have a specific memory of F6Bs either! And have a conflict of interest to boot.

I will say however to not read too much into my brief subjective words in my reviews.

 

[nathan]

Yes

Can you clarify what you mean by this?

Sighted listening tests are notoriously unreliable. If I know what I am listening to, that will skew my impression of the gear involved.

 

[The French]

Ranking speakers is complicated.
Ranking speakers and adding their prices in the equation, is very complicated.

As a materialist, I would turn the problematic differently (F6Bs Vs Elac DBR62)

Can I afford any of those two ? Yes.

Can I live knowing that I've chosen the cheapest pair, which objectively do not measures as well as the Ascilab : no I can't.

=> I know I would ended up buying the Ascilab.

Even if I'm not a fan of the plasticy / modern look of the Ascilab (I don't really like the Elac either), and I think they can look really nice with the right configuration
ie :

(the stands on that video are Solidsteel stands, around 500€...)

 

[DearSX]

Ranking speakers is complicated.
Ranking speakers and adding their prices in the equation, is very complicated.

As a materialist, I would turn the problematic differently (F6Bs Vs Elac DBR62)

Can I afford any of those two ? Yes.

Can I live knowing that I've chosen the cheapest pair, which objectively do not measures as well as the Ascilab : no I can't.

=> I know I would ended up buying the Ascilab.

Even if I'm not a fan of the plasticy / modern look of the Ascilab (I don't really like the Elac either), and I think they can look really nice with the right configuration
ie :

(the stands on that video are Solidsteel stands, around 500€...)

I may end up with both a AudioFirst and Ascilab speakers and XMechano DIY, just cause its so great to have and appreciate these small designers producing top notch products.

 

[DeadAndCompany]

The preference score for the ELAC DBR6.2 is 4.97 (7.22 with sub) while the score for the Ascilab F6B is 6.52 (8.47 with sub).

The preference score of the DBR62 is 5.7, not 4.97.

 

[Curvature]

The preference score of the DBR62 is 5.7, not 4.97.

Accidentally used: https://www.spinorama.org/speakers/Elac Debut 2.0 B6.2/ASR/index_asr.html

Should have used: https://www.spinorama.org/speakers/Elac Debut Reference DBR-62/ASR/index_asr.html Which has the scores below. I'm not sure where you are getting 5.7.

 

[eyke]

@AsciLab is there any significant benefit to high-passing the signal to the F6Bs, ie. do they perform "better" given a high-passed signal eg. >60Hz only vs full range signal, since i have a sub that takes care of the low end?

 

[Curvature]

@AsciLab is there any significant benefit to high-passing the signal to the F6Bs, ie. do they perform "better" given a high-passed signal eg. >60Hz only vs full range signal, since i have a sub that takes care of the low end?

https://audiore.kr/ascilab-f6b-sealed-%eb%a6%ac%eb%b7%b0/