Role Audio Skiff Speaker Review

[JSmith]

Besides that, what's the point of having a box where the Serial Number should be written if nothing is written in that box?

I'm fairly sure @amirm has removed the serial number, as other pics show a serial number;

... hey at least the stickers are straighter than the tested model. Thanks Amir for posting these results.


JSmith

 

[respice finem]

A round sticker would mitigate the problem

 

[DrCWO]

A good loudspeaker and crossover design ensures that this happens also that crossover region of a multi-way speaker.

This is not true!
Passive LR or other passive crossovers always cause phase errors. Even if you use a DSP that has the option of time delay it is impossible. There are only a few ways to achieve linear phase crossovers:

• Use first order crossovers (passive or DSP) but with this you have weak suppression of unwanted driver resonances.
• Use subtractive delayed crossovers.
• Use FIR crossovers with linear phase.

For educational purposes I already submitted something here:

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

Thank you for the info on Fq dependent smoothing in Acourate; it is interesting for me knowing the difference from psychoacoustic smoothing* in REW which I cited in my post #708; I wrote there, "nowadays, I always do not like psychoacoustic smoothing* ...." *In REW user manual: "Psychoacoustic...

audiosciencereview.com

I hope this puts an end to the discussion that phase errors can be avoided with passive crossovers

 

[DanielT]

This is not true!
Passive LR or other passive crossovers always cause phase errors. Even if you use a DSP that has the option of time delay it is impossible. There are only a few ways to achieve linear phase crossovers:

• Use first order crossovers (passive or DSP) but with this you have weak suppression of unwanted driver resonances.
• Use subtractive delayed crossovers.
• Use FIR crossovers with linear phase.

For educational purposes I already submitted something here:

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

Thank you for the info on Fq dependent smoothing in Acourate; it is interesting for me knowing the difference from psychoacoustic smoothing* in REW which I cited in my post #708; I wrote there, "nowadays, I always do not like psychoacoustic smoothing* ...." *In REW user manual: "Psychoacoustic...

audiosciencereview.com

I hope this puts an end to the discussion that phase errors can be avoided with passive crossovers

Does it matter so much in a listening room where the sound bounces around? There is a lot of reflected sound that comes later than the direct sound, and no filters or time compensation can do anything about that. Other than if would listen in an Anechoic Chamber but who can and wants to do that?
Maybe if you listen outdoors in an open field it might also work. Remove reflexes that is.

 

[Rja4000]

Thanks for this review.

Not sure if those are available outside of US, though ?

By the way, the "Skiff" name reminds me speakers I used as my main speakers for years (and I still use and like them)

Not the same kind of beast, for sure

(If you happen to read french, you'll see they state that they tested it with 126dB SPL peaks)

 

[DrCWO]

Does it matter so much in a listening room where the sound bounces around?

If reverberation time is too high you are right. But go and listen to a linear phase system.
Especially with percussion I can clearly hear the difference in a "normal" listening room.

Edit:
I added a Master theses that also researched on this topic. In the conclusion you find this sentence:
"As a general conclusion, phase errors due to crossover filter simulations are audible."

Sure, this is all theory, of course. The crucial question is: Do you even care?

 

[thewas]

This is not true!
Passive LR or other passive crossovers always cause phase errors. Even if you use a DSP that has the option of time delay it is impossible. There are only a few ways to achieve linear phase crossovers:

• Use first order crossovers (passive or DSP) but with this you have weak suppression of unwanted driver resonances.
• Use subtractive delayed crossovers.
• Use FIR crossovers with linear phase.

For educational purposes I already submitted something here:

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

Thank you for the info on Fq dependent smoothing in Acourate; it is interesting for me knowing the difference from psychoacoustic smoothing* in REW which I cited in my post #708; I wrote there, "nowadays, I always do not like psychoacoustic smoothing* ...." *In REW user manual: "Psychoacoustic...

audiosciencereview.com

I hope this puts an end to the discussion that phase errors can be avoided with passive crossovers

We are talking about two different things, I am talking about significant phase differences between the two drivers at the crossover region and you are talking about total linear phase. Also I didn't mention in my above post anything that the crossover is or should be passive.

If reverberation time is too high you are right. But go and listen to a linear phase system. Especially with percussion I can clearly hear the difference in a "normal" listening room.

I had directly compared the K+H O500C with and without linear phase FIR filters and the difference was very small despite the acoustically optimised room at the Neumann monitors R&D headquarters.

 

[Sokel]

That response!
If I was ever interrogated there would be no need to push me hard to confess,a half an hour listening to this would do the job just fine

Thanks Amir!

 

[amirm]

Besides that, what's the point of having a box where the Serial Number should be written if nothing is written in that box?
View attachment 312808

I erased that.

 

[AudioSceptic]

500 dollars a pair? This is nota a typo?
I don't get it.

Indeed. £/$100 for a pair would be more like it for a toy speaker.

 

[AudioSceptic]

You wrote your post before mine. I missed it. We basically say the same thing.

If you don't want to build something, as I mentioned in my last post, and want some nice small satellite speakers at a good price, buy a couple of used Kef Eggs instead. By the way, it would be really fun if Amir tests a couple of such eggs.

Or how about a pair of Apple Homepod Minis for £/$200?

 

[AudioSceptic]

A round sticker would mitigate the problem

They would still get the text crooked.

 

[AndreaT]

Waste of good money, waste of space, waste of your time. The speaker will falsify your perception of recorded music, and your overall listening experience: lack of bass, high distortion, poor FR. Thank you Amir.

 

[wwenze]

This is the kind of world we lived in 15 to 20 years ago.

Back then if you wanted a small speaker, Role was one of the few brands that even existed. Audioengine was another, and their A2 objectively (and most likely subjectively) performs better than this by a lot. But Role was the "better" speaker because it was "more expensive".

 

[Geert]

I added a PhD theses that also researched on this topic. In the conclusion you find this sentence: "As a general conclusion, phase errors due to crossover filter simulations are audible."

Did you attach the right paper because it's not a thesis and I can't find that quote in the document?

 

[ocinn]

True point source
Magnetically shielded
Perfect time coherence
Superior ElectricTM five-way terminals
True minimum phase
Cast frame full-range transducer
Omnimount 10.0 Series bracket retainer

To be fair, none of these are necessarily inaccurate.

Shame it’s completely broken on every other front though.

 

[Tangband]

This is not true!
Passive LR or other passive crossovers always cause phase errors. Even if you use a DSP that has the option of time delay it is impossible. There are only a few ways to achieve linear phase crossovers:

• Use first order crossovers (passive or DSP) but with this you have weak suppression of unwanted driver resonances.
• Use subtractive delayed crossovers.
• Use FIR crossovers with linear phase.

For educational purposes I already submitted something here:

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

Thank you for the info on Fq dependent smoothing in Acourate; it is interesting for me knowing the difference from psychoacoustic smoothing* in REW which I cited in my post #708; I wrote there, "nowadays, I always do not like psychoacoustic smoothing* ...." *In REW user manual: "Psychoacoustic...

audiosciencereview.com

I hope this puts an end to the discussion that phase errors can be avoided with passive crossovers

Yes, thanks for pointing out these facts.

 

[Tangband]

Does it matter so much in a listening room where the sound bounces around? There is a lot of reflected sound that comes later than the direct sound, and no filters or time compensation can do anything about that. Other than if would listen in an Anechoic Chamber but who can and wants to do that?
Maybe if you listen outdoors in an open field it might also work. Remove reflexes that is.

Do yourself a favour and take a listen to a really good single driver , like Markaudio chn110. The first arriving sound will dominate the listening experience, because of the precedence effect.

Most loudspeakers with many drivers and crossovers have issues thats audible, just like many single driver constructions have issues with frequency response and high spl etc…

There are no perfect speakers , all are compromises.

 

[peniku8]

@amirm would you mind adding normalized waterfall plots to your reviews? Or replace the unnormalized ones, as we already have the magnitude response in the on-axis FR graph and want the waterfall graph just to look at temporal anomalies, which a normalized plot would better showcase/isolate.

 

[DrCWO]

Did you attach the right paper because it's not a thesis and I can't find that quote in the document?

OMG, really got the wrong one, thank you
I updated the post.