KEF Blade 2 vs Ascend ELX Towers as End Game

[a4eaudio]

I have seen studies like that, but don't find them interesting. If you compare something like the SB26STAC and SB26ADC - same company, motor, etc. and listen to them they sound different. They have different on-axis, off-axis, phase, distortion characteristics. If you EQ them, phase match them, SPL match them, etc., etc., the sound same. Umm...okay.

I could probably prove that there is not difference between a Ferrari and a Chevy. As long as I put a turbo on the Chevy, a limiter on the Ferrari, change the tires, shocks and steering.

 

[PristineSound]

I have seen studies like that, but don't find them interesting. If you compare something like the SB26STAC and SB26ADC - same company, motor, etc. and listen to them they sound different. They have different on-axis, off-axis, phase, distortion characteristics. If you EQ them, phase match them, SPL match them, etc., etc., the sound same. Umm...okay.

I could probably prove that there is not difference between a Ferrari and a Chevy. As long as I put a turbo on the Chevy, a limiter on the Ferrari, change the tires, shocks and steering.

I think this test is simply testing for audible differences in the diaphragm material, which in itself has measurable physical properties in mass, transients, breakup, damping, etc.

Such eq and DSP is taking nonlinearities and timing distortions out along with taking room, baffle refractions, etc. out. I personally am ok with this approach.

But I am still hung up on the fact that the investigators recorded the tweeters playback and then played it back on an electrostatic STAX headphones to the test subjects for the experiment. Plus the demographics, hearing abilities and if trained or not was not disclosed in this paper.

I really want to say that this experiment sounds as stupid as Ted Denney selling cables, but just in case I'm the stupid one who is missing or ignorant about something, I will refrain my reactions.

Also let's not forget about Kunchur's AES paper on cables.

 

[jonfitch]

And here is a video demonstrating the physical properties and behaviors of different tweeter materials; again, will there be any audible difference? Hard to proof either way.

EDIT: Yes, I know this is a marketing video, but the demonstrations of the physical properties of various materials here are factual.

Even if they were audible differences the Personas are horribly designed with weird B&W-style V-shaped tuning and BBC dip due to a directivity mismatch between the tweeter and woofer. It's like they looked at the TAD Reference series and said "we can use beryllium for the tweeter and mids too!" and then designed a completely nonsensical product to the point that their cheaper Founder's series are their actual flagship now.

 

[Pearljam5000]

Does it make sense to compare $6K speakers to $28K speakers?

 

[mj30250]

Does it make sense to compare $6K speakers to $28K speakers?

I can think of at least 22,000 reasons why it might be helpful for some.

 

[traind]

Does it make sense to compare $6K speakers to $28K speakers?

Isn't it measurement driven to some degree? Published measurements look 'comparable' despite the vast difference in price. Then a Blade 2 Meta owner generously lent his pair to Ascend to compare at their showroom and people were invited to come and listen. Which is undeniably cool but also somewhat low risk for Ascend, being the far lower cost offering in the comparison.

 

[PristineSound]

Even if they were audible differences the Personas are horribly designed with weird B&W-style V-shaped tuning and BBC dip due to a directivity mismatch between the tweeter and woofer. It's like they looked at the TAD Reference series and said "we can use beryllium for the tweeter and mids too!" and then designed a completely nonsensical product to the point that their cheaper Founder's series are their actual flagship now.

I agree.

 

[Chanur]

Even if they were audible differences the Personas are horribly designed with weird B&W-style V-shaped tuning and BBC dip due to a directivity mismatch between the tweeter and woofer. It's like they looked at the TAD Reference series and said "we can use beryllium for the tweeter and mids too!" and then designed a completely nonsensical product to the point that their cheaper Founder's series are their actual flagship now.

The Founder 100F's are gorgeous.

 

[PristineSound]

Paradigm has it's start with Toole's NRCC influence. Now, it appears that they have deviated from science or just have bad engineers. What a shame and waste of Be for highs AND mids.

 

[MAB]

Am I reading this right? The various tweeters were recorded and then played back on a headphone to the test subjects?

I don't want to come across like an idiot, just in case there is something I am not aware and ignorant about, but at face value this experiment sounds as stupid as recording a system playback and then playing it back on YouTube.

Someone please tell me that I am missing something here.

Yes you are missing something here.

Videos and recordings of physical phenomena are not the problem. Most of the experiments I do professionally are recordings of phenomena.

At home, I use a microphone all the time to get valid results under controlled conditions. People on YouTube could too, but they typically don't, instead we get either amateur attempts at 'science', with mics and speakers changing position, the supposed investigator talking over the top of the sound, telling us what we are supposed to hear, etc. Or we get slickly curated marketing videos (one of which you offered as evidence above). Many of the videos are outright manipulated.

The problem is controls. As stated in the second paragraph of the paper, binaural recordings are often used to apply controls to an audio experiment. In this case, there is need to control for a variety of factors and was discussed from the outset of the paper. It seems you didn't read or understand the paper. Or maybe think you could get test subjects to all repeatedly orient their ears near tweeter, or maybe believe all of the research into head and ear shapes don't matter. Or maybe you are just obsessing on materials, which this industry fosters. Or possibly you reject controlled observations if they don't align with your sighted opinions.

Interestingly, this result is aligned with research by Toole and others, where frequency response is a prime factor in perceived sound.

I wonder what happened to that one guy who got it right but couldn't on the next day. Was it just pure luck?

This was explicitly discussed in the paper, is part of sample statistics on a study of this size, that 1 out of 20 trials will report a significant difference even while guessing. You don't understand sample statistics? Are you serious or are you sea-lioning?

People have failed to identify materials used in drivers for ages. Marketing drags us back into confusion. so they can sell exotic materials. Driver integration in the speaker system is way more important, which members on this forum are trying to tell you.

edit: fixed a sentence and typo

 

[PristineSound]

Yes you are missing something here.

Videos and recordings of physical phenomena are not the problem. Most of the experiments I do professionally are recordings of phenomena.

At home, I use a microphone all the time to get valid results under controlled conditions. People on YouTube could too, but they typically don't, instead we get either amateur attempts at 'science', with mics and speakers changing position, the supposed investigator talking over the top of the sound, telling us what we are supposed to hear, etc. Or we get slickly curated marketing videos (one of which you offered as evidence above). Many of the videos are outright manipulated.

With professionalism and utmost bluntness, you are merely saying: "Yes, this method of recording the actual playback and playing back the recording as a substitute of the real event is perfectly fine and legit for a scientific experiment, trust me, bro."

No, I don't trust you, nor should I or anyone else. If you are commanding trust in the method, provide the science and the data that using a recording to substitute the actual event is sufficiently sound for a scientific experiment that ought to carry weight.

I can think of multiple variables off the top of my head, the microphone used, the headphone used, the dispersion sound field.

A better method, that would not have been at all controversial would have been putting that test subjects inside the anechoic chamber, covering the speaker with an acoustically transparent fabric and listen to the speaker live instead of recording it.

You don't understand sample statistics? Are you serious or are you sea-lioning?

This sounds like you are getting snippy with me, I don't appreciate the tone.

This was me thinking out loud that I am curious what happened.

 

[MAB]

With professionalism and utmost bluntness, you are merely saying: "Yes, this method of recording the actual playback and playing back the recording as a substitute of the real event is perfectly fine and legit for a scientific experiment, trust me, bro."

No, I don't trust you, nor should I or anyone else. If you are commanding trust in the method, provide the science and the data that using a recording to substitute the actual event is sufficiently sound for a scientific experiment that ought to carry weight.

I can think of multiple variables off the top of my head, the microphone used, the headphone used, the dispersion sound field.

A better method, that would not have been at all controversial would have been putting that test subjects inside the anechoic chamber, covering the speaker with an acoustically transparent fabric and listen to the speaker live instead of recording it.


This sounds like you are getting snippy with me, I don't appreciate the tone.

This was me thinking out loud that I am curious what happened.

Let me be blunt. You keep sealioning here, even more so with your new handle.

I didn’t have patience back then when you used to badger for proof, and still don’t have.

 

[kyuu]

With professionalism and utmost bluntness,
A better method, that would not have been at all controversial would have been putting that test subjects inside the anechoic chamber, covering the speaker with an acoustically transparent fabric and listen to the speaker live instead of recording it.

Why would you do that? That eliminates the contribution of reflected sound, which is one of the primary things that might actually be different between the tweeters due to their different dispersion patterns.

 

[aarons915]

Floyd Toole chimed in to a similar discussion on AVS forum years ago and mentioned a blind study he conducted decades ago with a ribbon and dome with people making the same claims about ribbons, no differences were noted that couldn't be explained by the frequency response. Also that a heavier moving mass requires a heavier motor, it's just F=Ma where a higher mass requires a higher force to achieve the same acceleration. Here are a few of his quotes on the subject:

Right, but frequency response graphs don't usually tell the whole story. That's why Dave Fabrikant displays measurements such as Impulse Response, Energy Time Curve, and Cumulative Spectral Decay for his Sierra Towers: http://www.ascendacoustics.com/pages/products/speakers/SRT/Ascend Sierra Ribbon Tower.pdf

These can help reveal differences between, for example, dome and ribbon tweeters.

The time domain differences I see are correlated with frequency response irregularities, as is expected. Transducers are minimum-phase devices and the time domain performance can be calculated from the amplitude vs. frequency response. Waterfalls and the like add no new information, just different perspectives on the same information. Note that prolonged ringing in waterfalls always starts at a peak in the frequency response. Note also that the waterfall does not start declining at t=0. It should if properly windowed.

Further, there is persuasive evidence that humans do not hear the ringing. Our perceptions are dominated by spectral cues.
Toole, F. E. and Olive, S.E. (1988). “The modification of timbre by resonances: perception and measurement”, J. Audio Eng. Soc., 36, pp. 122-142.

This is not saying that the speaker at issue is inferior, just that the portrayal of performance is not balanced. There are much better dome tweeters.

and another later response:

The "quick decay" you speak of is totally predictable from amplitude vs. frequency responses because transducers are minimum-phase devices. Impulse response plots are impressive, but the reality is that humans do not respond to phase shift so these time-domain plots are misleading. We humans do not hear waveforms. I am an engineer and had to learn this myself, fortunately at an early stage.

It is interesting that one of the first loudspeakers I tested in the anechoic chamber at the NRCC in 1967 (52 years ago!!!!) was a ribbon tweeter, the Kelly Ribbon from the UK. I just looked at the measurements in my personal archive. At the time it was promoted for the same reasons that ribbons and electrostatic loudspeakers are today: "massless" diaphragms, "Instant" transient response, etc. etc. It was a learning experience for me, and only definitively provable in double-blind listening tests. Of course, nothing back then was as good as things are now, but the implied superiority simply was not heard. As time passed and measurements and knowledge improved, nothing has changed. There are good loudspeakers and not-so-good loudspeakers, and the method of moving the air has not revealed itself as being the dominant factor. If any one method was obviously superior, it would dominate the products we can buy.

As far as the "massless" consideration is concerned, I have a Tesla S P90D in my garage that accelerates from 0-60 mph in 3.2 s. It is a heavy four door, potentially 7-passenger, sedan that is much quicker than most two seat sports cars - and I have owned a few. The secret is in the motor, and that is also the relevant factor in loudspeaker transducers. Ribbons and electrostatic speakers typically have relatively weak motors compared to what can be put behind a cone or dome. In the end, frequency response is king.

 

[MAB]

Why would you do that? That eliminates the contribution of reflected sound, which is one of the primary things that might actually be different between the tweeters due to their different dispersion patterns.

He didn’t understand the paper, or how a binaural recording controls many of the difficulties in an experiment like this. Demands proof from members here while overlooking that the paper specifically cited the method in papers by Olive and Martens. So he didn't read or understand the citations either.

Instead he proposes experiments using anechoic chamber that introduce completely new and larger issues.

 

[Chromatischism]

Dave said that about 80% of people who audition the ELX in person in his factory, chose the Ribbon.

Yes and while I haven't heard them, the measurements posted on Ascend's website are better for the Titan Dome version of the speaker. If I were buying without hearing them, I would pocket the difference and get the one that measures better.

 

[PristineSound]

Let me be blunt. You keep sealioning here, even more so with your new handle.

I didn’t have patience back then when you used to badger for proof, and still don’t have.

You are hostile, confrontation, and eager to attack others. My sentiment towards you is mutual.

 

[Mort]

Yes and while I haven't heard them, the measurements posted on Ascend's website are better for the Titan Dome version of the speaker. If I were buying without hearing them, I would pocket the difference and get the one that measures better.

There are also differences in directivity if I remember. Harder to spot the 'better' one there. I expect that accounts for Dave's report that 80% prefer the wider ribbon.

 

[PristineSound]

Why would you do that? That eliminates the contribution of reflected sound, which is one of the primary things that might actually be different between the tweeters due to their different dispersion patterns.

As stated in the paper, the recording of the various tweeters were done in an anechoic chamber to eliminate reflection and directivity.

Some people not only didn't understand the paper, they didn't even read the entire paper and then like to claim others not understanding.

 

[PristineSound]

As stated in the paper, the recording of the various tweeters were done in an anechoic chamber to eliminate reflection and directivity.

Some people not only didn't understand the paper, they didn't even read the entire paper and then like to claim others not understanding.

And in case you are curious why the investigators of that experiment did this is because they are only testing for audible differences in tweeter material.