KEF Blade 2 vs Ascend ELX Towers as End Game

[witwald]

Larger sound stage, sharper imaging and better bass. . .much better. . .by miles.

Much of that perceived sound quality seems to potentially be largely due to the differences in the on-axis frequency response between the two loudspeakers. Also, it's presently somewhat unclear as to how much larger the "larger sound stage" is. Can you offer some further insights as to what you meant by that observation? For example, 10% higher and 20% wider?

 

[kyuu]

That's not surprising, as the ELX Ribbon Tower speakers have a much lower F3 than do the F228Be speakers.

As shown below, the frequency content of that track has got lots of energy in the 40–48 Hz range. This is in a region where the F228Be speakers are rolling off, while the ELX Ribbon Towers are still relatively flat in their response.

F3 is highly misleading. The F228Be does not "roll off" until right around 30Hz, which is where the tuning frequency of the port plainly is:

From about 80Hz down to 30Hz there is a bass shelf. It's not rolling off. The bass response of the Ascend ELX is actually quite similar; they are both about 8dB down at 30Hz.

 

[PristineSound]

Much of that perceived sound quality seems to potentially be largely due to the differences in the on-axis frequency response between the two loudspeakers. Also, it's presently somewjhat unclear as to how much larger the "larger sound stage" is. Can you offer some further insights as to what you meant by that observation? For example, 10% higher and 20% wider?

I would be a fool if I try to quantify it, it's been many years. I just remembered the 3 distinct characteristics that separated the two speakers. But if I go by memory, the biggest difference is the bass.

 

[witwald]

F3 is highly misleading. The F228Be does not "roll off" until right around 30Hz, which is where the tuning frequency of the port plainly is:

As shown in the excerpt below, using the thick magenta line, the roll-off is quite linear between 25 Hz and 95 Hz, with a reduction in low-frequency output of around 10 dB over that frequency range. That's an initial roll-off rate of approximately 5.2 dB/octave, followed by the classical steeper roll-off of a vented enclosure below Fb.

 

[Mark609]

Unplugged, the ELX goes deep and is rich and is perfect for music. Dave worked with Seas and tweaked the LX OEM Seas woofer with a bigger motor and longer excursion. He told me the Fs is 32 Hz FWIW.

I have it crossed over at 80 Hz to SEALED subs, I stuffed the ports with closed cell foam. Here's a DATS sweep of that tweak. Impedance drops to about 3.2.

Stock, open port

 

[kyuu]

As shown in the excerpt below, using the thick magenta line, the roll-off is quite linear between 25 Hz and 95 Hz, with a reduction in low-frequency output of around 10 dB over that frequency range. That's an initial roll-off rate of approximately 5.2 dB/octave, followed by the classical steeper roll-off of a vented enclosure below Fb.
View attachment 550689

That's a bass shelf, not a "roll off". It's done intentionally to counteract the natural tendency of that frequency range to suffer from room gain and modes that result in boomy, muddy bass.

Anyone sensible should be using DSP/room correction in this frequency range so the anechoic response of the speaker is not super relevant.

 

[07avivo]

So there wasn't a big difference between your revels and the elx? That's what im hearing from most non-biased opinions.

I don't think I can judge. The Revels are gone since 6 weeks or so.
Listening through my HIFI-Test playlist, I had the impression that I heard some details for the first time with the ELX and sometimes I thought I am missing bass. But going for deep bass tracks (sub bass test track "Fading Sun") everything is there. So obiously ELX is able to reproduce everything on the low end.
But the more bass heavy tracks I play, the more I get the impression that ELX is doing something different in a positive way. E.g. doors of my cupboards are rattling, this was definitely not the case with F228Be. All that without being boomy at all. Interesting, as F228Be with 2x 8" bass drivers is not a typical competitor.
I am looking forward to whatever dirac will to on EQ side. A/B testing is almost immediate and it always added a lot detail and clearity.

 

[PristineSound]

the more I get the impression that ELX is doing something different in a positive way. E.g. doors of my cupboards are rattling, this was definitely not the case with F228Be. All that without being boomy at all.

My windows rattle depending on track and if I hit a room mode at a certain distance.

And that rattle is in a good high school way.

 

[witwald]

That's a bass shelf, not a "roll off". It's done intentionally to counteract the natural tendency of that frequency range to suffer from room gain and modes that result in boomy, muddy bass.

Below is a simulation of a room frequency response curve. Notice the room gain rising smoothly at 12 dB/octave from below the lowest room mode, which is at 38 Hz for this 4.5x3.5x2.8 meter room. The room gain does not appear to be a particularly step-like function, so it is a good design choice that the speaker in question does not have a step-like low-frequency response.

 

[mj30250]

Sorry. I find that hard to believe. Erinsaudiocorner has a review of the F226Be. He has stated that the soundstage width encompasses the side walls and he believes stage width doesn't go further than that. So best chance the soundstage is the same with a difference in imaging and bass. Revel f226be also has a waveguide, hard to imagine it imaging worse than the ELX.

I had both the F226Bes and ELXs for several months, and directly compared them (sighted) for many hours over the course of several days while they were positioned AB/AB. I occasionally swapped out each set and would use them as normal, just to see if my opinions of each developed differently over time. In the end, I didn't find there to be a notable difference in "soundstage width"; both filled the room and were capable of strong and pleasant spatial cues. I did find the ELXs to be more dynamic overall, especially in the bass and treble. The Revels were excellent, but I kept the ELXs because in my opinion they sounded less like speakers and more like the actual sound sources that they were reproducing.

 

[Narek]

I had both the F226Bes and ELXs for several months, and directly compared them (sited) for many hours over the course of several days while they were positioned AB/AB. I occasionally swapped out each set and would use them as normal, just to see if my opinions of each developed differently over time. In the end, I didn't find there to be a notable difference in "soundstage width"; both filled the room and were capable of strong and pleasant spatial cues. I did find the ELXs to be more dynamic overall, especially in the bass and treble. The Revels were excellent, but I kept the ELXs because in my opinion they sounded less like speakers and more like the actual sound sources that they were reproducing.

Were these the Ribbon ELX or Titans?

 

[mj30250]

Were these the Ribbon ELX or Titans?

Ribbons.

 

[Matias]

Below is a simulation of a room frequency response curve. Notice the room gain rising smoothly at 12 dB/octave from below the lowest room mode, which is at 38 Hz for this 4.5x3.5x2.8 meter room. The room gain does not appear to be a particularly step-like function, so it is a good design choice that the speaker in question does not have a step-like low-frequency response.
View attachment 550858

Yes but the room gain in your simulation is below 30 Hz, and not below 80 Hz like some manufacturers do. Huge difference.

 

[Narek]

Ribbons.

Yeah that matches up with what I've heard from others. F226Be already encompasses the room and the stage width touches the side walls. Saying the ELX is wider is nonsense. At that point you can't even objectively determine if its wider. Your ears don't hear a width further than your side walls.

What about the imaging? Would you say one was better than the other?

 

[witwald]

Yes but the room gain in your simulation is below 30 Hz, and not below 80 Hz like some manufacturers do. Huge difference.

Be that as it may, for the room gain to kick in below 80 Hz (the first room mode) requires a room whose largest dimension is of the order of 2.1 metres. That seems to make for quite an exciting time with a very-nearfield listening experience. I'd rather be sitting a bit further away from the speakers.

 

[Mark609]

Be that as it may, for the room gain to kick in below 80 Hz (the first room mode) requires a room whose largest dimension is of the order of 2.1 metres. That seems to make for quite an exciting time with a very-nearfield listening experience. I'd rather be sitting a bit further away from the speakers.

Being that I have a large HT (3,900 cu. ft.) that opens into a hallway with 2 closed off rooms and one open room at the end, I don't worry about room gain anymore. I do bump Room Gain after an Anthem ARC run about 3.16 dB. Am running three 15" sealed subs. Sound is excellent.

 

[mj30250]

Yeah that matches up with what I've heard from others. F226Be already encompasses the room and the stage width touches the side walls. Saying the ELX is wider is nonsense. At that point you can't even objectively determine if its wider. Your ears don't hear a width further than your side walls.

Yes, but all else being equal, a wider dispersion speaker will still throw more acoustic energy at your sidewalls which will be reflected back to the listener in some fashion (distance and absorption dependent), so there could still be audible differences which might effect your perception of both soundstage and imaging, for better or worse.

What about the imaging? Would you say one was better than the other?

Both imaged very well, but with certain material the ELXs sounded more defined and realistic/"in-the-room". Shortly before selling off the 226s, I set up both pairs of speakers and played "There Is Probably Fire" by Dominique Fils-Aimé very loudly, and with the subs on. This is a highly impressive recording that includes several spatial effects by way of voices and instruments placed in a variety of positions across the soundstage. On the Revels it sounded amazing, but played through the ELXs, the imaging and depth and dynamics were absolutely incredible. From a sound quality perspective, I don't think I've ever had a more impressive experience listening to recorded music.

 

[Mark609]

.....but played through the ELXs, the imaging and depth and dynamics were absolutely incredible. From a sound quality perspective, I don't think I've ever had a more impressive experience listening to recorded music.

You want dynamic, broad and deep soundstage, percussion attacks, etc.? Fire up Al Di Meola "Kiss My Axe" on ELX.

Mouth dropping, audio engineering at its finest via the CD and speakers.

 

[Dumdum]

That's a bass shelf, not a "roll off". It's done intentionally to counteract the natural tendency of that frequency range to suffer from room gain and modes that result in boomy, muddy bass.

Anyone sensible should be using DSP/room correction in this frequency range so the anechoic response of the speaker is not super relevant.

It’s still a roll off… regardless of if by design or not… and then it gets to port tuning and falls off like a rock as per ported designs… and rolls off even steeper

 

[07avivo]

According to port and woofer measurements of F228 Be it is clearly by design. HPF cut-off of woofer way steeper compared to woofer roll off. If it is called shelf or not, in the end less bass energy into the room, isn't it?

Edit: I missed my main question: I remeber some speakers, I think Blade 2 or F328Be, didn't fit vertically onto the NFS and measurements had to be carried out sideways, resulting in bass artefacs. I can't recall what the isue was unfortunately.