In-Room: KEF LS 50 meta vs. Palmer Orbit 11 vs. JBL CBT 70-J

[Paffi]

KEF LS 50 meta vs. Palmer Orbit 11 vs. JBL CBT 70-J

or you can say

Hifi vs. Studio. vs. PA

This keeps me thinking for years and finally led to the question: Why do Hifi-Speakers exist, when there are Studio and PA-Speakers which measure also well? I googled, I asked but never came to a final conclusion.

So, in the process of replacing my KEF LS50 meta I already tried many different speakers and also studio and now also PA. Let´s say, I combined the question of "how can I upgrade" with "hifi vs. studio vs. pa". After that and to keep it short: I don´t know, why Hifi-Speakers exist anymore, apart from looks which is the only valid argument for me. The more I´m testing at home, the more I´m wondering.

So I´m curious what do you guys think about that.

In the following some in-Room measurements, always with psychoacustic smoothing, because they all measure well (it´s a deciding buying factor for me) but I´m interested in how does it sound & measure at my listening spot (so this should not be seen as a speaker review!).

LS50 meta vs Orbit 11 (always with Bass PAD -6db):

SPL, left channel; Brown Orbits, yellow KEFs

right channel: blue Orbits, yellow KEFs

Summary: The Orbits extend much further down AND are brighter.

Let´s compare distortion:

Left channel, brown Orbits, yellow KEFs

Right channel, blue Orbits, yellow KEFs

Summary: Orbits distort a lot more between 40 and 120hz.

The Palmer Orbits 11 sound way different to the KEF LS50 meta and I was wondering, if they to as well, wenn I EQ them to match the KEFs:

left channel: brown Orbits, yellow KEFs without EQ, lila KEF with EQ

right channel: blue Orbits, yellow KEFs without EQ, orange KEF with EQ

Summary & Listening impression: Despite frequency response is nearly the same in the mids and highs, the Orbits sound TOTALLY different compared to the KEFs. And I like the Orbits 11 much much more. The KEFs are more distant, less precise, the Orbits sound way more forward, the acoustic stage is nearer and more defined at it´s left and right side. I´m wondering if the cardioid response is responsible for that and hence there is more direct spl in the frequency response than room related reflections, which can´t be seen in the SPL measurements.

For me: Studio 1 : Hifi 0, studio clear winner.

So I then was curious how constant beamwith transducers compared to that. So I ordered the JBL CBT-70J and compared it to the Orbit 11 from Palmer.

Left channel, green Orbit 11, Orange JBL CBT 70J

Left channel, green Orbit 11, Red JBL CBT 70J

Summary: Look how nice the in-room bass response of the JBL is - even without EQ it is simply flat, with no slope. With this room gain someone may even say, there is no sub for music necessary.

Now distortion:

Left channel, green Orbits, orange JBL

Right channel, green Orbits, red JBLs

Summary & Listening impressions: nearly no disortion of the JBL compared to the Orbits between 40 and 120hz. And now it get´s tricky: Where the Orbits (Studio) are the clear winner compared to the KEFs (Hifi), until know I can´t cleary say, which one of those two I like better. JBL (PA) oder Orbit (Studio).

Both sound totally different and now it get´s very subjective. The JBLs play so effortless and livelike, I appreciate them very much for live concerts and bad produced music with much compression, which are fun to listen to at the JBLs but not so much at the Orbits. The Orbits on the other hand really shine with intimate music and high quality productions. Which doesn´t mean they sound bad at the JBLs, but a lot different. And despite they measure nearly the same in mids & highs and both should have less room influence (cardioid vs. cbt), the JBLs sound mouch more brighter, lighter and way more open. So now I have two speakers which I really like and can´t decide. The studio speaker is much more easily integrated in a living room (simply put them on stands), with the JBLs it´s really hard (there are no nice stands).

I thought I share my thought´s and experiments with you and hope you find that interesting.


At the end, for people reading till here and to bringt it all together, all three Speakers overlaid:

Right channel, Green Orbits, yellow KEFs, red JBL

Left channel, green Orbits, yellow KEFs, Orange JBLs.

 

[90sNintendoKid]

more to come, clicked to fast

Are you using EQ?

 

[Paffi]

Are you using EQ?

yes, I´m experimenting also with EQ. And the room is acoustically treated and there are 4 PSI Avaas working. I also have 4 wall mounted subs EQed with MSO, but they are turned off here.

 

[90sNintendoKid]

I think this analysis is awesome.

I have the Orbits too and I am planning on doing something similar, especially distortion measurements.

 

[Jim Taylor]

Your experiment MIGHT indicate that ...

1) People are very sensitive to slight changes in frequency response, including changes so slight that they were previously thought to be insignificant.
2) Dispersion characteristics (including cardioid response) might cause a greater difference in perceived in-room differences than commonly supposed.

This supposes, of course, that your listening was done with matched levels. From your post, it seems that there was no attempt to control sighted listening ... unfortunately.

 

[ernestcarl]

Why not match the bass response of the three speakers to eliminate at least one huge variance in your listening comparison?

 

[kemmler3D]

Good test here, an interesting result raises further questions. To me this report probably demonstrates something we already repeat very often - that off axis response is very important to how a speaker sounds. It's interesting to see how close the graphs are, but your subjective notes are pretty different. How close is your listening position, and how did you orient the mic?

 

[Paffi]

It's interesting to see how close the graphs are, but your subjective notes are pretty different. How close is your listening position, and how did you orient the mic?

Yes, I'm surprised as well and attribute it to the differences in off-axis, that was the reason I wanted to try the cbt thing, after the difference between ls50meta and orbit 11 was that big.

Listening distance is 2,5 meters, mic is pointed to the ceiling.

 

[Paffi]

Why not match the bass response of the three speakers to eliminate at least one huge variance in your listening comparison?

Valid point - the bass simply wasn't in my focus, because usually I use my subs. While the listening comparison, sometimes I turn them on (with same crossover), but the overall impression doesn't change.

 

[Blueprint]

What’s the SPL level? For distortion measurements and listening impressions.

 

[Randolf]

Nice work, of course frequency response and distortion is not the complete story. On the other hand frequency response of these 3 speaker do look rather different to me, depending on scale an alignment. On a more zoomed in and a litte bit different aligned view your data look like this:

Amir's in room estimantion for these speakers looks like this:

I guess the significant lower SPL between 2 and 10Khz of the LS50 Meta explains why you prefer the the other two. A few dB in the most sensitive frequency range of human hearing can make a huge difference.

BTW, which measurement technique did you used (single measurement at listening position, average of multiple measurements around listening position, moving microphone measurement with RTA, ....)?

 

[jeffaegrim]

The biggest difference - besides output - between these after EQ will be the width of the HF beam and the different reflection profile they create in-room. I have measurements of the CBT 70‘s overlaid with measurements of high DI horn speakers that look almost exactly the same - but both speaker pairs sound very dissimilar; there is much more direct sound from the horns but with a smaller soundstage, less direct sound from the CBTs but with a huge soundstage. The difference is easy to spot by looking at the impulse response.

I said it before, but the JBLs are „real“ speakers, in a sense that they provide almost unlimited dynamics over their playing range. Keeping them might necessitate buying 1(2) sub(s) though. The Orbits look good. They just won’t play loud at 3m.

 

[illusonic]

And the room is acoustically treated and there are 4 PSI Avaas working.

That seems like a lot of money for bass performance as a single AVAA C20 costs more than 3 Orbits. Not sure if this is really what you are using?
Where the AVAAs active during your measurements?

The CBT is horizontally much wider than the Orbits. -6dB is mostly close to 100° up to 4kHz while the Orbits are more like 60°.
Vertically, the CBT is extremely narrrow, basically avoiding any floor/ceiling reflections.
Both increase direct to reverberant sound ratio but in very different ways.
So they should sound quite different based on dispersion angles alone - interesting comparison though.

In addition, the huge membrane area and high sensitivity of the CBT probably give them massive SPL headroom over the Orbits (or any midfield monitor).
So these are simply very different speakers. Why not keep both if you like them? Cost seems small compared to the bass traps.

 

[illusonic]

I guess the significant lower SPL between 2 and 10Khz of the LS50 Meta explains why you prefer the the other two. A few dB in the most sensitive frequency range of human hearing can make a huge difference.

He EQed the KEFs to the same levels in mid/treble as the other speakers...

 

[Trinkle]

@Paffi

I am not a measurement expert, but I guess we need to look at intermodulation distortion instead of harmonic distortion. The palmer is supposed to be better at it than the KEF given it has extra drivers.

Secondly, the palmer is made out of aluminium, probably it has less resonance than the KEF and would help to compare CSD plots ?

 

[Paffi]

What’s the SPL level? For distortion measurements and listening impressions.

The distortion measurements correlate to the SPL measurements. For listening impressions: various levels over many days (to exclude different mood at different days), from loud to silent.

Nice work, of course frequency response and distortion is not the complete story. On the other hand frequency response of these 3 speaker do look rather different to me, depending on scale an alignment. On a more zoomed in and a litte bit different aligned view your data look like this:

I guess the significant lower SPL between 2 and 10Khz of the LS50 Meta explains why you prefer the the other two. A few dB in the most sensitive frequency range of human hearing can make a huge difference.

BTW, which measurement technique did you used (single measurement at listening position, average of multiple measurements around listening position, moving microphone measurement with RTA, ....)?

That´s pretty interesting, thanks! It was single measurement at the listening position, as I wanted to have the room influence measured as well and not a pure speaker measurement. However, I can´t correlate the different impressions to the (anechoic) SPL between 2 and 10Khz, I´m very sure that this is not the reason, I think it´s the composition of the spl (direct + reflected) at the listening spot. I support that, what @jeffaegrim says:

The biggest difference - besides output - between these after EQ will be the width of the HF beam and the different reflection profile they create in-room. I have measurements of the CBT 70‘s overlaid with measurements of high DI horn speakers that look almost exactly the same - but both speaker pairs sound very dissimilar; there is much more direct sound from the horns but with a smaller soundstage, less direct sound from the CBTs but with a huge soundstage. The difference is easy to spot by looking at the impulse response.

I said it before, but the JBLs are „real“ speakers, in a sense that they provide almost unlimited dynamics over their playing range. Keeping them might necessitate buying 1(2) sub(s) though. The Orbits look good. They just won’t play loud at 3m.

Just a simple experiment, which worked very well comparing these three speakers. While listening to music, put the hands behind your ears, to get an estimation of the effect of the room acoustics. With the KEFs, der Difference is HUGE. I mean really huge. With the Palmers, the difference is very small. With the JBLs, the difference is completely gone. Pure direct sound. Where it stands out pretty clear is Anette Askviks "April", but this works with nearly every track.

That seems like a lot of money for bass performance as a single AVAA C20 costs more than 3 Orbits. Not sure if this is really what you are using?
Where the AVAAs active during your measurements?

I think I´m pretty sure which devices are in my listening room The AVAAs where active. Room acoustics always comes first to me, so comparing the price of the avaa to the loudspeaker is not appropriate for me.

The CBT is horizontally much wider than the Orbits. -6dB is mostly close to 100° up to 4kHz while the Orbits are more like 60°.
Vertically, the CBT is extremely narrrow, basically avoiding any floor/ceiling reflections.
Both increase direct to reverberant sound ratio but in very different ways.
So they should sound quite different based on dispersion angles alone - interesting comparison though.

Both increase the sound ratio, as you said, but nevertheless sound different - it is confirmed by the simple test, described above. I found that pretty interesting too, leading to the question, if chasing the perfect measuring loudspeaker is really the way to go. Based on this experience, for me it´s not. I´d rather sacrifice some fine measurements in exchange for less room effect, hence I would prefer the Orbit over a Genelec.

Why not keep both if you like them? Cost seems small compared to the bass traps.

Because I´m willing to invest as much money as necessary for good sound - but not more. Hence I spent much more money on room acoustics then the technical equipment. I don´t need two speaker setups, I´m not rich Therefore I´ve decided to let the Orbits go.

As a conclusion: with the JBLs I get the dynamics of a horn with the soundstage of a KEF / Genelec Coax. That´s how I would describe that and what I was looking for, for 2 years already.

 

[illusonic]

yes, I´m experimenting also with EQ. And the room is acoustically treated and there are 4 PSI Avaas working. I also have 4 wall mounted subs EQed with MSO, but they are turned off here.

I think I´m pretty sure which devices are in my listening room The AVAAs where active. Room acoustics always comes first to me, so comparing the price of the avaa to the loudspeaker is not appropriate for me

Obviously it was me not being sure if you really spent 10k on active bass traps. Would be intresting to see the frequency response without basstraps.

Considering your efforts in bass traps I guess acoustic treatment may also be quite substantial. Combined with vertically constraint dispersion of the CBT you may be listening with very low room interaction.

 

[Jazigo]

Out of curiosity: do you have the step response for each?

My predictions are these:
- The KEF LS50 Meta will have longer trails of swings after the step.
- The Orbit will be dead flat
- More unsure about the JBL, but somewhere in between.