Rogers LS3/5a (BBC) Speaker Review

[Haskil]

An expensive audiophile joke...

Used to listen to music at home, it does not allow you to listen to a symphony orchestra without the risk of hearing the bass-midrange loudspeaker dragging... the same goes for a piano disc. Lived experiences... little more stirring than a Mozart sonata listened to pianissimo...

With a sensitivity of 83 dB and such a low admissible power it is logical... This speaker can only be used for close-range listening in a small, quiet space where the residual background noise must not exceed 30 to 33 dB... suffice to say that this doesn't happen every day... this speaker is totally outdated.

 

[AudioSceptic]

And the versions keep coming it seems:

Musical Fidelity LS3/5a Loudspeaker Review

Musical Fidelity LS3/5a Review

www.stereonet.com

View attachment 329473

David Price. Wow. That guy is still alive...

Yes. <https://www.stereonet.com/uk/news/david-price-joins-stereonet-as-editor-in-chief> and <https://www.youtube.com/channel/UCq45fRHZ3xQPYryZl_Z_r0A>

 

[Westsounds]

He does youtube as well. Here's a subjective review. Not of the Rogers but same speakers, they should all sound the same in theory.

 

[DanielT]

He does youtube as well. Here's a subjective review. Not of the Rogers but same speakers, they should all sound the same in theory.

I see audiophiles all the time, they only see (hear)...what they want to hear. They don't know what they are imagining.

Having said that. They can imagine as much as they want if they feel like it. Fun for them, but their imagination is their subjective one so it doesn't say much.

 

[mtmpenn]

I have the crossover schematic for the original continuum, but it uses the dayton original rs28a tweeter which is NLA.

Shame about Salk, I had no idea.

I have the schematic AND the driver's actually... I just haven't gotten around to building them in the past decade.
I keep thinking I will... but work and two young kids has left precious little time.

I have never heard any LS3/5a variant, but I have a question for the group about these measurements.


Klippel measurements, predicted in-room response, and preference score are all really interesting and can clearly show poor design (like the woofer resonance here popping up).

However, the process of taking these near field measurements and translating them into predicted in-room measurements and then assigning a preference scores must rely on a wide range of assumptions (speaker placement, room shape/size, etc). If you were custom designing a speaker for your own specific use it seems you might purposefully deviate from klippel perfect to take into account distance from back wall, room size/shape, tweeter height relative to ear height, intended listening dB, toe-in, etc.

Is it really fair then to assume that all speakers should adhere to to the same ideal? It is striking to me for example that so many of the computer style speakers have a similar shape to their frequency curve. It looks terrible on the klippel, but isn't it at least possible that we would want a different frequency response for a speaker intended to be placed on a desk and listened to at low spl without toe and 24 inches from the listener's ears in a tiny room compared to something on stands, 3 feet from the rear wall in a large room, with toe-in, 10 feet from the listener?

 

[GXAlan]

Klippel measurements, predicted in-room response, and preference score are all really interesting and can clearly show poor design (like the woofer resonance here popping up).

Yes, all good points. First, the Klippel NFS can be replaced with a massive military-grade anechoic chamber. The power of the Klippel NFS is that you can get speaker measures in a regular room through math.

There is enough data now to show that there’s extremely good correlation to that theoretical super anechoic chamber.

When you get into predicted room response, it’s true that it’s only valid for one simulated scenario. Your room and my room will certainly be different. That said it’s nice to see how the spinorama can be integrated into a single curve for easier understanding.

Is it really fair then to assume that all speakers should adhere to to the same ideal?

This is where the data is the data and the interpretation of the data is what is fair or not.

The Sonos Roam has a spectacular spinorama

Measurements for speaker Sonos Roam

www.spinorama.org

It deservedly is probably one of the best jumbo candy bar style speakers.

But it obviously is not going to sound as good as a Meyer Sound Amie or JBL 708P, both of which have worse Preference Scores.

This is where the balance of near field and far field may not be perfect. The original patent even shows the spread of predicted score and actual score. As you get into different scenarios you can imagine the spread being larger. Of course, the correlation is not random chance and it’s good.

… isn't it at least possible that we would want a different frequency response for a speaker intended to be placed on a desk and listened to at low spl without toe and 24 inches from the listener's ears in a tiny room compared to something on stands, 3 feet from the rear wall in a large room, with toe-in, 10 feet from the listener?

Look at the Meyer Sound Amie review for an example where the speaker doesn’t match the standard preference score metrics but sound great.

You actually see a similar strategy to reduce vertical dispersion with the Genelec S360 which initially looks like a directivity error but surprisingly is similar to the approach taken by Meyer Sound.

 

[theREALdotnet]

The Sonos Roam has a spectacular spinorama

Measurements for speaker Sonos Roam

www.spinorama.org

It deservedly is probably one of the best jumbo candy bar style speakers.

But it obviously is not going to sound as good as a Meyer Sound Amie or JBL 708P, both of which have worse Preference Scores.

If a simulation and mathematical model don’t match reality then the creator of the simulation and model (Klippel GmbH in this case) need to take this feedback on board and improve what they’ve created. Otherwise their creation will soon lose relevance.

 

[GXAlan]

If a simulation and mathematical model don’t match reality then the creator of the simulation and model (Klippel GmbH in this case) need to take this feedback on board and improve what they’ve created. Otherwise their creation will soon lose relevance.

The Preference Score has nothing to do with the Klippel NFS. They are two very different things. That spin of the Sonos Roam was done traditionally with outdoor gated measurements, for example.

The preference score is known to have its limitations and as Harman's research has moved toward headphones, the same research is now being kept internal trade secret, so the don't really care about the errors in the Preference Score because anything the know now to improve the computation is kept secret.

When talking about Klippel NFS vs dedicated anechoic differences, I am talking about things like temperature differences in a home environment versus dedicated test environment with specialized HVAC or the effect of the stand or microphone boom which is minor but discussed by amir in various historical posts.

 

[mtmpenn]

Interesting discussion.

For my part, I am not questioning the validity of klippel’s ability to estimate an anechoic measurement.

I’m questioning whether an anechoically flat speaker is the preferred speaker for every use case.

A speaker designed to be used at 70db, placed 6” from the wall behind it, on a reflective desk, 2 feet from the listener with the tweeters 6” below ear height seems like it should have a different mandate than a speaker designed to be used at 85db placed on a stand, 3’ from the rear wall, tweeter at ear height.

These two theoretical speakers might measure differently in an anechoic chamber and they might be perceived as sounding very different if you used them in a situation for which they were not designed.

 

[Vladimir Filevski]

... they might be perceived as sounding very different if you used them in a situation for which they were not designed.

So please use them as they are designed. Use airplane for flying and submarine for diving (as they are designed), not vice versa.

 

[GXAlan]

I’m questioning whether an anechoically flat speaker is the preferred speaker for every use case.

When you state anechoically flat speaker, do you mean on-axis or off-axis? An anechoically flat speaker in my opinion is preferred in most use cases, but the directivity is a different story. The highest directivity score is the most predictably good speaker in many homes, but we definitely have data showing that something that is on-axis flat but off-axis not can be a good idea.

Likewise, we know we like loudness contour and you can imagine a non-flat speaker having a superior inherent smiley face response that works at some specific SPL.

A speaker designed to be used at 70db, placed 6” from the wall behind it, on a reflective desk, 2 feet from the listener with the tweeters 6” below ear height seems like it should have a different mandate than a speaker designed to be used at 85db placed on a stand, 3’ from the rear wall, tweeter at ear height.

It does -- but you might find that the goal is still on-axis flat at the listening position, weighing the SPL and loudness contour, but the directivity may be very different.

 

[Purité Audio]

Interesting discussion.

For my part, I am not questioning the validity of klippel’s ability to estimate an anechoic measurement.

I’m questioning whether an anechoically flat speaker is the preferred speaker for every use case.

A speaker designed to be used at 70db, placed 6” from the wall behind it, on a reflective desk, 2 feet from the listener with the tweeters 6” below ear height seems like it should have a different mandate than a speaker designed to be used at 85db placed on a stand, 3’ from the rear wall, tweeter at ear height.

These two theoretical speakers might measure differently in an anechoic chamber and they might be perceived as sounding very different if you used them in a situation for which they were not designed.

A designer can only really create a really fine sounding loudspeaker they possibly account for every scenario in which it will be used and of course a fine sounding loudspeaker will still sound better than a poor one whatever the scenario.
Having said that increasingly contemporary designs include placement options the little KEF LXs for example.
Keith

 

[Murray A]

As a very early convert to the LS3/5A speaker I was immediately drawn to the thorough measurements utilizing the Klippel system.

In the late 70's through early 80's the buzz within the industry at the time, was very favorable to the sonic characteristics of this dimunative monitor.

Group members from the Toronto audio society participated in a large number of level matched double blind listening sessions at a technologically aware audio retailer in Toronto.

A personal friend was involved in in the design, construction and implementation of the switching system closely adhering to what was in use in the research at Canada's NRC.

A session setup consisted of four popular speakers from the marketplace in mono, positioned behind an acoustically transparent black scrim cloth.

The speakers involved from my memory were Dynaco A25, EPI 100, JBL L100, AR3Ax? the Advent speaker, Spendor BC1 and a smaller Infinity with Emit tweeter. There were others involved but the aformentioned stand out in my memory.

In every listening session I attended, the consensis winner was always the LS3/5A. As far as know an LS3/5A was always one of the 4 in every session. A rumour passed on to me by a reviewer that participated in the NRC tests was that the LS3/5A scored highly there as well.

I enjoyed four sets of LS3/5As over the years until release of the KEF LS50, Neumann KH80, 120 and 150, as well as the KEF R3.

I gave up mine in 2006 due the noticeable response peak in the 1KHz region due to age.

The LS3/5A was well reviewed in its time and was a standout amongst its peers in blind level matched comparisons.

 

[mtmpenn]

A designer can only really create a really fine sounding loudspeaker they possibly account for every scenario in which it will be used and of course a fine sounding loudspeaker will still sound better than a poor one whatever the scenario.
Having said that increasingly contemporary designs include placement options the little KEF LXs for example.
Keith

I think that this is likely true in most scenarios, but it may not be true in all scenarios.

There could be a rationale for designing a speaker for a specific purpose.

For example, if a company knew that 95% of their customers were going to put the speaker 3 inches from the wall without toe-in, they could very rationally choose to design a speaker with less baffle step compensation and more high frequency energy than would look optimal on a klippel.

Said speaker would not be optimized for the audiophile who insisted on setting it on stands 3’ from the wall with toe in to the main listening position, but would still perform better for its intended purpose.

I’m not trying to defend this particular speaker. Also there are clearly lots of examples of the klippel revealing major flaws in speakers on this site which is very informative.

I just worry that we have an inclination here to apply the same standard to all speakers when there may be a real rationale for deviation because they may be designed for a different purpose.

Regardless, I going to let it go now!

 

[Purité Audio]

I don’t see any problem designing a loudspeaker specifically for a particular application, corner placement, close to the front wall etc.
Keith

 

[sergeauckland]

I don’t see any problem designing a loudspeaker specifically for a particular application, corner placement, close to the front wall etc.
Keith

Exactly. There are lots of loudspeakers that are designed for specific applications. Fortunately most don't make it out in the world as 'audiophile' loudspeakers in the way the LS3/5a has.

S.

 

[audio_tony]

I don’t see any problem designing a loudspeaker specifically for a particular application, corner placement, close to the front wall etc.
Keith

Surely most speakers are designed with placement in mind? Whether it be on stands, on the floor, x feet away from walls / corners etc.

Otherwise manufacturers wouldn't provide placement advice in their manuals?

 

[Purité Audio]

Pro manufacturers do , close to front wall within 60cm or more than a metre and a half away but do many domestic manufacturers?
Keith

 

[audio_tony]

Pro manufacturers do , close to front wall within 60cm or more than a metre and a half away but do many domestic manufacturers?
Keith

A few domestic speakers I've owned over the years have contained recommendations for placement in the manuals (generally a pamphlet!) - however I will concede that there are quite a few that don't offer any placement recommendations either.

 

[Purité Audio]

I remember an ardent ATC fan being enraged when someone suggested that his speakers and he ‘should’ be more or less placed in an equilateral triangle ( his speakers were too close to each other) when he asked who suggested such a thing the poster pointed to the ATC manual.
Keith