About Room compensation in BL50/90

[MarioA]

Hi everyone, does anyone know whether room compensation process should be run with the speakers set in a specific directivity mode and repeated for each one (narrow, wide, omni)? Or is just a set of readings enough for the system to create the proper filters for the different Modes? I don’t believe I see that mentioned in the manual.
Thank you for reading this.

[Curly]

Great question! I’d like clarity on that too!

[NQVHNWI]

Ok.

You run calibration on your BL50/90 (and I’m talking specifically BL90 from my experience – I assume 50’s are the same) via the internal loudspeaker calibration process. This is distinct and separate from the BS Theatre calibration. The calibration(s) is/are generic for the number of listening points and calculates Wide/Narrow and Omni all at the same time.

When calibrating the room response for the Theatre processing, I have found that you must ensure the internal BL90 switching is/or was left in wide mode.

Then you run RoomSense with the calibration speaker preset set in Auto/zer0 (0)

This seems to give good results. I initially did this with the Speaker Preset in Theatre set to 2 (I assumed wide – as per the BL90 calibration – to give the latency time) but this was very echoey.

 

[NQVHNWI]

To add further.

If you wish to run music only (Wide/Narrow/Omni) – then repeat the Theatre with preset 1/2/3.  IIRC, The BeoApp has a toggle for high/low latency during setup. I’m not going there to check as it will bugge*-up my calibration (again). I also dont play any critical music via the Theatre to the BL90s, preferring a 3rd party source via XLR input.

NB: I think the app may also allow you to select W/N/O mode with the latency?? As I say…im not going there for you as it messes up my settings which im happy with.

[MarioA]

…Oh my goodness, I don’t want you to mess up your settings. I know what you mean, the software is still somewhat buggy, and I always think twice before getting into the system. I can see why the calibration on the 90s could be “generic”, take one and be done, but the 50s do have a movable lens to project the sound in wide or narrow mode. And so, I suspect that because of the two modes the calibration should be run individually, I just wanted to know if there is any official word about it.
Thank you so much for your kind reply.

[Yossi]

What I understood is that wide and narrow (for BL50) is a completely different behavior for the speakers (thanks to dsp), like having 2 different pairs of speakers.

So I run a room compensation with narrow selected and 3 points closely spaced (one chair no friends).This is for critical listening

And I run another room compensation with wide selected and 3 points “around my sofa” as explained by Geoff. This is for sharing music or a movie with my family.

[kawo]

Intersting! Did not know that you can let the BL50 run in wide mode (which means the acustic lens opens up) during calibration. I recall it automatically runs calibration in the “narrow” mode when you start the process.

[NQVHNWI]

Yossi wrote:

What I understood is that wide and narrow (for BL50) is a completely different behaviour for the speakers (thanks to dsp), like having 2 different pairs of speakers. So I run a room compensation with narrow selected and 3 points closely spaced (one chair no friends).This is for critical listening And I run another room compensation with wide selected and 3 points “around my sofa” as explained by Geoff. This is for sharing music or a movie with my family

As I understand it….with the Bl90, is that there are two distinct processes going on here:-

1. Narrow mode uses high latency calculations to 3 points in space (one chair, no mates, around where your head is). The narrow mode calculations minimise in-room reflections and amongst other things give “the sharpest imaging”. Wide mode on the otherhand can be used with lower latency, reduce the in-room calculations to reduce room reflections BUT offer amongst other things, a better lip-sync when watching a Beovision or less critical music listening.
2. The Second principle is of the number of listening positions and merging the separate calculations together. You can “blend” multiple one chair, no mates, listening positions in the room in narrow mode and merge them all together. (Not sure what the benefit will be but you can do it – or you just use wide or omni mode)

[geoffmartin]

Maybe I can help to clarify this…

The Beam Width / Beam Direction control settings (which I normally group under the term “directivity”) and the Room Compensation perform completely separate tasks for different reasons, although the directivity has an influence on the Room Compensation. The Latency mode is also completely independent, but has an effect on the directivity and therefore the room compensation.

When the room compensation measurements are being performed, the loudspeakers are not in any Beam Width mode, because each loudspeaker driver is measured individually. Those individual measurements can be used to calculate the room’s response to the loudspeaker for any of the available directivity modes.

This is why on the Beolab 90, if you are in a Preset with a chosen Directivity (Beam Width / Beam Direction) and a Room Compensation filter and you change the directivity, a new filter is calculated using the measurements of the loudspeaker drivers.

The reason this is necessary is that each of Beolab 90’s 14 Directivity/Latency modes (or each of Beolab 50’s 4 modes) couple differently to the room – so each mode requires a different filter.

If you’re wondering what those 14 (or 4) modes are: it’s like this:

Beolab 90:
(Narrow + 5 Wide directions + Omni) * 2 Latency modes = 7 * 2 = 14

Beolab 50:
(Narrow + Wide) * 2 Latency modes = 4

Note that you can have any Beam Width or Beam Direction in either latency – they’re independent.

If we discuss WHY these features are there:

• Narrow mode is intended to reduce the amount of sidewall reflections
• Wide mode has the double-duty of ensuring that the beam width is wider for a wider listening area as well as matching the directivity of the Beolab 5 more closely – for someone who uses both Beolab 90/50 and Beolab 5 in a surround setup
• Omni is a background music setting – for hearing music without listening to it actively.

The Room compensation controls the resonances in the room, the behaviour of which change with changes in directivity – but which are a separate artefact that requires correcting.

(The reason for High and Low latency modes  is explained in the Technical Sound Guide. I won’t repeat that here.)

 

The microphone measurements are ONLY used to create the Room Compensation filters. The filters that control the loudspeakers’ directivity modes were created using the measurements we did here in Struer and are unchanged by the measurements done after installation.

Hope that helps.

[Yossi]

Oh thank you for clarifying !

 

when I do the room compensation measures for the BL50 , the acoustic lens can be open (wide) or narrow, so the position of the acoustic lens does nothing  on the  measures? Or is it a bug and I should measure in narrow ?

[geoffmartin]

Hi Yossi,

It’s not a bug – the position of the acoustic lens can be either narrow or wide during the measurement.

[MarioA]

geoffmartin wrote:

Hi Yossi, It’s not a bug – the position of the acoustic lens can be either narrow or wide during the measurement.

Hi Geoff, The fact that room correction measurements can be performed with the lens in wide or narrow mode, is because the process does not involve high frequencies in the tweeter territory? Interesting to know and thank you for clarifying and for everything you do.

[geoffmartin]

Hi MarioA,

Generally speaking, the Room Compensation filter does not extend to the frequency range covered by the tweeters; nor should it, since at higher and higher frequencies, the room’s response is increasingly non-linear (on a long-term time scale). For example, small changes in the room (like someone walking in) create large changes in the response, small changes in measurement and listening position cause large changes in response, and we increasingly rely on the direct sound to determine timbre at high frequencies – so the room plays an increasingly smaller role.

Put this together and you can say that, as you go up in frequency, the room plays less and less of a role in the filter applied to the audio signal from the loudspeakers at the listening position, and you’ll get more and more of an error in your measurement as a prediction of “how it sounds”. Diminishing returns…

 

Finally, just to be clear – everything I’ve said in this thread is focused on the original question and heading. I’ve only been talking about Beolab 50/90.

[MarioA]

All very interesting, thank you again!

[Author]

Posts