Sonos Roam Portable Speaker: Spinorama and measurements (more proof mainstream speakers can be real good)

[napilopez]

Measured the Sonos Roam today, the company's new, smaller bluetooth speaker.

Here's the spin at 80% volume (measured at ~ 0.6m). So this is around 76dB@1m. Please read the disclaimer at the end of the post

And here an estimate of it at 60% volume (i spliced a nearfield woofer measurement made at 60% volume and then spliced it to the 80% farfield measurements, so it might not be totally representative). This is really about 68 db @1m:

More detailed measurements later, but basically: a very good speaker with limited SPL and a few flaws, but clearly a lot of attention was paid to the sound. As it's a portable speaker it seems emphasis was made on the in-room and off-axis response.

Important disclaimer: I'm not 100% sure about the bass here. As I couldn't remove the grille, I couldn't get very close to the woofer as I'd normally do, so I did not apply baffle step correction. This is because the measurements lined up closely with far-field measurements even with the closest I could get the microphone. So take the bass with a grain of salt.

For example, here's the my on-axis bass splice with the actual non-gated on-axis response in my backyard:

But there you go

 

[sweetchaos]

Many thanks.
Any chance of providing spinorama data?

 

[napilopez]

Many thanks.
Any chance of providing spinorama data?

Full data or just the spin curves?

I can provide the merged angle files when I'm on my PC but I prefer to analyze directivity sans merging since that kinda muddies the behavior in the low mids. So just need to keep that disclaimer in mind.

Anyway, going to try to measure five speakers tomorrow... Wish me luck.

JBL 4309
Polk Reserve r200
Genelec 8341B
Focal Solo 6 Be
KEF KC62

Edit: Oh a couple more disclaimers/notes: I definitely aimed the microphone a little wrong on this one. The speaker was measured in its vertical orientation, with the Sonos logo on the top. I thought the tweeter was at the center of the speaker but it's actually a little below that.

also, measuring very small speakers is tricky y'all. My normal setup causes too many small reflections that rather dramatically altered the frequency response. So I brought out my tripod like I did for the Nest Audio to minimize reflections, but even that was tricky.

 

[sweetchaos]

So we can calculate Olive Preference Score and gents can work their auto-EQ magic.

Pikachu face ...
You're spoiling us man!

 

[Valentin R]

Thanks

Google nest
Sonos Roam
Etc etc

It’s definitely a trend yes not very hi levels but in their class it’s great for such a divice

 

[napilopez]

Thanks

Google nest
Sonos Roam
Etc etc

It’s definitely a trend yes not very hi levels but in their class it’s great for such a divice

Yeah, from 70Hz to 5kHz I think the Roam is incredibly flat at low volumes, flatter than the KH80 even in this region, assuming the bass is correct. So it's nice for background music. You'll lose bass as you turn it up (the bass stops being flat between roughly 60% and 70% volume), but it's a really nice little speaker.

 

[pierre]

Quizz: which speaker is that?

 

[Valentin R]

Yeah, from 70Hz to 5kHz I think the Roam is incredibly flat at low volumes, flatter than the KH80 even in this region, assuming the bass is correct. So it's nice for background music. You'll lose bass as you turn it up (the bass stops being flat between roughly 60% and 70% volume), but it's a really nice little speaker.

Look @ Presonus and Mackie 3” ers Erin just measured

Seems Sonos and Google nest to be a better option at low volumes

 

[Vladimir Filevski]

@napilopez
Many thanks for the measurements of Sonos Roam!
However, listening test is also very important, and Oluv (from Oluv's gadgets / clavinetjunkie) is less than satisfied (which is a huge understatement!):

 

[napilopez]

@napilopez
Many thanks for the measurements of Sonos Roam!
However, listening test is also very important, and Oluv (from Oluv's gadgets / clavinetjunkie) is less than satisfied (which is a huge understatement!):

I always listen before measuring, but I haven't tested too many portable Bluetooth speakers so I cant say how it compares to others, and I'm not really making a value judgement. I liked how it sounds but oluv is welcome to his opinion of course! I did not notice the excessive distortion he described, but that could very well be because of different settings.

For these measurements, I turned off Auto TruePlay and Loudness EQ, while it seems Oluv had both of those engaged (sorry if I'm mistaken, I skimmed through the video).

He says that TruePlay wasn't doing anything but in my case it had a very obvious and measureable effect. In fact, my original intention was to create a spin both with TruePlay as that's how most people will use it, but it kept on slightly modifying the frequency response with every turn of the speaker.

Still, you can compare the on-axis responses. When measuring outdoors it appeared to be having a detrimental effect, perhaps because my backyard is very noisy, but indoors I've generally liked the effect.

 

[q3cpma]

Directivity is impressive (not too hard to get when using woofers so small they're midrange sized), but otherwise, it's just using DSP while eating the headroom, really. Wouldn't spend 180€ on this, personally.

 

[mononoaware]

Thanks for sharing your measurement of Sonos Roam.
I listen at mouse whisper quiet volume's so it makes me think of the Roam as a good portable speaker option.
I was interested in the Roam when I first heard it was announced, and your measurements help with adding confidence in the product.

For these measurements, I turned off Auto TruePlay and Loudness EQ, while it seems Oluv had both of those engaged (sorry if I'm mistaken, I skimmed through the video).

As I understand it Sonos "Trueplay" is their version of auto DSP room correction.
So does it do a frequency sweep through the App and uses your "phone/device's" microphone to tune the DSP for the environment of the Sonos Roam?

Seems odd to have this feature when it is a portable speaker, meaning it is likely to be in a different environment every time it is used. . .
Your conclusion is it is best to leave TruePlay "off" anyway so maybe that is the best option.

 

[napilopez]

Thanks for sharing your measurement of Sonos Roam.
I listen at mouse whisper quiet volume's so it makes me think of the Roam as a good portable speaker option.
I was interested in the Roam when I first heard it was announced, and your measurements help with adding confidence in the product.



As I understand it Sonos "Trueplay" is their version of auto DSP room correction.
So does it do a frequency sweep through the App and uses your "phone/device's" microphone to tune the DSP for the environment of the Sonos Roam?

Seems odd to have this feature when it is a portable speaker, meaning it is likely to be in a different environment every time it is used. . .
Your conclusion is it is best to leave TruePlay "off" anyway so maybe that is the best option.

Regular TruePlay plays a sweep through your phone and uses your phone's microphone. I didn't test this as I didnt have an iPhone handy. Auto TruePlay is completely self contained, using the speaker's built in microphone to analyze the sound of reflected sounds while music is being played.

So in theory it does make a lot of sense for a portable speaker to have it; the tuning is completed automated so whenever you reposition the speaker the sound is, in theory, optimized after a few seconds. Note that I'm not saying not to use trueplay, but for the purposes of these outdoor measurements, it was modifying the frequency response in a way that wasnt optimal for measurements. It's possible the speaker was adjusting the sound based off what it thought was the predicted in room ( well, outdoor, in this case) response.

Loudness EQ mainly affects the bass level at different SPLs

 

[Zensō]

Measured the Sonos Roam today, the company's new, smaller bluetooth speaker.

Here's the spin at 80% volume (measured at ~ 0.6m). So this is around 76dB@1m. Please read the disclaimer at the end of the post

View attachment 129934

And here an estimate of it at 60% volume (i spliced a nearfield woofer measurement made at 60% volume and then spliced it to the 80% farfield measurements, so it might not be totally representative). This is really about 68 db @1m:

View attachment 129940

More detailed measurements later, but basically: a very good speaker with limited SPL and a few flaws, but clearly a lot of attention was paid to the sound. As it's a portable speaker it seems emphasis was made on the in-room and off-axis response.

Important disclaimer: I'm not 100% sure about the bass here. As I couldn't remove the grille, I couldn't get very close to the woofer as I'd normally do, so I did not apply baffle step correction. This is because the measurements lined up closely with far-field measurements even with the closest I could get the microphone. So take the bass with a grain of salt.

For example, here's the my on-axis bass splice with the actual non-gated on-axis response in my backyard:

View attachment 129949

But there you go

Thanks very much for sharing these results. I’d love to see more “consumer” gear tested here on ASR. Do you have a website or YouTube channel?

 

[Ralph_Cramden]

An old story, but shows that Sonos has been measuring their products for a long time, with test facilities much better than some "high end" speaker companies.

https://vowe.net/archives/015354.html

 

[abdo123]

So we can calculate Olive Preference Score and gents can work their auto-EQ magic.

Pikachu face ...
You're spoiling us man!

this is alreadly EQed though.

 

[mononoaware]

Auto TruePlay is completely self contained, using the speaker's built in microphone to analyze the sound of reflected sounds while music is being played.

I remember now some "portable" bluetooth speakers have a built-in microphone (so people can take phone-calls with the speaker), I had completely forgotten about it.

I see so the Roam's built-in microphone monitors the speaker's output and adjusts accordingly, and it does so while music is being played so a "frequency sweep" tone is not necessary.

So in theory it does make a lot of sense for a portable speaker to have it; the tuning is completed automated so whenever you reposition the speaker the sound is, in theory, optimized after a few seconds.

Interesting, so it is constantly automated.

Note that I'm not saying not to use trueplay, but for the purposes of these outdoor measurements, it was modifying the frequency response in a way that wasnt optimal for measurements.

This is the first time I have heard of this "DSP tuning" which is constantly automated, so maybe it is new technology which is not fully optimised yet.

Loudness EQ mainly affects the bass level at different SPLs

Thanks for confirming this. I typically associate "loudness" EQ with both bass level and high frequency level (V-shaped) adjustment.

 

[napilopez]

Thanks for confirming this. I typically associate "loudness" EQ with both bass level and high frequency level (V-shaped) adjustment.

it often/usually is, but I think Sonos' is mainly a bass adjustment. As it should be. A little bit of treble adjustment can help but it's mostly the bass that really needs tweaking at low spls

 

[mononoaware]

at low spls

Ok thanks.

 

[jonfitch]

Very impressive. Sonos speakers perform amazing and certainly punch well above their price point. I wish they offered full bluetooth (stereo pairing, software support through BT) or physical connectivity with their speakers so you aren't limited to a tiny ecosystem, just like the Apple Homepod. The Move is nice but doesn't support stereo pairing in bluetooth even though a lot of other BT speakers do, it's kind of baffling.

The sad thing is the best sounding bluetooth speakers (Homepod, Sonos) are the most limited in use, whereas the most versatile functioning BT speakers are the crappiest sounding ones