• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Sonos Five Smart Speaker Review

Rate this smart speaker:

  • 1. Poor (headless panther)

    Votes: 13 4.0%
  • 2. Not terrible (postman panther)

    Votes: 46 14.2%
  • 3. Fine (happy panther)

    Votes: 174 53.5%
  • 4. Great (golfing panther)

    Votes: 92 28.3%

  • Total voters
    325
I believe those scores are heavily impacted by the difficulties in measuring the Five's performance about 10KHz. As a result (I think) the Five's preference scores are deceptively low.
What one may consider the advantage other one will consider as a curse. I don't think they did highs coverage very good either as I wouldn't go over 70° with two regular wave guided tweeters and even that is much. Other part for all such speakers with two tweeters and so wide radiation you can't tame them putting them off axes. Combine that with very low correction margin which will stick in highs and you get a picture.
 
Thanks Amir. Welcome change of pace and in a category that is broader than the usual "audiophile" stuff and more towards a broader consumer interest. Best twist, nice measurements and detailed, insightful listening commentary. Good job as always.
 
They sound better than what the very vast majority of people (outside of this forum) have at home these days, so I find the “only good for casual listening” comments a tad dismissive. I have Neumann monitors at home, but I could live with a pair of Sonos 5 if I had to.

I don’t like how they phase out the support for older models though, so Sonos is a no go for me.
I OFTEN live with an older Sony boombox, on my deck, outside, in my garage and in my basement and while not "High" fidelity, I am quite able to listen to it and enjoy it. In fact, at times, I can even "enjoy" it more than my main systems. Same as with how a car stereo can simply be more fun or enjoyable even if it lacks in absolute fidelity.
alowybeftjqbnymrdaxh.jpg
 
This is a review, listening tests, and detailed measurements of the SONOS Five streaming wired/wireless smart multi-room speaker. It was kindly drop shipped by a member and costs US $549.

View attachment 342206
The Five is solidly built and feels rather substantial in hand. Packaging is always is apple-like. Usability however, is quite poor with cryptically labeled buttons with various icons. And blinking lights you are supposed to decode like a world war II Morse operator. Needless to say, the app is must have and this time unlike previous SONOS products I have tested, it found the Five quickly and proceeded to update its firmware. As with physical controls, I find the usability of the app quite poor, requiring me to fumble around to get things done. Touching a two bar icon for example, asked me if I want to "end a session." Say what?

Volume control jumps in big increments of 4 (or more?). Worst thing is the very long latency. Feed it analog audio which you think would play instantly and it takes it good 2+ seconds before it plays. Use airplay and the same long pipeline delay exists. The delay was so long that it confused Klippel standard measurement method, forcing me to opt for a longer scan using asynchronous mode. I wonder if they are using long FIR filters and hence the long latency.

Speaker has an impressive array of three woofers and three tweeters:
View attachment 342207

There is a forward firing tweeter which I used as the acoustic center when scanning using Klippel NFS. There are two side firing tweeters, likely designed to spread the sound and not having it sound like a focused mono speaker. The interference from all three tweeters made for a complicated soundfield. Using my standard measurement scan of 1000 points, I was not able to get high accuracy much beyond 10 kHz (see below). As it is, the scan took five+ hours so I was in mood to double it just to get more accuracy there.

SONOS Five Measurements
Let's start with our array of frequency response measurements per CEA-2034 standard:
View attachment 342208
I must say, with all six drivers whaling, I didn't expect to see such a flattish on axis response. The listening window as indicated in green shows what it averages to which is pretty nice. Even nicer is the impressively low response. It goes deep to 30 Hz! It is boosted in that region but likely to compensate for non-declining high frequency energy due to extra side-firing two tweeters. We can see this effect in early window and predicted in-room responses:

View attachment 342209

View attachment 342210

We will have to listen to assess the tonality as our models don't fit this type of speaker well.

Horizontal directivity is rather chaotic but wide as you can imagine:
View attachment 342211

View attachment 342212

Vertically it looks more typical although we have tall excursions vertically:
View attachment 342213


Distortion is quite high due to two areas of concern:
View attachment 342214

I could not hear them much though as I was listening to the sweep which was cleaner and deeper than many speakers.
View attachment 342215

Note that the graph on the right represents the loudest I could get the unit to play. Bass is around 96 but the rest of the response is closer to 92 dBSPL.

Waterfall display shows a number of resonances:
View attachment 342216

And step response shows the 2+ second latency:
View attachment 342217
A lot of post ringing which I could after the sweep was finished (this happens with a number of active speakers).

SONOS Five Listening Test
I chose to listen to the Five in near-field at about 2+ meters/5 feet. First impression was most impressive. We are talking almost the same accuracy of a studio monitor! Sound was clean, and tonality was right on the money. Bass notes were deep and so much so that they activated the room modes, sounding a bit tubby at times. I made an attempt to reduce that by lowering the hump at 180 Hz and that helped a bit. Forgot to save it though.

At moderate volume, sub-bass reproduction was excellent, way beyond any bookshelf speaker. At max level though, it got distorted. Outside of that region, the sound was extremely nice even when maxed out, producing not only accurate but very pleasing tonality.

I covered the two side firing tweeters with my hand and the sound was still quite good. Taking my hand off gave a wider sense of space without the highs getting bothersome.

Conclusions
What is the old pilot line? Any airplane landing that you can walk out of is a success? Applied here, any smart speaker that doesn't sound like garbage is a major accomplishment. SONOS however, goes way beyond that requirement and produced an excellent sounding speaker that is powerful with very deep bass. Its response is not perfect but comes close to being accurate especially for this class. Side-firing tweeters provide the spacious and wider dispersion that such a speaker needs to have. And triple woofers provide the impressive dynamics. Clearly strong engineering and acoustic design has been used in develop of the Five.

I am happy to recommend the Sonos Five smart speaker.

------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Super surprised by the results! I've thought of Sonos in the same category as Bose - overpriced with poor sound.
 
Not surprised it tested better than expected. I've got two Play 5's in stereo in my library that sound fantastic, very satisfying. Trueplay made a big difference in this room. Also have a single Era 300 in the bedroom that sounds really good with Dolby Atmos, fairly good with 2.0. Can't really tell the difference pre/post Trueplay with the Era 300. Did a quick subjective comparison to a couple cheaper setups I have. Compared to Sonos Amp + Elac Debut 2.0 6.2s and that cheap, simple system is vastly better than 2x Play 5's or single Era 300. My teenagers setup - Wiim Pro Plus + Fosi V3 + old Epos ESL-3 speakers and ESL sub - also beats both systems. Well-done Dolby Atmos on Era 300 is really good though (like Coltrane's Blue Train) so tempted to get a second Era 300 for the bedroom and see how that sounds.
 
Great results but will it get bricked in the future like some previous Sonos speakers?
 
Probably. But anything with this type of embedded technology inside will eventually get left behind, so not unique to this product. The more reliant on technology a device is, the more exposed to redundancy it becomes.
 
Volume control jumps in big increments of 4 (or more?). Worst thing is the very long latency. Feed it analog audio which you think would play instantly and it takes it good 2+ seconds before it plays. Use airplay and the same long pipeline delay exists. The delay was so long that it confused Klippel standard measurement method, forcing me to opt for a longer scan using asynchronous mode. I wonder if they are using long FIR filters and hence the long latency.
You can drop the latency on the analog line-in from 2 seconds to 75 milliseconds in the app. (See image below for the app screen where you make the adjustment (Spanish version).) It sounds the same, I think, on any of the latency setting; my understanding is that there's no difference when playing line-in on a single speaker between the 75 ms and 2000 ms (I don't know this for 100% certainty, and Sonos can be opaque about these things, but that seems to be the consensus in Sonos forums -- it would be interesting to see it tested). The purpose of the latency, as I understand it, is not to add extra DSP, but to allow for wireless grouping of Sonos speakers across a house, where data transfer and network speeds might require the lag to keep the speakers in sync.

Personally, I have a single Five hooked to a turntable and quite like it. The room correction software works wonders in my oddly shaped apartment (lofted bedroom above living room, high ceilings, super reflective walls). But rather than buying a second Five to make a stereo pair, I have decided to switch to passive speakers/amp/streamer I found through Amir's reviews here on ASR: a pair of Polk ES20 bookshelf speakers, a Fosi V3 amp (48 volt), and a Wiim pro plus. I'll try room correction with REW and a UMIK. That assemblage of separates costs a bit more than a pair of Sonos Fives here in Mexico, with all the the import fees and taxes (in the US, it would cost less than the Sonos Fives).

But I'm happy to see this review. If the Polk/Fosi/Wiim combo doesn't sound good in my space, it's nice to see the golfing panther and know that I can return the Polks and get a second Sonos Five and not feel too ashamed to be lurking on these forums while listening to a lifestyle speaker. :)



IMG_5699.PNG
 
I auditoned Sonos when I was looking for speakers for my kitchen, but ended up going for tower speakers with a streamer instead. This was pre-pandemic though, so probably an earlier version? I thought it sounded tubby and boxy, for lack of technical terms...
 
Probably. But anything with this type of embedded technology inside will eventually get left behind, so not unique to this product. The more reliant on technology a device is, the more exposed to redundancy it becomes.
In the past Sonos intentionally bricked older devices. I think they have reversed this decision in the past couple of years. Maybe while they are at it they could improve repairability as well.
 
This is the first speaker review where I've felt like the Olive score might actually be adding something to the picture.
 
In the past Sonos intentionally bricked older devices. I think they have reversed this decision in the past couple of years. Maybe while they are at it they could improve repairability as well.
They gave the owner's the option to brick their speakers for a 50% off coupon on new ones, rather than having them send it back to the factory for destruction.

They moved to a new eco-system because they couldn't support newer standards with the old system. Especially Airplay2.

Eventually they gave owners the option of keeping their old speakers and downgrading newer speakers to work on the original Sonos network if desired.
 
My "serious" office set up are a couple of Sonos Fives + Sub. Trueplay tuning is a dead simple way to help with some placement/room issues. Sonos has full Roon support and so I can EQ additionally if needed. Also have more Sonos speakers throughout the house and so it was an easy decision to stay in the world of Sonos when buying something for my office. I'm pretty happy and not really itching to upgrade.
 
I have some Sonos, it’s fine for background music. They sounded a lot better than rando monitors and speakers I had laying around. my wife can work them without opening a ticket with me…
 
Two things:
1) I wonder how different the measurements in an anechoic chamber would be compared to the kippel given the side angled drivers? Would that matter to how the Klippel does the calculations?

2) I’d love to see some in room measurements of how well the Trueplay works? I’ve always thought that if it works well and is easy it should dramatically improve response for the average consumer. My only experience is the yoga studio I attend that has a couple of Era100s I think. They sound really good with good management of room modes, although they don’t really go that low.
 
Last edited:
Two things:
1) I wonder how different the measurements in an anechoic chamber would be compared to the kippel given the side angled drivers? Would that matter to how the Klippel does the calculations?

2) I’d love to see some in room measurements of how well the Trueplay works? I’ve always thought that if it works well and is easy it should dramatically improve response for the average consumer. My only experience is the yoga studio I attend that has a couple of Era100s I think. They sound really good with good management of room modes, although they don’t really go that low.
I believe that Sonos actually has access to an anechoic chamber. I'm trying to remember if I heard this from my engineering friend who works at Sonos. I'll have to see.
 
I have a Sonos Five under the roof of a gazebo we have on our pool deck, and plan to put another up to have stereo. I went this route because it I wanted something that would work with Roon, and it was easier than putting an Sonos amp and speakers out there. The Five is “humidity resistant” and so far is holding up just fine in a pretty humid environment. For the purpose it is great and should be excellent when I finally put a second speaker up. We spend a lot of time under the gazebo listening to music.
 
Two things:
1) I wonder how different the measurements in an anechoic chamber would be compared to the kippel given the side angled drivers? Would that matter to how the Klippel does the calculations?
In theory, Klippel NFS doesn't care what the sound source looks like, nor does it make any assumptions about it. You can have a thousand drivers pointing in any direction and it could quantify it.

As a practical matter, the more complex the soundfield, the more samples you need to make to maintain high accuracy. I use about 1000 points which was sufficient to achieve high accuracy (less than 1% error) up to about 10 kHz. How do I know this? Klippel NFS makes redundant measurements which it then uses to compare those actual measurements against the computed ones. So the error is known. The actual error on axis is actually quite small above 10 kHz. It is the off axis which suffered more.
 
Back
Top Bottom