Canon S-50 Wide Directivity Speaker Review

[richard12511]

I think this is good evidence that @amirm has a very strong preference for wide dispersion. We shall see going forward, but I expect well measuring wide dispersion designs to do much better than well measuring narrow dispersion designs.

I believe @amirm would have been among those who preferred the Salon 2 over the M2, despite the M2 having better spinorama.

 

[napilopez]

Preference Rating
SCORE: -0.1
SCORE w/ sub: 2.4

Thanks for the review Amir!

I think this is probably a good example of a speaker with measurements that are worse to the eye than the ear, though this can't be good any way you cut it (and I wouldn't expect that from such an old speaker!). In a more modern example, I had a somewhat similar experience with the Sonos Move which has a very uneven top half of its response yet it sounds very nice in balance, with an in-room response that's much smoother than the measurements would have you think:

In-room compression tests at 1.3m show the actual single-location in-room response. Despite the squiggles, it's clear Sonos was aiming for a smoothly descending in-room response, suggesting the company was following the science while going for a semi-omni design.

This is perhaps predicted best by the total horizontal reflections in this case:

 

[amirm]

What's causing the horizontal dispersion to vary above 3K?

Maybe the two wide vertical supports start to cause reflections once the wavelength gets small enough for their dimensions to become significant.

 

[richard12511]

Thanks for the review Amir!

I think this is probably a good example of a speaker with measurements that are worse to the eye than the ear, though this can't be good any way you cut it (and I wouldn't expect that from such an old speaker!). In a more modern example, I had a somewhat similar experience with the Sonos Move which has a very uneven top half of its response yet it sounds very nice in balance, with an in-room response that's much smoother than the measurements would have you think:

In-room compression tests at 1.3m show the actual single-location in-room response:

View attachment 68269

This is perhaps predicted best by the total horizontal reflections in this case:

View attachment 68270

The Toole/Olive research is pretty old at this point. Perhaps it doesn't apply well to the omni devices like the Alexa/Homepod that we see today?

 

[MZKM]

If you 1/6 octave smooth and then look at estimated in-room response vs. the original Harman Curve, it doesn't look nearly as bad.


View attachment 68265

View attachment 68267

What does the psychoacoustic smoothing look like?

 

[jam]

https://musicandmiscellany.com/2019/06/08/canon-audio-part-1/

That link tells a lot of behind the scenes. Canon had Dennis Ward (ex-Bowers & Wilkins), Mike Jewitt (Mordaunt Short), Stan Curtis (Cambridge Audio) and Alan Boothroyd (Meridian) working as consultants on this project.

Thanks for the link! Pretty interesting info on the project. Those are very big names that collaborated on the project.

 

[richard12511]

I think this is good evidence that @amirm has a very strong preference for wide dispersion. We shall see going forward, but I expect well measuring wide dispersion designs to do much better than well measuring narrow dispersion designs.

I believe @amirm would have been among those who preferred the Salon 2 over the M2, despite the M2 having better spinorama.

Possible counterpoint:

This is a super wide dispersion design, and listening in mono gives a huge advantage to wide dispersion that doesn't exist in stereo.

Based on the double blind listening tests I've done, both of my friends preferred the narrow dispersion speaker(sharper imaging) in stereo, but in mono they preferred the wide dispersion speaker.

 

[oldsysop]

With the backlog I suppose you have what's the point in measuring this 30-years old weird stuff?

The things of 30 years are what interests me that Amir measures.
Thanks Amir.

 

[GXAlan]

What does the psychoacoustic smoothing look like?

Looks pretty similar. Notice the Y-axis too.

 

[tjf]

The things of 30 years are what interests me that Amir measures.
Thanks Amir.

I'd second that, it would be cool to see how the Omni designs hold up today, such as this remake of the old Stu Hegeman Omni speaker from Don Morrison:

https://www.morrisonaudio.com/

Or the attempt to copy these designs made by Mirage in Canada:

http://www.soundstagenetwork.com/revequip/mirage_omni60.htm

Or even how some of the Carlsson designs (Sonab, etc.) would fare....but the Carlsson types are not Omni, but try to accommodate room boundary acoustics and controlled dispersion patterns with corner placement, special driver baffle mounting angles, etc. -- prob hard to measure on the Klippel...

https://teenage.engineering/products/od-11/carlssonstory

 

[TimVG]

I hope Amir gets to test a Sausalito speaker at one point if he like this one, warts and all.

 

[Xyrium]

I came here expecting a mess, and I got it....

I almost didn't post in order to prevent the thread from getting more attention. LOL

Next!

 

[Billy Budapest]

Canon S-35 adds a tweeter and conventional crossover, with a cast zinc structure atop a plastic base.

Measurements here:

https://www.stereophile.com/content/canon-s-35-loudspeaker-measurements

 

[theshakomaster]

@amirm it would seem from the listening test that you likely had buzzing in the speaker from the helmet during the Klippel test, which seems like it would destroy any reliability of the measurements. Are you sure the helmet was stable during the testing?

 

[Billy Budapest]

There is an awesome five part history of Canon Audio here. It pretty much tells the whole story of this small British speaker company owned by a large Japanese camera, printer, and copier company:

https://musicandmiscellany.com/2019/06/08/canon-audio-part-1/

 

[Dogen]

Thanks for reviewing this odd, old speaker. Not denying their issues, but something about the sound of omnidirectional speakers really appeals to me. Are there any that are good performers, or is the concept just too flawed to deliver good measurements?

 

[sam_adams]

There's also a recent thread on ASR here:

Fun series of articles about Canon's foray into loudspeaker design

It includes an article of a test of the S-25 from Sterophile and a pic of the S-70.

These were quite an interesting design, but I think that Duevel has them beat from a design perspective. The Duevel Orion looks like the alien machine from Kronos, Destoyer of the Universe:

 

[GXAlan]

Thanks for reviewing this odd, old speaker. Not denying their issues, but something about the sound of omnidirectional speakers really appeals to me. Are there any that are good performers, or is the concept just too flawed to deliver good measurements?

MBL speakers sound great even though they have the same diffraction errors at the high frequencies. The problem is that they are outrageously expensive and inefficient.

Magnepan speakers are dipoles, but as very flat panels, they end up delivering very good sound, although again, they do not measure well. They are spectacular for “singer with a guitar” type music.

In both cases, I think it may be challenging to have them as your only speaker for all of your music if you have eclectic tastes

 

[AudioTodd]

Brutal! I have to say it doesn't sound like -0.1 speaker. The equal weighting of high frequencies with others is a mistake I think in the algorithm. Response is actually pretty good up to 3 kHz or so.

Your inner subjectivist sneaking out??

 

[amirm]

@amirm it would seem from the listening test that you likely had buzzing in the speaker from the helmet during the Klippel test, which seems like it would destroy any reliability of the measurements. Are you sure the helmet was stable during the testing?

I was sitting on it during the measurement so no distortion was coming there. I had eaten something that may have produced some "gas." So perhaps that was the issue with high frequencies.