Devialet Expert 200 Amplifier, DAC and Streamer Review

[Soniclife]

View attachment 85184

The TV program was surprisingly fun, only partly because of VC.

 

[Thomas savage]

I am a newcomer here, but far from that as regards high fidelity audio. I do not know Frank as you do, nor do I have any motive to denigrate Devialet. Quite the reverse, I find the actual situation as opposed to what I belived to be the case, shocking and sad. It is never a pleasure to see a promising firm fail. Perhaps I misread Frank's comments, however there seems to be a constant theme that Devialet is misunderstood. Some of that has been dispatched by my replies, which I stand by. As to SAM, the burden of proof is on the original party making extraordinary claims. In this case Devialet and by proxy, Frank. I have sugested a common--sense method of doing so, however we all know this will not occur.

Is it that unusual for a fully active speaker to use some kind of signal processing to match a speaker drivers response ? I think that's all SAM is , you take know response albeit from a complete speaker system and change the output of the amp to best drive that speaker .

Might well be misappropriated and complete bollocks used with passive speaker systems but to claim its sole purpose is to fix a broken amp design ? I dont see any evidence of this other than your musings .

Anyway we probably best move on and accept our differing positions .

 

[Soniclife]

Is it that unusual for a fully active speaker to use some kind of signal processing to match a speaker drivers response ? I think that's all SAM is , you take know response albeit from a complete speaker system and change the output of the amp to best drive that speaker .

It's very similar to what DSP actives do, including the protection. It's also similar to what Dirac etc do, but they do bespoke speaker / room correction, where Sam does not address the room part. The basic technologie is indeed basic these days, the specific implementation is innovative, as no one else has measured speakers and produced correction files for them in the same sort of way. Well not before before Amir started up, and people created correction based on his measurements, what is being created here in things like the ls50 thread is next level, as it factors in more than just the on axis.

 

[Soniclife]

It's very similar to what DSP actives do, including the protection. It's also similar to what Dirac etc do, but they do bespoke speaker / room correction, where Sam does not address the room part. The basic technologie is indeed basic these days, the specific implementation is innovative, as no one else has measured speakers and produced correction files for them in the same sort of way. Well not before before Amir started up, and people created correction based on his measurements, what is being created here in things like the ls50 thread is next level, as it factors in more than just the on axis.

P.S. The nearest similar system is probably what Audeze do in providing roon with a bunch of correction files for each of their models that you can then select in roon. That's for headphones not speakers, and the files are provided by the manufacturer not a 3rd party, but it's similar tech.

 

[T-NYC]

Is it that unusual for a fully active speaker to use some kind of signal processing to match a speaker drivers response ? I think that's all SAM is , you take know response albeit from a complete speaker system and change the output of the amp to best drive that speaker .

Might well be misappropriated and complete bollocks used with passive speaker systems but to claim its sole purpose is to fix a broken amp design ? I dont see any evidence of this other than your musings .

SAM is for passive speakers; I have suggested a scientific & common-sense method of proving their claim. It will not happen.

One can know much about a person by the company they choose. Similarly, one may know a good deal about a firm through observation.
* ADH not class D except the current amplier side of the circuit is
* ADH new and revolutionary except it is based on a 1971 patent (as mentioned Hegel, Linn and others make claims similar to Devialet)
* Phantom omnidirectional by virtue of it's curved baffle -- except it isn't -- try standing exactly behind one while it is playing and see for yourself)
* Phantom suberb imaging -- in MONO -- see Bell Labs, Blumlein, Linkwitz for why this is impossible
* The most amazing performance -- the tests here, at Stereophile, and by Paul Miller show otherwise, to the point perhaps of fraud
* radical product with Belkin -- excpt read the reviews, it is junk
* radical product with Huawei -- ditto
* left the Apple Store because of sales and went to Microsoft -- maybe true, however Apple quietly dispatched all audio vendors from it's stores including B&W. A move to Microsoft spells desperation loudly, not choice
* see this thread for the imbecilic and shady statements about patents and intellectual property by the co-founder and former CEO
* see the Devialet forum for experiences by users -- many users -- which are extraordinaily bad, particularly for a luxury goods firm
* more if I had more time and interest

SAM -- based on their history immedaitely above, I'd say the burden is on them.

I feel badly for their founding engineer who clearly fell in with the wrong crowd, for their customers except those who purchased the original product and did not "upgrade" it (it will still be very fine today), and for their last round shareholders who purchased a pig in a poke.

 

[Shorty]

SAM is correction. How is it that no other firm has noticed the need for "correction" of this type?

SAM was not a new idea at all. For one, Perpetual Technologies pioneered the idea 20 years ago with its P1A, and DEQX has also been around since about that time.

 

[T-NYC]

SAM was not a new idea at all. For one, Perpetual Technologies pioneered the idea 20 years ago with its P1A, and DEQX has also been around since about that time.

http://www.soundstagenetwork.com/revequip/perpetual_p1a_p3a.htm does mention loudspeaker correction although it is never defined. As for DEQX the products I'm aware of are sophisticated active crossovers.

I have to ask this again -- why have no other amplifier manufacturers noticed, defined and corrected the errors alluded to; why haven't the loudspeaker manufacturers corrected their errors? It would seem most logical for them to do so if these errors existed.

 

[Dialectic]

http://www.soundstagenetwork.com/revequip/perpetual_p1a_p3a.htm does mention loudspeaker correction although it is never defined. As for DEQX the products I'm aware of are sophisticated active crossovers.

I have to ask this again -- why have no other amplifier manufacturers noticed, defined and corrected the errors alluded to; why haven't the loudspeaker manufacturers corrected their errors? It would seem most logical for them to do so if these errors existed.

All passive speakers have errors that are not easily corrected except with DSP.

SAM, which is limited to the speaker models in Devialet's database, uses DSP to correct frequency and time domain problems below 200 Hz, while implementing hard-limiting to prevent woofer bottoming. It works reasonably well with music on most speakers. Devialet staff happened to be at the Adelphi Mall (one of only a few places in the world where hundreds of high-end speakers are in the same location) measuring speakers for SAM when I was there.

That is all SAM does. It does not (and could not possibly) correct any limitation that the Devialet has in delivering large amounts of power at tweeter frequencies. This limitation is unlikely to be a practical problem unless the user has speakers with unusual, power hungry tweeters, such as exotic ribbons.

 

[T-NYC]

All passive speakers have errors that are not easily corrected except with DSP.

SAM, which is limited to the speaker models in Devialet's database, uses DSP to correct frequency and time domain problems below 200 Hz, while implementing hard-limiting to prevent woofer bottoming. It works reasonably well with music on most speakers. Devialet staff happened to be at the Adelphi Mall (one of only a few places in the world where hundreds of high-end speakers are in the same location) measuring speakers for SAM when I was there.

That is all SAM does. It does not (and could not possibly) correct any limitation that the Devialet has in delivering large amounts of power at tweeter frequencies. This limitation is unlikely to be a practical problem unless the user has speakers with unusual, power hungry tweeters, such as exotic ribbons.

Your reply is sane and sensible, thank you. The hard limiting aspect makes a good deal of sense, and is a point I had entirely forgotten. I do not understand how outboard DSP correction of a passive system would function, but that is my limitation. Your reasoning is why I use active speakers. However, based on their track record as partly inventoried above, I remain entirely skeptical of Devialet's many other claims.

 

[Shorty]

http://www.soundstagenetwork.com/revequip/perpetual_p1a_p3a.htm does mention loudspeaker correction although it is never defined. As for DEQX the products I'm aware of are sophisticated active crossovers.

I have to ask this again -- why have no other amplifier manufacturers noticed, defined and corrected the errors alluded to; why haven't the loudspeaker manufacturers corrected their errors? It would seem most logical for them to do so if these errors existed.

The answer is obvious: 1) it would be a Herculean task and ridiculously costly to analyse every loudspeaker under the sun and create individual correction filters for each of them, only then to experience how your perfect speaker behaviour gets murdered by the room, and so 2) it’s much simpler and cheaper to measure loudspeakers in the room where they perform and correct for in-room errors as well.

DEQX are indeed, among other things, sophisticated active crossovers. It first measures and corrects each individual speaker and then measures and corrects its behaviour in the room.

 

[T-NYC]

The answer is obvious: 1) it would be a Herculean task and ridiculously costly to analyse every loudspeaker under the sun and create individual correction filters for each of them, only then to experience how your perfect speaker behaviour gets murdered by the room, and so 2) it’s much simpler and cheaper to measure loudspeakers in the room where they perform and correct for in-room errors as well.

DEQX are indeed, among other things, sophisticated active crossovers. It first measures and corrects each individual speaker and then measures and corrects its behaviour in the room.

I agree with your statement however perhaps I was not clear in mine. If it is possible for a relative start up, Devialet, to do this, then why is it impossible, not done, etc by well-established and larger firms? A rhetorical question.

I'll have to take a look at DEQX again, as I recall most but not all of the capabilities you mention. Several years back one of their products was on my potential acquistions list, however I've had no reason to look since.

 

[RichB]

The answer is obvious: 1) it would be a Herculean task and ridiculously costly to analyse every loudspeaker under the sun and create individual correction filters for each of them, only then to experience how your perfect speaker behaviour gets murdered by the room, and so 2) it’s much simpler and cheaper to measure loudspeakers in the room where they perform and correct for in-room errors as well.

DEQX are indeed, among other things, sophisticated active crossovers. It first measures and corrects each individual speaker and then measures and corrects its behaviour in the room.

It is not disputed that microphones are not accurately measuring what we hear. If speaker data is to be used, it should be from spinorama in a controlled environment. As you said, PEQ/REQ is better handled to handle room issues and without an amp second guessing a speaker design. Correcting on-axis does can make thing worse when there are issues off axis.

If bass rolls-off, amplifier boost can help at the margins but there is a limit and may cause distortion and port-chuff.

An amplifier's job is the be load invariant (as much as possible), providing linear response with minimal distortion.
The poor performance below 1 watt makes it poor match for efficient speakers, this cannot be DSP'ed away.

- Rich

 

[T-NYC]

I agree with your statement however perhaps I was not clear in mine. If it is possible for a relative start up, Devialet, to do this, then why is it impossible, not done, etc by well-established and larger firms? A rhetorical question.

I'll have to take a look at DEQX again, as I recall most but not all of the capabilities you mention. Several years back one of their products was on my potential acquistions list, however I've had no reason to look since.

Just had a quick look and firstly their down-to-earth approach to describing their technology is a welcome change from the hyperbole of a certain competitor. The whitepaper is future reading, but a quick skim is convincing enough to be worth my allocating time to do so. "Several' -- my wife reminded me this was in 2011 when we were thinking of moving and my sole interest was in finding a very competent DSP crossover that would have been used in a bespoke active system ;-)

 

[StefaanE]

'Absolute piffle! ' last seen in 1978..

Codswallop and hogwash are feeling lonely and would like some attention too.

 

[Frank Dernie]

The poor performance below 1 watt makes it poor match for efficient speakers, this cannot be DSP'ed away.

I have never tried this model but I have tried a different one into 109dB/watt horns.
At a volume control position at which music was at a normal level but nothing playing there was no audible noise or hiss with my ear at the exit of any of the horns, my usual check for adequate snr, quite surprisingly quiet. Definitely the quietest amp I have tried with these speakers.
Plus it has a rotary volume control on the remote

 

[RichB]

I have never tried this model but I have tried a different one into 109dB/watt horns.
At a volume control position at which music was at a normal level but nothing playing there was no audible noise or hiss with my ear at the exit of any of the horns, my usual check for adequate snr, quite surprisingly quiet. Definitely the quietest amp I have tried with these speakers.
Plus it has a rotary volume control on the remote

Looking at the 109 dB/watt horns and this amplifier, .1 watt has S/N of about 69 dB or 30 dB of THD. If inaudible at the exit of horns, that would lead to a conclusion (though unlikely) that this amp is producing significant distortion which IMO is worse than noise.

- Rich

 

[Soniclife]

Looking at the 109 dB/watt horns and this amplifier, .1 watt has S/N of about 69 dB or 30 dB of THD. If inaudible at the exit of horns, that would lead to a conclusion (though unlikely) that this amp is producing significant distortion which IMO is worse than noise.


View attachment 85368

- Rich

That's the measurement from the dodgy one just before it broke, there is no repeat of that measurement, but the dashboard of the repaired one had lower noise so you would expect that test to also show lower noise.

 

[RichB]

Codswallop and hogwash are feeling lonely and would like some attention too.

codswallop
[ˈkädzˌwäləp]

NOUN
BRITISH
informal
codswallop (noun)

1. nonsense.
   "I think that's a right load of old codswallop"
   synonyms:
   prattle · chatter · twitter · babble · talk · prating · gabble · jabber · blether · rambling · nonsense · balderdash · gibberish · claptrap · rubbish · yackety-yak · yabbering · yatter · tripe · twaddle · hogwash · baloney · drivel · bilge · bosh · bull · bunk · guff · eyewash · piffle · poppycock · phooey · hooey · malarkey · dribble · rot · wittering · nattering · chuntering · cobblers · stuff and nonsense · tosh · cack · havers · garbage · flapdoodle · blathers · wack · bushwa · applesauce · bunkum · tommyrot · cod · gammon · toffee · clack · twattle

I think we have just scratched the surface

- Rich

 

[Soniclife]

codswallop
[ˈkädzˌwäləp]

NOUN
BRITISH
informal
codswallop (noun)

1. nonsense.
   "I think that's a right load of old codswallop"
   synonyms:
   prattle · chatter · twitter · babble · talk · prating · gabble · jabber · blether · rambling · nonsense · balderdash · gibberish · claptrap · rubbish · yackety-yak · yabbering · yatter · tripe · twaddle · hogwash · baloney · drivel · bilge · bosh · bull · bunk · guff · eyewash · piffle · poppycock · phooey · hooey · malarkey · dribble · rot · wittering · nattering · chuntering · cobblers · stuff and nonsense · tosh · cack · havers · garbage · flapdoodle · blathers · wack · bushwa · applesauce · bunkum · tommyrot · cod · gammon · toffee · clack · twattle

I think we have just scratched the surface

- Rich

This is the enimology I was told as a kid, and never really believed even then.
https://www.phrases.org.uk/meanings/codswallop.html

 

[Soniclife]

In a surprising turn of events they have added PEQ (via a clunky text file) in a new firmware.
https://help.devialet.com/hc/en-us/articles/360016719679